Developments and Advances in Defense and Security: Proceedings of MICRADS 2019 [1st ed.] 978-981-13-9154-5;978-981-13-9155-2

Table of contents :
Front Matter ....Pages i-xix
Front Matter ....Pages 1-1
Reducing the Attack Surface of Dynamic Binary Instrumentation Frameworks (Ailton Santos Filho, Ricardo J. Rodríguez, Eduardo L. Feitosa)....Pages 3-13
Single Sign-on Implementation: Leveraging Browser Storage for Handling Tabbed Browsing Sign-outs (Lokesh Ramamoorthi, Dilip Sarkar)....Pages 15-28
Cybernetic Dependency Capacity (Jorge Barbosa)....Pages 29-38
Portuguese Concerns and Experience of Specific Cybercrimes: A Benchmarking with European Citizens (João Vidal Carvalho, Álvaro Rocha, António Abreu, Avelino Victor)....Pages 39-50
Classification of Phishing Attack Solutions by Employing Deep Learning Techniques: A Systematic Literature Review (Eduardo Benavides, Walter Fuertes, Sandra Sanchez, Manuel Sanchez)....Pages 51-64
Is Cyber Warfare an Alternative? (Jorge Barbosa)....Pages 65-75
Memory Auditing for Detection of Compromised Switches in Software-Defined Networks Using Trusted Execution Environment (Filipe Augusto da Luz Lemos, Rubens Alexandre de Faria, Paulo Jose Abatti, Mauro Sergio Pereira Fonseca, Keiko Veronica Ono Fonseca)....Pages 77-85
Mobile Communication Systems: Evolution and Security (Teresa Guarda, Maria Fernanda Augusto, Isabel Lopes, José Avelino Victor, Álvaro Rocha, Lilian Molina)....Pages 87-94
Front Matter ....Pages 95-95
Evaluating Trail Formation in Collaborative UAV Networks with Lethal Threats (Nícolas Pereira Borges, Cinara G. Ghedini, Carlos Henrique Costa Ribeiro)....Pages 97-107
Visual Analytics for the Reduction of Air Pollution on Real-Time Data Derived from WSN (Dorys Quiroz, Byron Guanochanga, Walter Fuertes, Diego Benítez, Jenny Torres, Freddy Tapia et al.)....Pages 109-119
Front Matter ....Pages 121-121
Toward the Development of Surveillance and Reconnaissance Capacity in Ecuador: Geolocation System for Ground Targets Based on an Electro-Optical Sensor (Zurita C. Marco Antonio, Aguilar C. Wilbert Geovany, Enríquez C. Victor Xavier)....Pages 123-133
Fuzzy Logic for Speed Control in Object Tracking Inside a Restricted Area Using a Drone (Richard Navas Jácome, Harley Lovato Huertas, Patricia Constante Procel, Andrés Gordón Garcés)....Pages 135-145
Front Matter ....Pages 147-147
Micro-controlled EOG Device for Track and Control Military Applications (Nayana L. M. Viana, José Ailton L. Barbosa Junior, Francisco A. Brito-Filho)....Pages 149-157
Front Matter ....Pages 159-159
Multilevel Military Leadership Model: Correlation Between the Levels and Styles of Military Leadership Using MLQ in the Ecuadorian Armed Forces (Celio Humberto Puga Narváez, Alex Fernando Jimenez Vélez, Rafael Caballero Fernández, Osvaldo Fosado Téllez)....Pages 161-170
E-leadership Using WhatsApp, A Challenge for Navy Organizations: An Empirical Study (Rolando P. Reyes Ch., Luis Recalde Herrera, Galo Andrade Daza, Victor Gómez Bravo, Hugo Pérez Vaca)....Pages 171-181
Front Matter ....Pages 183-183
Career Anchors for the Portuguese Army’s Volunteers and Contract Personnel: Using the Career Orientations Inventory (Lúcio Agostinho Barreiros dos Santos, Maria Manuela Martins Saraiva Sarmento Coelho)....Pages 185-201
Front Matter ....Pages 203-203
Implementation of Dubin Curves-Based RRT* Using an Aerial Image for the Determination of Obstacles and Path Planning to Avoid Them During Displacement of the Mobile Robot (B. Daniel Tenezaca, Christian Canchignia, Wilbert Aguilar, Dario Mendoza)....Pages 205-215
Machine Learning and Multipath Fingerprints for Emitter Localization in Urban Scenario (Marcelo N. de Sousa, Rafael L. Cardoso, Henrique S. Melo, José W. C. Parente Jr., Reiner S. Thomä)....Pages 217-230
Virtual Rehabilitation System Using Electromyographic Sensors for Strengthening Upper Extremities (Z. Andrea Sánchez, T. Santiago Alvarez, F. Roberto Segura, C. Tomás Núñez, P. Urrutia-Urrutia, L. Franklin Salazar et al.)....Pages 231-241
Front Matter ....Pages 243-243
The Portuguese Special Operations Forces as Instrument of Foreign Policy: The Case Study of Afghanistan (João Reis, Sofia Menezes)....Pages 245-255
The Internal–External Security Nexus: EU Operation Sophia Through the Lens of Securitization (Ana Paula Brandão)....Pages 257-267
The Evolution of EU’s Maritime Security Thinking (João Almeida Silveira)....Pages 269-280
The Obstacles Women Face in Gaining Access to Special Operations Forces (João Reis, Rafael Gonçalves, Sofia Menezes, Manuela Kaczynska)....Pages 281-291
The Transformation of the Defense and Security Sector to the New Logistics 4.0: Public–Private Cooperation as a Necessary Catalyst Strategy (Manuel A. Fernández-Villacañas Marín)....Pages 293-303
Front Matter ....Pages 305-305
An Autonomous Airship Swarm for Maritime Patrol (Constantino G. Ribeiro, Luciano Santos Constantin Raptopoulos, Max Suell Dutra)....Pages 307-320
Assessing the Location of Search and Rescue Stations on the Portuguese Coast (Anacleto Correia, Ricardo Moura, Miguel Fonseca)....Pages 321-331
Operational Scenarios Identification and Prioritization in Systems of Systems: A Method and an Experience Report in the Defense Domain (Carlos Eduardo de Barros Paes, Valdemar Vicente Graciano Neto)....Pages 333-344
Science Diplomacy: Strategic Initiative to Create a Buffer Zone in the Caribbean Colombian Marine Protected Area Seaflower (Fabián Ramírez-Cabrales, Sergio Iván Rueda Forero)....Pages 345-359
Front Matter ....Pages 361-361
Quantum Well Infrared Photodetector for the SWIR Range (Pedro Pereira, Lesslie Guerra, G. M. Penello, M. P. Pires, L. D. Pinto, R. Jakomin et al.)....Pages 363-370
Independent Feeding of People Affected with Osteoarthritis Through a Didactic Robot and Visual Control (Arturo Jiménez, Katherine Aroca, Vicente Hallo, Nancy Velasco, Darío Mendoza)....Pages 371-381
Adjustable Structure in Height for Assisted Feeding in Patients with Osteoarthritis (Katherine Aroca, Arturo Jiménez, Nancy Velasco, Vicente Hallo, Darío Mendoza)....Pages 383-395
Voice-Controlled Assistance Device for Victims of Gender-Based Violence (Miguel A. Domínguez, David Palomeque, Juan M. Carrillo, José Mª Valverde, Juan F. Duque, Bruno Pérez et al.)....Pages 397-407
Front Matter ....Pages 409-409
Analysis and Evaluation of the Positioning of Autonomous Underwater Vehicles Using Acoustic Signals (Enrique V. Carrera, Manolo Paredes)....Pages 411-421
Front Matter ....Pages 423-423
Evaluation of 3D Printing Parameters on the Electrochemical Performance of Conductive Polymeric Components for Chemical Warfare Agent Sensing (Joseane R. Barbosa, Pedro H. O. Amorim, Mariana C. de O. Gonçalves, Rafael M. Dornellas, Robson P. Pereira, Felipe S. Semaan)....Pages 425-435
Front Matter ....Pages 437-437
Interactive System to Improve the Skills of Children with Dyslexia: A Preliminary Study (Jorge Buele, Victoria M. López, L. Franklin Salazar, Jordan-H. Edisson, Cristina Reinoso, Sandra Carrillo et al.)....Pages 439-449
Use of Information and Communication Technologies in the Classroom: An Exploratory Study in Professional Military Education (Nuno Alberto Rodrigues Santos Loureiro, Lúcio Agostinho Barreiros dos Santos)....Pages 451-473
Using Augmented Reality for Learning Naval Operations (Anacleto Correia, António Gonçalves, Marielba Zacarias)....Pages 475-485
Communication of the Quality of Higher Education Institutions of Ecuador (Guillermo Santa María, Carina Rey, Lilian Molina, Arturo Clery)....Pages 487-495
The Four Dimensions of the GDPR Framework: An Institutional Theory Perspective (Isabel Maria Lopes, Teresa Guarda, Pedro Oliveira)....Pages 497-506
Metrics and Indicators of Information Security Incident Management: A Systematic Mapping Study (Alyssa Cadena, Franklin Gualoto, Walter Fuertes, Luis Tello-Oquendo, Roberto Andrade, Freddy Tapia et al.)....Pages 507-519
Virtual Environment for Remote Control of UGVs Using a Haptic Device (F. Roberto Segura, Pilar Urrutia-Urrutia, Z. Andrea Sánchez, C. Tomás Núñez, T. Santiago Alvarez, L. Franklin Salazar et al.)....Pages 521-531
Back Matter ....Pages 533-534

Citation preview

Smart Innovation, Systems and Technologies 152

Álvaro Rocha Robson Pacheco Pereira Editors

Developments and Advances in Defense and Security Proceedings of MICRADS 2019

Smart Innovation, Systems and Technologies Volume 152

Series Editors Robert J. Howlett, Bournemouth University and KES International, Shoreham-by-sea, UK Lakhmi C. Jain, Faculty of Engineering and Information Technology, Centre for Artificial Intelligence, University of Technology Sydney, Broadway, NSW, Australia

The Smart Innovation, Systems and Technologies book series encompasses the topics of knowledge, intelligence, innovation and sustainability. The aim of the series is to make available a platform for the publication of books on all aspects of single and multi-disciplinary research on these themes in order to make the latest results available in a readily-accessible form. Volumes on interdisciplinary research combining two or more of these areas is particularly sought. The series covers systems and paradigms that employ knowledge and intelligence in a broad sense. Its scope is systems having embedded knowledge and intelligence, which may be applied to the solution of world problems in industry, the environment and the community. It also focusses on the knowledge-transfer methodologies and innovation strategies employed to make this happen effectively. The combination of intelligent systems tools and a broad range of applications introduces a need for a synergy of disciplines from science, technology, business and the humanities. The series will include conference proceedings, edited collections, monographs, handbooks, reference books, and other relevant types of book in areas of science and technology where smart systems and technologies can offer innovative solutions. High quality content is an essential feature for all book proposals accepted for the series. It is expected that editors of all accepted volumes will ensure that contributions are subjected to an appropriate level of reviewing process and adhere to KES quality principles. ** Indexing: The books of this series are submitted to ISI Proceedings, EI-Compendex, SCOPUS, Google Scholar and Springerlink **

More information about this series at http://www.springer.com/series/8767

Álvaro Rocha Robson Pacheco Pereira •

Editors

Developments and Advances in Defense and Security Proceedings of MICRADS 2019

123

Editors Álvaro Rocha Departamento de Engenharia Informática Universidade de Coimbra Coimbra, Portugal

Robson Pacheco Pereira Military Engineering Institute Rio de Janeiro, Brazil

ISSN 2190-3018 ISSN 2190-3026 (electronic) Smart Innovation, Systems and Technologies ISBN 978-981-13-9154-5 ISBN 978-981-13-9155-2 (eBook) https://doi.org/10.1007/978-981-13-9155-2 © Springer Nature Singapore Pte Ltd. 2020 This work is subject to copyright. All rights are reserved by the Publisher, whether the whole or part of the material is concerned, specifically the rights of translation, reprinting, reuse of illustrations, recitation, broadcasting, reproduction on microfilms or in any other physical way, and transmission or information storage and retrieval, electronic adaptation, computer software, or by similar or dissimilar methodology now known or hereafter developed. The use of general descriptive names, registered names, trademarks, service marks, etc. in this publication does not imply, even in the absence of a specific statement, that such names are exempt from the relevant protective laws and regulations and therefore free for general use. The publisher, the authors and the editors are safe to assume that the advice and information in this book are believed to be true and accurate at the date of publication. Neither the publisher nor the authors or the editors give a warranty, expressed or implied, with respect to the material contained herein or for any errors or omissions that may have been made. The publisher remains neutral with regard to jurisdictional claims in published maps and institutional affiliations. This Springer imprint is published by the registered company Springer Nature Singapore Pte Ltd. The registered company address is: 152 Beach Road, #21-01/04 Gateway East, Singapore 189721, Singapore

Organization

Honorary Chair Álvaro Rocha, University of Coimbra, Portugal

Honorary Co-chair Lakhmi C. Jain, University of Canberra, Australia

General Chairs Robson Pacheco Pereira, Instituto Militar de Engenharia—IME/RJ, Brazil José Avelino Moreira Victor, University Institute of Maia, Portugal

Local Organizing Chair Luiz Eduardo Pizarro Borges, Instituto Militar de Engenharia—IME/RJ, Brazil Luiz Augusto Cavalcante Moniz de Aragão Filho, Instituto Militar de Engenharia— IME/RJ, Brazil Filipe Silva Semaan, Universidade Federal Fluminense, Brazil Bárbara Dias de Castro Pacheco, Universidade Federal Fluminense, Brazil

v

vi

Organization

Scientific Committee Abderrazak Sebaa, University of Bejaia, Algeria Alex Fernando Jimenez Velez, Fuerza Aérea Ecuatoriana (FAE), Ecuador Amita Nandal, UIST—University for Information Science and Technology “St. Paul The Apostle”, Ohrid, Macedonia Anacleto Correia, CINAV/Escola Naval, Portugal Andrea D’Ambrogio, University of Rome Tor Vergata, Italy Andrea Visconti, Università di Milano, Italy Angelo Borzino, Brazilian Army Technological Center, Brazil Angelo Brayner, Federal University of Ceará, Brazil António Abreu, ISCAP/IPP, Portugal Antonios Andreatos, Hellenic Air Force Academy, Greece Asanka Pallewatta, University of Kelaniya, Sri Lanka Bruno de Pinho Silveira, ESPE—University of the Armed Forces, Ecuador Bruno Madeira, CTEx, Brazil Calogero Vetro, University of Palermo, Italy Chiara Braghin, Università degli Studi di Milano, Italy Chinenye Ajibo, University of Nigeria, Nsukka, Nigeria Damir Blazevic, Faculty of Electrical Engineering, Computer Science and Information Technology Osijek, Croatia Daniel Sampaio, University Institute of Maia, Portugal Daniela Suzuki, Institute of Biomedical Engineering (IEB-UFSC), Brazil Darwin Manolo Paredes Calderon, ESPE—University of the Armed Forces, Ecuador Derya Yiltas-Kaplan, Istanbul University, Turkey Diana Patricia Arias Henao, Universidad Militar Nueva Granada, Colombia Diego Paes de Andrade Peña, Universidade Federal do Maranhão, Brazil Dimitrios Dalaklis, World Maritime University (WMU), Sweden Eduardo Izycki, Presidência da República do Brasil, Brazil Eduardo Siqueira Brick, Universidade Federal Fluminense—Instituto de Estudos Estratégicos, Brazil Emanuele Bellini, University of Florence, Italy Enrique Carrera, ESPE—University of the Armed Forces, Ecuador Fabian Ramirez Cabrales, Colombian Naval Academy “Almirante Padilla”, Colombia Felipe Torres Leite, Universidade do Estado do Rio Grande do Norte—UERN, Brazil Fernando Almeida, University of Campinas (UNICAMP), Brazil Filipe Sá, Câmara Municipal de Penacova, Portugal Flávio de Barros Vidal, University of Brasilia, Brazil Franklin Johnny Dominguez Ruiz, Universidad da Coruna, Ecuador Franklin Manuel Silva Monteros, Universidad de las Fuerzas Armadas—ESPE, Ecuador

Organization

vii

Gabriele Virzi’ Mariotti, University of Palermo, Italy Galo Ricardo Andrade Daza, AGUENA Centro de Estudios Estratégicos de la Armada, Ecuador Geert de Cubber, Royal Military Academy, Belgium Giovana Garrido, Universidad Tecnológica de Panamá, Panamá Glauber Cruz, Federal University of Maranhão, Brazil Guilherme Pussieldi, Universidade Federal de Viçosa—Campus Florestal, Brazil Hachem Slimani, University of Bejaia, Algeria Helio de Oliveira, Universidade Federal de Pernambuco, Brazil Henrique Miguel Gouveia Silva, University Institute of Maia, Portugal Henry Omar Cruz Carrillo, Centro de Investigacíon de Aplicaciones Militares— CICTE, Ecuador Hiroshika Premarathne, FCT, University of Kelaniya, Sri Lanka Hugo Peixoto, Algoritmi Research Center, University of Minho, Portugal Ioannis Chatzigiannakis, Sapienza University of Rome, Italy Isabel Lopes, Instituto Politécnico de Bragança, Portugal Ivan Machado, Federal University of Bahia, Brazil Jaime Meza Hormaza, Universitat Politecnica de Catalunya, Spain Jeimy Cano, Universidad del Rosario, Colombia Jiahn-Horng Chen, National Taiwan Ocean University, Taiwan João Reis, Portuguese Military Academy, Portugal João Vidal de Carvalho, ISCAP/IPP, Portugal José Antonio Apolinario Junior, Military Institute of Engineering (IME—SE/3), Brazil José Avelino Moreira Victor, University Institute of Maia, Portugal José Carlos Dias Rouco, Portuguese Military Academy, Portugal José Luís Reis, IPAM—Instituto Português de Marketing, Portugal José Miguel de Carvalho Cerqueira, Instituto Universitário Militar/Academia Militar/Swedish Defence University, Portugal Jurij Mihelič, University of Ljubljana, Slovenia Kalinka Kaloyanova, Sofia University, Bulgaria Lúcio Agostinho Barreiros dos Santos, Instituto Universitário Militar, Portugal Luis Alvarez Sabucedo, University of Vigo, Spain Luis Dieulefait, Universitat de Barcelona, Spain Luis Eduardo Palacios Aguirre, Fuerza Aerea Ecuatoriana, Ecuador Luis Anido Rifón, Universidade de Vigo, Spain Luiz Goncalves Junior, Sao Paulo State University—UNESP, Brazil Manolo Paredes, Universidad de Las Fuerzas Armadas, Ecuador Manuel Francisco González Penedo, Centro de Investigación CITIC, UDC, Spain Manuel Tupia, Pontificia Universidad Catolica del Perú, Perú Manuel Vilares Ferro, University of Vigo, Spain Marcelo Henrique Prado da Silva, Instituto Militar de Engenharia—IME, Brazil Marcos Barreto, Universidade Federal da Bahia, Brazil María Carolina Romero Lares, World Maritime University, Sweden

viii

Organization

Maria Manuela Martins Saraiva Sarmento Coelho, Military University Institute, Portugal Mario Bernabé Ron Egas, Universidad de las Fuerzas Armadas—ESPE, Ecuador Maroi Agrebi, University of Polytechnique Hauts-de-France, France Martín López Nores, University of Vigo, Spain Matheus Pinheiro Ferreira, Instituto Militar de Engenharia, Brazil Mauricio Loachamín Valencia, Universidad de las Fuerzas Armadas—ESPE, Ecuador Maximo Jr. Q. Mejia, World Maritime University, Sweden Mohammed Mahmood Ali, Muffakham Jah College of Engineering and Technology, Osmania University, India Muhammed Ali Aydin, Istanbul University, Turkey Nina Figueira, Brazilian Army, Brazil Oscar Barrionuevo Vaca, Naval High School, Ecuador Pastor David Chávez Muñoz, Pontificia Universidad Catolica de Peru, Peru Paulo Afonso Silva, Military Institute of Engineering, Brazil Pedro Ferreira, Centre for Marine Technology and Ocean Engineering (CENTEC), Portugal Rafael Timoteo de Sousa Junior, University of Brasilia, Brazil Rashed Mustafa, University of Chittagong, Bangladesh Regilberto Girão, Ministério Público Federal, Brazil Renato Jose Sassi, Universidade Nove de Julho, Brazil Reza Malekian, University of Pretoria, South Africa Robert Beeres, Netherlands Defence Academy, The Netherlands Robson Pacheco Pereira, Instituto Militar de Engenharia—IME/RJ, Brazil Ronaldo Salles, Military Institute of Engineering—IME, Brazil Rosalba Rodríguez Reyes, ESPE—University of the Armed Forces, Ecuador Rui Carreira, Maia Polytechnic Institute, Portugal Rui Jorge Palhoto Lucena, Military University Institute—Military Academy, Portugal Sarat Mohapatra, Centre for Marine Technology and Ocean Engineering (CENTEC), Instituto Superior Tecnico, University of Lisbon, Portugal Sebastião Alves Filho, Universidade do Estado do Rio Grande do Norte—UERN, Brazil Sidnei Alves de Araújo, Universidade Nove de Julho, Brazil Silvio Melo, Universidade Federal de Pernambuco, Brazil Simona Safarikova, Palacky University, Department of Development and Environmental Studies, Czech Republic Sonia Cárdenas Delgado, Universidad Politécnica de las Fuerzas Armadas—ESPE, Ecuador Suzana Paula Gomes Fernando da Silva Lampreia, Portuguese Navy, Portugal Taybi Chakib, Faculty of Sciences, Mohammed First University, Morocco Telmo Bento, ISMAI/IPMAIA, Portugal Teresa Guarda, ESPE—University of the Armed Forces, Ecuador Uma N. Dulhare, MJCET, Hyderabad, India

Organization

ix

Veronica Rossano, Department of Computer Science—University of Bari Aldo Moro, Italy Wagner Tanaka Botelho, Federal University of ABC, Brazil Zohra Bakkoury, Ecole Mohammadia d’Ingénieurs, Morocco

Preface

This book contains a selection of papers accepted for presentation and discussion at The 2019 Multidisciplinary International Conference of Research Applied to Defense and Security (MICRADS’19). This Conference had the support of IME (Military Institute of Engineering) of Brazil, ESPE (University of Armed Forces) of Ecuador, and AISTI (Iberian Association for Information Systems and Technologies). It took place at Rio de Janeiro, Brazil, during May 8–10, 2019. The 2019 Multidisciplinary International Conference of Research Applied to Defense and Security (MICRADS’19) is an international forum for researchers and practitioners to present and discuss the most recent innovations, trends, results, experiences, and concerns in the several perspectives of defense and security. The Program Committee of MICRADS’19 was composed of a multidisciplinary group of more than 100 experts from 35 countries around the world and those who are intimately concerned with Research Applied to Defense and Security. They have had the responsibility for evaluating, in a ‘double-blind review’ process, the papers received for each of the main themes proposed for the Conference: (A) systems, communication and defense; (B) strategy and political–administrative vision in defense; and (C) engineering and technologies applied to defense. MICRADS’19 received about 100 contributions from 11 countries around the world. The papers accepted for presentation and discussion at the Conference are published by Springer (this book) and by AISTI and will be submitted for indexing by ISI, Ei Compendex, Scopus, and/or Google Scholar, among others. We acknowledge all of those that contributed to the staging of MICRADS’19 (authors, committees, workshop organizers, and sponsors). We deeply appreciate their involvement and support that was crucial for the success of MICRADS’19. Rio de Janeiro, Brazil May 2019

Álvaro Rocha Robson Pacheco Pereira

xi

Contents

Part I

Cybersecurity and Cyberdefense

Reducing the Attack Surface of Dynamic Binary Instrumentation Frameworks . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Ailton Santos Filho, Ricardo J. Rodríguez and Eduardo L. Feitosa

3

Single Sign-on Implementation: Leveraging Browser Storage for Handling Tabbed Browsing Sign-outs . . . . . . . . . . . . . . . . . . . . . . . Lokesh Ramamoorthi and Dilip Sarkar

15

Cybernetic Dependency Capacity . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Jorge Barbosa Portuguese Concerns and Experience of Specific Cybercrimes: A Benchmarking with European Citizens . . . . . . . . . . . . . . . . . . . . . . . . João Vidal Carvalho, Álvaro Rocha, António Abreu and Avelino Victor Classification of Phishing Attack Solutions by Employing Deep Learning Techniques: A Systematic Literature Review . . . . . . . . . . . . . Eduardo Benavides, Walter Fuertes, Sandra Sanchez and Manuel Sanchez Is Cyber Warfare an Alternative? . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Jorge Barbosa Memory Auditing for Detection of Compromised Switches in Software-Defined Networks Using Trusted Execution Environment . . . . Filipe Augusto da Luz Lemos, Rubens Alexandre de Faria, Paulo Jose Abatti, Mauro Sergio Pereira Fonseca and Keiko Veronica Ono Fonseca Mobile Communication Systems: Evolution and Security . . . . . . . . . . . . Teresa Guarda, Maria Fernanda Augusto, Isabel Lopes, José Avelino Victor, Álvaro Rocha and Lilian Molina

29

39

51

65

77

87

xiii

xiv

Part II

Contents

Computer Networks, Mobility and Pervasive Systems

Evaluating Trail Formation in Collaborative UAV Networks with Lethal Threats . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Nícolas Pereira Borges, Cinara G. Ghedini and Carlos Henrique Costa Ribeiro

97

Visual Analytics for the Reduction of Air Pollution on Real-Time Data Derived from WSN . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 109 Dorys Quiroz, Byron Guanochanga, Walter Fuertes, Diego Benítez, Jenny Torres, Freddy Tapia and Theofilos Toulkkeridis Part III

Defense Engineering

Toward the Development of Surveillance and Reconnaissance Capacity in Ecuador: Geolocation System for Ground Targets Based on an Electro-Optical Sensor . . . . . . . . . . . . . . . . . . . . . . . . . . . . 123 Zurita C. Marco Antonio, Aguilar C. Wilbert Geovany and Enríquez C. Victor Xavier Fuzzy Logic for Speed Control in Object Tracking Inside a Restricted Area Using a Drone . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 135 Richard Navas Jácome, Harley Lovato Huertas, Patricia Constante Procel and Andrés Gordón Garcés Part IV

Health Informatics in Military Applications

Micro-controlled EOG Device for Track and Control Military Applications . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 149 Nayana L. M. Viana, José Ailton L. Barbosa Junior and Francisco A. Brito-Filho Part V

Leadership and e-Leadership

Multilevel Military Leadership Model: Correlation Between the Levels and Styles of Military Leadership Using MLQ in the Ecuadorian Armed Forces . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 161 Celio Humberto Puga Narváez, Alex Fernando Jimenez Vélez, Rafael Caballero Fernández and Osvaldo Fosado Téllez E-leadership Using WhatsApp, A Challenge for Navy Organizations: An Empirical Study . . . . . . . . . . . . . . . . . . . . . . . . . . . . 171 Rolando P. Reyes Ch., Luis Recalde Herrera, Galo Andrade Daza, Victor Gómez Bravo and Hugo Pérez Vaca

Contents

Part VI

xv

Planning, Economy and Logistics Applied to Defense

Career Anchors for the Portuguese Army’s Volunteers and Contract Personnel: Using the Career Orientations Inventory . . . . . . . . . . . . . . . 185 Lúcio Agostinho Barreiros dos Santos and Maria Manuela Martins Saraiva Sarmento Coelho Part VII

Simulation and Computer Vision in Military Applications

Implementation of Dubin Curves-Based RRT* Using an Aerial Image for the Determination of Obstacles and Path Planning to Avoid Them During Displacement of the Mobile Robot . . . . . . . . . . . 205 B. Daniel Tenezaca, Christian Canchignia, Wilbert Aguilar and Dario Mendoza Machine Learning and Multipath Fingerprints for Emitter Localization in Urban Scenario . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 217 Marcelo N. de Sousa, Rafael L. Cardoso, Henrique S. Melo, José W. C. Parente Jr. and Reiner S. Thomä Virtual Rehabilitation System Using Electromyographic Sensors for Strengthening Upper Extremities . . . . . . . . . . . . . . . . . . . . . . . . . . . 231 Z. Andrea Sánchez, T. Santiago Alvarez, F. Roberto Segura, C. Tomás Núñez, P. Urrutia-Urrutia, L. Franklin Salazar, S. Altamirano and J. Buele Part VIII

Strategy, Geopolitics and Oceanopolitics

The Portuguese Special Operations Forces as Instrument of Foreign Policy: The Case Study of Afghanistan . . . . . . . . . . . . . . . . . 245 João Reis and Sofia Menezes The Internal–External Security Nexus: EU Operation Sophia Through the Lens of Securitization . . . . . . . . . . . . . . . . . . . . . . . . . . . . 257 Ana Paula Brandão The Evolution of EU’s Maritime Security Thinking . . . . . . . . . . . . . . . . 269 João Almeida Silveira The Obstacles Women Face in Gaining Access to Special Operations Forces . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 281 João Reis, Rafael Gonçalves, Sofia Menezes and Manuela Kaczynska The Transformation of the Defense and Security Sector to the New Logistics 4.0: Public–Private Cooperation as a Necessary Catalyst Strategy . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 293 Manuel A. Fernández-Villacañas Marín

xvi

Part IX

Contents

Safety and Maritime Protection

An Autonomous Airship Swarm for Maritime Patrol . . . . . . . . . . . . . . 307 Constantino G. Ribeiro, Luciano Santos Constantin Raptopoulos and Max Suell Dutra Assessing the Location of Search and Rescue Stations on the Portuguese Coast . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 321 Anacleto Correia, Ricardo Moura and Miguel Fonseca Operational Scenarios Identification and Prioritization in Systems of Systems: A Method and an Experience Report in the Defense Domain . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 333 Carlos Eduardo de Barros Paes and Valdemar Vicente Graciano Neto Science Diplomacy: Strategic Initiative to Create a Buffer Zone in the Caribbean Colombian Marine Protected Area Seaflower . . . . . . . 345 Fabián Ramírez-Cabrales and Sergio Iván Rueda Forero Part X

Wearable Technology and Assistance Devices

Quantum Well Infrared Photodetector for the SWIR Range . . . . . . . . . 363 Pedro Pereira, Lesslie Guerra, G. M. Penello, M. P. Pires, L. D. Pinto, R. Jakomin, R. T. Mourão, M. H. Degani, M. Z. Maialle and P. L. Souza Independent Feeding of People Affected with Osteoarthritis Through a Didactic Robot and Visual Control . . . . . . . . . . . . . . . . . . . . 371 Arturo Jiménez, Katherine Aroca, Vicente Hallo, Nancy Velasco and Darío Mendoza Adjustable Structure in Height for Assisted Feeding in Patients with Osteoarthritis . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 383 Katherine Aroca, Arturo Jiménez, Nancy Velasco, Vicente Hallo and Darío Mendoza Voice-Controlled Assistance Device for Victims of Gender-Based Violence . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 397 Miguel A. Domínguez, David Palomeque, Juan M. Carrillo, José Mª Valverde, Juan F. Duque, Bruno Pérez and Raquel Pérez-Aloe Part XI

Analysis and Signal Processing

Analysis and Evaluation of the Positioning of Autonomous Underwater Vehicles Using Acoustic Signals . . . . . . . . . . . . . . . . . . . . . 411 Enrique V. Carrera and Manolo Paredes

Contents

Part XII

xvii

Chemical, Biological and Nuclear Defense

Evaluation of 3D Printing Parameters on the Electrochemical Performance of Conductive Polymeric Components for Chemical Warfare Agent Sensing . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 425 Joseane R. Barbosa, Pedro H. O. Amorim, Mariana C. de O. Gonçalves, Rafael M. Dornellas, Robson P. Pereira and Felipe S. Semaan Part XIII

Information and Communication Technology in Education

Interactive System to Improve the Skills of Children with Dyslexia: A Preliminary Study . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 439 Jorge Buele, Victoria M. López, L. Franklin Salazar, Jordan-H. Edisson, Cristina Reinoso, Sandra Carrillo, Angel Soria, Raúl Andrango and Pilar Urrutia-Urrutia Use of Information and Communication Technologies in the Classroom: An Exploratory Study in Professional Military Education . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 451 Nuno Alberto Rodrigues Santos Loureiro and Lúcio Agostinho Barreiros dos Santos Using Augmented Reality for Learning Naval Operations . . . . . . . . . . . 475 Anacleto Correia, António Gonçalves and Marielba Zacarias Communication of the Quality of Higher Education Institutions of Ecuador . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 487 Guillermo Santa María, Carina Rey, Lilian Molina and Arturo Clery The Four Dimensions of the GDPR Framework: An Institutional Theory Perspective . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 497 Isabel Maria Lopes, Teresa Guarda and Pedro Oliveira Metrics and Indicators of Information Security Incident Management: A Systematic Mapping Study . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 507 Alyssa Cadena, Franklin Gualoto, Walter Fuertes, Luis Tello-Oquendo, Roberto Andrade, Freddy Tapia and Jenny Torres Virtual Environment for Remote Control of UGVs Using a Haptic Device . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 521 F. Roberto Segura, Pilar Urrutia-Urrutia, Z. Andrea Sánchez, C. Tomás Núñez, T. Santiago Alvarez, L. Franklin Salazar, Santiago Altamirano and Jorge Buele Author Index . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 533

About the Editors

Álvaro Rocha holds a Habilitation (postdoctoral degree) in Information Science, Ph.D. in Information Systems and Technologies, M.Sc. in Information Management, and B.C.S. in Computer Science. He is currently a Professor of Information Systems and Software Engineering at the University of Coimbra, researcher at CISUC (Centre for Informatics and Systems of the University of Coimbra) and a collaborator researcher at LIACC (Laboratory of Artificial Intelligence and Computer Science) and CINTESIS (Centre for Research in Health Technologies and Information Systems). His main research interests are in Information Systems Planning and Management, Maturity Models, Information Systems Quality, Online Service Quality, Intelligent Information Systems, Software Engineering, e-Government, e-Health, and Information Technology in Education. He is also President of the AISTI (Iberian Association for Information Systems and Technologies), Chair of the IEEE Portugal Section Systems, Man, and Cybernetics Society Chapter, and Editor-in-Chief of both JISEM (Journal of Information Systems Engineering & Management) and RISTI (Iberian Journal of Information Systems and Technologies). Professor Rocha has also served as a Vice-Chair of Experts in the European Commission’s Horizon 2020 initiative, as an Expert at the Italian Ministry of Education, University and Research, and as a General Chair of the WorldCIST (World Conference on Information Systems and Technologies). Robson Pacheco Pereira holds a degree in Chemistry (2000) and a doctorate in Physics/Chemistry (2005) from the Federal University of Rio de Janeiro. He is currently a Professor at the Military Institute of Engineering (IME/RJ), where he coordinates the Nanotechnology Group and Physics and Chemistry, and develops projects in the areas of nanotechnology, materials, electrochemistry, biosensors and theoretical chemistry, intended for use in the area of Defense. The Group chiefly pursues work on nanotechnology, physicochemistry, polymeric materials and inorganic and nano-structured nanoparticles, kinetics, electrochemical solids, corrosion, and power conversion and storage devices. Professor Pereira has also published extensively in international journals.

xix

Part I

Cybersecurity and Cyberdefense

Reducing the Attack Surface of Dynamic Binary Instrumentation Frameworks Ailton Santos Filho, Ricardo J. Rodríguez and Eduardo L. Feitosa

Abstract Malicious applications pose as one of the most relevant issues in today’s technology scenario, being considered the root of many Internet security threats. In part, this owes the ability of malware developers to promptly respond to the emergence of new security solutions by developing artifacts to detect and avoid them. In this work, we present three countermeasures to mitigate recent mechanisms used by malware to detect analysis environments. Among these techniques, this work focuses on those that enable a malware to detect dynamic binary instrumentation frameworks, thus increasing their attack surface. To ensure the effectiveness of the proposed countermeasures, proofs of concept were developed and tested in a controlled environment with a set of anti-instrumentation techniques. Finally, we evaluated the performance impact of using such countermeasures. Keywords Anti-instrumentation · Analysis-aware · Malware · Dynamic binary instrumentation · Anti-analysis

1 Introduction The number of software specially crafted with malicious intentions (commonly referred to as malware) has increased in quantity and complexity during the last years [2]. This fact supposes an issue for the anti-virus companies that need to dispose of an up-to-date database of known malware to protect their customers. Thus, malware analysts face every day to an increasing number of malware samples. For A. S. Filho (B) · E. L. Feitosa Instituto de Computação, Universidade Federal do Amazonas (UFAM), Manaus, Brazil e-mail: [email protected] E. L. Feitosa e-mail: [email protected] R. J. Rodríguez Centro Universitario de la Defensa, Academia General Militar, Zaragoza, Spain e-mail: [email protected] © Springer Nature Singapore Pte Ltd. 2020 Á. Rocha and R. P. Pereira (eds.), Developments and Advances in Defense and Security, Smart Innovation, Systems and Technologies 152, https://doi.org/10.1007/978-981-13-9155-2_1

3

4

A. S. Filho et al.

instance, Kaspersky (a well-known anti-virus company) stated that they analyzed more than 360.000 malware samples per day in 2017 [11]. The process of analyzing a piece of software and determining if it is malicious can be done manually or automatically. A manual analysis requires to first analyze the program assembly code in a static way (without executing it) and then to execute it to analyze the program behavior when interacting with the operating system (internal behavior) and with the network (external behavior). However, this manual approach is a very intensive and time-consuming process (in fact, it is usually referred to as an art instead of science). To cope with the increasing trend of malware samples, automation of malware analysis tasks in isolated analysis environments as sandboxes or hypervisors has emerged in recent years [9]. This automatic approach allows the anti-virus companies to keep updated their databases in a timely manner. Unfortunately, malware writers have also started to incorporate small pieces of code into their software to detect analysis environments. This is mainly motivated because the longer the malware is undetected, the more revenue the cybercriminals achieve. The kind of malware that behave differently depending on where they are executed is referred to as evasive malware, analysis-aware malware, or split personality malware [3, 13, 19, 21, 23]. The incorporation of those evasion techniques allows a malware to check where it is being launched and thus behave benignly when an analysis environment is detected. As a consequence, the malware that is wrongly identified as benign software can easily disseminate and penetrate a target system [5]. Special care has to be taken when analyzing an evasive malware to prevent detection of the analysis environment. Some authors proposed to execute them first on a bare-metal system and then compare their behavior when executed on other emulation- and virtualization-based analysis systems [12]. Other authors proposed tools such as PinVMShield [21] or Arancino [19] detect or circumvent the evasion techniques used by malware to hinder their analysis. Both tools are based on dynamic binary instrumentation, which allow the malware analyst to insert arbitrary code during malware execution. Although this kind of analysis is feasible, it can also be detected through different techniques, as widely reported in the literature (we reviewed all these works in Sect. 2). In this paper, we evaluate three techniques to demonstrate how a framework for dynamic binary analysis (in particular, Intel Pin) can be detected. Those techniques were previously proposed in the literature, but only from a theoretical perspective. Here, we have implemented them to test their detection efficiency. Those evasion techniques increase the attack surface of a dynamic binary instrumentation framework, since they can be used by malware to detect that analysis environment. Moreover, we also provide the corresponding countermeasures to mitigate those evasion techniques. Those countermeasures have been integrated in PinVMShield [21], a plugin-based tool specially designed to circumvent evasive techniques performed by malware. Furthermore, they have also been integrated in the benchmark-like tool eXait [7], which serves to verify if a DBI framework is recognizable. Finally, to prove the effectiveness of our countermeasures, we evaluate their efficacy and performance using SPEC CPU 2006 [6]. The results showed that although the evasive behavior

Reducing the Attack Surface of Dynamic Binary Instrumentation Frameworks

5

can be circumvented, the use of current frameworks of dynamic binary analysis has a great overhead, as already claimed by other authors [20]. This paper is organized as follows. Section 2 reviews the literature. Section 3 is devoted to previous concepts needed to follow the rest of this work, as the DBI framework developed by Intel or the architecture of the PinVMShield tool. Section 4 introduces the new techniques for detecting those analysis environments, as well as the proper countermeasures. Finally, Sect. 5 concludes the work and states future work.

2 Related Work This section presents in chronological order the works related to techniques for detecting a DBI framework, detailing also the countermeasures proposed in each work. We first review the related works in the industry and then in the academy. Most of the existing studies come from industry-related security conferences. Falcón and Riva introduced in [7] the first techniques to detect the presence of the Intel Pin DBI framework in a Windows OS. In particular, thirteen techniques with the corresponding proofs of concepts (PoCs) were presented. Those PoCs were distributed as a sort of a benchmark-like tool called eXait. This tool is useful to verify if a DBI framework is recognizable. Two years later, Li and Li [14] showed that the DynamoRIO DBI framework can be detected in both Windows and Linux environments through ten new evasion techniques. Furthermore, they remarked the transparency problem of DBI frameworks in some cases. Similarly, Hron and Jermáˇr proposed in [10] six new evasive techniques against the Pin and DynamoRIO, providing also PoCs of those techniques. Sun et al. also presented in [22] other six new evasive techniques aimed at the DBI Pin and DynamoRIO frameworks. In addition, they introduced the idea of escaping of the sandbox-like environment provided by a DBI framework, thus compromising the analysis environment. Regarding the academic literature, Rodríguez et al. [21] reviewed and provided a taxonomy of the existing anti-DBI and countermeasures techniques up to that time. Moreover, they released an extensible Pin-based tool called PinVMShield that enabled to analyze applications with those evasive techniques by means of circumventing them. Later, Polino et al. [19] introduced a review and classification of some anti-DBI techniques mainly focused on the Pin DBI framework. Furthermore, they developed a set of approaches to mitigate evasion techniques and used the eXait tool [7] to validate their effectiveness. They also introduced a Pin-based tool named Arancino, which incorporates the proposed countermeasures and allows to retrieve the original binary form of malware programs protected by software packers. Recently, Zhechev published a Master’s thesis [24] raising the question whether DBI frameworks are appropriate tools for analyzing malware and other potentially evasive artifacts. Moreover, he introduced thirteen new techniques to detect the presence of Intel Pin DBI framework in a Linux OS and also demonstrated that escaping

6

A. S. Filho et al.

of such a DBI framework is feasible due to the shared memory model used by DBI frameworks. In this regard, Zhechev stated that the isolation and the stealthiness of the analysis code under DBI frameworks are not guaranteed, and thus, DBI frameworks are unsuitable for building any security-related application. As shown, DBI frameworks are gaining popularity among security researchers as a way to insert arbitrary code during program execution. Following this trend, in this paper, we provide PoCs of three evasion techniques already documented in the literature (but without PoCs) and we give the proper countermeasures.

3 Previous Concepts 3.1 Intel Pin DBI Framework The Pin DBI framework (or Pin for short) enables to build easy-to-use, portable, transparent, and efficient dynamic instrumentation tools. Pin was designed by Intel in 2005 and gives support for the three major desktop operating systems (i.e., Windows, Linux, MacOS X). Pin is composed of the three typical DBI components (depicted in Fig. 1): (1) the application to be instrumented; (2) a dynamic binary analysis tool developed with Pin, normally termed as Pintool; and (3) the DBI engine. The DBI engine consists of a virtual machine (VM), a code cache, and an instrumentation application programming interface (API) invoked by the Pintool. The VM takes as input the native executable code of the application to be instrumented and uses a just-in-time (JIT) compiler to insert the instrumentation code, prior to execution. Then, the resulting instrumented code is saved in the code cache and the execution is transferred to it. After execution, the JIT compiler fetches the next sequence of instructions to be executed and generates more code. The emulator unit is in charge of instructions that cannot be directly executed, such as system calls that require special handling from the VM [16]. As shown in Fig. 1, the instrumented application, the Pintool, and the DBI engine are executed in the same memory address space. Pin has been widely used in the scientific community. As a result, a lot of interesting Pintools have been released. In this regard, it is worth mentioning Pintools that enable debugging for instrumented applications, such as PinADX [15] or the tool introduced in [18].

3.2 PinVMShield PinVMShield [21] is a tool to detect and circumvent evasion techniques used by analysis-aware malware. The tool follows a plugin architecture, enabling an easy extension for covering other evasion techniques than the currently supported ones.

Reducing the Attack Surface of Dynamic Binary Instrumentation Frameworks Fig. 1 Pin’s software architecture (adapted from [16])

Pin Application

JIT Compiler Dispatcher

same address space

Pintool Code cache

Virtual Machine (VM)

7

Instrumentation APIs

Emulation Unit

Operating System (OS)

It was released under GNU GPL version 3 license and its source code is publicly available at https://bitbucket.org/rjrodriguez/pinvmshield/. PinVMShield is mainly focused on Windows OS and uses two granularity instrumentation levels, at the routine level and at the instruction level. This allows the tool to detect evasive behavior based on Windows APIs, such as checking the presence of a software debugger though IsDebuggerPresent or CheckRemote Debugger, or based on specific assembly code instructions (e.g., sidt, sgdt, or sldt [8]). PinVMShield currently addresses the evasive behavior performed by software binaries to detect virtual environments (in particular, VirtualPC, VMWare, and Virtualbox), debuggers (WinDBG, OllyDBG, and ImmunityDebugger), and sandboxes (WinJail, Cuckoo Sandbox, Norman, Sandboxie, CWSandbox, JoeSandbox, and Anubis).

4 Evasive Techniques and Countermeasures In this section, we address some of the evasion techniques that a malware can incorporate into its code to detect when it is being executed inside a DBI framework. Furthermore, to the best of our knowledge, we are the first to provide countermeasures for those evasion techniques, which are introduced next.

4.1 Evasive Techniques An exhaustive list of techniques to detect DBI frameworks is discussed in [22, 24]. There, small pieces of code are given as PoCs, as well as proper countermeasures,

8

A. S. Filho et al.

specially designed for the Intel Pin DBI framework and Linux OS running on top of Intel x86-64 architectures. Among those evasive techniques for which no countermeasures are provided, there are few of them particularly relevant, since they may be used as an attack vector by malware against the analysis environment. In particular, those techniques are the neglecting of the No-eXecute bit, TLS detection, and code cache signatures detection. In the rest of this section, we describe each of those evasion techniques in detail. Furthermore, we also provide PoCs and countermeasures. Unlike [24], we focus on the Windows OS, as it is the most prevalent system attacked by evasive malware [21]. To demonstrate the effectiveness of these evasion techniques against Pin, our PoCs have been integrated as plugins for eXait [7]. Those plugins are publicly released under GNU GPLv3 and freely available online1 Moreover, our proposed countermeasures have been integrated in PinVMShield [21]. Neglecting No-eXecute Bit. The No-eXecute (NX) bit is a defense mechanism added at hardware level to prevent the execution of data memory pages by the processor. Roughly speaking, the main idea is that no memory zone is simultaneously writable and executable. This feature is incorporated by almost all the processor’s manufacturers, although referred to under different nomenclature (e.g., execute disable bit in Intel, enhanced virus protection in AMD, or execute never bit in ARM). Recall that a DBI framework uses a JIT compiler that “recompiles” the application code in conjunction with the instrumented code (see Sect. 3.1). That is, it needs to writethen-execute certain memory zones. Therefore, as suggested by Zhechev [24], any program instrumented by a DBI framework in JIT mode has this protection disabled. Therefore, an application can detect a DBI framework by allocating a new heap space, placing the valid code on it, and then executing it. When the application is not being instrumented, then the execution crashes since the heap memory page has no permission to execute. However, the execution under a DBI framework will continue normally. TLS Detection. Thread Local Storage (TLS) is a feature that allows a developer to provide unique data for each thread, in vector format, accessed by the process using a global index. Roughly speaking, TLS variables can be seen as global variables only visible to a particular thread and not the whole program. In the case of Windows, those per-thread global variables are maintained in the TLS directory, which is a part of the Portable Executable (PE) header of an executable image. The PE header is the header of any Windows executable file. The minimum number of positions in the directory is guaranteed to be at least 64 for any system, while the maximum number is 1088 [17]. According to [22], DBI frameworks as Pin allocate and use positions of that data structure for internal purposes.

1 See

https://github.com/ailton07/eXait_Plugin_PinDetectionByDEPNeglect, https://github.com/ ailton07/eXait_Plugin_CodeCacheDetectionByFEEDBEAF, https://github.com/ailton07/eXait_ Plugin_PinDetectionByTLS.

Reducing the Attack Surface of Dynamic Binary Instrumentation Frameworks

9

Since allocated indexes are shared by any thread in the process, an application can inspect the number of positions allocated in the TLS directory, revealing the presence of a DBI framework. Code Cache Signatures Detection. As reported in [19], there are several artifacts that a DBI tool unavoidably leaves in memory. In this regard, the authors in [22] shown that those memory artifacts can be easily detected by any application under analysis. In the case of Pin, one of those memory artifacts is the hexadecimal pattern 0xFEEDBEAF. This byte sequence is repeated through the memory zones used as code cache by Pin, as an identification fingerprint. Therefore, an application can detect those memory patterns by means of memory scans. The presence of those byte patterns in the memory allocated by a process can be used as a criterion to indicate the presence of Pin in this particular case.

4.2 Countermeasures For the development of the proposed countermeasures, we used the tool PinVMShield since its plugin-based architecture makes easier for a developer to incorporate new mitigation techniques against an evasion technique. To foster research in this area, the developed countermeasures are also publicly released an available online.2 In the following, we briefly describe each countermeasure. Neglecting No-Execute Bit. As countermeasure, we can rely on the set of Windows APIs that allow a developer to check for specific memory addresses and permissions. The developed countermeasure detects when an attempt to execute a piece of code from an invalid region (non-executable memory) occurs, and allows the binary analyst to take action afterward. A more transparent implementation would raise an exception with status STATUS_ACCESS_VIOLATION, imitating the behavior outside the DBI framework. TLS Detection. Since the TLS access is made exclusively using APIs provided by Windows OS, the proposed countermeasure relies on monitoring those APIs and logically isolating the positions used by Pin. Thus, an attempt to read or modify a position already allocated by the DBI tool is redirected to a position unrelated to it. The PinVMShield tool already provides facilities for intercepting and modifying Windows APIs, through the ReplaceWinAPI method. Code Cache Signatures Detection. As suggested in [19], a process can scan its entire memory space performing an inspection of every allocated memory page. In Windows OS, this can be done through APIs as VirtualQuery or NtQueryVirtualMemory. To mitigate this evasion technique, it is necessary to establish a memory access policy, telling which memory addresses are protected memory addresses. This can be done by creating a blacklist of memory addresses related to the DBI framework which 2 See

https://github.com/ailton07/PinVMShield.

10

A. S. Filho et al.

cannot be accessible by the instrumented application. In this regard, Pin provides a set of APIs, such as CODECACHE_AddTraceInsertedFunction, which allows an analyst to monitor the creation of code caches and thus get their addresses, enabling the creation of that blacklist upon execution. Once the blacklist is initialized, those Windows APIs that allow memory page inspection can be monitored to bypass the queries of any protected memory regions. Our developed countermeasure intercepts those calls when an attempt to obtain information from a protected memory region is done and returns a zero value, indicating that the function failed.

4.3 Efficacy of the Countermeasures and Performance Impact These countermeasures were incorporated in PinVMShield to evaluate their effectiveness. We developed the small pieces of code implementing evasive techniques and execute them with PinVMShield and our countermeasures enabled. As a result, TLS Detection and Code Cache Signature Detection were mitigated successfully. Similarly, Neglecting No-Execute Bit was detected during the execution attempt, allowing the analyst to take appropriate action at run time. To measure the impact caused by the use of the PinVMShield tool in the application performance, we used the benchmark tool SPEC CPU2006 [6] widely used to evaluate performance measurements on DBI tools [1, 4, 16]. While executing SPEC with PinVMShield, features not related to the countermeasures presented here were disabled to avoid performance degradation by other means. As experimentation environment, we used a virtualized environment on top of a KVM processor 8 cores 3.41 GHz, with 16GB of RAM memory. The virtual machine was running a Windows 7 SP1 x64 with Intel Pin 2.14-71313 and Microsoft 32bit C/C++ compiler v16 for SPEC2006 compilation. Table 1 summarizes the results obtained for each one of benchmark tools of SPEC. The first column lists the names of the benchmarks. The second column tells the execution time (in seconds) of Pin, while the third shows the execution time (in seconds) of Pin with the PinVMShield tool. Finally, the fourth column presents the overhead when PinVMShield is used. On average, the overhead was 59.73% with a standard deviation of 98.68%.

5 Conclusions and Future Directions Malicious software able to detect an analysis environment is a potential threat, since they can behave differently to evade the classification as malware when they detect an analysis environment.

Reducing the Attack Surface of Dynamic Binary Instrumentation Frameworks

11

Table 1 Overhead introduced by PinVMShield with our countermeasures Benchmark Instrumentation time PinVMShield time PinVMShield (s) (s) overhead (%) 400.perlbench 401.bzip2 403.gcc 429.mcf 445.gobmk 456.hmmer 458.sjeng 464.h264ref 471.omnetpp 473.astar 483.xalancbmk

484 560 443 192 521 680 606 1069 294 339 287

534 562 1150 209 537 718 607 4071 819 326 326

10.3305 0.3571 159.5936 8.8541 3.0710 5.5882 0.1650 280.8231 178.5714 –3.8348 13.5888

In this paper, we have studied three evasion techniques of DBI frameworks. Those techniques are used as attack vectors by malware to detect an analysis environment. We have provided small pieces of code as proof of concepts of those techniques, as well as their countermeasures to reduce the attack surface of DBI frameworks. Those countermeasures were developed on top of the PinVMShield tool. Both proof of concepts and countermeasures have been evaluated in a virtualized environment to prove their effectiveness. Our experiments showed a performance overhead close to 60% by the use of developed countermeasures. Unlike the work in [24], we have shown that DBI frameworks are suitable for security purposes: when equipped with the appropriate tools, the requirements needed in a security analysis context (i.e., stealthiness and isolation) are achieved. However, the performance overhead introduced by the use of DBI frameworks is still an issue. Nonetheless, we argue that it is necessary to keep observing the growth of evasion techniques focused on DBI tools. Despite efforts done to develop DBI evasion techniques countermeasures, there are still evasion techniques that threaten the stealthiness of DBI frameworks and pose a challenge to system security professionals. For instance, the intrinsic problem of overhead detection that basically affects to any dynamic analysis tool. Acknowledgements The research of A. Santos Filho and E. L. Feitosa supported in part by the FAPEAM Proc. No. 009/2017 and by the Federal University of Amazonas (UFAM). The research of R. J. Rodríguez was supported in part by the University, Industry and Innovation Department of the Aragonese Government under Programa de Proyectos Estratégicos de Grupos de Investigación (project references T21-17R).

12

A. S. Filho et al.

References 1. Arafa, P.: Time-aware dynamic binary instrumentation. Ph.D. thesis, University of Waterloo (2017) 2. AV-TEST GmbH: The AV-TEST Security Report 2017/2018 (2018) 3. Balzarotti, D., Cova, M., Karlberger, C., Kirda, E., Kruegel, C., Vigna, G.: Efficient detection of split personalities in malware. In: Proceedings of the Network and Distributed System Security Symposium (NDSS) (2010) 4. Bruening, D., Zhao, Q., Amarasinghe, S.: Transparent dynamic instrumentation. ACM SIGPLAN Not. 47(7), 133–144 (2012) 5. Carpenter, M., Liston, T., Skoudis, E.: Hiding virtualization from attackers and malware. IEEE Secur. Priv. 5(3), 62–65 (2007) 6. CPU2006, S.: Standard performance evaluation corporation. https://www.spec.org/cpu2006/ (2006) (Online) 7. Falcón, F., Riva, N.: Dynamic binary instrumentation frameworks: I know you’re there spying on me (2012) 8. Ferrie, P.: Attacks on virtual machine emulators. Symantec Adv. Res. Threat. Res. 1–13 (2007) 9. Greamo, C., Ghosh, A.: Sandboxing and virtualization: modern tools for combating malware. IEEE Secur. Priv. 9(2), 79–82 (2011) 10. Hron, M., Jermáˇr, J.: SafeMachine malware needs love, too. https://www.virusbulletin.com/ uploads/pdf/conference_slides/2014/sponsorAVAST-VB2014.pdf (2014) (Online) 11. Kaspersky lab: Kaspersky lab detects 360,000 new malicious files daily—up 11.5% from 2016. https://www.kaspersky.com/about/press-releases/2017_kaspersky-lab-detects-360000new-malicious-files-daily (2017) (Online) 12. Kirat, D., Vigna, G., Kruegel, C.: Barecloud: Bare-metal analysis-based evasive malware detection. In: 23rd USENIX Security Symposium (USENIX Security 14), pp. 287–301. USENIX Association, San Diego, CA (2014) 13. Kumar, A.V., Vishnani, K., Kumar, K.V.: Split personality malware detection and defeating in popular virtual machines. In: Proceedings of the 5th International Conference on Security of Information and Networks (SIN), pp. 20–26. ACM (2012) 14. Li, X., Li, K.: Defeating the transparency features of dynamic binary instrumentation. BlackHat US (2014) 15. Lueck, G., Patil, H., Pereira, C.: PinADX: An interface for customizable debugging with dynamic instrumentation. In: Proceedings of the 10th International Symposium on Code Generation and Optimization (CGO), pp. 114–123. ACM, New York, NY, USA (2012) 16. Luk, C.K., Cohn, R., Muth, R., Patil, H., Klauser, A., Lowney, G., Wallace, S., Reddi, V.J., Hazelwood, K.: Pin: Building customized program analysis tools with dynamic instrumentation. In: Proceedings of the 2005 ACM SIGPLAN conference on Programming Language Design and Implementation, PLDI ’05, pp. 190–200. ACM, New York, NY, USA (2005) 17. Microsoft: Thread Local Storage. https://msdn.microsoft.com/en-us/library/windows/ desktop/ms686749(v=vs.85).aspx (2018) (Online) 18. Pan, H., Asanovi´c, K., Cohn, R., Luk, C.K.: Controlling program execution through binary instrumentation. SIGARCH Comput. Archit. News 33(5), 45–50 (2005) 19. Polino, M., Continella, A., Mariani, S., D’Alessio, S., Fontata, L., Gritti, F., Zanero, S.: Measuring and Defeating Anti-Instrumentation-Equipped Malware. Detection of Intrusions and Malware and Vulnerability Assessment, pp. 73–96. Springer International Publishing, Cham (2017) 20. Rodríguez, R.J., Artal, J.A., Merseguer, J.: Performance evaluation of dynamic binary instrumentation frameworks. IEEE Lat. Am. Trans. (Rev. IEEE Am. Lat.) 12(8), 1572–1580 (2014) 21. Rodríguez, R.J., Gaston, I.R., Alonso, J.: Towards the detection of isolation-aware malware. IEEE Lat. Am. Trans. 14(2), 1024–1036 (2016) 22. Sun, K., Li, X., Ou, Y.: Break Out of The Truman Show: Active Detection and Escape of Dynamic Binary Instrumentation. Black Hat Asia (2016)

Reducing the Attack Surface of Dynamic Binary Instrumentation Frameworks

13

23. Vishnani, K., Pais, A.R., Mohandas, R.: Detecting & defeating split personality malware. In: Proocedings of the 5th International Conference on Emerging Security Information, Systems and Technologies (SECURWARE), pp. 7–13 (2011) 24. Zhechev, Z.: Security evaluation of dynamic binary instrumentation engines. Master’s thesis, Department of Informatics Technical University of Munich (2018)

Single Sign-on Implementation: Leveraging Browser Storage for Handling Tabbed Browsing Sign-outs Lokesh Ramamoorthi and Dilip Sarkar

Abstract Organizations provide multiple web-based services to its users. To manage easy provisioning and deprovisioning of users, they offer single sign-on (SSO) service as an access control mechanism. To access any SSO service, users need to remember only one set of credentials. These credentials are managed by organization’s identity and access management system (IAM). The identity provider (IDP) is the federated identity management platform to authenticate users, for service providers (SPs), to access and use the services. Users may access the services provided by organization via web browsers. Modern web browsers provide a feature called tabbed browsing, where tabs are widgets within the browser window, so that the users can stay organized when browsing multiple Web sites. In this paper, we analyzed how SSO works in a tabbed browsing environment. Our analysis shows that, in some scenarios, a user may not sign out from some services that he or she was using. This situation may lead to information security attacks. Also, we propose a solution that can be implemented on the browser to ensure a safe sign-out process. Keywords Information security · Access control · Federated identity management · Authentication · Authorization · Identity provider · Service provider · Browser security

1 Introduction Today, every entity (university, organizations, and businesses, etc.) provides multiple web-based services. The user population of an entity changes dynamically, that is, existing users leave and new users are added. For user conveninces, the Information Technology (IT) team of the organization provides Single Sign-on (SSO) service to its L. Ramamoorthi (B) · D. Sarkar University of Miami, Coral Gables, FL 33146, USA e-mail: [email protected] D. Sarkar e-mail: [email protected] © Springer Nature Singapore Pte Ltd. 2020 Á. Rocha and R. P. Pereira (eds.), Developments and Advances in Defense and Security, Smart Innovation, Systems and Technologies 152, https://doi.org/10.1007/978-981-13-9155-2_2

15

16

L. Ramamoorthi and D. Sarkar

users. A user of an entity that provides SSO service needs only one set of credential (say, username and password) to access all available services.1 In a SSO service system, a (physical or logical) Identity Provider (IDP) verifies credentials of all users who want to receive any web-based service provided by the entity. While a user is verified for one of the many web-based services, the user can access all other services authorized to the user because the IDP maintains an active IDP session of the user in the user’s browser. Currently, SSO service has expanded to individual Web sites, where popular social networking sites (such as Facebook, Google, or LinkedIn) work as IDP. This concept is referred to as social login. All the security issues common to SSO are also present in social login system. In this paper, we investigate how single sign-on implementation on tabbed browsing scenarios may risk the information security. The rest of the paper is organized as follows, key components of SSO, how tabbed browsing works, common browser storage mechanisms, how multiple tabs handle information during SSO process, security pitfalls, possible solutions to avoid information security incidents during SSO.

1.1 Key Components of SSO Implementation Many commercial implementations of SSO follow OAuth 2.0 standard [1], a Security Assertion Markup Language (SAML)-based reference architecture. SAML is a XML-based framework [2] for communicating user authentication, entitlement, and attribute information between Identity Provider (IDP) and Service Provider (SP) via web browser. Three main components in a SAML-based SSO implementation are: (i) Web browser of the user, (ii) Identity provider—(often) authentication server of the user’s organization, and (iii) Service provider—the application that the user gets service. Figure 1 illustrates the interaction model of the components in the SSO implementation. Details about SSO components and related information security issues have been presented in earlier work (see [3] and references therein).

2 Tabbed Browsing Observing users’ tendency of accessing multiple Web sites simultaneously by opening multiple browser windows, one tab for each Web site, the browser software companies introduced tabbed browsing. Tabs are browser widgets, where in a single 1 For

instance, federated identity management is a widely used SSO mechanism.

Single Sign-on Implementation: Leveraging Browser Storage …

17

Fig. 1 Browser-centric SSO process

Fig. 2 A tabbed browser window

browser window users can open multiple web pages simultaneously, one for each web page. With the help of tabbed browsing, users can stay organized and it provide them with an organized and a convenient browsing experience. When a user signs in to a Web site on one tab and opens another page from the same Web site on a different tab, both the tabs share the same session information and the user is not required to sign in again. The user adaptation for tabbed browsing is also very positive, because he/she can keep all their browsing activities organized in a single window [4]. Figure 2 shows how a tabbed browser appears to users. Whenever the user clicks on browser icon, such as Mozilla Firefox, Internet Explorer, or Google Chrome, the browser application opens with the default tab. The default tab is highlighted in the Fig. 2 as ‘Browser Tab 1.’ If the user would like to open additional web pages, it can be accomplished by opening more tabs. Each tab displays a web page. If the user decides to have any of the tabs separated in a new window, he/she can do so by dragging it as out of the main window; this transforms the selected tab to a new browser window.

18

L. Ramamoorthi and D. Sarkar

3 Client-Side Storage To provide an enhanced personalized web browsing experience, the service providers send some information to the web browsers about the users’ activities. The browser may store this information on the client, (user’s computer or other devices used for browsing) and retrieves the information when requested. This storage is called clientside storage or browser storage. The information storage and retrieval is enabled by Javascript API functions. Table 1 gives a summary of various browser storage options. A good understanding of all the Web site storage options helps in selecting the most secure and efficient storage solutions for the browsers and their users. In the

Table 1 Standard browser storage options Type of storage Description Data format stored Cookies

Flat file storage

WebStorage API—Local storage

Browser inbuilt Key value [5] storage, data persisted in the system even after session Browser inbuilt Key value storage, data is lost after the tab/browser closure Browser inbuilt No SQL storage, NoSQL format, multimedia supported

WebStorage API—Session storage

IndexedDB storage

CacheAPI and service workers

Key value

Advantages

Disadvantages

Traditional, widely used, simple format

Highly susceptible to information security attacks such as cross-site scripting (XSS) and cross-site request forgery (CSRF) Not encrypted. Complex datatypes not supported

Stored within the browser, scope restricted to the Web site

Stored within the Not encrypted. browser, scope Complex datatypes restricted to the not supported browser session Multimedia objects can also be stored, indexed for faster access

Effective for Multiple formats Useful for offline offline access experiences, HTML, CSS can be stored

Encryption is not inbuilt, but can be integrated with WebCryptoAPI [6], data retrieval may depend on the indexing and implementation Encryption not inbuilt, efficiency and performance is dependent on the implementation

Single Sign-on Implementation: Leveraging Browser Storage …

19

forthcoming sections, we will discuss how current SSO implementations rely more on cookies and the problems associated with it. We will also discuss few possible solutions leveraging other client-side storage methods such as IndexedDB API.

4 Process Sequence in Sign-in, Sign-out Activities In this section, we discuss the process flow during a single sign-on, sign-out activity as seen from a web browser’s perspective. For the discussion here, we assume that the SPs and the IDP belong to different platforms (either physically or logically). Whenever a user attempts to sign-on to any of the SPs, the user is redirected to the IDP. For ease of description, complete discussion is divided into cases: (i) the user has no active session with any of the SPs and (ii) the user has one or more active session with some SPs. .

4.1 User Sign-in Activity: No Active Signed-in Sessions Here, the user has no active session with any SP and the browser has no record of active SSO. Once a user requests for a service via the browser, the SP checks for possible active service session with it; because it is an initial request, the SP requires that the user be authenticated. For this, the SP finds the source of request (Where are you from?—WAYF) and redirects the browser to the user’s IDP. As soon as the browser is redirected from the SP to the IDP for authentication, the IDP checks the browser for existing active IDP session. Since no IDP session exists, the IDP challenges the user for valid credentials and displays the sign-on page; then, the user provides sign-in credentials for possible sign-on. Once the user provides valid credentials (which could be a simple user ID and password or a complicated multi-factor authentication), IDP creates a new active session and asserts it on the browser. IDP stores the information in the browser cookies. In addition, the IDP creates a SAML record response that contains assertion information of the authenticated user for the browser. The assertion information in the SAML record contains information about users organization as well as authorization level. The browser forwards the IDPs SAML record to the service provider. After receiving the SAML assertion info, the SP creates a session for the user; at this point, a service session is initiated. The service provider stores the information either in local storage or in the browser cookies. Figure 3 shows a logical view of the browser storage after the service session is initialized. The tab shown in the figure is the default tab, where the SP session is accessed by the user.

20

L. Ramamoorthi and D. Sarkar

Fig. 3 Browser storage structure with IDP and session info

As presented, in the SSO approach for an initial sign-in, the browser coordinates communication between the SP and IDP. Moreover, browser maintains session information for both the SP and IDP. As we discussed in Sect. 2, the user can access multiple pages of the same SP via. different browser tabs.

4.2 User Sign-in Activity: Tabbed Browsing Let us consider two scenarios where a user accesses services in a tabbed browsing fashion, (1) Scenario 1: the user access to a new web page of the SP where she is currently signed-in and (2) Scenario 2: the user tries to access a new web page of a SP where she is not currently signed-in, but IDP of the new SP is same as the IDP of the SP where she is currently signed-in. Scenario 1: Once a user signs in to a service provider’s Web site, she can access another page of the same Web site opening a new browser tab. Figure 4 shows storage structure when two browser tabs of one SP is accessed. Both the tabs Tab1 and Tab2 access the SP1 session information.

Fig. 4 Browser storage structure for one SP and two tabs

Single Sign-on Implementation: Leveraging Browser Storage …

21

Fig. 5 Browser storage structure for two SPs and three tabs

Scenario 2: In this tabbed browsing scenario, the user signs in for a service and opens a new tab for another service. Let us consider services SP1 and SP2 , both leverage authentication of the same IDP. Initially, user has signed on to service SP1 , and a while later, user wishes to open another browser tab and access another service say SP2 . The service SP2 also leverages the authentication provided by the same IDP of the organization. Figure 5 shows the logical view of the browser storage and how the session information is accessed by multiple tabs. Even though the browser allows access to SP2 without requiring user to provide credentials, there is a risk in this situation, where the user might sign out of one service provider and assume that she will be signed out from all other service providers. The critical consideration here is the status of IDP session may remain active even after user signs out from SP. First, we consider possible situations that arise, because of sign-out by the user.

4.3 Sign-out Activity—Tabbed Browsing with Parallel Sessions In this section, we will discuss the sign-out activity by the user in a tabbed browsing scenario. The key reason we discuss the sign-out activity for parallel tabbed sessions is that, the user does not explicitly sign in when she opens a request in new tab. It is possible that the user can assume sign-out activity and also does not require explicitly signing-out from all the related tabs. Here, there are two cases of sign-out, the user might encounter in the tabbed browsing session. Case 1: When a user signs out from a service from a browser tab, the SP session is invalidated. However, when the user accesses the same service via another tab, the browser does not require user to sign in again, because the IDP session is still active. So during the second time request of a service via a different tab, the service retrieves

22

L. Ramamoorthi and D. Sarkar

SAML from the IDP session and does not explicitly challenges the user for sign-in credentials. Hence, the scenario is bit confusing for the user because the user may assume that the sign-out from one tab of a service signs out all the active sessions, whereas in the actual scenario of SSO implementation as long as IDP session is active, the service is still accessible without requiring user credentials. Case 2: Assuming two services SP1 and SP2 leverages the service of the same IDP, when the user requests for SP2 after signing-out from SP1 , the browser does not prompt for user sign-in credentials. This is because SP2 could get the information from the IDP session and will not prompt the user for sign-in credentials. As shown in Fig. 5, irrespective of how many different services being accessed, the IDP session is shared by all the services.

5 Solutions to Avoid Undesired Active Sessions There are many advantages of SSO. But still there are concerns about the implementation that revolves around the security and privacy of user information [7]. In this section, we discuss how application developers and SSO service developers prevent hidden active sessions.

5.1 Notifying the User During Sign-Out During a SP sign-out activity, the SP can display a notification to the users, You are signing-out from the Service, but you may have other active sessions, including the Identity Provider sessions that was created by your authentication server.

This would forewarn the user that there might be other active and open SP sessions which the user has to manually sign out. This notification would be very helpful in a tabbed browsing scenario where when the user signs out from one tab the service provider can notify that there might be other tabs that are active. The sign-out confirmation link can also include a trigger or link that invalidates the IDP session. This will reduce the risk to a greater extent when the user opened multiple tabs for multiple service sessions.

5.2 Triggering IDP Session Invalidation During Service Provider Sign-out Another solution is, whenever the user signs out of a service, the SP application sends a notification to the IDP that triggers a sign-out process of the IDP session. So

Single Sign-on Implementation: Leveraging Browser Storage …

23

in this case, the IDP session becomes inactive. Any other users trying to access the service has to sign in with their own set of credentials.

5.3 Automatically Sign-out at Browser Window Closing Event With modern browsers and advanced browser scripting technologies, the applications can be implemented where in case if the user closes the browser window, the SP session variables are invalidated. Not many users are aware of the fact that closing a browser window is not the same as a sign-out activity. Implementing autoinvalidation of a session during a window closing event helps where the users do not perform explicit sign-out. This scenario is very appropriate implementation in situations where multiple users accessing a common computer, such as in public library. The auto-trigger IDP sign-out during a service provider sign-out is very effective during tabbed browsing scenario as well, where the browser can give a quick notification to the user and triggers a IDP sign-out. This way the user can feel secure about the sign-out activity that no one can access any services without a user sign-in credentials. So, when new users open the browser, they are prompted for their credentials.

6 Specification for Handling SSO Sign-out Scenarios via. A Browser-Based SSO Manager Module The following proposal emphasizes on browser to handle the security gaps discussed above. In this specification, we create a new module called SSO Manager Module. This module can be implemented as a browser extension. This module does not manipulate or add any additional features to IDP or SP applications. This module provides a browser-based mechanism that ensures a safe sign-out process. SSO Manager module uses a data object named Active SSO SessionsInfo that contains essential information about SP and IDP sessions. Whenever a sign-out event happens, the module will effectively handle the event either by automatically signing-out the user from the session, or displaying a notification for input from the user.

6.1 SSO Manager Module : Data Structure The SSO Manager module is a web browser extension. It saves the information about the active IDP Sessions and the SP sessions that leverage the IDP. To store the information, SSO manager leverages IndexedDB storage of the web browser.

24

L. Ramamoorthi and D. Sarkar

Fig. 6 Active SSO sessions—schema

The main advantage of IndexedDB is that it helps in organizing the IDP-SP hierarchy. Following is the proposed schema for the SSO Manager application. The SSO Manager stores the IDP Session metadata in the IndexedDB. Figure 6 shows us the schema of data stored by the SSO Manager module. Every record holds metadata about the IDP, name of the IDP, IDP’s sign-out URL, and IDP’s timeout threshold. Each IDP Info record contains an array of service provider’s meta data. Every SP metadata contains the information about the service provider such as name or id, SP’s timeout threshold, and SP’s sign-out URL. As we discussed in Sect. 2, users may access the services via multiple browser tabs. Hence, the SP Info data contains the list of tabs that the SP is currently accessed. The structure Tabs contains tab id, parent window id, and the URL accessed by the user. Whenever a user tries to sign in via. SSO, the SSO Manager gathers information from the browser and stores it in its schema. The functionality of the SSO Manager module is described in the following sections.

6.2 SSO Manager Module Implementation : Initial Sign-in Scenario Listing 1.1 Database Entry after IDP Session Initiation 1

{ " IDP Name " : " U n i v e r s i t y A - ID and Auth " , " IDP T i m e o u t T h r e s h o l d " : " 4 5 m " " IDP Sign - out URL " : " https : // univA . com / auth / s i g n o u t " , " SP Info " : [ ]

2 3 4 5 6

}

Listing 1.2 Database Entry after SP Session Initiation 1 2 3 4

{ " IDP Name " : " U n i v e r s i t y A - ID and Auth " , " IDP T i m e o u t T h r e s h o l d " : " 4 5 m " , " IDP Sign - out URL " : " https : // univA . com / auth / s i g n o u t " ,

Single Sign-on Implementation: Leveraging Browser Storage … " SP Info " : [ { " SP Name " : " S m a r t S t u d y LMS " , " SP T i m e o u t T h r e s h o l d " : " 3 5 m " , " SP Sign - out URL " : " https : // lms . s s t u d y . com / s i g n o u t " , " Tabs " : [ { " Tab ID " : " 0 0 1 " , " P a r e n t W i n d o w ID " : " 0 0 1 " , " URL " : " https : // u n i v A . s s t u d y . com / h o m e " } ] } ]

5 6 7 8 9 10 11 12 13 14 15 16 17 18 19

25

}

Let us consider a scenario of user X, a student of university A. To access the services such as learning management system, enrollment financial systems, shared file repository, X has to use the university id. Here, the university is the IDP. X logs into a computer’s web browser and tries to access the learning management system (LMS). When X tries to access the university’s LMS URL, the LMS as a service provider redirects the browser to university’s Identity and Auth Platform (IDP). Once the IDP session is created, the SSO Manager creates a new record in the browser’s IndexedDB. The SSO Manager records the following information in the IndexedDB. And during the sign-in process, after the SP session is created, in Sect. 4.1, the record is updated as follows. Hence after the sign-on is complete, we could infer that SSO Manager recorded the information in IndexedDB of the web browser. A sample entry for the above student sign-on scenario is depicted in Listing 1.2.

6.3 SSO Manager Module Implementation : Additional Tabs In continuation of above scenario, user X would like to access assignments and course information simultaneously on two browser tabs. So she opens the LMS via another tab. The SSO Manager recognizes the event of open tab and modifies the existing entry to add the information about newly opened tab. Listing 1.3 shows the additional information that has been added about the newly opened browser tab. Listing 1.3 Database Entry with additional tabs 1 2 3 4 5 6 7 8 9 10 11

{ " IDP Name " : " U n i v e r s i t y A - ID and Auth " , " IDP T i m e o u t T h r e s h o l d " : " 4 5 m " , " IDP Sign - out URL " : " https : // univA . com / auth / s i g n o u t " , " SP Info " : [ { " SP Name " : " S m a r t S t u d y LMS " , " SP T i m e o u t T h r e s h o l d " : " 3 5 m " , " SP Sign - out URL " : " https : // lms . s s t u d y . com / s i g n o u t " , " Tabs " : [ {

26

L. Ramamoorthi and D. Sarkar " Tab ID " : " 0 0 2 " , " P a r e n t W i n d o w ID " : " 0 0 1 " , " URL " : " https : // u n i v A . s s t u d y . com / c o u r s e s "

12 13 14

}, {

15 16

" Tab ID " : " 0 0 3 " , " P a r e n t W i n d o w ID " : " 0 0 1 " , " URL " : " https : // u n i v A . s s t u d y . com / a s s i g n m e n t s "

17 18 19

}

20

]

21

}

22

]

23 24

}

Listing 1.4 Database Entry with multiple SPs 1

{ " IDP Name " : " U n i v e r s i t y A - ID and Auth " , " IDP T i m e o u t T h r e s h o l d " : " 4 5 m " , " IDP Sign - out URL " : " https : // univA . com / auth / s i g n o u t " , " SP Info " : [ { " SP Name " : " S m a r t S t u d y LMS " , " SP T i m e o u t T h r e s h o l d " : " 3 5 m " , " SP Sign - out URL " : " https : // lms . s s t u d y . com / s i g n o u t " , " Tabs " : [ { " Tab ID " : " 0 0 2 " , " P a r e n t W i n d o w ID " : " 0 0 1 " , " URL " : " https : // u n i v A . s s t u d y . com / c o u r s e s " }, { " Tab ID " : " 0 0 3 " , " P a r e n t W i n d o w ID " : " 0 0 1 " , " URL " : " https : // u n i v A . s s t u d y . com / a s s i g n m e n t s " } ] } { " SP Name " : " E n r o l l m e n t A p p l i c a t i o n " , " SP T i m e o u t T h r e s h o l d " : " 3 0 m " , " SP Sign - out URL " : " https : // u n i v A . e n r o l l . com / s i g n o u t " , " Tabs " : [ { " Tab ID " : " 0 0 4 " , " P a r e n t W i n d o w ID " : " 0 0 1 " , " URL " : " https : // u n i v A . e n r o l l . com / h o m e " } ] } ]

2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36

}

6.4 SSO Manager Module Implementation : Additional Services Similar to the additional tabs, additional services are handled by the SSO Manager. In the scenario, where the user would like to access another service leveraging the same IDP, the information is updated.

Single Sign-on Implementation: Leveraging Browser Storage …

27

If the user X opens the another service (SP2 ) on a new tab, the service application does not requires the user to sign-in again. It leverages the information available from existing active IDP session. Here, in this scenario, the SSO Manager module adds the information of SP2 to the SSO Sessions Info entry. The sample entry is listed in Listing 1.4.

6.5 SSO Manager Module Implementation : Sign-out Scenario The key purpose of the SSO Manager is to efficiently manage the sign-out scenarios. If the user signs out of a service from a tab, SSO Manager interprets the event and displays a prompt to the user about other active tabs of the same service provider. In the student user example, if the student chooses to sign out from the assignments tab, SSO Manager prompts a message for input from the user regarding all the active tabs and services for the IDP. Once the user confirms about signing-out of all the services, SSO Manager removes the SSO Sessions Info entry, once the user signs out from the IDP and SPs. SSO Manager can also handle situations that arise from accidental closing of browser or tab windows. If there are no browser windows in open state, the SSO Manager can be configured to invoke the sign-out routine of the IDP and all the respective services.

7 Conclusion With growing cybersecurity attacks in the digitized world, everyday hackers and malicious actors are looking for exploiting access credentials of naive users. Getting an access to one user’s account opens a multitude of possibilities for hackers, such as phishing attacks, ransomware attacks, data corruption, and so on. In this paper, we analyzed single sign-on (or federated identity management) process is handled in tabbed browsing environment, identified potential security compromise scenarios, and proposed a solution to avoid such scenarios. Therefore, it is essential to completely understand the process of single sign-on and eliminate the gaps for a safer access control.

References 1. Hardt, D.: The OAuth 2.0 authorization framework, RFC 6749. Internet Engineering Task Force (2012). https://tools.ietf.org/html/rfc6749 2. Kemp, J., Cantor, S., Mishra, P., Philpott, R., Maler, E.: Assertions and protocols for the OASIS security assertion markup language (SAML) v2.0. OASIS (2015). http://saml.xml.org/samlspecifications

28

L. Ramamoorthi and D. Sarkar

3. Ramamoorthi, L., Sarkar, D.: Single sign-on demystified: security considerations for developers and users. In: Rocha, Á., Adeli, H., Reis, L.P., Costanzo, S. (eds.) Trends and Advances in Information Systems and Technologies, pp. 185–196. Springer International Publishing, Cham (2018) 4. Dubroy, P., Balakrishnan, R.: A study of tabbed browsing among Mozilla Firefox users. In: Proceedings of the SIGCHI Conference on Human Factors in Computing Systems, CHI ’10, pp. 673–682. ACM, New York, NY, USA (2010) 5. Mozilla developer network - web storage API. https://developer.mozilla.org/en-US/docs/Web/ API/Web_Storage_API 6. W3c docs - web cryptography API (2017). https://www.w3.org/TR/WebCryptoAPI/ 7. Sun, S.T., Pospisil, E., Muslukhov, I., Dindar, N., Hawkey, K., Beznosov, K.: What makes users refuse web single sign-on?: An empirical investigation of openid. In: Proceedings of the Seventh Symposium on Usable Privacy and Security, SOUPS ’11, pp. 4:1–4:20. ACM, New York, NY, USA (2011)

Cybernetic Dependency Capacity Jorge Barbosa

Abstract The state of development of the information society of a particular country can have a great influence on the outcome of a cyberwarfare in which it is involved. This influence, in the possible outcome of a war of this type, can be positive or negative. In addition, contrary to what might initially be thought, a low development of the information society can be advantageous to obtain a victory, if one strictly considers the scenario of cyberwar and the countries in confrontation. This way we analyze this influence and relationship between the information society and cyberwarfare in the scope of security and defense. We designate this dependency as Cybernetic Dependency Capacity. This dependence is a very important factor in the results of cyberwarfare actions and can be positive or negative for both technologically developed and less-developed countries. For these reasons, the destiny of any country, within the scope of a cyberwar, is very influenced by the dependence of this country of their information society. As a major development of the IT society in one country will not make it theoretically victorious in a cyberwarfare with a country less IT developed. This fact, which is apparently paradoxical, is because the larger information society of a country is more informatics systems have to exist. Then, if much more informatics systems exist, more cyber-doors are open for cyber-attacks, because the probability of these systems will be unprotected, from the point of view of cyber-attacks, increases with the growth in the number of computers systems. For the same reason, the countries with less dependence on the information society do not have this cyber-permeability and may be more cyber-resilient. In the limit, theoretically, the least dependent cyber-country can be the winner and the most dependent and as such, technologically more developed, be the loser in a cyberwar between the two. These scenarios are also applicable to non-states and terrorist organizations and for these reasons should be analyzed and considered in the definition of the cyber-defense policies of any country, namely in the factors like a Cybernetic Attack Capacity and a Cybernetic Dependency Capacity. J. Barbosa (B) Coimbra Polytechnic – ISEC, Coimbra, Portugal e-mail: [email protected] Portuguese National Defense Institute, Lisbon, Portugal © Springer Nature Singapore Pte Ltd. 2020 Á. Rocha and R. P. Pereira (eds.), Developments and Advances in Defense and Security, Smart Innovation, Systems and Technologies 152, https://doi.org/10.1007/978-981-13-9155-2_3

29

30

J. Barbosa

Keywords Cyber Dependence · Cyber-Warfare · Cyber-Defense · Cybersecurity · Cyber Resilience · Information Society

1 Introduction The benefits to civil societies resulting from the massive use of information technology in all aspects of that society, namely service provision, government, central and local administration, banking and finance, transport management, logistics, food and fuel distribution, etc. are now an undisputed reality in all minimally developed countries. It can even be said that it is no longer possible at the present for those countries to survive without such means and facilities provided by the information society. The maintenance of the information society and the means necessary for its operation are therefore of paramount importance. Their allocation or destruction is therefore unthinkable in the current development context. Cyber-criminal activities are usually associated with common cybercrime, i.e., activities related to the cybersecurity of individuals or companies. However, similar actions, with a larger scale and scope, can be used in cyber-actions against state sovereignty, known as cyberwarfare actions. Actions, even relatively small and isolated in their range and effects, such as acts by hackers, which either individually or in small groups, that trigger cyber-actions against individuals or companies, can have relatively serious consequences. The issues of great concern to many security and defense agencies and services are that they do not come from these actors, but are orchestrated and unleashed by other actors, notably countries, which have specialized units. If concerted and large-scale action is taken against a country with the exclusive use of computers means, it can profoundly affect the IT systems of that other country, in particular, those linked to its critical infrastructures. The situation could become quite serious for the country under attack and may even lead to its paralysis, subjugation and loss of national sovereignty. Cyberwarfare [1, 2] designates this type of action. Given its intimate connection with the information society, we will then analyze the implications that a lesser or greater dependence and development of the IT may have in the countries, in case to be targeted of a cyberwar. Those cyberwarfare actions can affect not only to military targets and to installations but also to civilian targets, such as critical infrastructures linked to the functioning of modern civil societies, seriously affecting the lives of such societies. Even if there is no great physical destruction, both civilian and military, and therefore the losses do not seem to be significant because they are not visible, for the common citizen this “modern” war can be as harmful as the classic wars that occurred in past centuries. Their daily and habitual experiences may be deeply affected, in particular, by taking away the benefits and comforts they currently enjoy.

Cybernetic Dependency Capacity

31

2 Known Cyber and CyberWarfare Actions In what concerns to security and defense of states, and in one conventional perspective, the most relevant actions are those that are carried out with the aim of triggering actions that undermine the designated sovereignty of the states. In this perspective, two main groups of actions are considered: Those that can be considered actions of common criminality, although they can be quite disturbing of the daily experience, like actions of domestic terrorism; or the actions that disturb the order and the civil life of the society of that country to the point of endangering the sovereignty of the states. The border between these two types of actions may be tenuous and so it is often difficult to establish. In most democratic states, according to this separation, the first set of these actions are dealt with by police or internal security agencies (and in the case of actions with external origin in collaboration with other external security organs). For the second set of actions, as they are detrimental to the national sovereignty of the states, are treated by the organs that defend this sovereignty, that is, by the armed forces of the states. This model is also usually followed with regard to actions triggered by the use of cybernetic means, i.e., actions triggered in cyberspace. For the particular characteristics of this environment, there are also very particular and specific considerations that have to be addressed. First, the difficulty of separating and classifying the actions initiated soon in the establishment of the classification frontier of such actions. Physical borders, as in the case of conventional external aggressions, do not limit actions in cyberspace. In addition to the possible use of entities directly linked to foreign governments, organizations or individuals acting on their own may also trigger cyber-actions that threaten national sovereignty. They will not be simple cybersecurity actions, but cyberwarfare actions. The means used in common cybersecurity actions and in cyberwarfare actions may be similar, but the scope and latitude of cyberwarfare put the sovereignty of the states into question, hence being classified in this particular way. Thus, many states have developed within their conventional armed forces cyber corps to deal with such cyber-actions. At the same time, the internal intelligence and security forces have developed specialized units to deal with common cybersecurity actions. In the case of national sovereignty, it may be understood by the states affected by such actions, not to divulge much the occurrence of cyber-actions of which they were targeted. In this way, there is not much information about such acts and what there is may not be reliable. In the literature, by its impact and apparent, there are references to some of these actions, indicated as having been perpetuated by groups or individuals against other states or perpetuated states against other states or even regions of the same state, in situations of, for example, attempts of political separation by these regions. Many other cyber-attacks are may be due to the rise of “no state” hackers groups. The most referred attack methods used in the actions identified with “no state” groups as Traffic Redirection and Web sites defacement. The most attack methods

32

J. Barbosa

identified with the actions of states are distributed denial of service (DDoS), viruses and trojans and a specific cyberweapon, such as those used in the famous action against Iran known as Stuxnet, [3]. The DDoS attacks are also used by “no state” groups. In the literature, [1, 4, 5] are reported many cyber-attacks actions: (1) USA and South Korea: On the July 4, 2009; (2) Iran: During the disputed Iranian presidential elections of June 14, 2009 and the most cyber-action know by Stuxnet; (3) Tatarstan: In June 2009, the president of Tatarstan’s Web site was knocked offline and Internet access was lost; (4) USA: On April 21, 2009, the Wall Street Journal reported that security around the Pentagon’s multi-billion-dollar Joint Strike Fighter project was compromised; (5) Kyrgyzstan: On January 18, 2009, a DDoS attack shuttered two to three of the nation’s four ISPs for several days; (6) Israel and the Palestinian National Authority: In December 2008 along with Israel’s military action designated Operation Cast Lead against Hamas; (7) Zimbabwe: In December 2008, a action reported by Concerned Africa Scholars; (8) Myanmar: On September 23, 2008, the government launched DDoS attacks against three Web sites that support the monks and many others. The above actions can be considered at the border between common cybersecurity actions and cyberwarfare actions. Better separation will be achieved if it is precisely determined those as put in danger the national sovereignty of the affected states. Many other cyberwarfare actions are allegedly attributed to Russia against Estonia, Georgia and Ukraine. These cyberwarfare actions were triggered using DDoS, SQL Injection, SQLi, and cross-site scripting (XSS) attacks. In the strict sense of cyberwarfare, are used specific cyberweapons. These cyberweapons are developed based on the exploration of computer systems vulnerabilities, namely not yet known vulnerabilities. These vulnerabilities are known as zero-day vulnerabilities, ZDV, and based on these vulnerabilities are developed specific applications named zero-day exploits, ZDE, which can affect and even paralyze the computer system affected by this ZDV. The most known cyberwarfare action, known as Stuxnet, used cyberweapons, which exploit ZDV’s in the computer systems of nuclear centrifuges of Iran, [3].

3 Cybernetic Dependency Capacity As far as conventional wars are concerned, on a primary level, the military and political strategists and leaders of a country equate the potential, in the face of a possible and determined enemy, through two factors: the “Attack Capacity” and the “Defense Capacity” of both your country and that of the eventual enemy country. From the consideration of these capacities, they infer who, in the case of a conventional war between these two countries, theoretically, can become the winner or the loser of the same. In the case of a cyberwar, these considerations have to be different, given the much more complex scenarios. The key factors to be taken into account should be

Cybernetic Dependency Capacity

33

not only the two above-mentioned, adapted to the cyberwar, but three factors, namely: “Cybernetic Attack Capacity,” “Cybernetic Defense Capacity” and a third and new factor regarding conventional warfare, the “Cybernetic Dependency Capacity” of the countries involved [6]. Then, the Cybernetic Dependency Capacity of a country, as the dependency that this country and its society have on the IT organization and the complexity and density of connections and interoperability between their IT systems. The larger this organization and the complexity and density of connections of its computer systems, the greater the country’s dependence on IT and the greater vulnerability to eventual computer crashes. A serious computer fault in an important critical computer system can be very severe for any modern country and, in extreme cases, can even paralyze the country or sectors thereof. Modern and developed Western societies, as well as those of some Asian countries, have a very high Cybernetic Dependency Capacity and as such are extremely sensitive to problems in their computer systems, in particular, those resulting from cyberwar. Less-developed societies are normally often underdeveloped in terms of the massive use of information technology in the various organizational aspects of that society. This causes them to have a very low dependency cybernetic capacity. Because they do not rely heavily on computing for their day-to-day operations, they are not so sensitive to computer glitches, nor do they have computer “targets” to attack.

4 Influence of Information Society in a Cyber Warfare It can even be said that it is no longer possible, at present, for the countries to survive without such means and facilities provided by the information society. Then, the maintenance of the information society and the means necessary for its operation are, therefore, of paramount importance. Their allocation or destruction is therefore unthinkable in the current development context. Actions, even relatively small and isolated in their range and effects, such as acts by hackers, which either individually or in small groups, trigger cyber-actions against individuals or companies, can have relatively serious consequences, [1, 7]. The issue of great concern to many security and defense agencies and services is that they do not come from these actors, but are orchestrated and unleashed by other actors, notably countries, which have specialized units. If concerted and large-scale action is taken against a country by means of the exclusive use of computerized means, it can profoundly affect the IT systems of that other country, in particular, those linked to its critical infrastructures, [5, 8]. Given its close dependence, the level of development of the information society in a given country may make it very likely to be heavily affected by cyberwarfare. This aspect is amplified by the fact that even in technologically advanced countries and, as such, very dependent on computer systems for the most diverse and banal activities, a large part of its population still has a very big illiteracy in computer

34

J. Barbosa

operation. This makes many of the computer systems unprotected or poorly protected by seemingly harmless actions of their users and as such permeable to cyberwarfare. The vulgar actions of hacking are often minimized and typical considered as actions in the scope of cybersecurity, related to attacks against personal or corporate’s computers. However, they may be hiding real cyberwarfare actions. In particular, they may be masking pre-preparation actions for future cyberwarfare actions, such as DDoS attacks on government, military or critical infrastructure installations. The computer systems thus affected stay then in standby. If they necessary to use them for a cyberwar will already be prepared. Due to these minimizations of the importance of the attacks or their misclassification, this preparation is made without the knowledge of the legitimate owners of these systems. This failure to correctly consider or classify such acts, further increases the criticality and permeability of systems greatly affecting the Cybernetic Dependency Capability. Thus, a country’s less or greater dependence on the information society and the computer skills of its population, namely your education and sensitivity for the computer security, has a direct implication on the degree of cyberwarfare permeability of the cyber-attacked countries. The situation could become quite serious for the country under attack and may even lead to its paralysis, subjugation and loss of national sovereignty, [4].

5 Cybernetic Dependency Capacity and Cybernetic Resilience Considering only the cyberwarfare, i.e., not considering conventional wars at all, due this great dependence of the information society, an underdeveloped country can be a cyberwar winner and a highly developed country can be a loser in the same war, [6]. This is a very interesting paradox. This may occur if the less-developed country has minimal cyber-attack and cyberdefense capabilities in such a way that it can inflict significant damage on the computer systems, especially the critical ones, in the opposing country. This may also have good cyber-attack capabilities, but as the other country does not have a minimum of IT structures, they do not have what to attack in the other less-developed country in order to paralyze it from a computer point of view. However, the developed country, attacked in the cyberwar, can get their systems deeply affected and since it is very dependent on these systems, it can even paralyze. Of course, the consideration of the conventional war component in this equation can greatly alter this end result, especially if, as usual, the most developed country has minimum conventional warfare capabilities and so can retaliate conventionally and inflict significant damage on its opponent. However, this question about the use of conventional means of warfare can be critical if the two countries are not geographically close. The displacement and positioning of conventional media can

Cybernetic Dependency Capacity

35

be difficult, time-consuming and late in the theater. Military logistics may have even been affected by cyber-attacks, which would further affect conventional operations. From a mere speculative point of view and considering some recent developments in relations between these two countries, such a situation could hypothetically occur in a strictly cybernetic confrontation between the USA and North Korea. American society is a lot, if not totally, dependent on information technology for its functioning. For its part, and to the best of our knowledge, North Korean society is not. In the country, the informatics is considered incipient and even the small use of computing means is practically done by the governmental members, which are very controlled. In addition, the computer connected to the outside of this country is practically non-existent, which means that there are practically no “doors” to the entry of cyberattacks. According to intelligence, North Korea has relatively developed cyberwarfare capabilities and there are several specialized and highly trained military units for cyberwarfare, i.e., cyber-acts. The hypothetical considerations made in respect of these two countries could be made on a few others. An obvious example of the influence of Cybernetic Dependency on cyberwarfare is the case of Estonia. Shortly after Estonia became independent of the Soviet Union, it promoted a policy of development and modernization of the country. One of the biggest stakes was the massive introduction of information and knowledge technologies. With this policy, it has become one of the countries with most modern computerized public services. Numerous public administration services, banks and services have emerged, based on computer systems with which Estonian citizens can interact directly. Considering Cybernetic Dependency, Estonia became a country with a high cybernetic capacity for dependency, i.e., a country potentially very vulnerable to cyber-attacks. On 27 April 2007, cyber-actions was launched in the form of violent DDoS attacks targeting a number of IT systems in Estonia, including important government Web sites, parliament, banks and the communications system. These actions became known as the “Estonian Cyberwar.” It is assumed that for these DDoS attacks, a net bot was created consisting of 85,000 servers having the attacks lasted three weeks and having been attacked sixty Web sites and the almost stoppage of the country, [3]. These cyber-actions have never been officially assigned to any country. However, the Estonian Foreign Minister attributed the attacks to Russia, which always denied their official participation in them and considered them as actions of individuals who acted as “Patriotic Hackers.” Eventually, if the country had not promoted this development policy of its IT, it would not have become a desirable and possible target for these actions, which were facilitated by the fact that the country had a high Cybernetic Dependency Capacity. The use of computerized means in cyberwars could also be considered by nonstates or by terrorist organizations. Theoretically, these states or organizations do not have the capabilities they can use to trigger conventional attacks to countries that are minimally capacitated from a conventional point of view. This lack of conventional capabilities, on the part of such countries and organizations, could result from the

36

J. Barbosa

lack of means, structures, organization, and even logistics to trigger conventional attacks, conduct, and sustain them for as long as such conventional wars lasted. On the contrary, it may theoretically be relatively easy for them to launch cyberwars against some countries, even large conventional powers. The fact that they are not highly structured and not very dependent states or organizations leads, according to the previous analysis, to a very low or even zero Cybernetic Dependency Capacity because they do not have behind a civil society and even military organization dependent on the modern information technologies. As such, they are not very likely to be, as well, cyber attacked. Another feature that could create interest in this type of non-states and especially in terrorist organizations is that their targets, in addition to their immediate areas of operation and influence, usually extend to other countries, especially Western ones. As these latter countries are not generally geographically close to the theaters of action of such organizations, the use of cyber-media would be more viable for these organizations as well. From the operational point of view, however, there would be many constraints that we consider impeding or at least very restrictive in triggering large-scale cyberwarfare against these countries, by these organizations. First of all, the need for human and technological resources, which are fundamental to trigger these large-scale actions, would be very large, and as such it would be very difficult for them to operationalize such actions. In other words, such states and organizations theoretically have a very low or even zero cyber-attack capacity. However, this aspect will not be a deterrent or an obstacle to the launching of small-scale cyberwarfare, which will nonetheless be disturbing in the societies of the countries under attack. Cyber-attacks well targeted and directed at critical infrastructure in any country can be quite nefarious. We could compare these types of actions to nuclear attacks. For similar reasons, non-states or terrorist organizations do not have the capacity to launch large-scale nuclear attacks. However, theoretically, they could launch isolated and limited critical actions, for example, using the so-called dirty bombs, which are not necessarily intended to cause large-scale physical destruction, but only to contaminate an area use for a long period and hinder the normal functioning of the country. These organizations can theoretically and with the same objective launch cyber-“dirty bombs” actions. These non-states and organizations have to be monitored from the point of view of their cybernetic activities and cannot be neglected by the cyber-defense structures of the countries. Policies to define the cyber-defense capacity should consider these possibilities and be a concern to be taken into account, in order to make more resilient societies organized and very dependent on the information society given the great Cybernetic Capacity of Dependence of such societies. In the same way as in common cybersecurity actions, protective measures can be used in computer systems and action and prevention plans established [2, 5, 8]. The difficulty with regard to cyberwarfare is the unpredictability of such actions and the need to, quickly, take decisions that stop or at least minimize the effects of such actions, in particular, if are attacked Critical Infrastructures, such as hydroelectric dams, banking systems and so on.

Cybernetic Dependency Capacity

37

However, the main problem concern with cyberwarfare, and what the usual measures of protection are difficult to prevent, is the use of specific weapons against the computer systems that control these critical structures, in particular the cyberweapons developed to exploit the vulnerabilities not known in those systems, i.e., the zero-day exploit. It may even be impossible to take preventive and protective measures for these cyberweapons. The use of ZDE is a major challenge to the cybernetic resilience of any country.

6 Conclusion The advantages to civil societies of countries, resulting from the massive use of information technology, are now an indisputable reality in all minimally developed countries. One can even say that it is no longer possible for modern civil societies to survive without the means and facilities provided by the information society. Then, the maintenance of the information society and of the means necessary for its operation is therefore of paramount importance, since the countries could be paralyzed by their affectation or even their destruction. This dependency has created what we call as Cybernetic Dependency Capacity. This factor, in our approach, allows us to consider the influence in the daily experience of a civil society due to the affectation, by cyberwarfare actions, of country’s computer systems, namely those related to the control of its critical infrastructures. The greater this Cybernetic Dependency Capacity, the more exposed that country is to the effects of actions affecting those systems. Generally, and only from the point of view of cyberwarfare, a country very dependent on the information society is, maybe, much more vulnerable to these actions than others less dependent. This conclusion leads us to another, which translates into a very interesting paradox: strictly considering cyberwar, that is, not considering conventional war at all, an underdeveloped country can be a winner in a cyberwar and a developed country may be losing in this same war. The possible use of cyberweapons by non-states or by terrorist organizations was analyzed. From this analysis, we concluded that its use, by such organizations is quite possible and may be desirable. The use of these means to attack certain countries would be more viable to them, than the use of means of conventional warfare. As the Cybernetic Dependency Capacity of such non-states and organizations is very low if not null, cyber-retaliation against these countries and organizations in order to cause them harm would be practically impossible. Then, the very low Cybernetic Dependency Capacity becomes an advantage for such groups. However, such countries would have to overcome many constraints in order to trigger cyber-actions on a large scale, which is virtually impossible for them. In this way, their cyber-actions, to occur, should be limited to isolated attacks on well-selected targets, namely critical infrastructures, which would nevertheless create difficulties for modern societies, especially Western ones, strongly based on information society and as such with a great Cybernetic Capacity of Dependency. Then, the cybernetic activities of these “no

38

J. Barbosa

states” groups and organizations have to be permanently monitored by cyber-defense intelligence. The possibility of using specific cybernetic weapons based on the vulnerabilities of the computer systems, namely vulnerabilities known as zero-day vulnerabilities, ZDV, which allow the development of applications, called zero-day exploit, ZDE, that, using these vulnerabilities, possibility the attack and even take the control of the affected computer systems, are the main concerns in the field of cyber-defense. It may even be impossible to take preventive and protective measures for these cyberweapons, making the use of ZDE with cyberweapons the biggest challenge to the cybernetic resilience of any country. This way, due to this high dependency, in the civil and military areas, of the information society, the policies of any country should include in the cyber-defense strategies the consideration of your Cybernetic Capacity of Dependence, as a concern to be taken into account in order to make the societies more cybernetic resilient. Acknowledgements I thank the Portuguese National Defense Institute as well as the ISEC/IPC for the support and facilities granted. This paper is based on the work Final Report of National Defense Course, done by the author in the Portuguese National Defense Institute.

References 1. Carr, J.: Inside Cyber Warfare, 2nd edn. O’Reilly Media Inc., Sebastopol, CA, USA (2012) 2. Ranger, S.: What is cyberwar? Everything you need to know about the frightening future of digital conflict. https://www.zdnet.com/article/cyberwar-a-guide-to-the-frightening-futureof-online-conflict/ (2018). Acedido em 25 Setembro 2018 3. Zetter, K.: Countdown to Zero Day: Stuxnet and the Launch of the World’s First Digital Weapon. Crown Publishers, NY (2014) 4. Clark, R.A., Knake, R.K.: Cyber War: The Next Threat to National Security and What to do About. HarperCollins Publishers Inc, New York (2010) 5. Kostopoulos, G.K.: Cyberspace and Cybersecurity. CRC Press, Taylor & Francis Group, New York (2013) 6. Barbosa, J.:Pequenas potências militares convencionais, Grandes potências militares cibernéticas - Abordagem da utilização de meios informáticos na defesa/ataque militar moderno. IDN, Lisboa (2018) 7. Singer, P.W., Friedman, A.: Cybersecurity and Cyberwar—What Everyone Needs to Know. Oxford University Press, NY (2014) 8. Wittkop, J.: Building a Comprehensive IT Security Program. Apress, Bolton, USA (2016) 9. Ayala, L.: Cyber-Physical Attack Recovery Procedures: A Step-by-Step Preparation and Response Guide. Apress, New York (2016)

Portuguese Concerns and Experience of Specific Cybercrimes: A Benchmarking with European Citizens João Vidal Carvalho, Álvaro Rocha, António Abreu and Avelino Victor

Abstract The increasing use of a variety of devices that can access the Internet has contributed to the overall growth of Internet use. The growth of Internet use also leads to a general increase in cybercrime concerns as well as the number of users’ victims of this permanent threat. This reality is also very present in Portugal where there is a sense of lack of control and combat of this kind of cyber-threats. In this context, it is important to know some indicators associated with cybercrime, in order to take the best measures/actions in a reasoned way, to combat this phenomenon. This article presents the results of a study carried out in 2017 by the European Commission, whose analysis helps to understand the Portuguese positioning in the concerns and experiences about cybercrime, in relation to the average of the European Community citizens. Analysing this study, it was possible to perceive that the Portuguese’s concerns about cybercrime are in line with the average of the European citizens, although the percentage of Portuguese experienced these attacks is, for now, below average. Based on these results, it is possible to identify some public policies that aim to mitigate this phenomenon. Keywords Cyber security · Cybercrime · Cyber-threats · Information security · Online fraud · Scam · Cyber extortion

J. V. Carvalho (B) · A. Abreu Politécnico do Porto, ISCAP, CEOS.PP, São Mamede de Infesta, Portugal e-mail: [email protected] A. Abreu e-mail: [email protected] Á. Rocha Departamento de Engenharia Informática, Universidade de Coimbra, Coimbra, Portugal e-mail: [email protected] A. Victor Instituto Universitário da Maia, Instituto Politécnico da Maia, Maia, Portugal e-mail: [email protected] © Springer Nature Singapore Pte Ltd. 2020 Á. Rocha and R. P. Pereira (eds.), Developments and Advances in Defense and Security, Smart Innovation, Systems and Technologies 152, https://doi.org/10.1007/978-981-13-9155-2_4

39

40

J. V. Carvalho et al.

1 Introduction It is generally accepted that computerization and the introduction of information systems and technologies (IST) in all social activities significantly increase the efficiency and effectiveness of the activities of each person or organization. In addition, this process enhances the quality of life of an individual through the ability to access from any place and at any time, essential information in services such as electronic administration, electronic health, education, leisure and shopping. However, the consequent opening of databases and information systems of public administration and other business entities inevitably leads to the possibility of such openness being exploited by malicious organized individuals or groups. The possibilities for abuse are large ranging from unauthorized access to system data and resources, to identity theft of users to actions that significantly or completely degrade the functionality of information systems. Such activities not only affect users and information systems that are the target of the attack, but to a greater or lesser extent undermine the confidence of users and business entities, which ultimately leads to a slowdown in technological development. Indeed, the advancement of IST and digital economy has great potential for the well-being of citizens and for business growth, but this new paradigm also entails cyber security challenges that can have high economic impacts. Cybercrime is estimated to cause the loss of billions of euros per year and is placing an increasing strain on law enforcement response capability [1]. Cybercrime is a borderless problem, consisting of criminal acts that are committed online by using electronic communications networks and information systems, including crimes specific to the Internet, online fraud and forgery, and illegal online content such as child pornography [2]. It is estimated that more than one million people worldwide become victims of cybercrime every day [3]. The European Union is leading the effort to regulate defence against cyber threats in both normative/legal and strategic areas. Through this strategy, the European Commission sought to create legislation to criminalize such crimes, increase cyber security capacity and promote the exchange of information between countries (in particular on recorded incidents) [4]. In fact, given the development of cybercrime in recent years, the European Commission has designed a coordinated policy in close cooperation with European Union member states and the other European Union institutions. As part of the European Community, Portugal is highly involved in this European strategy. In this sense, also in Portugal there is a growing concern about cybercrime, both at the level of the state and at the citizen level. Nevertheless, it is important to understand how the Portuguese citizens face this new paradigm and especially to understand how far they are compared to other Europeans. This comparison is made at the level of concerns about cybercrime (and some of the most recurrent threats associated with this concept) as well as the experiences they have experienced in the recent past. The analysis of this comparison will allow Portugal to position itself against the other European countries in this sensitive and important area for today’s citizens and businesses. In sum, the study presented here examines the experience

Portuguese Concerns and Experience of Specific Cybercrimes …

41

of cybercrimes that EU citizens and the Portuguese in particular have and the level of concern they feel about this type of crime. In this article, we will initially present the description of a study carried out in Europe in 2017 about cyber security. In the next section, the values of concerns and experiences of ten cybercrimes will be compared between Portuguese and the European citizen’s average. Finally, we will present the discussion about the results of the study, with a view to identify measures that have mitigated this phenomenon.

2 Cyber Security Study in Europe The Special Eurobarometer series on cyber security is the most important resource for learning about cybercrime in Europe [1]. The importance of this resource stems from the treatment and analysis of representative data of different types of cybercrimes collected in the last five years of the 28 member states of the European Community. The most recent report covers a wide range of threats and aims to understand European citizens’ experiences and perceptions of cyber security issues. That is, the survey adopted in this report analyses the nature and frequency of Internet use by citizens; their awareness and experience of cybercrime; and the level of concern they feel about this type of crime. In this paper, we analyse only the perceptions of Internet users to cyber security, whether they have experienced or been a victim of cybercrime and the level of concern they feel about it. This survey adopted in this report was carried out between the 13th and the 26th June 2017, by TNS opinion & social,1 carried out the wave 87.4 of the Eurobarometer survey, on request of the European Commission. The wave 87.4 covers the population of the respective nationalities of the European Union member states, resident in each of the 28 member states and aged 15 years and over. In total, 28,093 respondents (1075 from Portugal) from different social and demographic groups were interviewed face-to-face at home in their mother tongue on behalf of the Directorate-General for Home Affairs. The methodology used is that of Eurobarometer surveys as carried out by the Directorate-General for Communication (“Strategic Communication” Unit).2 The findings from this survey update a previous survey which was carried out in 2013 [5] and 2015 [2]. The 2017 survey repeats most of the questions asked in 2015 in order to provide insight into the evolution of knowledge, behaviour and attitudes towards cyber security in the European Union.

1 TNS opinion

& social is a consortium created between TNS political & social, TNS UK and TNS opinion. 2 http://ec.europa.eu/public_opinion/index_en.htm.

42

J. V. Carvalho et al.

2.1 Questions Used in the Study for Cyber Security Concerns and Experience In order to identify the cyber security indicators applied in the survey, we have relied on the most important types of cybercrime. Table 1 presents these different types of cybercrime and a summary description of each. It should be noted that these descriptions were presented to the respondents in the same terms. To assess the levels of concern of respondents, they were asked how concerned they were about falling victim to cybercrimes. Thus, the following question was asked: “Cybercrimes can include many different types of criminal activity. How concerned are you personally about experiencing or being a victim of the following situations?” Having been asked about their concerned of cybercrimes, respondents were then asked whether they had been the victim of any such crimes. Thus, the following question was asked: “How often have you experienced or been a victim of the following situations?”

Table 1 Description of cyber security indicators Cyber security indicator

Description

Identity theft

Somebody stealing your personal data and impersonating you

Scam emails or phone calls

Receiving fraudulent emails or phone calls asking for your personal details (including access to your computer, logins, banking or payment information)

Online fraud

Online fraud where goods purchased are not delivered, are counterfeit or are not as advertised

Offensive material and child pornography

Accidentally encountering child pornography online

Material promoting racial hatred or religious extremism

Accidentally encountering material which promotes racial hatred or religious extremism

Access to online services

Not being able to access online services like banking or public services because of cyber-attacks

Email account hacking

Your social network account or email being hacked

Online banking fraud

Being a victim of bank card or online banking fraud

Cyber extortion

Being asked for a payment in return for getting back control of your device

Malicious software

Discovering malicious software (viruses, etc.) on your device

Portuguese Concerns and Experience of Specific Cybercrimes …

43

In any of the questions, the “following situations” refer to the ten different types of cybercrime that were applied in this survey and are presented in the following table.

2.2 Concerns with Cybercrimes Europeans’ concerns about cyber security have been growing over the last few years. In fact, almost nine in ten (87%) respondents see cybercrime as an important challenge, a significant increase in the 80% recorded in March 2015. The rise is even more significant when looking at the proportion of respondents who see cybercrime as a very important challenge: 56% compared with 42% in 2015. Less than half (49%) of the respondents agree or mostly agree that law enforcement is doing enough to combat cybercrime, with the proportion of respondents who totally agree being generally low across member states. Moreover, a significant proportion (14%) does not know if enough is being done to combat cybercrime. In this last study of 2017, we can see that the percentage of respondents worried about any type of threat is always higher than those that do not have concerns (see Table 2). Respondents have become more concerned about “Identity theft”, “Malicious software” and “Online banking fraud” that present percentages close to 70%. On the other hand, “Material promoting racial hatred or religious extremism” presents only a little more than 50% concern by the average European respondents. Comparing the levels of concern of the Portuguese, we can see that they are in line with the average of the Europeans (Fig. 1). However, there are threats that concern more the Portuguese, namely “Offensive material and child pornography” and “Material promoting racial hatred or religious extremism”.

2.3 Experience of Cybercrimes Europeans’ experience about cyber security has been growing over the last few years. In this last study of 2017, we can see that the percentage of respondents with concrete experiences with these types of threats associated with cyber security has increased (see Table 3). However, there is a significant variation among member states on the proportion in this type of experience. For instance, in general the Portuguese present a percentage of cases in which they were victims, inferior to the majority of the member states. The average European respondents report more cases in which they were victims associated with “Malicious software” and “Scam emails or phone calls” that present percentages close to 50%. On the other hand, “Identity theft”, “Offensive material and child pornography” and “Cyber extortion” present percentages below 10%.

44

J. V. Carvalho et al.

Table 2 Concerns of Portuguese and Europeans average in cybercrimes Very concerned (%)

Fairly concerned (%)

Not very concerned (%)

Not at all concerned (%)

Don’t know (%)

Identity theft

EU

33

36

22

8

1

PT

35

34

15

16

0

Scam emails or phone calls

EU

26

34

26

13

1

PT

27

33

18

22

0

Online fraud

EU

22

36

28

12

2

PT

21

38

22

17

2

Offensive material and child pornography

EU

27

26

26

19

2

PT

30

31

16

23

0

Material promoting racial hatred or religious extremism

EU

21

30

29

18

2

PT

24

31

19

25

1

Access to online services

EU

22

35

28

13

2

PT

25

30

18

25

2

Email account hacking

EU

27

36

24

11

2

PT

30

36

18

15

1

Online banking fraud

EU

32

34

21

11

2

PT

31

31

18

18

2

Cyber extortion

EU

25

30

28

15

2

PT

28

30

16

24

2

Malicious software

EU

29

40

22

8

1

PT

29

42

22

7

0

Comparing the experience of the Portuguese with the average of the European respondents (Fig. 2), we can see that in some cases the percentage of cases in which respondents are victims are very similar, namely “Identity theft”, “Offensive material and child pornography”, “Access to online services” and “Cyber extortion”. However, there are cyber security threats whose percentages of cases are more divergent between the Portuguese and the European average. Indeed, “Material promoting racial hatred or religious extremism” presents a 10% difference and especially “Scam emails or phone calls” which present a 22% difference.

Portuguese Concerns and Experience of Specific Cybercrimes …

45

Fig. 1 Comparison between Portuguese and Europeans average concerns in cybercrimes indicators Table 3 Experience of Portuguese and Europeans average in cybercrimes Has been a victim (%) Identity theft

Never (%)

Don’t know (%) 1

EU

8

91

PT

4

96

0

Scam emails or phone calls

EU

38

61

1

PT

11

89

0

Online fraud

EU

16

83

1

PT

7

92

1

Offensive material and child pornography

EU

7

92

1

PT

7

93

0

Material promoting racial hatred or religious extremism

EU

18

81

1

PT

8

91

1

Access to online services

EU

11

87

2

PT

7

92

1

Email account hacking

EU

14

85

1

PT

6

93

1

EU

11

8

1

PT

3

96

1

Cyber extortion

EU

8

91

1

PT

5

95

0

Malicious software

EU

42

57

1

PT

33

66

1

Online banking fraud

46

J. V. Carvalho et al.

Fig. 2 Comparison between Portuguese and Europeans average concerns in cybercrimes indicators

It should be noted that only a minority of respondents have experienced any of these crimes. Most of the attacks described here were experienced by less than 15% of respondents, with the exception of “Scam emails or phone calls” and “Malicious software” that presented more representative values.

2.4 Institutions Perceived as Responsible to Provide Assistance to Citizens for the Different Types of Cybercrimes In face of the concerns and cases experienced by European citizens, it is important to know which entities they resort to, if they are involved in these threats. In this context, citizens were questioned for each of the indicators, witch entity that would resort to if they were victims of one of these threats. The applied question would be “If you experiences or were a victim of the following situations, who would you contact?”. Regarding the possibilities of choice, these would be: Police; Website/Vendor; Consumer protection Organisation; Your Internet service provider; Other; No one; and Don’t know. The vast majority of respondents would go to the police if they were victims of any kind of cybercrime represented here, as can be seen in Table 4. The only exception is related to “Malicious software” in which the average European citizens prefers to use the Internet service provider (29%) and only 25% resort to the Police. All threats of cybercrime considered in this study have considerable percentages associated with the police as institutions perceived as responsible to provide assistance to citizens for the different types of cybercrimes. However, we can also

Portuguese Concerns and Experience of Specific Cybercrimes …

47

Table 4 Institutions perceived as responsible to provide assistance to citizens Police (%) Identity theft

EU

85

PT

88

EU

53

PT

58

EU

52

PT

61

Offensive material and child pornography

EU

76

PT

67

Material promoting racial hatred or religious extremism

EU

59

PT

50

Access to online services

EU

40

Scam emails or phone calls Online fraud

PT

47

Email account hacking

EU

40

PT

38

Online banking fraud

EU

76

PT

84

EU

70

Cyber extortion Malicious software

PT

76

EU

25

PT

32

Internet service provider (%)

29

conclude that the Portuguese resort to this entity in a percentage higher than the average of the other Europeans. The following figure (Fig. 3) presents the spider chart showing the percentage of the Portuguese and European citizens that have been victims of one of cyber threats and go to the police. It is observed in Fig. 3 that in most of the types of threats the Portuguese resort to police in a greater percentage, except “Offensive material and child pornography” and “Material promoting racial hatred or religious extremism”.

3 Discussion The increasing use of a growing variety of devices that can access the Internet has contributed to the overall growth of Internet use. This reality also leading to a general increase in cybercrime concerns as well as the number of cases associated with this phenomenon where more and more users are involved. In this context, there is an increase in citizens’ concerns in most of the different types of threats, with particular focus on the “Identity theft”, “Malicious software” and “Online banking fraud” that

48

J. V. Carvalho et al.

Fig. 3 Comparison between Portuguese and Europeans average about institutions perceived as responsible to provide assistance to citizens

present percentages close to 70%. Comparing the experience of the Portuguese with the average of the European respondents, we can see that the European citizens have experienced these threats more often than the Portuguese in most of the different types of cybercrimes. However, only a minority of respondents have experienced any of these crimes. Anyway, “Malicious software” stands out as the threat that both the Portuguese and the Europeans more mentioned. In the case where European citizens have been victims of one of these threats, the majority of them have resort to the police. The Portuguese have the same approach, although with slight differences in the various types of threats. Based on the data analysed by this study, there is a growing need to combat this phenomenon. In fact, knowing that there is a growing concern about cybercrimes and an increase in cases where people are victims justifies intervention through public policies to take action to combat these threats. In this sense, new and more restrictive legislation has been implemented, which in the medium term may contribute to the improvement of these results. In 2013, the European Commission presented a proposal for a Directive concerning measures to ensure a high common level of network and information security across the Union. The Directive (EU) 2016/1148 of the European Parliament and of the Council was approved on the 6 July 2016 and is currently undergoing the transposition phase to the internal legal systems of the EU countries [6]. In addition to new legislation, the European Commission has established in 2013 a European Cybercrime Centre (EC3)3 to help protect European citizens and businesses against these growing cyber-threats. This centre leads European entity to combat cybercrime and focus on illegal online activities by organized 3 https://www.europol.europa.eu/about-europol/european-cybercrime-centre-ec3.

Portuguese Concerns and Experience of Specific Cybercrimes …

49

crime groups, particularly those that generate large criminal profits, such as online fraud involving credit cards and bank credentials [3]. However, the most important institutions in the European Union in the field of network and information security are European Agency for Network and Information Security (ENISA), established in 2004, and European Computer Reaction Initiative (CERT-EU), formed in 2012. Also, Portugal creates a Centre (CNCS),4 that is the operational coordinator and the Portuguese national authority specialized in cyber security working in this field with state entities, operators of critical infrastructures, operators of essential services and digital service providers [6]. Another way of improving the results of this study is related to the level of knowledge and information of the citizens themselves. Some of the threats most experienced by citizens may be reduced by an additional caution that they must have. Actually, users can do more to protect themselves, because the two most frequently experienced forms of cybercrime, infection with malicious software and fraudulent obtaining of personal information, are actions that a well-informed public can do much to prevent themselves [1].

4 Conclusion Despite the permanent development of information technologies and new forms of potential abuse, Internet users expect that every information system be able to reject attacks that could jeopardize system data [7]. Additionally, these users, on their own initiative, publish a large number of their personal information in cyberspace, without thinking that this information could be used against them. Such behaviour inevitably leads to a marked lack of trust in systems and an ongoing concern with security issues. These concerns are well present in the results of the survey analysed here. The Portuguese and the European citizens in general have the same degree of concern about cyber security and the threats associated with cybercrime. It is noted that these concerns have not diminished over time, and therefore, actions are justified to reverse this situation. Actions should be taken in a coordinated manner to combat large-scale organized cybercrime and impose data protection standards on providers of online services. Besides that, the findings of this survey also highlight the importance of greater public education on types of cybercrime, their consequences, and ways in which their impact can be avoided or mitigated. In the near future, it will be important to carry out a new survey with the same format, which allows to know the evolution of these cybercrime indicators over time, as well as, to check if the combat measures implemented are taking effect or require a reinforcement or adjustment.

4 CNCS—Centro

Nacional de Cibersegurança (National Cybersecurity Centre).

50

J. V. Carvalho et al.

References 1. European Commission: Europeans attitudes towards cyber security. Special Eurobarometer 464a (2017) 2. European Commission: Cyber security. Special Eurobarometer 423 (2015) 3. European Commission: Cyber security. Special Eurobarometer 390 (2012) 4. Barros, G.: A Cibersegurança em Portugal. Gabinete de Estratégia e Estudos - Estudos Económicos, p. 56 (2018) 5. European Commission: Cyber security. Special Eurobarometer 404 (2013) 6. CNCS: Centro Nacional de Cibersegurança, 13 Dec 2018. Available from https://www.cncs. gov.pt/en/about-us/ (2018) 7. Vilic, V.: International and Serbian legal framework of the right to privacy in cyberspace. MEST J. 6(1), 119–131 (2018)

Classification of Phishing Attack Solutions by Employing Deep Learning Techniques: A Systematic Literature Review Eduardo Benavides, Walter Fuertes, Sandra Sanchez and Manuel Sanchez

Abstract Phishing is the technique by which the attacker tries to obtain confidential information from the user, with the purpose of using it fraudulently. These days, three ways to mitigate such attacks stand out: Focus based on awareness, based on blacklists, and based on machine learning (ML). However, in the last days, Deep Learning (DL) has emerged as one of the most efficient techniques of machine learning. Thus, this systematic literature review has been aimed to offer to other researchers, readers and users, an analysis of a variety of proposals of other researchers how to face these attacks, applying Deep Learning algorithms. Some of the contributions of the current study include a synthesis of each selected work and the classification of antiphishing solutions through its approach, obtaining that the uniform resource locator (URL)-oriented approach is the most used. Furthermore, we have been able to classify the Deep Learning algorithms selected in each solution, which yielded that the most commonly used are the deep neural network (DNN) and convolutional neural network (CNN), among other fundamental data. Keywords Phishing · Deep Learning · Social Engineering · Machine Learning · Cybersecurity

1 Introduction Phishing is the most commonly used attack on social engineering. Through such attacks, the phisher tries to obtain confidential information from the user, with the purpose of using it fraudulently against himself or its organization [1]. The most E. Benavides (B) · W. Fuertes Escuela Politécnica Nacional, Quito, Ecuador e-mail: [email protected] E. Benavides · W. Fuertes · S. Sanchez Universidad de Las Fuerzas Armadas, Sangolqui, Ecuador M. Sanchez Universidad de Alcalá de Henares, Madrid, Spain © Springer Nature Singapore Pte Ltd. 2020 Á. Rocha and R. P. Pereira (eds.), Developments and Advances in Defense and Security, Smart Innovation, Systems and Technologies 152, https://doi.org/10.1007/978-981-13-9155-2_5

51

52

E. Benavides et al.

common way to become a victim to these attackers is by entering a Web site with a fraudulent URL [2]. Currently, three ways to mitigate social engineering attacks stand out: The first way is through user awareness and training to deal with these attacks [1]. The second and most commonly used way is to have a blacklist of phishing sites [3]. However, the disadvantage of this approach is that it must first know the sites that have already been detected as phishing. The third way to prevent these attacks, being more effective than the previous one, is to detect them in their early appearance, using machine learning algorithms, which are able to predict if a page is phishing, based on the previously analyzed characteristics of many other pages that already have been adulterated. The problem with the ML approach is that, its need to involve the knowledge of an expert, and a considerable period of time, in order to select the characteristics that dictate that a page is phishing or not. A particular technique of machine learning is Deep Learning [4]. With this approach, based on Deep Learning algorithms, it is not necessary that the user or an expert indicates the characteristics of a contaminated page, because the algorithm is able to detect them automatically and independently [5]. Currently, ML anti-phishing approaches can be classified into two groups: Phishing email-based approaches and those based on phishing Web sites. Among the solutions approaches to phishing Web sites, the one based on URLs stands out. Thus, our work is oriented to give a complete vision to the techniques of mitigation of phishing URLs by means of Deep Learning algorithms. Based on the aforementioned, our study aims to analyze the state of the art related to the detection of social engineering attacks associated with phishing URLs, using Deep Learning techniques. In order to fulfill such attempt, a systematic literature review (SLR) has been conducted [6], with the purpose of analyzing the studies related to the subject and evaluating the proposals of other researchers, in order to determine their impact and find potential solutions. Some of the contributions of the current study include a SLR, whose preliminary result provides a general approach on the application of Deep Learning techniques in the process of detection and mitigation of social engineering attacks, in order to reduce the given risk. The rest of this article has been organized as follows: Sect. 2 sets the background of this research. Next, in Sect. 3, the phases of the SLR are explained, and a synthesis of the most relevant articles is presented. In Sect. 4, the obtained results are discussed. Finally, the conclusions and future work are described in the last section.

2 Background 2.1 Phishing Phishing is the combination of social engineering and technical exploitations, designed to convince a victim to provide personal information such as account

Classification of Phishing Attack Solutions by Employing …

53

numbers, passwords and credit card details, and with the usual purpose created in order to obtain a monetary gain by the attacker. Most phishing attacks are realized by fake emails, containing a uniform resource locator (URL). In case of its activation by a click, such URL leads to a fake malicious Web site. Despite the important attention, it has received over the years, there is still no definitive solution to solve this problem [2]. In [3], authors define “a Phishing page, like any web page, that without permission, claims to act on behalf of a third party; with the intention of confusing the spectators in the performance of an action.” Anti-phishing studies may be classified; these may be classified, according to their objective, in two large groups being those aimed at mitigating phishing emails, and those aimed at mitigating attacks on Web sites. Those oriented toward Web sites may be classified mainly into four different classes: (1) focused on the URL of the Web site; (2) focused on the HTML content; (3) focused on behavior; and (4) focused on a hybrid approach.

2.2 Deep Learning According to [4], Deep Learning is defined as a machine learning technique, where many layers of information processing stations are exploited, by classification patterns and characteristics, or by learning by representation. In fact, Deep Learning is implemented by neural networks, with many layers that are not something new. However, it has become popular recently, due to three factors: First, there is a notable increase in processing capabilities (e.g., video cards, graphical processors, etc.); second, by affordable computer hardware; and third, due to recent advances and developments in Deep Learning research. Deep learning algorithms may be classified into three subgroups, which depend on the fact if the algorithms are trained to yield the obtainable results or not. The three subgroups are classified as unsupervised, supervised and hybrid. In [5], the consistency of each of the Deep Learning algorithms has been described in detail (Fig. 1).

3 Methodology Our study followed the SLR methodology proposed by Barbara Kitchenham [6]. Five consecutive steps have been conducted: (1) define the research questions; (2) search the relevant documents; (3) select the primary studies; (4) analyze the abstracts and extract keywords and data; and (5) map the selected primary studies. First, the search of the related articles has been accomplished and as a consequence, the relevant results have been listed. Then, the data extraction has been achieved and the reports of the encountered information have been finally produced. A brief description of each phase is described below:

54

E. Benavides et al.

Deep Learning

Unsupervised

Autoencoder (AE)

Hybrid

Sum Product Network

Recurrent Neural Network

(SPN)

(RNN)

Boltzman Machine (BM)

Deep Neural Network

ConvoluƟonal Neural Network

(DNN)

(CNN)

Stacked AE

Deep BM

Restricted BM

(SAE)

(DBM)

(RBM)

Denoising AE (DAE)

Supervised

Deep Belief Network (DBN)

Fig. 1 Classification of deep learning techniques [5]

3.1 Define the Research Questions What are the techniques of Deep Learning that are currently used in primary studies and how do researchers use these Deep Learning techniques, in order to mitigate phishing attacks?

3.2 Search the Relevant Documents Our strategy consisted of first defining the search chain, which consisted of two parts, being URL phishing and Deep Learning. Next, the search chain has been modified by synonyms of each of the two parts. With this chain, we proceeded to search in all articles related to the given subject.

3.2.1

Inclusion Criteria

Once the search keywords have been defined, only the following documents have been included being those: (1) whose main topic is phishing and Deep Learning; (2) which have been written exclusively in English; (3) its publication has been only of at most the last five years; (4) its solution method is about URL-oriented.

Classification of Phishing Attack Solutions by Employing …

3.2.2

55

Exclusion Criteria

(1) articles whose subject has been only phishing or only Deep Learning, without combing both topics; (2) being in a language other than English; (3) articles published prior to 2010; (4) lacking of methods applied to URL-oriented.

3.2.3

Quality Criteria

In order to comply with the quality criteria of the search, it has been decided to conduct the scrutiny only on pages recognized at the scientific level, which are also from the area of information technology. The indexed databases chosen have been: IEEE Explore, Taylor and Francis, Springer Linker, ACM Portal and Science Direct.

3.3 Selecting the Primary Studies After searching for the needed information, 59 candidate articles have been encountered, and a group of 19 primary studies has been established. Based on the synthesis performed in the previous step, we proceeded to conduct the report, and to demonstrate the information gathered by the analysis.

4 Primary Studies

Id

Used technique

Used algorithm

Applied methodology

Phishing objective

Applied technique

Main features

[7]

a TML

support vector machines (SVM), neural networks, selforganizing maps and K-means

Comparative Analysis

Malign Email Features

N/A

Authors evaluate some algorithms: Support vector machines (SVM), neural networks, self-organizing maps and K-means. Among the studied methods, we observed that the most accurate has been SVM with an accuracy of 97.99% (continued)

56

E. Benavides et al.

(continued) Id

Used technique

Used algorithm

Applied methodology

Phishing objective

Applied technique

Main features

[8]

b HMDL

Deep Learning Algorithms and Machine learning algorithms

Comparative Analysis and A new proposal

Malign Phishing Software

Multi-stage Malwaredetection ML → DL

Normal or abnormal behavior of the programs executed in the operating system has been monitored. If for some reason and based on the rules of machine learning, this software has been suspicious, then it passed through an evaluation phase using Deep Learning

[9]

c DL

CNN

A new proposal

Homoglyph at URL. i.e., svchost.exe replaced with svch0st.exe

CNN is trained with the Euclidean distance between their respective features

This solution uses a technique applied in the rendering of images. In this way, a CNN learns features that are used to detect visual similarities of the rendered strings. Methodologically, first the domain name text strings are transformed into images and then these are passed through the SCNN

[10]

DL

CNN

A new proposal

URL obfuscation

eXpose neural network

With this technique, we emphasize that the data entered are short strings of raw data (data without format or without classification). Then, it learns to extract characteristics and classify them, using incrustations at the character level with a convolutional neural network

[11]

DL

CNN

A new proposal

URL obfuscation

Bag of Bytes

The researchers had previously assigned a hexadecimal value to each character in the URL string. Then, those values have been matched with each value on the right. Later, these datasets in hexadecimal configurations are passed through the algorithm of the neural network model for a subsequent classification. eXpose is improved (continued)

Classification of Phishing Attack Solutions by Employing …

57

(continued) Id

Used technique

Used algorithm

Applied methodology

Phishing objective

Applied technique

Main features

[12]

ML DL

a TML algorithms logistic regression with bigrams CNN CNNLSTM

Comparative Analysis

URL obfuscation

N/A

First, traditional methods of machine learning are evaluated, and then it compares them with the methods of Deep Learning. In conclusion, it is deduced that the Deep Learning LSTM method is the most accurate, with an accuracy of 98%

[13]

N/A

N/A

User Awareness

N/A

N/A

The researchers propose to mitigate the attacks by teaching how to prevent these. This proposal is rather practical, applying the concept of big data analytics, in order to mitigate the attacks of social engineering

[14]

ML

AdaBoost, bagging. random forest, sequential minimal optimization (SMO)

A new proposal

URL obfuscation

(1) A syntactic analysis of the text; and (2) A statistical analysis of the Web site. Semantic features

The premise is used that a set of characters on a phishing page is similar to a legitimate page. It also performs an evaluation of the statistics of the characteristics of the phishing pages. That occurs if it is a new page or, for example, if it has a bad reputation, etc.

[15]

ML DL

Support vector machines (SVM), random forest, Naive Bayes, decision trees, K-nearest neighbors and multilayer perceptron

Comparative Analysis and a new proposal

URLs in general

This system recommended the use of numerical characteristics for training, in order to obtain a better rate in prediction accuracy.

While these algorithms used a public dataset of around 2.4 million URLs, this system recommended the use of numerical characteristics

(continued)

58

E. Benavides et al.

(continued) Id

Used technique

Used algorithm

Applied methodology

Phishing objective

Applied technique

Main features

[16]

DL

Long Short-Term Memory Recurrent Neural Network LSTMRNN

A new proposal

URL obfuscation

Feature extraction

Phishing pages are collected, and then the LSTM recurrent neural network algorithm is trained with 12 characteristics. This finally produces a 99.14% effectiveness in the detection of new false sites

[17]

DL

Deep Boltzmann Machine (DBM)

A new proposal

Unbalanced data

BorderlineSMOTE (Synthetic Minority Oversampling Technique) algorithm

First, three groups of characteristics of the pages of phishing are extracted being these the characteristics of URL, of Web pages and features of the images. With this data, the algorithm is trained and afterward an effectiveness over 99% in the detection of new affected sites is obtained

[18]

DL

Deep belief network (DBN)

Framework to detect phishing pages

Internet Protocol Flow

[19]

DL

Neural networks, support vector machine, decision tree DNN. Stacked Autoencoder technique (SAE)

Comparative Analysis

URL obfuscation

It focuses mainly on the application of a framework to detect phishing pages. Firstly, two types of characteristics of the false sites are classified, being the original ones and the interaction ones. Hereby, a model based on Deep Belief Network is proposed Before being analyzed, the URLs are translated into the ASCII code. In addition, it creates a determination of the incidence percentage of each characteristic in a phishing page

This is a comparative study among traditional ML approaches and DL approaches at phishing detection. As a result, the DL technique achieved the best detection rate with an accuracy of 80%, regardless of the amount of data entered for training

(continued)

Classification of Phishing Attack Solutions by Employing …

59

(continued) Id

Used technique

Used algorithm

Applied methodology

Phishing objective

Applied technique

Main features

[20]

DL

RNN

A new proposal

URLs in General

Gated recurrent neural networks (GRUs)

An advantage of such study is that it has been able to also identify which type of attack on URLs is being attempted. Hereby it classifies that URL in Legitim, SQL Injection, XSS Attack, Sensitive File Attack, Directory Traversal, or another type of attack

[21]

DL

CNN

– Comparative study and – A new proposal

Malicious URLs Malicious DNSs

– Feature Selection Scheme and – CNN at the word and character level

This approach uses a CNN scheme based on character level. After using this scheme, a comparison is performed with other schemes, obtaining by using a feature selection scheme, finding some 282 failures. Later, using CNN at the word level, 158 failures have been encountered, while, using CNN at the character level, some 40 failures were detected

[22]

DL

CNN LSTM

A new proposal

DNS flow

D-FENDS (DNS Filtering & Extraction Network System)

D-FENS identifies malicious domain names in real time. This system runs inside a DNS server. D-FENS opt for a Deep Learning approach that learns the characteristics automatically from the input data. An unusual method is where detection is applied directly to traffic in a DNS

[23]

ML DL

Random forest CNN

A new proposal

Domain generation algorithms (DGAs) and fraudulent URLs

WordGraph

The authors propose to use a tool called WordGraph, which generates dictionaries similar to those used by DGAs (continued)

60

E. Benavides et al.

(continued) Id

Used technique

Used algorithm

Applied methodology

Phishing objective

Applied technique

Main features

[24]

ML

Random forest J48, logistic regression, Bayes network, multilayer perceptron, sequential minimal optimization, AdaBoostM1, SVM

Comparative Study

• URL obfuscation features • Thirdpartybased features • Hyperlinkbased features

To identify the phishing sites based on the Features extracted from URL, Web site content and third-party services

This study has been included, because it explains in detail the different characteristics that may be extracted from a deceptive site

[25]

DL

DNN

Behavior People Study

People

DeepSeq

First, you get the characteristic profile of people who are commonly pro-thought to be phishers. For this, based on the obtained logs, the intrinsic data of the people is compared (age, sex, occupation, etc.), versus the data of the visited sites (business, art, social media, etc.). Finally, after performing an analysis by means of DNN, a profile is obtained

[26]

DL

DNN

Address bar-based, abnormalbased features, HTML- and JavaScriptbased, domainbased features

Bat Algorithm feedforward neural networks

The researchers propose the TDLBA method (Tuning Deep Learning using Bat Algorithm). This method combines swarm intelligence approaches by configuring the parameters of Deep Learning networks

a TML

= traditional machine learning. b HMDL = hybrid of machine and Deep Learning. c DL = Deep Learning

5 Discussion of the Selected Primary Studies After performing the analysis on the initially found 59 studies, only 19 relevant works related to phishing attacks have been selected through the use of Deep Learning algorithms. Table 1 documents that most of them have been published in Springer (See Fig. 2). We also encountered that most research about deep learning applied to combat phishing attacks has been conducted in China (See Fig. 3).

Classification of Phishing Attack Solutions by Employing … Table 1 Articles versus approaches

Article

URL

Content

Behavior

[5]

No

No

Yes

[8]

No

No

Yes

[9]

Yes

No

No

[10]

Yes

No

No

[11]

Yes

No

No

[12]

Yes

No

No

[14]

Yes

Yes

No

[15]

Yes

No

No

[16]

No

No

No

[17]

Yes

Yes

No

[18]

Yes

Yes

No

[19]

Yes

No

No

[20]

Yes

No

No

[21]

Yes

No

No

[23]

Yes

No

No

[24]

Yes

Yes

Yes

[26]

Yes

Yes

Yes

Fig. 2 Articles published by indexed database

Fig. 3 Articles published by the country

61

8 7 6 5 4 3 2 1 0

6 5 4 3 2 1 0

62 Table 2 Articles per deep learning algorithm

E. Benavides et al.

Category

Deep learning algorithm

Unsupervised

SAE—stacked AE

Articles found

SDAE—stacked DAE SPN—sum product network RNN—recurrent neural network

[12, 16, 20]

DBM—deep BM

[5, 17]

DBN—deep belief network

[18]

Hybrid

DNN—deep neural network

[19, 23, 25, 26]

Supervised

CNN—convolutional neural network

[9, 10, 11, 21]

In addition, we observed that the main anti-phishing approaches applying Deep Learning, has been able to be classified according to the place of the Web page, in which the analysis of the data has been performed. For instance, in: URL, Content (HTML Content), Behavior (Behavior of the website) (see Table 1). In Table 2, we observed and classified the proposed solutions according to the use of Deep Learning algorithms that are Not Supervised, supervised and hybrid.

6 Conclusions and Future Work The current study is a systematic literature review on current Deep Learning solutions in order to combat phishing attacks, even latest threats, such as spear phishing attacks. To achieve such attempt, the different Deep Learning solutions have been characterized, classified and analyzed. As a result, it has been evident that there is still a great gap in the application of Deep Learning algorithms in the detection of cyber threats in general that has not been exploited yet. As future work, a comparative study is expected to determine which Deep Learning algorithm is the most effective in determining malicious sites. In addition, it has been planned to design models or algorithms to combat phishing attacks, analyzing the similar characteristics in the pages generated by recurring bots. A new algorithm will also be proposed to accelerate performance in the execution of Deep Learning algorithms, in phishing detection.

Classification of Phishing Attack Solutions by Employing …

63

References 1. Hajgude, J., Ragha, L.: Phish mail guard: Phishing mail detection technique by using textual and URL analysis. In: 2012 World Congress on Information and Communication Technologies, pp. 297–302 (2012) 2. Marchal, S., Armano, G., Grondahl, T., Saari, K., Singh, N., Asokan, N.: Off-the-Hook: An Efficient and Usable Client-Side Phishing Prevention Application. IEEE Trans. Comput. 66(10), 1717–1733 (2017) 3. Whittaker, C., Ryner, B., Nazif, M.: Large-Scale Automatic Classification of Phishing Pages 4. Deng, L.: A Tutorial Survey of Architectures, Algorithms, and Applications for Deep Learning. APSIPA Trans. Signal Inf. Process. (2014) 5. Selvaganapathy, S., Nivaashini, M., Natarajan, H.: Deep belief network based detection and categorization of malicious URLs. Inf. Secur. J. A Glob. Perspect. 27(3), 145–161 (2018) 6. Kitchenham, B., Pearl Brereton, O., Budgen, D., Turner, M., Bailey, J., Linkman, S.: Systematic literature reviews in software engineering – A systematic literature review. Inf. Softw. Technol. 51(1), 7–15 (2009) 7. Basnet, R., Mukkamala, S., Sung, A.H.: Detection of Phishing Attacks: A Machine Learning Approach. In Soft Computing Applications in Industry, pp. 373–383. Berlin, Heidelberg, Springer Berlin Heidelberg (2008) 8. Yuan, X.: PhD Forum: Deep Learning-Based Real-Time Malware Detection with Multi-Stage Analysis. In 2017 IEEE International Conference on Smart Computing (SMARTCOMP), pp. 1–2 (2017) 9. Woodbridge, J., Anderson, H.S., Ahuja, A., Endgame, D.G.: Detecting Homoglyph Attacks with a Siamese Neural Network 10. Saxe, J., Berlin, K.: eXpose: A Character-Level Convolutional Neural Network with Embeddings For Detecting Malicious URLs, File Paths and Registry Keys (2017) 11. Shima, K., et al.: Classification of URL bitstreams using Bag of Bytes (2018) 12. Vazhayil, A., Vinayakumar, R., Soman, K.: Comparative Study of the Detection of Malicious URLs Using Shallow and Deep Networks. In 2018 9th International Conference on Computing, Communication and Networking Technologies (ICCCNT), pp. 1–6 (2018) 13. Epishkina, A., Zapechnikov, S.: A syllabus on data mining and machine learning with applications to cybersecurity. In 2016 Third International Conference on Digital Information Processing, Data Mining, and Wireless Communications (DIPDMWC), pp. 194–199 (2016) 14. Zhang, X., Zeng, Y., Jin, X.-B., Yan, Z.-W., Geng, G.-G.: Boosting the phishing detection performance by semantic analysis. In 2017 IEEE International Conference on Big Data (Big Data), pp. 1063–1070 (2017) 15. Vanhoenshoven, F., Napoles, G., Falcon, R., Vanhoof, K., Koppen, M.: Detecting malicious URLs using machine learning techniques. In 2016 IEEE Symposium Series on Computational Intelligence (SSCI), pp. 1–8 (2016) 16. Chen, W., Zhang, W., Su, Y.: Phishing Detection Research Based on LSTM Recurrent Neural Network, pp. 638–645. Springer, Singapore (2018) 17. Zhang, J., Li, X.: Phishing Detection Method Based on Borderline-Smote Deep Belief Network, pp. 45–53. Springer, Cham (2017) 18. Yi, P., Guan, Y., Zou, F., Yao, Y., Wang, W., Zhu, T.: Web Phishing Detection Using a Deep Learning Framework. Wirel. Commun. Mob. Comput. 2018, 1–9 (2018) 19. Aksu, D., Turgut, Z., Üstebay, S., Aydin, M.A.: Phishing Analysis of Websites Using Classification Techniques, pp. 251–258. Springer, Singapore (2019) 20. Zhao, J., Wang, N., Ma, Q., Cheng, Z.: Classifying Malicious URLs Using Gated Recurrent Neural Networks, pp. 385–394. Springer, Cham (2019) 21. Jiang, J., et al.: A Deep Learning Based Online Malicious URL and DNS Detection Scheme, pp. 438–448. Springer, Cham (2018) 22. Spaulding, J., Mohaisen, A.: Defending Internet of Things Against Malicious Domain Names using D-FENS. In 2018 IEEE/ACM Symposium on Edge Computing (SEC), pp. 387–392 (2018)

64

E. Benavides et al.

23. Pereira, M., Coleman, S., Yu, B., DeCock, M., Nascimento, A.: Dictionary Extraction and Detection of Algorithmically Generated Domain Names in Passive DNS Traffic, pp. 295–314. Springer, Cham (2018) 24. Rao, R.S., Pais, A.R.: Detection of phishing websites using an efficient feature-based machine learning framework. Neural Comput. Appl., 1–23 (2018) 25. Sur, C.: DeepSeq: learning browsing log data based personalized security vulnerabilities and counter intelligent measures. J. Ambient Intell. Humaniz. Comput., 1–30 (2018) 26. Vrbanˇciˇc, G., Fister, I., Podgorelec, V.: Swarm Intelligence Approaches for Parameter Setting of Deep Learning Neural Network. In Proceedings of the 8th International Conference on Web Intelligence, Mining and Semantics—WIMS ’18, pp. 1–8 (2018)

Is Cyber Warfare an Alternative? Jorge Barbosa

Abstract The possibility of using computerized means in offensive action against civilian or military installations from enemy countries is a real. Given the potential of this type of action, they can be seen by many countries as interesting, and as an alternative to conventional kinetic warfare. Despite its many theoretical advantages, obtaining the cyber capabilities needed to implement cyber military units is not easy and may not be within reach of any country. The necessary financial means may not be a major obstacle to this achievement because they are relatively low, when compared to those necessary to obtain conventional means. Similarly, obtaining a support infrastructure is also not theoretically difficult. However, obtaining both human and technological resources, other than those related to the support infrastructure, may not be available to any country, namely the designated zero-day exploits, ZDE, that are needed to explore computer vulnerabilities in informatics systems of the other countries. There are other types of constraints, time needed to prepare the means and the planning and development of cyber actions, type of targets to be attained, specificity of computer systems and even the possible need for cyber media complementarity with conventional means that can be limiting and impede their widespread use. Thus, cyber warfare is a very interesting alternative, but given these constraints and limitations, it may not be a real viable alternative to be used in all situations and by all countries. Keywords Cyber warfare · Cyber weapons · Cyber capability · Zero-day exploits

1 Introduction The use of cyber weapons, which enable cyber warfare to be triggered, is due to the weaknesses created by the existence of hardware and/or software vulnerabilities, which, in general, all computer systems have. When these vulnerabilities are J. Barbosa (B) Coimbra Polytechnic – ISEC, Coimbra, Portugal e-mail: [email protected] Portuguese National Defense Institute, Lisbon, Portugal © Springer Nature Singapore Pte Ltd. 2020 Á. Rocha and R. P. Pereira (eds.), Developments and Advances in Defense and Security, Smart Innovation, Systems and Technologies 152, https://doi.org/10.1007/978-981-13-9155-2_6

65

66

J. Barbosa

not known, in the sense that they are not publicly known but only known within a strict group, they are called zero-day vulnerabilities, ZDV. These vulnerabilities, because they are not publicly known, can be used to great advantage in the construction of applications called exploits, which if based on ZDV will be called zero-day exploit, ZDE, which can be used with great success as cyber weapons. These ZDEs by exploiting the vulnerabilities of computer systems can take control or even paralyze the systems affected by those vulnerabilities. If such systems are linked to the control of so-called critical infrastructures, i.e., systems such as power dams, banking systems and others, their allocation and even stoppage can have very great detrimental effects and hence their usefulness as cyber weapons. The interest of this type of cyber weapon is not, or not only, to be used in the vulgar cybercrime but essentially because they can be used in actions of cyber war by sovereign states, against other states. As this type of action may jeopardize state sovereignty, these actions are referred to as cyber warfare rather than cybersecurity actions, which will then only be related to vulgar cybercrime actions. The use of computerized means to trigger offensive actions against antagonistic countries made possible by the constant evolution of computing means, both from the point of view of the means itself and from the increase in the capacity of the computer networks and their increasing interconnection, is an indisputable reality. Due to this increased use and diversification of these IT resources, it can be considered that in modern societies, there are no sectors where they are not used. From critical infrastructure control systems, dams and other power plants, water treatment and supply systems, to the management and control of banking and financial operations and institutions, transport logistics, fuel logistics, etc. Today, practically everything is based on computer controllers, essentially of the industrial controllers like SCADA or PLC’s in enterprise installations or sophisticate databases and customer relationships informatics systems in financial institutions or public services. The reverse side of this beneficial use is that if such systems are unsafe and if, on the contrary, they are likely to be penetrated, it could result on catastrophic effects as the effects of the above systems, their destruction or even their alteration, would paralyze or would seriously affect modern societies. This dependence is therefore criticized from this point of view, and thus, these systems are tangible and passable targets of military action against them. These military actions can be triggered using conventional means, but also the computer itself, through the so-called cyber weapons. These cyber weapons would act directly on the computer systems destroying, paralyzing or altering them in order to stop or interfere with their normal functioning, and thus, the systems downstream that they control or manage. The theoretical possibilities offered by this military option may lead to the consideration of its use by a large number of countries and not just by the current conventional powers. Theoretically, the amount of resources required, whether human or material, in this case technological, are much lower than those required for the initiation of conventional military operations. Likewise, the economic resources necessary to acquire these technological means and to train the human resources adapted to this type of activity are substantial, but when compared to the economic resources

Is Cyber Warfare an Alternative?

67

required for conventional actions, they are much smaller. The combination of all these factors, therefore, makes this option very desirable, and this, theoretically, can lead to the spread and proliferation of cyber powers ready for cyber warfare as an alternative to kinetic war. However, its operationalization may not be easy and may even be impossible for certain states. These difficulties are essentially related to the aforementioned needs for human and technological resources. However, in quantitative terms, the needs are smaller. Compared to those required for a conventional war, the qualitative needs are much greater and much more demanding. These qualitative constraints are the most restrictive of any country being able to obtain the capabilities needed to launch cyber actions. Considering all these apparent advantages of using military actions based on computer systems together with the difficulties that may exist in the constitution of cyber military units, will this option be a true alternative for any country?

2 Critical and Essential Conditions There are factors critical and essential in constitution of cyber units so that the actions they bring can be successful. This applies to all countries seeking the acquisition of cyber warfare capabilities. It has already been mentioned that the financial requirements for the constitution of these units are relatively lower than those required for the constitution of modern conventional units. Then, perhaps for any country, those financial aspects are not impeditive to the formation of cyber armies. It can be considered that there are some critical and essential factors to obtain cyber capabilities for use in a cyber war. These factors are: (i) highly qualified human resources to integrate the cyber units; (ii) sophisticated technological resources; (iii) the existence of computer vulnerabilities in the systems of opposing countries and (iv) the cyber units having knowledge of these failures and have the capacity to exploit them. It is essential that the computer systems of opposing countries have vulnerabilities in the hardware and/or software, which allow the use of software specifically developed, informatics exploits, to control such systems using these failures. If those systems do not have vulnerabilities cannot be attacked using cyber means because cyber weapons consist in exploits which use the vulnerabilities to affect the systems. Time is another important factor in this scenario. Not only the time required for the preparation of the cyber troops, but also the time necessary for the preparation and development of the cyber actions, which has to be done with great anticipation. These specific developed informatics exploits have to be secretly placed into enemy computer systems, which require a lot of preparation and time. On the other hand, if there is a long time between the secret placement of these exploits and their possible use, the other country may have realized those vulnerabilities or may have changed the physical equipment or the software. This way, purposely or not, it can

68

J. Barbosa

frustrate and prevent the use of those exploits, which, because they are specific to the previous setup, would thus no longer be useful after these changes. On that case, the whole process would have to be restarted if that would be possible. Then, in the scenario of a cyber warfare, the time factor is very important in the preparation and development of cyber war actions.

2.1 Human Resources Cyber units cannot be formed with any “recruit” as it is the case of some conventional armies. There are armies in which the structure and preparation are very different when compared to the conventional armies of the more developed nations. Often these armies are no more than tribal or regional groupings that seized power and after became the army of these countries and did not necessarily evolve into a modern conventional army. In these conditions, it will be very difficult for these armies to evolve into the creation of modern cyber units, given the great capacity of organization and the technical knowledge that would be necessary for this. The requirements for cyber troops have nothing to do with those needed by most troops of conventional armies. Excluding, perhaps, the senior officers and some specific military units, the requirements for military capability of conventional units are primarily their physical and athletic abilities, which should enable a very good capacity for the execution of military operations, especially on the ground. The requirements for the elements of military cyber units are very different. Although the physical and athletic ability is also important, it is the intellectual capacity that is paramount. Maybe the most very important factor that may hinder, or at least make it difficult, the constitution of cyber units, is the need for highly trained members from the point of view of computer skills. These cyber troops must have the capacity to develop computer applications, the designated informatics exploits, capable of taking advantage of possible hardware or software failures, vulnerabilities, in the computer systems of the opposing countries. To be used in cyber warfare, these exploits have the ability to interfere with the normal activities of these systems or even paralyze them. Then, the knowledge and mastery of information technology in its various subareas, software, hardware, data networks, databases, programming, etc. of the cyber troops, should be excellent, and it is fundamental and indispensable. The above preparation requires a lot of time and maybe not is possible to give to anyone. The computer aptitude of those involved is fundamental, since the computer skills and abilities are very innate. On the contrary, the preparation of a substantial part of the troops for a conventional war is relatively quick, and it is even possible that, at the height of a conventional attack, a country will be able to form a large part of these conventional troops on an accelerated basis. In other words, it is relatively more easy to obtain the skills and abilities necessary for a conventional military unit than for a cyber unit. For the reasons given, replacing, in case of loss, the elements of a cyber unit may not be easy or fast and may even be impossible. This can make it

Is Cyber Warfare an Alternative?

69

impossible to launch cyber war operations, even hinder, or difficult the continuation of those already started. The loss of a single element, if it is a fundamental element due to its capacities and knowledge, can be disastrous for cyber unity, given its great value of knowledge. The number of these troops is not of the same order of values of those normally existing in a conventional army, being smaller. However, given the quantity, diversity and complexity of the tasks to be performed, have to be a reasonable size.

2.2 Technological Resources In the field of technical needs, and in addition to the needs of highly qualified human resources, technological means are necessary to enable such cyber troops to operate and to contribute to and enable the success of their missions. Fundamentally, there are two types of important technological means: (i) hardware, physical equipment and computer network necessary to set up a support structure for the missions; (ii) software, exploits, that allow the control or manipulation of the systems. The constitution of the support structure, computer hardware and computer network, is relatively easy to obtain, and it is almost just a financial question whether any country obtains them or not. These financial needs will not be very large, especially if compared to those required for the purchase of a modern conventional military medium, military aircraft, submarine or even a battle tank. Exploits, in particular ZDE, are much more difficult to obtain because they are specific to the systems and vulnerabilities in them. In the so-called Deep Web, there is, however, a vast offer of exploits. However, these exploits that are in the Deep Web have, in general, little interest for the cyber warfare exactly because already being publicly known, that is, to say not being ZDE. As such, at least theoretically, computer systems, especially those linked to critical infrastructures, may already be protected from attacks based on this failure. They are not, therefore, properly exploits of “cyber military quality”. Rather, they are only of interest to cybercriminals who want to launch attacks aimed at individuals and small businesses. We do not want to say that such attacks are unimportant because they can inflict great damage and are a great concern. Without these vital resources, the activity, usefulness and necessity of cyber military units become quite unimportant, and small nations or technologically less advanced ones are unlikely to succeed in building these units. The difficulty in obtaining and using exploits, such as the “quality” cyber military ZDE, could be too one greatest obstacle to the proliferation of cyber military units.

70

J. Barbosa

3 Interest and Need of Cyber Armies We can question the interest and/or necessity of any country to constitute cyber armies. “Why?” and “For what?” any country would want to create a cyber army. It is too required to analyze if this option is a valid alternative to the conventional means. Building cyber capacities is not easy for most country, and the conditions to maintain this capacity are not simple too. If the initial conditions that made possible obtain cyber capacities are lost, namely if during the cyber warfare get lost of human or technologically resources, the action can be turned against the aggressor country. Then, the decision on the constitution of a cyber army must consider not only the initial conditions and time to develop cyber-attack capacities but also cyber-defense capacity and the possibility of the countries to maintain those actions. Many countries may initially depend on other countries for the constitution of its cyber capacity. To form the necessary frameworks for the constitution of its cyber army, namely their human resources, they may be obliged to use foreign educational and research institutions. To get the technological means, for example, ZDEs, initially they can also have obtained them in foreign countries.

3.1 The Purpose of the Cyber Warfare In many less developed countries, the conflicts that arise are generally regional conflicts to dominate the exploitation of natural resources or tribal or ethnic conflicts with the same objective of exploiting natural resources in other parts of the same country or to obtain the supremacy of one ethnic group over another or others existing in that country. In this way, triggering a cyber war instead of a kinetic war does not make much sense in these types of scenarios and countries. Despite the advantages that a cyber warfare can have in general, capacity building takes a long time and requires resources, such as human resources, that will not be readily available for this type of conflict. Many of these conflicts are often triggered by reaction to close and recent actions, and as such, it is not possible to establish planning and obtain the necessary human and material resources for a cyber warfare. In the same way, the already mentioned conflicts triggered to obtain profits in the exploitation of natural resources as the primary objective takes the control of the respective mining complexes and this presupposes the physical and territorial domination of these places, which does not fit very well with the concept of cyber warfare. Not taking as an absolutely correct premise, we consider that in a large number of conflicts triggered with some frequency, the applicability of a cyber warfare is of little interest although isolated cyber actions may occur.

Is Cyber Warfare an Alternative?

71

Even regional powers, in general, have the concern that they are regional powers to protect themselves and not necessarily to militarily dominate their neighbors. Even if they develop cyber resources capable of endowing them with a cyber warfare, they would not trigger it in the same logic that even though they may have some conventional ability, they only use it for defense. The Federative Republic of Brazil and the Republic of South Africa will be examples of this. In the case of the great global military powers, the USA, Russia and China, cyber warfare has been strongly considered and even implemented for some time, being considered as another operational domain. In the case of the USA, one of its first units linked to cyber warfare, yours first “Cybercorp”, the “Air Force—1st Cyber Division”, was formed in the nineties of last century. However, with regard to their use in military attack, these global powers have considered this alternative as another military option of their vast arsenals, along with the other conventional means already existing. The philosophy will then be to employ or not, according to the military theaters and operative situations. It has been particularly considered as an initial use option, to be used to weaken the enemy and to prepare the entry of conventional means, for example, to blind air-control radars or to shuffle the enemy’s military logistics. In addition to these possible uses of computer means to military attack, these global powers have given great importance to yours cyber-defense. Much of the US initiative in cyber warfare area is in the way to increase and consolidate its cyberdefense capabilities. This perspective of cyber-defense rather than cyber-attacks is also more in keeping with the principles set out in the Charter of the United Nations. One other type of actors, may be capable of considering the employment of cyber means, is non-states and some terrorist organizations. However, we believe that in these cases too, the above-mentioned needs, essentially the preparation time and the necessary resources, would be great obstacles to its realization. This possibility is more difficult to employ, so that in these situations, such non-states or terrorist organizations would probably have opted instead for the use of conventional means of warfare, not excluding of course isolated cyber-attacks.

3.2 Applications of Cyber Weapons An important aspect related to the issue of the use of cyber weapons is the fact that cyber warfare may not be of general use. Then, is not suitable for certain scenarios where conventional warfare may be more advantageous. For example, if we consider that the initiation of a war against a country may have as its objective the territorial domination of that other country or part of it, cyber warfare alone is not adequate, having to be complemented by the use of conventional means, namely with ground troops who proceed to this territorial occupation. Obviously, even in these scenarios, cyber warfare is useful and can be used, but only as a complement to action, either in the preparation of conventional actions

72

J. Barbosa

preceding kinetic warfare, or during or a posteriori of conventional actions for aid or consolidation of the victory of conventional means. The cyber warfare can too be seen not as an alternative way to the kinetic war but as a subsidiarity mean and vice versa, i.e., the conventional means can complement a cyber offensive and cyber means can be used to complement the kinetic war. This way, conventional means may also be employed, by the cyber-attacked country, for a conventional retaliation to a cyber-attack.

3.3 Commitment of Cyber Means As already mentioned, for various reasons, maybe the use of cyber means is not to a general application. Soon, the countries to “attack” will have cyber structures that can be targets for the cyber weapons of the attacking country. On the other hand, these cyber structures must be permeable to the use of cyber weapons, that is, they must have vulnerabilities, hardware and/or software vulnerabilities, which allow the development of computer applications, the exploits, that using these vulnerabilities may affect those computer systems and thereby directly or indirectly affect the infrastructures that the systems control or manage. These conditions eliminate the use of cyber media in confrontations with many countries, if they do not have these cyber targets. Since many of the conventional wars are also small scale and between relatively undeveloped countries, cyber wars will not be an alternative to them. If the target is countries in these conditions of weak dependence on information technology in their society, even countries that are capable of having a high cyberattack capacity will be impeded or very limited in the use of their cyber means against them. One of the great difficulties for any country is the ability to gauge the cyber capabilities of another country. Unlike conventional benchmarking capacities, which can be minimally perceptible through espionage and the use of technical means of espionage, such as satellites or others, cybernetic capability is not easily measurable. This difficulty exists not only because cyber media are difficult to observe, namely software-based ones that are immaterial and therefore invisible, but even if one can perceive their existence one may not know or be able to determine the effectiveness of the same. If the determination of the number of human resources linked to existing cyber armies its possible, that determination alone may not be enough given that in this particular case, it is not only the number that is important, but also, mainly, its technical capacity. A large cyber army might not be a good cyber army if the human resources were not highly trained and cannot therefore know how to take advantage of the weaknesses of enemy systems. Another difficulty and possibility are if the country to attack initially does not have or have little capacity for cyber-defense capabilities, but after the attack can

Is Cyber Warfare an Alternative?

73

establish an alliance with some cyber-enabled country that can retaliate against the attack in favor of the country’s cyber ally. This possibility may totally change the initial considerations, which the attacking country may have taken into account. It should be noted that due to the possibilities of cyber media, the allied country does not have to be geographically close. The most referred attack methods used as traffic redirection, website defacement, distributed denial-of-service (DDoS), viruses and trojans and a specific cyber weapons using ZDE, such as those used in the famous action against Iran known as Stuxnet (Zetter, 2014). In the strict sense of cyber warfare, are used specific cyber weapons, the ZDE. The most known cyber warfare action, known as Stuxnet, used cyber weapons, which exploit ZDVs in the computer systems of nuclear centrifuges of Iran (Zetter, 2014). The very big difficulty in identifying the aggressor country is very restrictive and prevents any retaliatory action, whether using conventional means or using cyber media. Many cyber actions can be triggered by using computers in third countries that have nothing to do with cyber warfare. The including DDoS attacks it’s a very good example of this. Maybe can very hard to identify the authors whereas launched the DDoS attack. A bad identification and retaliation against other country them can escalate the war and increase exponentially cyber warfare because the country from which the attack was apparently launched may not be the primary perpetrator of the attack not having it launched and having nothing to do with it. It could only have been used as a launching platform for the attack, using computers located in its territory, whose legitimate owners were unaware of this use. A cyber-attack against this third country can also cause the same to retaliate, thus escalating the cyber war. A classic example of this is the so-called Estonian Warfare in 2007, in which apparently many of the computers used in these attacks would be located in the USA, an allied country of Estonia. In the literature, [1, 2] are reported many cyber-attacks actions against more countries, such USA, South Korea, Iran, Tatarstan, Kyrgyzstan, Israel and the Palestinian National Authority, Zimbabwe and Myanmar. More of these actions can be considered at the border between common cybersecurity actions and cyber warfare actions. Better separation will be achieved if it is precisely determined those as put in danger the national sovereignty of the affected states. Many others cyber warfare actions are allegedly attributed to Russia against Estonia, Georgia and Ukraine. These cyber warfare actions were triggered using not only the mentioned DDoS method but too SQL Injection, SQLi, and Cross-Site Scripting (XSS) attacks. On the other hand, cyber actions not being dependent or limited by physical distances or frontiers. This way the aggressor countries not need be geographically close to the country to attack. This characteristic can allow too, a type of military acts for small-scale actions only with intentions of disruption, may be liable to be launched against particular countries, notably against the large western countries.

74

J. Barbosa

4 Conclusion The possibility of using cyber weapons, which allows the launching of a cyber warfare, is due to the vulnerabilities existing in computer hardware and/or software, which, in general, all informatics systems have. When these failures exist in systems, not only in the system directly linked to the processing of vital activities of coordination or management of entities more directly related to civil society, but also in those related to the control of facilities, i.e., critical infrastructures, the negative consequences of cyber-attacks can be devastating for the country under attack. The possibility of the occurrence of cyber war is at present very real and that the total impossibility of its occurrence does not seem very likely very soon, but may instead increase, since this impossibility would have to stem from the existence of faultless computer systems, bugs, which seems to us to be utopian in the near future. Taking into account only the technological aspects referred to above, we can assume that any nation can acquire cyber military capacities because the creation of a computer infrastructure supporting cyber war activities, although not easy, is not financially significant enough to make it impossible for any country. However, the other necessary resources, namely the need for highly trained human resources, both for the development of cyber warfare and for the conduct of cyber warfare, are already much more difficult to obtain. The formation of these human resources is time consuming, the existence of people with skills for this area and for this type of activity is rare, and therefore, it is not possible to transform anyone into a qualified resource, by much training that can be given. In addition, the volatility and alteration of the systems require that this training be constant in order to be able to maintain the minimum level of knowledge that eventually could have been reached when the cyber unit was formed. Another significant obstacle, although it is theoretically possible to be overcome, is the need to know the vulnerabilities of the systems, both hardware and software systems, if you explore them. These vulnerabilities, which must necessarily exist in the systems in order to be possible to attack them, are referred to as vulnerability, and if they are not yet known and publicly disclosed, they are called zero-day vulnerability, ZDV. These vulnerabilities are not very common nor are they widely available. In order to be able to exploit these faults in the computers of the opposing countries, it is necessary to develop or obtain specific software for this, called exploits, and if they were developed for ZDV, they are called ZDE. In addition, if any country that claims to acquire cyber military capacities might not be able to obtain and use ZDE, this claim is impossible to achieve. Cyber warfare is a very interesting alternative, but given these constraints and limitations, it may not be a very viable alternative in certain situations and for certain countries. A cyber warfare, by itself, may not be enough to defeat an “enemy” or to subjugate it even if it is only economically. In addition, not all conditions and required resources to establish a cyber army are easy to obtain and maintain.

Is Cyber Warfare an Alternative?

75

Acknowledgements I thank the Portuguese National Defense Institute as well as the ISEC/IPC for the support and facilities granted. This paper is based on the work Final Report of National Defense Course, done by the author in the Portuguese National Defense Institute.

References 1. Carr, J.: Inside Cyber Warfare, 2nd edn. O’Reilly Media Inc., Sebastopol, CA, USA (2012) 2. Kostopoulos, G.K.: Cyberspace and Cybersecurity. Boca Raton, FL, CRC Press—Taylor & Francis Group (2013) 3. Barbosa, J.: Pequenas potências militares convencionais, Grandes potências militares cibernéticas—Abordagem da utilização de meios informáticos na defesa/ataque militar moderno. Final Report of National Defense Course, Portuguese National Defense Institute, Lisbon, Portugal (2018) 4. Clark, R.A., Knake, R.K.: Cyber War: The Next Threat to National Security and What to do About. HarperCollins Publishers Inc, New York (2010) 5. Halpin, E., Trevorrow, P., Webb, D., Wright, S. (eds.): Cyberwar, Netwar and the Revolution in Military. Palgrave Macmillan, NY, USA (2006) 6. Ranger, S.: What is Cyberwar? Everything you Need to Know About the Frightening Future of Digital Conflict. https://www.zdnet.com/article/cyberwar-a-guide-to-the-frightening-future-ofonline-conflict/ (2018), Last accessed in November 2018 7. Singer, P.W., Friedman, A.: Cybersecurity and Cyberwar—What Everyone Needs to Know. NY, Oxford University Press (2014)

Memory Auditing for Detection of Compromised Switches in Software-Defined Networks Using Trusted Execution Environment Filipe Augusto da Luz Lemos, Rubens Alexandre de Faria, Paulo Jose Abatti, Mauro Sergio Pereira Fonseca and Keiko Veronica Ono Fonseca Abstract Current solutions for detecting compromised switches in software-defined network (SDN) usually rely on the monitoring of the network traffic or conformance of the packets traversing through them and the rules defined by the controllers. Although satisfying, those solutions cannot detect a switch that has been compromised if it is not acting maliciously at the moment of the traffic monitoring as sleeper agents, which can pose as a national security risk when defense networks are the targets. An architecture capable of auditing the memory of switches in softwaredefined networks is proposed as a solution to detect compromised switches even when they are not acting maliciously and only leave micro-traces of its activities. This auditing should be able to verify the conformity between what is in the memory and the flow rules set by the controllers without overusing the system resources. A trusted execution environment is also proposed to improve the security of the auditing processes. Keywords Software-defined networks · Memory auditing · Compromised switches · Trusted execution environment · Sleeper agents

F. Augusto da Luz Lemos (B) · R. Alexandre de Faria · P. Jose Abatti · M. S. Pereira Fonseca · K. V. Ono Fonseca Federal University of Technology - Parana, Curitiba, Brazil e-mail: [email protected] R. Alexandre de Faria e-mail: [email protected] P. Jose Abatti e-mail: [email protected] M. S. Pereira Fonseca e-mail: [email protected] K. V. Ono Fonseca e-mail: [email protected] © Springer Nature Singapore Pte Ltd. 2020 Á. Rocha and R. P. Pereira (eds.), Developments and Advances in Defense and Security, Smart Innovation, Systems and Technologies 152, https://doi.org/10.1007/978-981-13-9155-2_7

77

78

F. Augusto da Luz Lemos et al.

1 Introduction Software-defined network (SDN) is a network architecture that proposes the separation of the control plane from the data plane [1]. The separation of the planes is made by software and introduces the centralizing figure of the controller [2], as exemplified in Fig. 1. The decoupling amplifies the awareness of the controller over the network topology under its supervision, providing a better control over the network traffic, for example, [2]. This greater awareness of the network has been successfully used to improve network security against attacks, such as in DDoS attacks [3]. The separation of planes has brought improvements to the network security; however, event detection of possible security incidents mostly relies on the monitoring of the data flows, the compliance between network flow rules and the network traffic [4]. This fact raises a problem of detecting a compromised network equipment: the detection requires it to be behaving maliciously at the moment of the monitoring [5]. Figure 2 exemplifies that case: the switch has an invalid flow entry, but the network data monitor only verifies the data flowing through it and would probably not detect the problem. Fitting to this case are the “sleeper agents” which are compromised network devices that only start to function maliciously when activated by their true controller [6]. Examples of hidden malware and malicious hardware installed without the user knowledge can be found in cases like tampering of servers that were sold to major Americans Telecommunication Companies [7], posing an issue to the national security. In this paper, we propose a solution to detect compromised SDN switches by auditing its memory to identify if the contents of the switch memory comply with the rules set by the network controller or if any unauthorized rule was added or removed from it, what we define as micro-traces. As a further improvement of our proposal, a trusted execution environment (TEE), in specific the Intel Software Guard Extensions (Intel SGX) [8], will be discussed as the execution environment of the aforementioned auditing process.

Fig. 1 Simplified example of a software-defined network architecture

Memory Auditing for Detection of Compromised Switches …

79

Fig. 2 Example of a sleeper agent: an invalid entry is not detected by the network data monitor

The remainder of this paper is organized as follows: Sect. 2 presents some assumptions for the software-defined network architecture described in this paper, Sect. 3 presents our proposal, Sect. 4 discusses the major points of the proposed architecture, and Sect. 5 presents the conclusions.

2 Assumptions on the Targeted SDN Architecture The SDN architecture targeted in this paper relies under two assumptions regarding its switches: the size of the switch-audited memory is relatively small and its memory content has little variance varying accordingly to the OpenFlow Protocol rules [9]. The concept behind this is that the auditing process needs to have the lowest impact possible on the system, so it does not deprive the controller and the switches from the system resources.

2.1 Audit Memory Size The memory size of the switches refers to the space needed to store the flow table that is set by the network controller. The table size varies with the number of flow entries, and each flow table entry content is defined under the OpenFlow Switch Specification [9] as having seven different fields, as shown in Table 1. That is a linear correlation between the memory size and the number of flow table entries.

80

F. Augusto da Luz Lemos et al.

Table 1 Seven different fields in a flow table entry according to the OpenFlow Switch Specification 1.5.1 [9] Match fields Priority

Counters

Instructions Timeouts

Cookie

Flags

2.2 Entries Contents As mentioned previously, the seven different fields of a flow table entry vary accordingly with the OpenFlow Protocol and are well documented under the Switch Specification [9], thus limiting the possible outcomes of the memory. Although this may look like a lot at a first glance, an entry always follows the same rules even on SDN controllers deployed by different developers, therefore limiting the number of combinations of values for the entries.

3 Proposed Architecture As stated in the previous section, the auditing process is concerned about the limitation of the system resources during its execution. Therefore, an online (real-time execution) verification of each entry in the flow table of a targeted switch would require processing and network resources that may be not available at the switch. Figure 3 presents a simplified flowchart of the envisioned auditing process. The figure shows that the auditing entity collects information from the network controller (NC) and makes a dump of the switch memory, so it can later calculate the hash values and compare them. Figure 4 exemplifies how the proposed architecture leverages the usual monitoring approach against sleeper agents: the auditing process can audit the memory, identify if there is an invalid entry, and generate a warning to the network controller, so it can take the necessary measures. A malicious attacker could intercept the memory dump data and change its contents to comply with valid entries as expected by the NC. An approach to overcome this attack is to exchange only encrypted data between the NC and the switch. However, if the encryption process itself were not protected, the auditing cannot be trusted. By using trusted execution environments (TEEs), additional protection can be provided aiming at improving data security, as discussed in the next section.

3.1 Trusted Execution Environments The introduction of the trusted execution environments (TEEs), like the Intel SGX [8], aims to prevent the auditing to become a new security threat. TEEs are protected areas of the processor that allow code to be executed inside of it from being

Memory Auditing for Detection of Compromised Switches …

Fig. 3 Example of flowchart of the auditing process Fig. 4 Under the proposed architecture, the auditing process is able to detect a problem even when the switch is not acting maliciously

81

82

F. Augusto da Luz Lemos et al.

Fig. 5 Representation of the execution of a code inside a secure enclave

disclosed or modified [8]. As shown in Fig. 5, an untrusted code calls a function to be executed inside the protected enclave (ecall). Another important point shown in Fig. 5 is the ability of a trusted code to call an outside function (ocall), for example, the printf function. Although the Intel SGX has presented some vulnerability issues [10], Intel’s TEE is still relevant when applied to secure processes like encryption, attestation, auditing, and others. From the steps shown in Fig. 3, the auditing process will utilize TEEs to encrypt the communication with the network controller, to protect the processes of preparing the retrieved information, to calculate the hash values, and to compare them. This usage of the enclave can impact the performance of the auditing process, and its use needs to be limited to only the necessary as shown in previous research [11].

3.2 Performance Evaluation The two main metrics for software-defined networks [12] are the throughput, how much information is sent per second, and latency, the time between a request and its response. Those two metrics combined with the analysis of the system resource usage (memory, processor, etc.) will be used to evaluate the process performance and impact on the system. Specifically to the Intel SGX TEE, it will be important to evaluate the time spent on e/ocalls and the processing time spent inside the enclaves. Previous research [13] has shown the high impact that e/ocalls causes to the system.

4 Discussion The proposed architecture aimed to enforce the auditing of running switches should consider security by design, that is, from its conception, a separation of trusted and untrusted software must be identified in order to decide which/what should run

Memory Auditing for Detection of Compromised Switches …

83

Fig. 6 Representation of a scenario where: 1 shows the network working properly; 2 shows the hijacking of the switch; 3 the auditing process warning the network controller; 4 the network controller isolating the malicious switch and diverts the host traffic to a new and trusted switch

inside trusted execution environments. Similarly as developed in previous researches on auditing of software [11], it was not possible to guarantee that the auditing process would not affect the system performance without a well-defined system where a benchmark could be performed. Other questions, such as what would be the periodicity of the memory auditing should be answered from demand performance tests of the implemented system. For performance tests, a few scenarios were envisioned, including single and/or multiple switches starting with and without invalid entries; single and/or multiple switches having an entry being changed randomly; and a switch being hijacked (not responding to its true network controller). The last scenario is exemplified in Fig. 6, where after a switch has been hijacked, the audit process warns the network controller that acts to isolate the compromised switch and divert the host traffic to a new and trusted switch.

84

F. Augusto da Luz Lemos et al.

5 Conclusion At this stage, the proposed architecture is theoretically adequate to cater the problem of identifying compromised switches in software-defined networks while not demanding much of the system resources, but, as a work in progress, the results of this work are still under development and analysis. The process is able to identify the conformity of the flow tables using a simple hash value method. The proof of concept under development utilizes a simplified SDN architecture running on the Intel SGX. As future study, we propose the implementation of an architecture able to audit the switch execution environment and to evaluate the impact of securing the controller-switch communication with Intel SGX to further ensure software-defined network security. Acknowledgements This research work explores possible applications for TEE and secure and scalable cloud applications as part of the EU-BR SecureCloud project. The project has been receiving funds granted from the 3rd EU-BR Coordinated Call (Brazilian Ministry of Science, Technology and Innovation, MCTIC/RNP, BR grant agreements 2550, 2549, 2553, 2552 and 2568) and European Union Horizon 2020 programme—EU Grant Agreement 690111). The project is also supported by the Swiss State Secretariat for Education, Research and Innovation (SERI). This research work also explores possible security solutions for forensics applications under the CAPES Pró-Forenses Project 025/2014.

References 1. Nunes, B.A.A., Mendonca, M., Nguyen, X., Obraczka, K., Turletti, T.: A survey of softwaredefined networking: past, present, and future of programmable networks. IEEE Commun. Surv. Tutor. 16(3), 1617–1634, (2014). (Third Quarter).https://doi.org/10.1109/SURV.2014.012214. 00180 2. Nadeau, T.D., Gray, W.K., SDN - Software Defined Networks: O’Reilly (2013). ISBN: 1449342426 3. Van Trung, P., Huong, T.T., Van Tuyen, D., Duc, D.M., Thanh, N.H., Marshall, A.: A multicriteria-based DDoS-attack prevention solution using software defined networking. In: 2015 International Conference on Advanced Technologies for Communications (ATC), pp. 308-313. Ho Chi Minh City (2015). https://doi.org/10.1109/ATC.2015.7388340 4. Zhou, H., et al.: SDN-RDCD: a real-time and reliable method for detecting compromised SDN devices. IEEE/ACM Trans. Netw. 26(5), 2048–2061 (2018). https://doi.org/10.1109/TNET. 2018.2859483 5. Dabbagh, M., Hamdaoui, B., Guizani, M., Rayes, A.: Software-defined networking security: pros and cons. IEEE Commun. Mag. 53(6), 73–79 (2015). https://doi.org/10.1109/MCOM. 2015.7120048 6. Price, D.: A guide to cyber intelligence. J. US Intell. Stud. 21(1) (2014–2015) 7. Robterson, J., Riley, M.: The big hack: how China used a tiny chip to infiltrate U.S. companies. Bloomberg Businessweek, 4 Oct 2018 8. Intel Corporation, Intel Software Guard Extensions (Intel SGX), Website, Accessed December 14 2018 9. Open Networking Foundation, OpenFlow Switch Specification, Version 1.5.1 (Protocol version 0x06), 26 March 2015

Memory Auditing for Detection of Compromised Switches …

85

10. Newman, L.H.: Spectre-Like Flaw Undermines Intel Processors’ Most Secure Element. Wired, 14 Aug 2018 11. Pereira, L., et al.: Using Intel SGX to enforce auditing of running software in insecure environments. In: The 10th IEEE International Conference on Cloud Computing Technology and Science (2018) 12. Gelberger, A., Yemini, N., Giladi, R.: Performance analysis of software-defined networking (SDN). In: IEEE 21st International Symposium on Modelling, Analysis and Simulation of Computer and Telecommunication Systems, San Francisco, CA 2013, pp. 389–393 (2013). https://doi.org/10.1109/MASCOTS.2013.58 13. Costa, R.S., Pigatto, D.F., Fonseca, K.V.O., Rosa, M.O.: Securing Video on Demand Content with SGX: A Decryption Performance Evaluation in Client-Side, Simposio Brasileiro em Segurança da Informação e de Sistemas Computacionais (SBSeg), [S.l.], pp. 127–140 (2018)

Mobile Communication Systems: Evolution and Security Teresa Guarda, Maria Fernanda Augusto, Isabel Lopes, José Avelino Victor, Álvaro Rocha and Lilian Molina

Abstract With the evolution and popularization of mobile communication systems, especially for access to data services, it becomes increasingly important to have a closer look at security systems. From analog systems to the fourth generation, many changes and evolutions occurred. This makes possible to question whether mobile networks have sufficiently secure systems, and what vulnerabilities we should be aware of. Due to the widespread risk of attacks on mobile communications, telecommunication companies are increasingly emphasizing telecommunication security T. Guarda (B) · M. F. Augusto · L. Molina Universidad Estatal Península de Santa Elena—UPSE, La Libertad, Ecuador e-mail: [email protected] M. F. Augusto e-mail: [email protected] L. Molina e-mail: [email protected] T. Guarda Universidad de Las Fuerzas Armadas-ESPE, Sangolqui, Quito, Ecuador T. Guarda · I. Lopes Algoritmi Centre, Minho University, Guimarães, Portugal M. F. Augusto BITrum-Research Group, Calle San Lorenzo 2, 24007 León, Spain I. Lopes UNIAG (Applied Management Research Unit), Polytechnic Institute of Bragança, Bragança, Portugal e-mail: [email protected] J. A. Victor Instituto Politécnico da Maia, Maia, Portugal e-mail: [email protected] Instituto Universitário da Maia, Maia, Portugal Á. Rocha Universidade de Coimbra, Coimbra, Portugal e-mail: [email protected] © Springer Nature Singapore Pte Ltd. 2020 Á. Rocha and R. P. Pereira (eds.), Developments and Advances in Defense and Security, Smart Innovation, Systems and Technologies 152, https://doi.org/10.1007/978-981-13-9155-2_8

87

88

T. Guarda et al.

technologies specifically in fourth generation (4G) technology, in order to ensure the main security functions to provide an efficient service to its customers. Keywords Telecommunications · Security · G technologies · Mobile communications

1 Introduction Networks that provide mobility services have evolved rapidly, going far beyond their original purpose of providing mobile voice services. With 4G/LTE (Long-Term Evolution), which are fully IP-based, it is possible to support data transmission at high speeds, in the order of Mbps, in addition to traditional voice services. This structure is critical to building new business models based on mobile services such as the Internet of things, cloud computing, video content distribution, application providers, and more [1]. With 4G technology, new security issues have emerged, gaining great relevance over LTE/4G networks because they are based on the IP protocol. As a result, new threats to a mobile service provider can be identified as a breach of confidentiality, breach of integrity, availability reduction, and fraud [1, 2]. Threats are becoming more important as mobile subscribers are having a more and more powerful smartphone because the 4G architecture considers the use of external networks not controlled by operators as new forms of access, and threats are spreading more and more in IP networks. In this scenario, mobile operators must prepare for challenges by building a robust and flexible security architecture, while at the same time ensuring at least interoperability with legacy systems (2G/3G/4G networks), a smooth transition to IPv6 and support for virtualizing infrastructure [3].

2 Background The first generation (1G) was a technology widely used in the 1980s by the Advanced Mobile Phone System (AMPS) network and was made up of analog systems, with only voice transmission possible. It was gradually replaced by the digital signal (2G) [4]. Second generation (2D) became popular in the 1990s when the mobile phone began to be more used by the population, and the signal changed from analog to a digital signal with the most widely used technologies Code Division Multiple Access (CDMA) and Global System for Mobile Communications (GSM). CDMA is a method of accessing communication systems that can be used for both mobile telephony and GPS. GSM—it is a method used to send and receive Short Message Service (SMS) [3, 4].

Mobile Communication Systems: Evolution and Security

89

Between the second generation (2G) and the third generation (3G), a number of developments have taken place: Deploy the General Packet Radio Service (GPRS) wireless data transmission system for telephones that send and receive short messages; the 3G pioneer standard, Enhanced Data rates for GSM Evolution (EDGE), is being used; set the High-Speed Circuit-Switched Data (HSCSD) as a specification for transferring data over GSM networks with digital voice signals and channels; and the Wireless Application Protocol (WAP), that is, the basic mobile Internet system for any mobile device, allowing easy access of content to original pages [1, 4]. As there was evolution in the technology of communication of wireless networks and evolution of the devices that until then only allowed to make analog connection, there was the advance in the digital technology that allowed not only the voice call, but also the communication of data through images, audio, and video. With the new form of wireless network communication, large companies in the global mobile device landscape began to vie for a growing market of users looking for devices capable of communicating over a more advanced wireless network, services ranging from sending messages to sending image, audio, or video and technology called 3G made it all possible. 3G technology emerged in the 2000s, when mobile phones finally became smartphones with their own operating systems, Wi-Fi Internet, better quality cameras, Bluetooth signal, and more. 3G technology has come to enable smartphones to function properly, as they have better capabilities. Highlights include data transmission over long distances in mobile environments. 3G mobile technology is capable of supporting a greater number of customers and data, especially in urban areas, by increasing data rates [4]. The fourth generation started in 2010, being entirely based on IP, it can be understood as a system and network at the same time, being much more than the mere extension of 3G technology. In this way, it is possible to converge cable, wireless, and computer networks with electronic devices, exceeding access speeds of 100 Mbps and reaching up to 1 Gbps [3] (Fig. 1). It is anticipated that mobile networks of 5G will be 10–20 times faster than those of previous generation (4G). The world does not stop, and 5G technology is already taking its initial steps. Mobile devices and computers are increasingly interconnected, transmitting more data, and soon they will need a new network to handle all that speed and information.

3 4G Technology It is not a defined technology or standard, but a collection of technologies and protocols that allow maximum performance. This convergence of technologies arises from the need to group different standards in use, in order to delimit the scope of operation of each of them, and integrate all communication possibilities in a single device, in a transparent way to the user. 4G is the fourth generation Internet connection technology; at the moment, it is the new technology present in the majority of mobile phones worldwide. Being a

90

T. Guarda et al.

Fig. 1 G technologies speed: from 1G to 4G

system based on packets containing fewer network elements, among its main features are providing high performance in terms of higher connection speed and decreased latency when accessing the connection compared to its predecessor technologies [5]. This technology works with the TCP/IP architecture and its set of protocols, and because of the fact that it works with the TCP/IP protocol group, this technology has greater vulnerabilities in its security compared to its previous technologies [4]. In 4G technology, it is a real challenge to ensure the security of communications, and one of its main factors to take into account is that wireless technology does not have fixed connection points and must focus on the mobility and flexibility requirements of the 4G technology [6]. The 4G technology was initially designed to support a bandwidth with large capacity, are designed based on flat architectures with fewer network elements, low latency, and fast convergence for adapt to new wireless network technologies. The 4G architecture was developed taking into account the principles of security from its inception and design based on the following security features: security of access to the network; security of the network domain; security of the user’s domain; and security of the domain of the application [1]. 4G technology is based on the use of IP technology in the core network and access in order to support all services. In comparison with the old technologies, the 4G technology has a smaller number of elements in its implementation, basically as all the technologies of the mobile communications have three networks where they must follow until finally reaching the service to the station: core network; transport network; and access network [6]. In the core network, we have the main controllers MME and SGW, where the security parameters of the different 4G nodes are configured, the transport network is responsible for transporting the encapsulated service that leaves the core network and it transports it to the access network through the different existing networks using

Mobile Communication Systems: Evolution and Security

91

fiber optic in most cases, the access network is responsible for bringing the service finally to the 4G nodes, the main technologies of the access network are optical fiber and microwave [7]. 4G technology is designed with strong cryptography techniques, mutual authentication between the elements of an LTE network, these techniques are embedded in its architecture, and industry regulators have evaluated and identified new security vulnerabilities that can be mitigated moment of network implementation. The main purpose of implementing the LTE standard was to provide a highperformance radiofrequency access that allows high transmission and reception speeds in mobile devices and allows coexistence with other systems [5, 6]. The LTE standard has enough flexibility to allow vendors to combine these different modules into a single device or multiple devices, for example, separating MME and SGW into different devices.

4 Vulnerabilities and Threats In just over a decade, there has been an extraordinary growth associated with mobile telephony, both in terms of number of users and technology, which went from the first to the fourth generation. Due to this phenomenon, the mobile devices have changed their status, becoming fewer and fewer phones and computers, and therefore gradually becoming more susceptible to various forms of attacks. An unprotected mobile device associated with an insecure network can result in an invasion of individual privacy [8], allowing for facts such as tampering with data or even theft, the intruder will be able to listen to your calls, read your text messages, track your activity on the Internet and not know your geographical location. A network to be minimally safe should follow some basic principles of security: confidentiality; integrity; availability; and authenticity [9]. Confidentiality ensures that access to information is linked only to legitimate entities, that is, those authorized by the owner of the information. In the case of integrity, all manipulated information maintains all original characteristics established by the owner of the information, including changes, birth, maintenance, and end of the information. Availability ensures that the information is always available for legitimate use, by those users authorized by the owner of the information. With regard to authenticity, it is guaranteed that the information was produced, modified, or discarded by a particular person, body, entity, or system in order to validate its origin. Among the most common types of attacks in mobile networks are: unauthorized access; denial of service; channel jamming; eavesdropping; session hijacking; manin-the-middle attack; and message forgery [10–12] (Table 1). Due to the dynamism of the technology of a mobile network, there is no way to offer any kind of definitive security certainty. However, there are many methods to minimize or in several cases eliminate these attempts to break down the security mechanisms of mobile networks [13].

92

T. Guarda et al.

Table 1 Most common types of attacks in mobile networks Attack type

Description

Unauthorized access

If an authentication method is not properly applied or is misconfigured, then an attacker can gain free access to the network and use its services even though it is not authorized to do so

Denial of service

Is caused by the excessive sending of data on the network, more than it can support, leaving users without the network resources available

Channel jamming

It is a technique used by the invaders whose objective is to destroy or degrade the signal of the air interface and thus to disable the access of the legitimate users of that network leaving them exposed to other networks

Eavesdropping

If traffic in the air interface is not heavily encrypted, an attacker could spy on or intercept important data or confidential phone calls

Session hijacking

A malicious user can hijack an already established session and act as a legitimate user of a particular base station

Man-in-the-middle attack

An attacker can be between a cell phone and a network access point to intercept messages

Message forgery

If the communication channel is not enough secure, an attacker can change the content of the messages in both directions without the actual recipients realizing it

The fourth generation of mobile communication systems had the development principle to be fully IP, that is, all elements, except the terminal, should communicate through purely IP interfaces [3]. As a result, various authentication, integrity, and confidentiality techniques existing and used in IP networks are used in 4G, such as IPSec, Certificate Management Protocol (CMP) and Transport Layer Security (TLS). A fourth generation network uses less physical elements, equipment, but each with more functions. It is also planned to use cells to cover small areas, such as homes or offices, called Femto Cells (HeNB-House and NB). In the LTE, the encryption is applied to the user and control data, and the integrity protection is applied only to the control data, through the Packet Data Convergence Protocol (PDCP) [3, 14]. Among the main vulnerabilities of 4G, we can mention the vulnerabilities due to the joint operation with 2G and 3G networks and attacks from the IP interfaces; as well as 3G networks where many users use smartphones, and attacks can exploit operating system vulnerabilities through malicious code hidden in applications, text messages, or multimedia messages [7, 14]. Breaches discovered by researchers in LTE networks, which can lead to problems of privacy of user data. According to the researchers at Ruhr-Universität Bochum and the University of New York, there are three types of attacks that can happen due to the security flaw found in the 4G LTE data link layer. The first two attacks are passive, meaning the attacker cannot interfere with the attacking network. In these two cases, they are responsible for doing an identity mapping and for a method that

Mobile Communication Systems: Evolution and Security

93

can perform fingerprints of Web sites. The third, in turn, is an active attack, which the researchers nicknamed aLTeR. ALTeR uses a separate device by which the cybercriminals can simulate the original network and trick the victim so that they unknowingly reveal information. This method, however, is quite complex since the attacker first redirects network connections by performing DNS spoofing and then exploits a specific fault in LTE. Apparently, the attack can happen to anyone, but because it is a complex method of exploiting vulnerability, it is likely that such invasions will occur with people of interest, i.e., politicians, journalists, and the like. The researchers further conclude that while the demand to perform these actions is exhaustive, current technology allows cybercriminals to evolve and even perform them outside of LTE; even those that support authenticated cryptography [2].

5 Conclusions First generation networks have almost no security features, gaining second generation attention from developers, creating systems of authentication and confidentiality, which were previously non-existent. In turn, 3G networks have brought integrity techniques as well as advances in existing security requirements, such as increasing the size of 128-bit security keys. Fourth generation networks, predominantly IP-based, add security techniques until then used only in the network core. In 4G, all the interfaces except the radio are IP, and this facilitates and makes flexible the installation and maintenance of the networks. Authentication and key exchange systems have been improved compared to systems applied to 2G and 3G networks using HMAC-SHA-256 switches. Encryption uses authentication vectors that are incompatible with those used in 3G networks and to maintain user compatibility, the USIM must check the separation bit. The algorithms used are AES and SNOW 3G, which use 128-bit keys capable of processing 256-bit keys in an upcoming release. The fifth generation (5G) is already in the test phase, with promises to be implemented by the middle of 2020, to revolutionize its services, uniting 3G technologies such as voice, data, videos; and services available in 4G such as digital broadband, IP communication, HD transmission, dynamism of information exchange, among others. 5G technology will better serve the radio spectrum and enable more devices to access the mobile Internet at the same time, promising faster data download and upload speeds, broader coverage, and more stable connections. The purpose of 5G is not to be a new technology, but to unify the best in 3G and 4G technology. In addition, it intends to improve the hardware part such as the devices themselves that will serve as base stations for connection between the devices, thus, there will be no loss of signal or interruption of services, thus managing to maintain the quality of connection.

94

T. Guarda et al.

References 1. Dahlman, E., Parkvall, S., Skold, J., Beming, P.: 3G evolution: HSPA and LTE for mobile broadband. Academic Press (2010) 2. Rupprecht, D., Kohls, K., Holz, T., Pöpper, C.: Breaking LTE on layer two. In: IEEE Symposium on Security & Privacy, pp. 2–16 (2019) 3. Rost, P., Banchs, A., Berberana, I., Breitbach, M., Doll, M., Droste, H., Sayadi, B.: Mobile network architecture evolution toward 5G. IEEE Commun. Mag. 54(5), 84–91 (2016) 4. Gupta, P.: Evolvement of mobile generations: 1G to 5G. Int. J. Technol. Res. Eng. 1, 152–157 (2013) 5. Ratasuk, R., Prasad, A., Li, Z., Ghosh, A., Uusitalo, M.A.: Recent advancements in M2M communications in 4G networks and evolution towards 5G. In: 18th International Conference on Intelligence in Next Generation Networks (ICIN) (2015) 6. Dahlman, E., Parkvall, S., Skold, J.: 4G, LTE-advanced Pro and the road to 5G. Academic Press (2016) 7. Sun, K., Kim, Y.: Gap analysis for adapting the distributed mobility management model in 4G/5G mobile networks. In: IEEE Conference on Network Softwarization (NetSoft) (2017) 8. Garba, A.B., Armarego, J., Murray, D., Kenworthy, W.: Review of the information security and privacy challenges in bring your own device (BYOD) environments. J. Inf. Priv. Secur. 11(1), 38–54 (2015) 9. Li, S., Tryfonas, T., Li, H.: The internet of things: a security point of view. Internet Res. 26(2), 337–359 (2016) 10. Maxim, M., Pollino, D.: Wireless Security. McGraw-Hill/Osborne (2002) 11. Yang, H., Ricciato, F., Lu, S., Zhang, L.: Securing a wireless world. Proc. IEEE 94(2), 442–454 (2006) 12. Welch, D., Lathrop, S.: Wireless security threat taxonomy. In: Information Assurance Workshop, pp. 78–83 (2003) 13. Liyanage, M., Abro, A.B., Ylianttila, M., Gurtov, A.: Opportunities and challenges of softwaredefined mobile networks in network security. IEEE Secur. Priv. 14(4), 34–44 (2016) 14. Mavoungou, S., Kaddoum, G., Taha, M., Matar, G.: Survey on threats and attacks on mo-bile networks. IEEE Access 4, 4543–4572 (2016)

Part II

Computer Networks, Mobility and Pervasive Systems

Evaluating Trail Formation in Collaborative UAV Networks with Lethal Threats Nícolas Pereira Borges, Cinara G. Ghedini and Carlos Henrique Costa Ribeiro

Abstract Aerial vehicles are widely used for missions in hostile environments. Some of these missions can benefit from the use of unmanned aerial vehicles (UAVs) working collaboratively to achieve a common goal. In such scenarios, the use of stealth approaches is essential for mission success, because it allows missions to be conducted in the proximity of threat radars and also reduces the probability of a UAV being spotted by enemies by enemies. In this direction, we extend a model that allows autonomous UAVs to dynamically and collaboratively reduce their exposure to threats by adopting a stealth policy in which they turn off their communication radars once a threat is detected. The model also encompasses strategies for UAVs leaving the threat regions as quickly as possible and estimating the future position of others UAVs, thus trying to regroup to formation. The model was evaluated in simulations using different hostility and weapon range distributions. The results demonstrated that the proposed model is able to significantly improve the number of UAVs that are able to complete the mission without being hit by any threat. In addition, the proposed approach for calculating the goal position when a UAV is exposed to a threat proves to be a key factor for missions performed in hostile environments as its significantly reduces the fraction of UAVs hit by threats. Keywords Unmanned aerial vehicles · Trail formation · Leave threat region · Hitting probability

N. P. Borges (B) · C. G. Ghedini · C. H. C. Ribeiro Computer Science Division, Aeronautics Institute of Technology, São José dos Campos, SP, Brazil e-mail: [email protected] C. G. Ghedini e-mail: [email protected] C. H. C. Ribeiro e-mail: [email protected] © Springer Nature Singapore Pte Ltd. 2020 Á. Rocha and R. P. Pereira (eds.), Developments and Advances in Defense and Security, Smart Innovation, Systems and Technologies 152, https://doi.org/10.1007/978-981-13-9155-2_9

97

98

N. P. Borges et al.

1 Introduction Advanced air defense radar systems significantly reduced the traditional electronic warfare effectiveness. Unmanned aerial vehicles (UAVs) are currently used as a replacement for traditional aircraft on potentially dangerous missions where threats have to be detected and avoided. In addition, considering that autonomous UAVs are able to make decisions faster than non-autonomous counterparts [1], it is possible to connect different types of autonomous UAVs to dynamically and collaboratively reduce their exposure to threats. When autonomous UAVs make decisions collaboratively, they compose a collaborative UAV network (CN) [2]. Since the UAVs act collaboratively, the features of formation flying become a key factor for the success of the mission. Formation flying is a technique that is implemented by more than one aircraft. The aircraft keeps its positions against a leader and maintain the formation shape during the flight. Different formations can be used with different objectives, such as maximizing coverage area for surveillance or rescue missions. Regarding offensive missions in hostile scenarios, one of the most used formation flyings is the trail. This formation minimizes the exposure of UAVs, thus improving their stealthiness [2]. Moreover, this formation also reduces the fuel usage during the flight [3]. In this work, we focus on a stealth approach for trail formation proposed in [2]. Their model performance was evaluated based on the minimization of the CN exposure to threats, disregarding how the formation would behave in scenarios where the UAVs can be hit by the enemies. In this sense, we extend this model to allow the UAVs, which are exposed to enemy radars, to evade the threat regions, thus reducing their overall time exposure to enemies and improving their chance to regroup with the other UAVs. Additionally, considering that the trail formation is an effective formation for attack missions and that UAVs may be hit during missions, we analyze the impact of lethal threats on the mission accomplishment by verifying the fraction of UAVs hit during the flight. Regarding lethal threats, we consider that our scenario contains ground-based air defense systems [4]. The remaining of this paper is organized as follows. Section 2 presents the model proposed in [2]. Section 3 details the proposed approach for UAVs to leave threat regions. Section 4 discusses the hitting probability of a UAV. Experiments are presented in Sect. 5 and the results are discussed in Sect. 6. Finally, Sect. 7 concludes the paper and outlines future research steps.

2 Collaborative UAV Networks In this section, we describe the general aspects of collaborative UAV networks using the model proposed in [2] as a reference. Thus, our scenarios are composed of UAVs, all flying at the same height, and threats. For convenience, we will refer to this model as reference model.

Evaluating Trail Formation in Collaborative UAV Networks …

99

Fig. 1 An instance of the proposed scenario

Threats are represented by circles, defined by their center position (physical radar position) and detection range. Considering that all enemy radars monitor critical locations, we extended the scenario of the reference model to make the threats lethal, that is, whenever a radar detects a UAV it automatically triggers the weapon systems that aim to destroy it. Notice that this does not imply that the UAV will necessarily be hit by the enemy. The likelihood of a UAV being hit by the weapon is addressed further. As in the reference model, UAVs flight from a start to a goal position using a collaborative approach to detect and avoid threats through a stealth policy. Similarly to [2], we adopt a simulation model over a time interval t = [t0 . . . tmax ]. However, considering that the proposed scenario is composed of lethal threats, the simulation for a single UAV ends when it reaches the goal position, is destroyed by the enemy or due to the maximum simulation time. If the simulation of all UAVs ends before tmax , the simulation also ends. A typical scenario consisting of UAVs performing collaborative tasks in hostile environments is presented in Fig. 1 with threats depicted by red circles.

2.1 Communication Model Considering that the communication between UAVs is commonly relying on highfrequency, very high-frequency or ultra-high frequency radars, it is assumed that each UAV is equipped with one of these devices [5]. In this sense, a UAV can communicate with others that are within its communication radius, establishing a network that can be modeled as an undirected graph G. Regarding the graph, each UAV is a vertex and each communication link between two UAVs is an edge.

100

N. P. Borges et al.

2.2 Threat Detection UAVs should be able to detect threats in flight. Therefore, as in the preliminary model, it is assumed that each UAV is equipped with a passive receiver intended to detect and identify enemy radar signals, named radar warning receiver (RWR). In addition to threat detection, it can also estimate the position of the radar [6]. Our model assumes that a threat is considered detected by a UAV if its range overlaps the RWR range of the UAV. Accordingly, if a UAV detects the presence of a threat, the threat position and detection range are precisely known.

2.3 Stealth Policy In military missions, surveillance radars are expected to monitor strategic spots. Thus, they are often paired with passive radars to enhance their detection capabilities. In this regard, when a UAV is exposed to a surveillance radar, its communication with neighbors can also be exposed to a passive radar, which may compromise the entire mission. In such a scenario, for stealth purposes, the UAV shuts off its communication radar while its RWR identifies a threat and connects it again when no threat is detected by the RWR. This approach, called stealth dissemination, was based on the Try-andBounce communication protocol described in [7]. Notice that the RWR is a passive receiver, so it does not have to be disabled.

2.4 Formation Flying According to [8], in a typical formation flying there are the roles of leader and wingmen. In the context of the trail formation, the wingmen try to follow the trajectory of the leader, taking the other aircraft as a reference to keep its own position in the formation. The leader defines the trajectory to the goal. For generating the trail formation, firstly, it is computed the path of all UAVs to the goal position. The UAV that is chosen to be the leader of a connected component of the network is the UAV that has the shortest path to the goal. The other UAVs’ ranking is generated based on the length of their path to the leader: The closer a UAV is to the leader, the lower is its ranking index, i.e., the higher is its ranking position.1 The formation ranking, however, cannot be static because the connection between two UAVs may be lost. This can occur because both the distance between the UAVs is greater than its communication range or a UAV shuts off its communication radar. Hence, it is essential for the formation flying to detect these situations so that UAVs can perform actions to reestablish the formation as quickly as possible [8]. 1 The

rank of the leader is 1.

Evaluating Trail Formation in Collaborative UAV Networks …

101

Another scenario where it is necessary to update the formation flying is when there are two connected components close enough to be merged. For updating a formation flying, the number of connected UAVs within a formation ranked as 1, i.e., the number of leaders, is verified. This is necessary because there can only be one leader in trail formation. In case of more than one leader, the one with the shortest path to the goal is designated as the leader. We choose as leader the UAV with the shortest path to the goal instead of the UAV that is actually closest to the goal because the scenario contains threats that must be avoided, and it is not possible to ensure that the UAV that is closest to the goal will reach the goal earlier [2]. In the context of this work, the formation flying is responsible for defining the goal position of each UAV. Once the goal position is defined, the next step consists in computing the path planning.

2.5 Path Planning There are several path planning algorithms for UAVs and, among those that consider obstacle or threat avoidance, one of the most frequently referenced in the literature is the Rapidly Exploring Random Tree (RRT) [9]. This algorithm is based on the construction of a tree of feasible trajectories through randomly generated intermediate points. The branches are created as long as they do not pass through any threat regions. One limitation of the RRT for UAV applications, especially for fixed-wings, is that there is no restriction on invalid movements to be performed. In this direction, it is possible to adapt the algorithm for only generating branches in the tree if the angle between the new intermediate goal point and the nearest node in the tree is in a specific range. Notice that the output of the RRT is a path composed of consecutive lines. These lines can be processed to reduce the length of the resulting path. Since the path of all UAVs is computed and reduced, they must fly according to a control law, referred here as displacement model [2].

2.6 Displacement Model Given the path of each UAV, defined by the path planning, it is possible to define the control law responsible for implementing the UAVs movements. Consider that each UAV is modeled as a single integrator system and its velocity can be directly controlled, namely: (1) p˙ i = u i , where u i ∈ Rm is a control input that must be set so that a global objective can be achieved.

102

N. P. Borges et al.

The objective of this control law is to drive the UAV i to follow a path avoiding threat regions. Thus, consider the path Pathγ (i) of the UAV i, which is located at the position pi at a moment t. By using interpolation, it is possible to estimate the position xγi in which the UAV should be at a moment t + 1, by following Pathγ (i). Thus, the control law that drives the UAV i to follow the path is defined by: x i − pi

u i = xγi − p  β (t) , i γ

(2)

where β (t) ∈ R is the linear velocity of the UAV. For performance purposes, the path planning algorithm does not have to be executed at each iteration t. For instance, considering a scenario where the UAV i does not shut down its communication radar and that the straight segment that passes through pi and xγi does not intercept any threat regions, instead of running RRT again it is possible to update the previously computed path. The case in which the UAV is inside a threat region is analogous. However, when the UAV turns on the communication radar again the path needs to be recalculated.

3 Leaving the Threat Region The stealth policy of disabling the communications radar, described in the preliminary model, considers that a UAV in a threat region cannot communicate with other UAVs. Besides, it should still act independently of the network and outline a path to leave the threat area toward the goal position. The reference model considers that a UAV that has entered a threat region will trace a path to the goal position and will fly until it leaves the threat region. We extend this approach, considering different strategies, described as follows. Regarding strategies to leave the threat region, one of the most intuitive is to leave it as quickly as possible. However, the collaboration between UAVs is crucial for stealthiness [2]. Therefore, it is also important providing UAVs with means that allow them to trace a path that improves their chance of regrouping with the others UAVs in the network. The procedure consists in generating candidate points that allow the UAV that is exposed to a threat to leave the threat region as quickly as possible. For this, it is assumed that a UAV has two candidate points to use as a reference, named here as Pi and Pi . Given that, the linear regression of the last position of the UAV L Pr oj and its current position Pi are computed. The candidate points Pi and Pi are the first points that are outside the threat region, considering the line L Per p which is perpendicular to L Pr oj and pass through the center of the threat. The procedure for computing a goal for the UAV in a threat region assumes that each UAV knows its own position Pi and its neighbor positions. Thus, the next candidate point Pi results from the estimation of future positions of connected UAVs. Considering that all wingmen follow the leader in the trail formation, it is possible

Evaluating Trail Formation in Collaborative UAV Networks …

103

Fig. 2 Candidate points for computing path planning to leave threat region

to estimate the future position of UAVs by computing the path of a UAV’s successor in formation. The UAV that is exposed to a threat computes the path assuming its successor’s position in the formation as its current position. Moreover, it is assumed that its successor knows about the threat. The candidate point Pi is the point in the successor path that is closest to Pi or Pi . Figure 2 illustrates a scenario in which this procedure is applied. Give the three candidate points, the last step consists in selecting which one will be the goal position to leave the threat area. Considering that the collaboration is crucial for CN, in case that the length of the path between the current UAV position Pi and Pi is lower than a threshold, Pi will be selected as the goal. Otherwise, the goal will be the candidate point (Pi or Pi ) which results in the shortest path length from Pi .

4 Hitting Probability For computing the probability of a UAV being hit by a threat, several factors must be taken into account. For instance, the position and characteristics of both the UAV and the weapon, as well as the weather conditions must be considered. A straightforward way to evaluate the hitting probability of a UAV is addressed in [10]. However, it is based only on computed path. We adapted it for computing the probability of a UAV being hit considering only its current position Pi , as follows:

104

N. P. Borges et al.

 Pweapon (Pi ) =

0 if dw > rweapon Dw2 /dw2 otherwise.

(3)

where Dw is the weapon attack range, dw is the distance between Pi and the weapon and rweapon is the hitting radius of the weapon. Notice that as the UAV approaches to a weapon its probability of being hit increases. We assume that the position of the weapon and the surveillance radar is the same. Besides, for validation purposes, we consider that a UAV that is inside a threat region can be hit by it, i.e., rweapon is the same value of the detection range of the corresponding surveillance radar. In addition, once a UAV is hit it cannot proceed the mission, i.e., the UAV is removed from the CN.

5 Experimental Setup The proposed model was evaluated through simulations, using MATLAB® R2018a. The UAVs fly from a start to a goal point in different scenarios. Each scenario is composed of a start and a goal position—same position for all the scenarios— approximately 17 km apart, as well as threats and UAVs, and last at most 3,000 s, with a 1 second time step. Three sets of scenarios were considered: low, medium and high hostility, with 2, 8 and 16 threats, respectively. The location of each threat in each scenario was randomly generated. Each UAV possesses a communication and RWR ranges of 200 m, while threats have 700 m of detection range and hitting radius. The model performance was evaluated considering networks with 10 UAVs. The initial network topology is always the same, since the variation of the threat displacement is more relevant to the scenario than the UAVs start positioning. Moreover, it is assumed that the trail formation starts established. For model validation, we evaluate the fraction of UAVs that have been hit during the simulation. This metric assumes values in the range [0, 1]. The closer the value is to 1, the higher the fraction of UAV hit during the mission. With regard to the limited weapon attack range distribution (Eq. 3), let N t be the total number of threats, [ri, r f ] be the possible values interval for the weapon range, and [ri  , r f  ] be a subset of [ri, r f ]. Moreover, let f d 2 be the fraction of threats that will have weapon range in the interval [ri  , r f  ]. For instance, setting N t = 20, ri = 100, r f = 600, ri  = 100, r f  = 250 and f d = 0.8, means that 16 threats will have weapon range between 100 and 250 ((80% of 20) and 4 threats will have weapon range between 250 and 600 (20% of 20)). For properly evaluating the impact of different weapon range distributions on the model performance, we predefined weapon ranges in three intervals: (100, 250] (ψ = 1), (250, 400] (ψ = 2) and (400, 550] (ψ = 3). The results were confronted with a uniform weapon range distribution (ψ = 0). 2 The

value of f d adopted in this work was 0.8.

Evaluating Trail Formation in Collaborative UAV Networks …

105

Simulations were carried out for 100 scenarios with 10 UAVs, for each hostility level and weapon range distribution. The results are then the average of these scenarios. Regarding the parameters of the CN, simulations were performed considering the stealth approach active. For convenience, we adopted ρ = 1 when this approach is active and ρ = 0 when it is inactive. For ρ = 0, the UAVs fly directly to the goal position without the threat detection and avoidance policy. Moreover, we performed simulations with and without the approach for leaving threat regions. For convenience, we adopted φ = 1 when this approach is active and φ = 0, when it is inactive. Thus, we investigate 3 combinations of parameters: (ρ = 0, φ = 0), (ρ = 0, φ = 1) and (ρ = 1, φ = 1). Notice that the combination of parameters (ρ = 1, φ = 0) corresponds to the reference model, while the combination (ρ = 1, φ = 1) simulates the extension of this model.

6 Results and Discussion This section presents and discusses the results of the experiments. The reference model results show that the exposure of UAVs reduces when the stealth policy is active [2]. Once we introduced lethal threats in the scenario, the fraction of UAVs that have been hit during the mission significantly reduces when the stealth policy is active, as shown in Fig. 3. This occurs because the exposure of UAVs to threats also reduces, as described in the preliminary model.

Fraction of UAVs hit ( =1) =0 =0 =1 =0 =1 =1

0.8 0.6 0.4 0.2 0

Low

Medium

High

Fraction of UAVs hit

Fraction of UAVs hit

Fraction of UAVs hit ( =0)

1

1

0.6 0.4 0.2 0

Medium

Hostility level of scenarios

(a) Results with ψ = 0

(b) Results with ψ = 1

Fraction of UAVs hit ( =2) =0 =0 =1 =0 =1 =1

0.8 0.6 0.4 0.2 Low

Medium

Hostility level of scenarios

(c) Results with ψ = 2

High

Fraction of UAVs hit ( =3)

High

Fraction of UAVs hit

Fraction of UAVs hit

Low

Hostility level of scenarios

1

0

=0 =0 =1 =0 =1 =1

0.8

1

=0 =0 =1 =0 =1 =1

0.8 0.6 0.4 0.2 0

Low

Medium

High

Hostility level of scenarios

(d) Results with ψ = 3

Fig. 3 Fraction of UAVs hit, varying the weapon range distribution

106

N. P. Borges et al.

Notice that as the hostility level of the scenario increases, the fraction of UAVs hit also increases, especially when the stealth policy is inactive (ρ = 0). In addition, for this case, the weapon range distribution does not result in significant variation of UAVs hit. The first reason for this relies on the fact that the fraction of UAVs hit, for the medium and high hostility level, is already close to 1. In the case of low hostility level, the variation was insignificant because the UAVs can be exposed, at most, to only 2 threats during the mission. When the stealth policy is active (ρ = 1), the weapon range distribution significantly affects the hitting probability of UAVs. The fraction of UAVs increases when the values of ψ were larger. For the case in which the weapon range distribution was uniform (ψ = 0), the fraction of UAVs hit was closer to the case in which weapon range distribution was intermediary, i.e., ψ = 2. The approach for leaving threat areas proposed in this paper proved to be a key factor regarding the hitting probability of UAVs. The fraction of UAVs hit by threats was always lower when this approach was active (ρ = 1 and φ = 1). The importance of this approach becomes more evident as long as the values of weapon range distribution and hostility level of the scenario increases. For instance, considering the high hostility level and ψ = 3, the fraction of UAVs hit decreases in approximately 300%, when this approach is active, in comparison with the approach ρ = 1 and φ = 0. This occurs for three reasons. The first is because the overall time of UAVs exposure to threats reduces. The second is due to the fact that our approach makes the UAV avoid the center of the threat, where the hitting probability is larger. The last one is because it increases the probability of the UAV that turned off the radar reconnecting to the network, increasing the stealth level of the network; consequently, reducing the number of UAV exposed.

7 Conclusions and Future Work In this paper, we presented an extended model to improve stealth in collaborative UAV networks under local communication constraints, using trail formation flying. We proposed an approach to evade threat regions, when a UAV is exposed to an enemy radar. Moreover, considering that the trail formation is an effective formation for attack missions, and that UAVs may be hit during missions, we also conducted an analysis of the impact of hit UAVs on the CN. The approach was validated through simulations, varying the hostility of the scenario and the weapon attack range of threats. The fraction of UAVs hit decreases when the stealth procedure is active. Also, it was shown the importance of the approach for leaving threat regions, when a UAV is exposed to a threat. The proposed approach provides promising results, considering the fraction of UAVs hit by threats was significantly lower when this approach was active. The proposed model does not consider the maximum operation time of the UAVs to perform a mission. Thus, the path length calculated by the UAVs may be impractical for real applications. In this context, we propose as future work modeling

Evaluating Trail Formation in Collaborative UAV Networks …

107

the maximum operation time of the UAVs during the computation of the path plan. Moreover, we are assuming that the threats position and ranges are precisely estimated. Evaluating the impact of the model for the cases which these are not correctly estimated can be exploited. Finally, 3D modeling of the scenario will allow exploring other stealth approaches considering control of yaw and pitch angles [10]. Acknowledgements The authors thank CAPES (Social Demands Program) and FAPESP (proc. no. 2017/02055-8) for the financial support to carry out this research.

References 1. Bekmezci, I., Sahingoz, O.K., Temel, S.: ¸ Flying ad-hoc networks (FANETs): a survey. Ad Hoc Netw. 11(3), 1254–1270 (2013) 2. Borges, N.P., Ghedini, C.G., Ribeiro, C.H.C.: A local communication model for improving stealth in collaborative UAV networks. An. do Comput. Beach 249–258 (2017) 3. Cetin, O., Yilmaz, G.: Real-time autonomous UAV formation flight with collision and obstacle avoidance in unknown environment. J. Intell. Robot. Syst. 84(1–4), 415–433 (2016) 4. Wållberg, K.: Method of fire control for gun-based anti-aircraft defence (2017). US Patent 9,625,236 5. Adamy, D.: EW 102: A Second Course in Electronic Warfare. Artech House, Boston (2004) 6. Gee, W.A.: Radar warning receiver (RWR) time-coincident pulse data extraction and processing. In: 2012 IEEE Radar Conference, pp. 0752–0757 (2012). https://doi.org/10.1109/RADAR. 2012.6212238 7. Turgut, D., Turgut, B., Bölöni, L.: Stealthy dissemination in intruder tracking sensor networks. In: 2009 IEEE 34th Conference on Local Computer Networks, pp. 22–29 (2009). https://doi. org/10.1109/LCN.2009.5355150 8. Giulietti, F., Pollini, L., Innocenti, M.: Autonomous formation flight. IEEE Control Syst. 20(6), 34–44 (2000). https://doi.org/10.1109/37.887447 9. LaValle, S.M.: Rapidly-exploring random trees: a new tool for path planning (1998) 10. He, P., Dai, S.: Stealth coverage multi-path corridors planning for UAV fleet. In: Proceedings of the 2013 International Conference on Mechatronic Sciences, Electric Engineering and Computer (MEC), pp. 2922–2926. IEEE (2013)

Visual Analytics for the Reduction of Air Pollution on Real-Time Data Derived from WSN Dorys Quiroz, Byron Guanochanga, Walter Fuertes, Diego Benítez, Jenny Torres, Freddy Tapia and Theofilos Toulkkeridis

Abstract Nowadays, the contaminated and poor air quality that is inhaled by the city population in industrialized cities around the world has led to one of the main causes of premature death due to respiratory diseases. Therefore, the improvement of air quality becomes a priority. In this sense, the current study aimed to design and implement a visual analytics tool, in order to process large data sets, which have been generated by wireless sensor networks (WSN), which measured different environmental pollutants in real time. Hereby, the phases of the CRISP-DM methodology have been applied as a reference to guide the process. In the data preparation phase, programs have been implemented using Python. Then, the results have been stored in collections within a MongoDB database. Furthermore, for the modeling and visual exploration of the data, the Tableau tool has been used. The evaluation of the results allowed to demonstrate certain behavior of air pollutants around the city, such as the D. Quiroz · B. Guanochanga · W. Fuertes (B) · F. Tapia · T. Toulkkeridis Departamento de Ciencias de la Computación, Universidad de las Fuerzas Armadas ESPE, 171-5-231-B Sangolquí, Ecuador e-mail: [email protected] D. Quiroz e-mail: [email protected] B. Guanochanga e-mail: [email protected] F. Tapia e-mail: [email protected] T. Toulkkeridis e-mail: [email protected] D. Benítez Colegio de Ciencias e Ingenierías El Politécnico, Universidad San Francisco de Quito USFQ, Campus Cumbayá, Casilla Postal, 17-1200-841 Quito, Ecuador e-mail: [email protected] J. Torres Departamento de Informática y Ciencias de la Computación, Escuela Politécnica Nacional, 17-01-2759, Quito, Ecuador e-mail: [email protected] © Springer Nature Singapore Pte Ltd. 2020 Á. Rocha and R. P. Pereira (eds.), Developments and Advances in Defense and Security, Smart Innovation, Systems and Technologies 152, https://doi.org/10.1007/978-981-13-9155-2_10

109

110

D. Quiroz et al.

increased air pollution levels during daylight hours. Similarly, we discovered that the presence of particulate material PM10 and PM2.5 is directly related to the increase of the Air Quality Index for the city of Quito (IQCA). This leads to the conclusion that our analysis may be useful as a support tool in the decision-making of public policies for the reduction of air pollution. Keywords Air pollution · Data mining · Wireless sensor networks · Visual analytics

1 Introduction Air pollution is caused by the presence of chemical substances or toxic compounds derived mainly by a variety of anthropogenic activities that trigger adverse effects on human well-being and the environment [1]. The presence of these pollutants decreases air quality and harms the health and quality of life of people [2]. Due to this, it is necessary to provide the corresponding agencies with tools that facilitate finding solutions that improve air quality. Therefore, both the industry and the scientific community have been developing several solutions using wireless sensor networks (WSN) in order to reduce the aforementioned negative effects. However, most of these implementations have focused on the design of the electronic circuits, the assembly of electronic sensor boards, the implementation of software bridges to store data and, finally, the development of interfaces, which allow to know the measurements. Nonetheless, all these efforts become unproductive, if the final users or those responsible for this issues are not provided with tools that benefit the descriptive analysis, in real time, depending on what the WSN determines. The aim of the current study was to analyze the information derived from a WSN, which we have assembled in order to determine air pollution. Hereby we used data mining (DM) and visual analytics techniques. The large volumes of data obtained from the WSN are stored in real time in a non-relational database. Sensor networks were deployed in four sectors of the cities of Quito and Sangolquí, within the province of Pichincha in north-central Ecuador. The CRISP-DM methodology has been used in order to guide the DM process. In the data preparation phase, programs were implemented in Python. Subsequently, the results have been stored in collections in a new database. For the modeling and visual exploration of the data, we used the Tableau tool. The main contribution of this study is the generation of a visual analytical tool that benefits the descriptive analysis to decision-making support and as an important reference for the generation of public policies within the control of air pollution. The results obtained demonstrate that the presence of particulate material PM10 and PM2.5 increases the IQCA determination. Furthermore, during daylight rush hours, with the highest density of vehicular traffic, we observed a significant increase in the level of air pollution. This certainly has an impact on people’s health leading directly to an increase of respiratory diseases such as asthma, bronchitis [2], among others.

Visual Analytics for the Reduction of Air Pollution …

111

The remainder of the article is structured as follows: Sect. 2 presents related works. Section 3 shows an insight of the conducted research process, while in Sect. 4 we explain the results obtained and the validation of this study. Section 5 finalizes with the obtained conclusions.

2 Related Work With the implementation of WSNs that monitor the concentration of air pollutants in real time, hundreds of thousands of data have been available that will continue to increase over time. Therefore, the design of visualization tools that facilitate the interactive analysis of these data becomes indispensable. Thus, Liao in [3] developed a visual analytics system for monitoring air quality data called AirVIS for China. This system has the availability of allowing three views: GIS for spatial analysis of data, scatter plots for temporal analysis, and a parallel coordinated view used for multidimensional analysis. Similarly, [4] presents a system that uses visual analysis for the study of air pollution in Honk Kong. The authors in such work propose several visualizations that allow tasks such as the detection of correlations of variables, comparative data, identification of trends, and predictions about climate and air quality. In [5], a semi-supervised learning approach was used to generate visualizations that allow researchers to model the spatial correlation of the air quality of different locations according to meteorological data, traffic flow, human mobility, structure of road networks, and points of interest. In [6], authors designed a system for visual analysis of air pollution data. They added cluster concepts, transforming the data into 2D plots views, allowing them to relate geographic areas, and identifying patterns among them. They also used the Voronoi diagram to allow users to iterate with the spatial clusters. Currently, they provide access to a large quantity of multidimensional spatiotemporal data and greater precision generated by environmental monitoring sensors. This has generated the need to create visualizations available to users with a more comprehensive and interactive approach. To facilitate the analysis in [7], a detailed study focused on the use of visual methods based on maps, calendars, and views of trends to identify lines and patterns was presented. Within this same context in [8], the identification of correlations between different regions where environmental pollution is determined was also presented. In order to allow such approach, meteorological data (wind and its direction) were used to calculate the lead/lag correlation. The Pearson correlation coefficient was used to determine the pollution ratio between the different regions monitored along different time spaces. The attempt toward the implementation of smart cities based on (IoT) has been described in [9], where an ecosystem of management and analysis of data was constructed, which ranges from processing, integration, analysis up to visualization. Also, in [10], an analysis of the air pollutants in Bahrain was conducted using the robust and nonparametric Theil–Sen method in order to study temporary quantitative

112

D. Quiroz et al.

variations of PM10 and PM2.5 particles, NO2 , SO2 , and O3 , as well as trends and effects of temperature, humidity, and wind. The correlation analysis between different parameters of air quality determination is the center of the study reported in [11], which demonstrates how to analyze distributed data in spatial temporal contexts. Additionally, the authors proposed a visualization tool using a Time Correlation Partition Tree (TCP). Similarly, in [12], authors presented an interesting approach in visual analytics in order to detect patterns in numerical time series, specifically environmental sciences. Furthermore, a guide for the correct design of large data visualizations has been analyzed in [13]. This is a fundamental guide for the visual management of different types of data. Finally, in [14], large stored volumes of fixed data derived in real time from our WSN have been analyzed using business intelligence (BI) and data mining (DM) techniques. All previous studies described in this section demonstrate the importance of providing managers with some descriptive analysis tools for decision-making support. Therefore, our project has been designed to provide visual analytics as an input in the improvement of public policies in this context.

3 Research Design This section describes the process conducted based on CRISP-DM, which has been chosen to provide a standardized description of the life cycle of a data analysis project, which is analogous to the way used in software engineering with software development life cycle models [15]. The CRISP-DM model covers the project phases, their respective tasks, and the relationships between these tasks. Additionally, CRISPDM considers the data analysis process as a professional project, thus establishing a much broader context that influences the development of the models. Descriptions of the phases executed are described below.

3.1 Business Understanding The air pollution study is important because of the possible negative effects on peoples health [1]. Therefore, the World Health Organization (WHO) has designed several air quality guidelines, which set the guidelines on the values of concentration of pollutants and their permissible limits [16]. The Metropolitan District of Quito, from the Ministry of the Environment, has implemented the Environmental Monitoring Network, establishing the Index of Air Quality for the city of Quito (IQCA), which is based on the Ecuadorian Standard of Air Quality [17, 18]. The established ranges to determine pollution levels are shown in Table 1. Based on these guidelines, the information to be used was identified and it was defined how this problem will be addressed through data mining.

Visual Analytics for the Reduction of Air Pollution …

113

Table 1 Index of Air Quality for the city of Quito (IQCA) [17] Ranges Condition Condition from the health point of view 0–50 50–100 100–200 200–300 300–400 400–500

Desirable

Air quality is considered satisfactory and environmental pollution has little or no risk to health Acceptable Air quality is acceptable. However, there may be minor effects on health for individuals highly sensitive to environmental pollution Caution Not healthy for individuals (chronic and convalescent) Alarm Not healthy for the majority of the population Alert Not healthy for the majority of the population and dangerous for sensitive individuals Emergency Dangerous for the entire population

Fig. 1 Example of a message (frame) in JSON format

3.2 Data Understanding The data collection has been conducted automatically. The sensor nodes send the determined information to each gateway node. The latter then sends the data to the application server in the cloud, where they are stored in a non-relational database such as MongoDB. Each message sent from the sensor nodes is composed by the concentration measurements of each gas, as well as for the particulate material concentrations, geographical location, time of measurement, temperature, pressure, humidity, and identifier of each sensor node. Figure 1, illustrates the JSON format of the frame that each sensor node sends.

3.3 Data Preparation In this phase, we accomplished an extraction, transformation and loading process (ETL) that included the cleaning, transformation, and storage of the data. The following tasks are specified below: (1) Based on the guidelines on air quality for the city of Quito, the pollutant concentration of gases: CO, SO2 , NO2 , and particulate material: PM2.5 and PM10, have been used in units of micro-grams per cubic meter (ug/m3). (2) The information has been transformed to one half per minute. (3) In the cleaning process, concentration values less than or equal to zero have been eliminated. (4) Ranges have been verified, and erroneous, meaningless, and duplicate values have been identified and corrected. (5) Fields with the air quality index for

114

D. Quiroz et al.

Fig. 2 Schematic architecture of the descriptive model

each gas have been added. Then, the total air quality index has been obtained. For these tasks, several programs for data cleaning, transformation and subsequent storage in several collections in MongoDB, have been implemented in Python.

3.4 Modeling Figure 2 illustrates the architecture of the descriptive model of the data analysis process conducted on air pollution data and meteorological variables. The model has been composed of three individual layers: (1) The data source layer. This is the information from the WSN, which has been stored in a central database. In the same way, the information derived from external sources that have been used in the preparation process. (2) The storage layer, composed of the base resulting from the ETL process, where the data has been purified, transformed, treated, and then stored in different collections of a database. Finally, (3) the visualization layer, for which the Tableau tool has been applied in order to visualize and understand the data through the generation of dashboards and interactive graphics, correlation and trend among the studied variables.

3.5 Visual Modeling With the execution of the tasks indicated in the data preparation, the obtained information was stored in collections in a database. This allowed the task of data analytics through the preparation of visualizations in order to study the behavior of pollutants measured in the sensed sectors. In order to provide important information about these data, a view model has been designed that follows the scheme shown in Fig. 3. The

Visual Analytics for the Reduction of Air Pollution …

115

Fig. 3 Elements and interrelations of visual modeling

used design criteria were the bar graphs that facilitate the comparison between the different values of the pollutants. The visualization of the behavior of the pollutants has been designed using time series in which trend lines were included. The identification of patterns is fundamental within this analysis; therefore, we have used dispersion graphs. Additionally, we have included correlation matrices in order to detect direct or inverse linear relationships between the different parameters; thus, we calculated the Pearson correlation. Furthermore, dashboards have been designed using the Tableau visualization tool. In each of them, the behavior of air pollutants in the monitored sectors has been reflected. Figure 4, for example, illustrates the first dashboard, where the behavior of gases in each of the concentrators is presented based on their averaged index. Specifically, it is the general visualization per concentrator and pollutant. From this panel, users may select a concentrator and/or a contaminant to see the details in the two remaining panels. The second dashboard, illustrated in Fig. 5, presents the categories in which the determinations have been framed. These categories have been those defined in the regulations on the IQCA. This consist on IQCA by concentrator: It demonstrates the index that has been calculated from the IQCA for each concentrator and the category according to the Quito standard. The identification of patterns and correlations is one of the most important points in visual analysis. Therefore, one visualization was created, which indicates the Pearson correlation values between the IQCA and the presence of pollutants (see Fig. 6).

4 Evaluation of Results The dashboard of Fig. 4, allows to analyze the behavior of the pollutants in the Alameda Concentrator. As it can be seen, the bar graph allows to identify that PM2.5 has the highest concentration. The dashboard of Fig. 5 shows the panel with the heat

116

D. Quiroz et al.

Fig. 4 General view of determinations per concentrator and pollutant

Fig. 5 The weekly behavior of the IQCA

maps, which allows, for example, to identify that the last week of October in the Villa Flora Concentrator the IQCA went from DESIRABLE to ACCEPTABLE. During the months of July, August, and September, in the sensors of Villa Flora and ESPE nodes, a high influence of NO2 on the IQCA points out, while in Alameda node a low correlation has been observed even being inverse due to its negative coefficient. While, for San Diego node, it lacks of a clear conclusion, since it does not provide of any significant correlation.

Visual Analytics for the Reduction of Air Pollution …

117

Fig. 6 Correlation of the IQCA versus contaminants

Nonetheless, the concentration of CO and SO2 in the concentrators of Villaflora and ESPE has a high correlation with the IQCA in the first months of the monitoring. Since, for November this correlation is stabilized although its value is not as high as PM10 and PM2.5, which has a certain degree of influence. For the La Alameda concentrator, the CO does not have a noteworthy increase over time, while SO2 marks a growing trend. At the San Diego concentrator, the data available for November indicate that the correlation of both gases with the IQCA is insignificant. All these analyses lead to the conclusion that PM10 and PM2.5 particulate matter are the pollutants that mostly influence the increase of the IQCA values. Of all the correlations analyzed, those of PM10 and PM2.5 have been the most significant and are evidenced in the four concentrators. This most likely means, that with any PM10 increase, the presence of PM2.5 also increases. Another significant correlation that was visualized in the first three months of monitoring in Villa Flora and ESPE concentrators is that of NO2 and SO2 . However, as of October, this correlation drops substantially. However, in Alameda concentrator, the given correlation is inverse. That is, the greater the presence of NO2 , the lower the presence of SO2 and vice versa. While in San Diego concentrator, the correlation for the month of November behaves as in Villa Flora and ESPE nodes, although there are no data for the month of December with which this correlation could have been corroborated. A high correlation between CO and SO2 occurs only in the ESPE concentrator, while in the other concentrators it remains to be insignificant.

118

D. Quiroz et al.

5 Conclusions This study has been conducted with the purpose of analyzing the determined data of air pollutants collected by a WSN in real time. The CRISP-DM methodology was followed for the DM process. In the preparation phase, an ETL process has been operated, in which data were cleaned, treated, and transformed, the results of this process were subsequently stored in collections within a non-relational database. For the analysis and visualization of the obtained data, we applied the Tableau tool. Views were generated, allowing pollutant behavior analysis as well as the correlation and patterns analysis of pollutants at every minute in the sensed sectors. Therefore, the descriptive analysis may be used as a significant decision-making support tool. This study may be used as an aid for the generation of public policies that certainly will better address environmental problems. Acknowledgements The authors would like to thank the financial support of the Ecuadorian Corporation for the Development of Research and the Academy (RED CEDIA) in the development of this study, within the Project Grant CEPRA-XI-2017-13.

References 1. Kampa, M., Castanas, E.: Human health effects of air pollution. Environ. Pollut. 151(2), 362– 367 (2008) 2. WHO: Ambient air pollution: A global assessment of exposure and burden of disease. World Health Organization, Geneva, Switzerland (2016) 3. Liao, Z., Peng, Y., Li, Y., Liang, X., Zhao, Y.: A web-based visual analytics system for air quality monitoring data. In: 2014 22nd International Conference on Geoinformatics (GeoInformatics), pp. 1–6. IEEE (2014) 4. Qu, H., Chan, W.Y., Xu, A., Chung, K.L., Lau, K.H., Guo, P.: Visual analysis of the air pollution problem in hong kong. IEEE Trans. Vis. Comput. Graph. 13(6), 1408–1415 (2007) 5. Zheng, Y., Liu, F., Hsieh, H.: U-air: When urban air quality inference meets big data. In: Proceedings of the 19th ACM SIGKDD International Conference on Knowledge Discovery and Data Mining, pp. 1436–1444. ACM (2013) 6. Zhou, Z., Ye, Z., Liu, Y., Liu, F., Tao, Y., Su, W.: Visual analytics for spatial clusters of airquality data. IEEE Comput. Graph. Appl. 37(5), 98–105 (2017) 7. Du, Y., Ma, C., Wu, C., Xu, X., Guo, Y., Zhou, Y., Li, J.: A visual analytics approach for station-based air quality data. Sensors 17(1), 30 (2016) 8. Du, Y., Abish, M., Zhou, L., Zhou, Y.: A correlation visual analytics system for air quality. Chin. J. Electron. 27(5), 920–926 (2018) 9. Ahlers, D., Kraemer, F., Braten, A., Liu, X., Anthonisen, F., Driscoll, P., Krogstie, J.: Analysis and visualization of urban emission measurements in smart cities. In: Proceedings of the 21st International Conference on Extending Database Technology (EDBT) (2018) 10. Jassim, M.S., Coskuner, G., Munir, S.: Temporal analysis of air pollution and its relationship with meteorological parameters in bahrain, 2006–2012. Arab. J. Geosci. 11(3), 62 (2018) 11. Fangzhou Guo Tianlong Gu, W.C.F.W.Q.W.L.S., Qu, H.: Visual exploration of air quality data with a time-correlation partitioning tree based on information theory. ACM Trans. Interact. Intell. Syst. 1(1) (2018). https://doi.org/10.1145/3182187.

Visual Analytics for the Reduction of Air Pollution …

119

12. Sips, M., Kothur, P., Unger, A., Hege, H.C., Dransch, D.: A visual analytics approach to multiscale exploration of environmental time series. IEEE Trans. Vis. Comput. Graph. 12, 2899–2907 (2012) 13. Keim, D., Andrienko, G., Fekete, J.D., Görg, C., Kohlhammer, J., Melançon, G.: Visual analytics: Definition, process, and challenges. In: Information visualization, pp. 154–175. Springer, Berlin (2008) 14. Fuertes, W., Alyssa, C., Torres, J., Benítez, D.S., Tapia, F., Toulkeridis, T.: Data analytics on real-time air pollution monitoring system derived from a wireless sensor network (2019) 15. Marbán, Ó., Mariscal, G., Segovia, J.: A data mining and knowledge discovery process model. In: Data Mining and Knowledge Discovery in Real Life Applications. InTech (2009) 16. WHO.: Air Quality Guidelines: Global Update 2005. Particulate Matter, Ozone, Nitrogen Dioxide and Sulfur Dioxide. World Health Organization (2006) 17. Secretaría del Ambiente: índice Quiteño de la Calidad del Aire (2004) 18. del Ambiente, Secretaría: Informe de la Calidad del Aire del Distrito Metropolitano de Quito 2017. Tech. rep, Secretaría del Ambiente Alcaldía (2018)

Part III

Defense Engineering

Toward the Development of Surveillance and Reconnaissance Capacity in Ecuador: Geolocation System for Ground Targets Based on an Electro-Optical Sensor Zurita C. Marco Antonio, Aguilar C. Wilbert Geovany and Enríquez C. Victor Xavier Abstract This paper presents the progress of a research work that seeks to determine the geographic coordinates of an object on the land of interest by applying the Denavit–Hartemberg methodology to the surveillance and recognition system developed in the CID FAE SEO-D1; this together with mathematical cartography will provide information in support of military operations. The results shown in the simulation present a good approximation of the geographic coordinates of interest. In light of the above, it is suggested that this kind of study provides a low-cost surveillance and reconnaissance alternative to existing advanced systems such as satellites for countries that do not have this type of technology, as is the case of Ecuador. Keywords Robot’s articulation · Kinematic chain · Denavit–Hartemberg Algorithm · Mathematical cartography · Space projection · Gimbal

1 Introduction The Armed Forces of Ecuador, as part of their doctrine, must guarantee Ecuadorian society the defense of sovereignty and territorial integrity and, in addition, support the integral security of the state [1]. This doctrine also establishes a relationship with the socioeconomic development of the population, a relationship that within its process and conception has had changes due to the appearance of organizations that have caused problems within the country. Therefore, the Ecuadorian state has seen the need to prioritize the border sector, specifically the northern border where illegal activities have been developing, which constitute serious threats to the security of the state [2]. Z. C. Marco Antonio (B) · A. C. Wilbert Geovany University of the Armed Forces-ESPE, Sangolqui, Ecuador e-mail: [email protected] E. C. Victor Xavier Ecuadorian Air Force Research and Development Center, Ambato, Ecuador © Springer Nature Singapore Pte Ltd. 2020 Á. Rocha and R. P. Pereira (eds.), Developments and Advances in Defense and Security, Smart Innovation, Systems and Technologies 152, https://doi.org/10.1007/978-981-13-9155-2_11

123

124

Z. C. Marco Antonio et al.

On the other hand, the strategic planning of the defense is intended to guarantee the security and defense of the state, conceiving from a comprehensive approach a correct decision making in everything related to sovereignty and territorial integrity; generating policies aimed at contributing to the safety and welfare of its citizens, through surveillance and control activities such as land, river and air patrol operations. Therefore, the Research and Development Center of the Ecuadorian Air Force CID FAE is an entity dedicated to the development of projects oriented to security and defense. Among the emblematic projects developed in this research center, there is the surveillance and reconnaissance system. This system arises from the need to capture images in real time for surveillance and reconnaissance applications [3]. The surveillance and reconnaissance system developed at the CID FAE consists of a ground station, the gimbal system and a camera. These sub-systems allow the operation of the gimbal in flight along with the camera and obtaining real-time information of images and parameters such as height, gimbal geographic location, pitch, yaw and roll degrees. The known geographical location (latitude and longitude) corresponds to the gimbal mounted on an aircraft in flight, but we cannot say that these coordinates are those corresponding to the surrounding objects on land. That is to say that we can visualize images on the ground without knowing their geographical location. To solve this problem, it is necessary to know the orientation and position of the lens of the gimbal’s camera in the space; this way we would determine the direction where the camera points and consequently know the location of surrounding objects on land. In robotics, it is a common need to know the orientation and location of the final effector of a robot (a tool normally) from the knowledge of the values taken by the articulations of the robot. This is called the direct kinematic problem [4] (Fig. 1). There are different approaches to solve the direct kinematic problem, but for the geolocation system developed the Denavit–Hartemberg methodology was used. Denavit and Hartenberg proposed in 1955 a matrix method to systematically establish a coordinate system which we will call Si linked to each link i of an articulated chain, and the kinematic equations of the whole chain can then be determined [5] which are the result of the multiplication of the homogeneous transformation matrices. The homogeneous transformation matrix is symbolized by i−1 Ai and represents the relative position and orientation of the system Si with respect to the system Si−1 . That is to say that 0 A1 describes the position and orientation of the solidary reference system to the first link with respect to the solidary reference system to the base.

Fig. 1 Direct kinematic problem. Source Robotics fundamentals [4]

Toward the Development of Surveillance and Reconnaissance … i−1

⎡

125

Ai = T(z, θi )T(0, 0, di )T(ai , 0, 0)T(x, αi )

cos(θi ) − sin(θi ) ⎢ sin(θi ) cos(θi ) i−1 Ai = ⎢ ⎣ 0 0 0 0

0 0 1 0

⎤⎡ 0 1 ⎢0 0⎥ ⎥⎢ 0 ⎦⎣ 0 1 0

0 1 0 0

0 0 1 0

⎤⎡ 0 100 ⎢0 1 0 0⎥ ⎥⎢ di ⎦⎣ 0 0 1 000 1

⎤⎡ 1 0 0 ai ⎢ 0 cos(αi ) − sin(αi ) 0⎥ ⎥⎢ 0 ⎦⎣ 0 sin(αi ) cos(αi ) 0 0 0 1

(1) ⎤ 0 0⎥ ⎥ 0⎦ 1

(2)

To represent the position and orientation of the solidary reference system to the final effector S6 with respect to the solidary reference system to the base S0 , it will only be necessary to multiply the homogeneous transformation matrices of all successive links of the robot. ⎡

nx ⎢ n y T = 0 A6 = 0 A1 1 A2 2 A3 3 A4 4 A5 5 A6 = ⎢ ⎣ nz 0

ox oy oz 0

ax ay az 0

⎤ px py ⎥ ⎥ pz ⎦ 1

(3)

Since the Denavit–Hartemberg algorithm gives us the position and orientation of the final effector in values relatives of the plane X, Y, Z, it is necessary the application of mathematical cartography for the representation of values of latitude (φ) and longitude (λ) of the globe to their respective values passed to a plane and vice versa. Cartography is the science that deals with the drawing of geographical charts. Geographical charts, also called maps, are representations on a plane of all or part of the earth’s surface. In order to make this representation, a certain projection of the aforementioned reference land surface on a plane will be used [6]. To obtain the cartographic representations of a terrestrial zone, it is necessary to know the formulas as that relate the values in the plane with the values of latitude and longitude. x = f 1 (φ, λ)

(4)

y = f 2 (φ, λ)

(5)

To these, formulas must exist a biunivocal correspondence for its inverse case. φ = F1 (x, y)

(6)

λ = F2 (x, y)

(7)

For the calculation of these values, we used the (UTM) Universal Transversal Mercator representation, because it is the most widely used and has less deformations in its representation to the plane. Whose formulas and calculations are detailed in “Bolletino de Geodesia e Science Affini” number 1 presented by Alberto Cotticia and Lucia-no Surace (Fig. 2).

126

Fig. 2 UTM zones in a plane. Source Antonio R. Franco [10]

Fig. 3 UTM zones for Ecuador. Source Geo-ingeniería [11]

Z. C. Marco Antonio et al.

Toward the Development of Surveillance and Reconnaissance …

127

Ecuador is inside four zones (15, 16, 17 and 18), so the formulas applied will have this condition (Fig. 3).

2 Design of the Geolocation System The geolocation system is composed of elements that have the electro-optical sensor such as the altimeter and inertial measurement unit (IMU), the gimbal, the ground station with its HMI screen and algorithms implemented by software. In this paper, we will focus on the algorithms to be implemented by software. First, it is necessary to establish the characteristics of the gimbal on which the Denavit–Hartenberg algorithm will be applied. Since the gimbal is mounted on an aircraft, this will have variations in height, pitch, yaw and roll product of the maneuvers executed in flight; these four variables will be considered as values equivalent to those taken by the joints of a robot (four degrees of freedom). These four degrees of freedom are added to two degrees of freedom corresponding to two servomotors that integrate the gimbal (Fig. 4). With these conditions, we proceed to put all systems associated with each joint in accordance with Denavit–Hartemberg (Table 1).

Fig. 4 Gimbal characteristics. Source Author

128

Z. C. Marco Antonio et al.

Table 1 Systems associated with each articulation S0 (Altimeter)

S1 (Yaw)

S2 (Pitch)

S3 (Roll)

S4 (Servomotor 1)

S5 (Servomotor 2)

S6 (Camera lens)

Table 2 Denavit–Hartemberg parameter table

θzi−1

dzi−1

axi

1

1

0

h1

0

2

2

q1

0

0

−90

3

3

q2 − 90

0

0

−90

−d0

−90 −90

Joins 0A 1A 2A 3A

4

4

q3 − 90

0

4A

5

5

q4 − 90

−d1

0

5A

6

6

q5

0

+d2

αxi 0

0

Once the systems associated with the joints have been established, we can obtain the values for the homogeneous transformation matrices (Table 2). T = 0 A6 ⎡ 10 ⎢0 1 ⎢ =⎣ 00 00

0 0 1 0

⎤⎡ 0 cos(q1 ) ⎢ 0 ⎥ ⎥⎢ sin(q1 ) h 1 ⎦⎣ 0 0 1

0 − sin(q1 ) 0 cos(q1 ) −1 0 0 0

⎤⎡ 0 cos(q2 − 90) ⎢ 0⎥ ⎥⎢ sin(q2 − 90) 0 ⎦⎣ 0 1 0

0 − sin(q2 − 90) 0 cos(q2 − 90) −1 0 0 0

⎤ 0 0⎥ ⎥ 0⎦ 1

(8)

Toward the Development of Surveillance and Reconnaissance … ⎡

cos(q3 − 90) ⎢ ⎢ sin(q3 − 90) ⎢ ⎣ 0 0

⎤⎡ cos(q4 − 90) 0 − sin(q3 − 90) −d0 cos(q3 − 90) ⎥⎢ 0 cos(q3 − 90) −d0 sin(q3 − 90) ⎥⎢ sin(q4 − 90) ⎥⎢ ⎦⎣ −1 0 0 0 0 0 1 0 ⎡ ⎤ cos(q5 ) − sin(q5 ) 0 d2 cos(q5 ) ⎢ ⎥ ⎢ sin(q5 ) cos(q5 ) 0 d2 sin(q5 ) ⎥ ⎢ ⎥ ⎣ 0 ⎦ 0 1 0 0 0 0 1

129 ⎤ 0 − sin(q4 − 90) 0 ⎥ 0 cos(q4 − 90) 0 ⎥ ⎥ −1 0 −d1 ⎦ 0 0 1

The T matrix will allow obtaining the orientation and position of the camera lens S6 with respect to the S0 system. In this way, a straight line in the x-axis (camera lens) of the system S6 can be projected into space using two points that are part of the x-axis (camera lens). The equation of the line in space with the continuous form is given by [8]: y − y0 z − z0 x − x0 = = v1 v2 v3

(9)

v1 = x1 − x0

(10)

v2 = y1 − y0

(11)

v3 = z 1 − z 0

(12)

From this projected line, the x and y values of the plane z = 0 are determined. Which we will call x0utm and y0utm , these are the differential values in X and Y with reference to the origin system S0 (Fig. 5).

Fig. 5 UTM target points. Source Author

130

Z. C. Marco Antonio et al.

As already mentioned, the surveillance and reconnaissance system already provide the geographic coordinates of the gimbal in real time. Then, it will be necessary to pass these coordinates to UTM values through the formulas presented in “Bolletino de Geodesia e Science Affini” to obtain the xutm and xutm values [12]. xtarget = xutm + x0utm

(13)

ytarget = yutm + y0utm

(14)

If it is required to obtain in geographic coordinates, it will be necessary to carry out the conversion of UTM coordinates to geographic coordinates. These values will be appropriate if the coordinates of both the gimbal and the ground target are within the same UTM zone.

3 Simulation and Test In order to carry out the simulations, the conditions in a certain period of time in which the gimbal is in flight are established. For this, it is imposed that the gimbal is up to 1000 m of height, in the coordinates (1.285309, −78.834736) and whose orientation is the result of the variation of q4 and q5 in 45°. With these parameters, the orientation given by the transformation T matrix will be as follows (Fig. 6). The values of the desire parameters are obtained by programing in MATLAB, and the formulas are presented or mentioned in this paper (Table 3). In order to appreciate the results, Google maps [9] are used, where the gimbals and target will be visualized (Fig. 7).

Fig. 6 UTM target points simulation. Source Author

Toward the Development of Surveillance and Reconnaissance … Table 3 Parameter results table

Parameter

Value

Xutm

7.409466249039110e + 05

Yutm

1.421692420845669e + 05

xoutm

7.070360705084230e + 02

youtm

7.070360705084230e + 02

xtarget

7.416536609744194e + 05

ytarget

1.428762781550754e + 05

Target coordinates

(1.291695931165799, − 78.828379873412587)

131

Fig. 7 UTM target points Google maps. Source Google maps [9]

As we can see, the geolocation system of targets on land shows a good approximation of the geographical coordinates of the target pointed by the camera. In order hand to test the system, preliminary experiments were carried out in controlled conditions and in locations with heights of no more than 13 m (Fig. 8).

132

Z. C. Marco Antonio et al.

Fig. 8 Geolocation test. Source Author

4 Discussion This paper presents the results of the simulation of the designed geolocalization system. This is a tool that can give information in real time of the geographic coordinates of the surroundings, specifically of the objects visualized by the camera of the gimbal. This information can be very useful as an input for the planning of military ground operations. The results of the preliminary tests of Table 4 show good results with a tolerable error; this error was product in greater proportion by the sensitivity that presented the servomotors of the gimbal which by their physical characteristics and the time of operation presented a resolution of 1.5° if this resolution improved the error will consequently reduce. It is important to emphasize that the system has the limitation of the UTM zone, since the formulas of the transformation of geographic coordinates to UTM and vice versa use as input to the central coordinate of the corresponding UTM zone, which is different for each zone. With the development of the geolocation system, a boost will be given to the development of operational capabilities in the Air Force by means of new technologies. The use of this technology will be attached to the nation’s security and

Table 4 Preliminary experiments table Test height (m)

Objective range (m)

Error x (m)

Error y (m)

Range variation (m)

5

22.5

1.4468

0.88633

1.6967

2

5

37.4

3.5672

4.3113

5.5958

3

13

37.3

4.7877

0.55752

4.8201

4

13

43.2

4.6758

1.2211

4.8326

5

10

46.6

1.7792

2.7675

3.2901

6

10

1.1

0.3338

0.2215

0.4007

1

Toward the Development of Surveillance and Reconnaissance …

133

defense activities. In addition, this technology can be extrapolated for use in civilian applications.

References 1. COMACO: ModeloEducativo de las Fuerzas Armadas del Ecuador. Dirección Ge-neral de Educación y DoctrinaMilitar (2016) 2. Gómez López, A.O.: Políticaspúblicas de seguridad de Colombia y de Ecuador: una visióncomparada. Fronteras: rupturas y convergencias, 63–75 (2013) 3. Rafael, M.O.: Informe De Avance De Componentes Del Avión Uav - Sistema Electrooptico Seo D1. Centro de Investigación y Desarrollo de la FAE, Ambato (2014) 4. Barrientos, A.: Fundamentos de robótica. McGraw-Hill, New York (1997) 5. Baturone, A.O.: ROBOTICA manipuladores y robots móviles. Mar-combo, Barcelona (2001) 6. Gamboa, J.M.: Fundamentos para CartografíaNáutica. JM ediciones, Cadiz (2009) 7. Maass, S.F., Pérez, M.E.: Principio Básicos de Cartografía y Cartografía au-tomatizada. Universidad Autónoma del Estado de México, Tolupa (2003) 8. Kindle, J.H.: Teoria y Problemas de GeometriaAnlítica Plana y del Espacio. McGraw-Hill, México (1987) 9. Google. Google maps (Internet) 10. Barcelona, P.O.:. Astronomíaesférica y mecánica celeste (noviembre de 2018) 11. Geo-ingeniería, R.P.: Coordenadas UTM en google earth (9 de Diciembre de 2018) 12. Cotticia, A., Surace, L.: Bolletino Di Geodesia E Science Affini

Fuzzy Logic for Speed Control in Object Tracking Inside a Restricted Area Using a Drone Richard Navas Jácome, Harley Lovato Huertas, Patricia Constante Procel and Andrés Gordón Garcés

Abstract This article presents the autonomous speed and positioning control for a drone using fuzzy logic to track a target within a restricted area. Classical controllers present a problem, as in general they only support one input and one output and a system model is always required. For this application, it is necessary to analyze two inputs, the position in “x” and the position in “y” of an object that will be recognized by the drone through artificial vision. The goal is to control the speed at which the drone moves according to the position of the object detected by the machine vision within a restricted area, resulting in a faster or slower movement that will improve the tracking of a moving target by delivering real-time object monitoring information to the user in order to take some action based on this information. Keywords Fuzzy control · Drone · Artificial intelligence

1 Introduction Fuzzy logic is a mathematical language that represents non-quantifiable values. In classical logic, an element belongs or does not belong to a set, whereas in fuzzy logic, each element has degrees of membership [1]. This type of control is typical of Artificial Intelligence (AI), where attempts are made to emulate human behavior. The design of the fuzzy controller uses the knowledge and experience of the designer; one of the advantages of this control is that it does not require a mathematical model of the process to be controlled [2]. In recent years, unmanned aerial vehicles, and more specifically drones, have been used for the development of multiple and varied applications, either because of their versatility, relative ease of flight control, or because of the ability to obtain informaR. N. Jácome · H. L. Huertas · P. C. Procel (B) · A. G. Garcés Universidad de las Fuerzas Armadas ESPE, Sangolquí, Ecuador e-mail: [email protected] A. G. Garcés e-mail: [email protected] © Springer Nature Singapore Pte Ltd. 2020 Á. Rocha and R. P. Pereira (eds.), Developments and Advances in Defense and Security, Smart Innovation, Systems and Technologies 152, https://doi.org/10.1007/978-981-13-9155-2_12

135

136

R. N. Jácome et al.

tion from the environment through cameras, sensors, etc. [3]. At Zayed University, they analyzed the potential applications of drones in environmental monitoring and urban space management, as well as their potential risks [4]. In the USA, researchers are looking for ways to supply basic necessities [5] and improve logistics in the event of natural disasters [6]. In agriculture, combining the benefits of working with unmanned aerial vehicles and introducing concepts of AI and the Internet of Things (IoT), so-called precision farming is sought after [7]. This shows that drones are an ideal tool to provide solutions to multiple problems in today’s world. An application for security and surveillance is the tracking of targets by means of several robots with sensors on board [8], and this evidences that the use of robots that provide data in real time is an advantage in the monitoring of restricted areas. Early detection of an intruder invading private or restricted property ensures that defense mechanisms take more efficient action to safeguard the integrity of the complex and the security of elements of force, occupying a network of technologies that help maintain order. This research proposes a fuzzy control system to improve the positioning of a drone in a satisfactory manner. There are classic control systems well known as the PID which requires precise mathematical models, used by Korean engineers to develop an intelligent tracking drone [9]. Hybrid controllers such as fuzzy PID control have also been used to avoid obstacles in enclosed environments, proposed by the Universidad de La Frontera (Chile) [10]. However, fuzzy logic control was chosen for its favorable performance in systems where it is intended to position itself in a point of interest, a field in which this research is focused; for example, the autonomous landing system of an unmanned aerial vehicle based on artificial vision, which, through the integrated camera, detects the landing point and positions it through a fuzzy control [11]. The implementation of the fuzzy control was carried out using a Python programming algorithm without the use of libraries. To form the fuzzy logic sets, the work of researchers at the University of Surabaya [12] was taken into account, showing how they built the sets for each of the axes of a drone.

2 Definition of the Control System Fuzzy logic allows ordinary language to be used as a variable for a computer. Knowledge is interpreted as a collection of fuzzy constraints on a group of variables. This logic is widely used in decision support systems. Fuzzy logic allows a computer to reason in linguistic terms and rules in a manner similar to that of human beings [13]. A fundamental part of the fuzzy logic is the selected membership functions. A fuzzy logic set is an ordered pair of an “x” element belonging to a U universe and its degree of membership, as shown at (Fig. 1). The functions of membership, which give the degree of membership, can be defined in a functional way, which has different forms such as trapezoidal, triangular, bell, etc. [14]. The type of controller chosen was the Mamdani [15], its architecture, which has the following structure [16]:

Fuzzy Logic for Speed Control in Object …

137

Fig. 1 Fuzzy set

• Determination of membership functions that will perform the fuzzy scaling. • Establishment of rules based on conditional proposals. • Selection of belonging functions for deburring and the method for finding the output value, which is the centroid of the geometric figure formed by the assemblies; it is used (1).   μ(x) · x μ(x) · x   (1) μ(x) μ(x)

3 Development 3.1 Membership Functions for the Fuzzy Process To form fuzzy sets and membership functions, it is necessary to know what input data and the range of values they can take. The system detects by artificial vision the position of the object’s center of interest so the data entry will be the location of pixels in the X- and Y-axes of the image; and the size in pixels of the input image is 640 in the X-axis and 480 in the Y-axis as shown in (Fig. 2a).

(a) image size

Fig. 2 Image acquisition

(b) Reference plane

138

R. N. Jácome et al.

(a) Fuzzy sets on the reference plane

(b) Linguistic variables

Fig. 3 Reference image

The purpose of the application is to move the drone to the location of the object in question; therefore, it must be located in the center of the input image, being this the set point. Thus, we have the image divided into four quadrants, sectioning the X-axis into two parts of 320 pixels each and the Y-axis into 240 pixels as shown in (Fig. 2b). Within the image, three regions are defined to determine in which area the object is located. “Close” corresponds to the area near the center of the image, i.e., when the distance between the drone and the object is relatively small. “Normal” refers to the middle zone between the center of the image and the peripheral zone of the image, and in this region, the distance will be intermediate. And “Far” corresponds to the farthest area located at the edges of the image, and at these points, the drone is at a long distance from the target, as shown at (Fig. 3a). The set point, located in the center of the image, divides the X- and Y-axes into two equal parts. The X-axis has its components right and left, and the Y-axis has its components above and down, as shown at (Fig. 3b). Once the input parameters and the way in which the data obtained is sorted are defined, the fuzzy sets are formed. As can be seen in (Fig. 4), both the X- and Y-axes are similar, so the number of sets will be the same, with a variation in the range of values. On the X-axis, the range of values is from −320 to 320 pixels forming five sets: “FarL,” region far to the left side, “NormalL” intermediate region to the left side, “Near” region near the center of the image, “NormalR” intermediate region to the right side and “FarR” distant region to the right side, as shown in (Fig. 4a).

Fig. 4 a Fuzzy sets on the “x” axis. b Fuzzy sets on the “y” axis

Fuzzy Logic for Speed Control in Object …

139

Similarly, on the Y-axis, the range of values ranges from −240 to 240 pixels, making five sets: “FarL,” far region on the left side, “NormalL” intermediate region on the left side, “Near” region near the centre of the image, “NormalR” intermediate region on the right side and “FarR” far region on the right side, as shown in (Fig. 4b). Once the sets have been formed, they are mathematically defined for later coding in programming language. The mathematical equation of the lines forming the polygon of each set and the range of values they are in are described in Tables 1 and 2.

3.2 Settings the Rules With the fuzzy process data dissemination, the fuzzy controller rules for speed variation are developed, where the considerations are as follows: For “far,” the drone will be at its maximum “fast” speed, and as it approaches the object it will pass at “normal” distance so that its speed will be “medium” and when it reaches a location near it “near,” its speed will be “slow” with respect to the center point. The rules are shown in Table 3.

Table 1 Equation for X pixel variation Name Near

Equation



N ear =

x 80 + 1, −80 < x < 0 −x 80 + 1, 0 < x < 80

Normal right N or mal R =

Normal left

Far left

⎪ ⎪ ⎩

x 80

− 0.5, 40 < x < 120 1, 120 < x < 200

−x 40

+ 6, 200 < x < 240

N or mal L = ⎧ −x ⎪ ⎪ ⎨ 80 − 0.5, −40 < x ← 120 1, −120 < x ← 200 ⎪ ⎪ ⎩ x + 6, −200 < x ← 240 40

Far right

⎧ ⎪ ⎪ ⎨



Far R =

− 5, 200 < x < 240 1, 240 < x < 320

 Far L =

x 40

−x 40

− 5, −200 < x ← 240

1. − 240 < x ← 320

140

R. N. Jácome et al.

Table 2 Equation for Y pixel variation Name

Equation

Near



N ear =

y 60 + 1. − 60 < x < 0 −y 60 + 1.0 < x < 60

Normal above N or mal R =

Normal down

⎧ ⎪ ⎪ ⎨ ⎪ ⎪ ⎩

y 60

− 0.5.30 < x < 90 1.90 < x < 150

−y 30

+ 6.150 < x < 180

N or mal L = ⎧ −y ⎪ ⎪ ⎨ 60 − 0.5, −30 < x ← 90 ⎪ ⎪ ⎩

1. − 90 < x ← 150

y 30

+ 6. − 150 < x ← 180  y − 5.150 < x < 180 Far R = 30 1, 180 < x < 240  −y 30 − 5, −150 < x ← 180 Far L = 1, −180 < x ← 240

Far above

Far down

Table 3 FAM matrix

Axis X

Axis Y

Far L

Normal L

Near

Normal R

Far R

Far L

F

F

F

F

F

Normal L

F

M

M

M

F

Near

F

M

S

M

F

Normal D

F

M

M

M

F

Far D

F

F

F

F

F

The speed parameters: F = fast, M = medium, S = slow

3.3 Membership Functions for Defuzzification For the deburring of the processed data, the membership functions for the fuzzy control are formed where the speed given for the drone is from 0 m/s to a maximum of 15 m/s, considering the three parameters for “slow,” “medium” and “quick,” as shown in (Fig. 5). Finally, the previously formed sets are mathematically defined for later implementation in programming language (Table 4).

Fuzzy Logic for Speed Control in Object …

141

Fig. 5 Fuzzy output sets

Table 4 For the output speed

Name Low speed Medium speed

High speed

Equation  Slow = − u7X + 17 , 0 < V < 7  5 ∗ u X + 3.3 < V < 8 Medium = −5 ∗ u X + 13.8 < V < 13  4 ∗ u X + 8.8 < V < 12 Quick = u X = 112 < V < 20

4 Implementation The implementation of the code was done in Python, since in this language it is possible to find the functions that allow control the air vehicle. The programming of the image acquisition, addressing and fuzzy controller for the speed was based on the structure shown in (Fig. 6). The drone used for the implementation is a 3DR SOLO (Fig. 7) with a PIXHAWK 2.1 controller, to encode the board is used Python programming language and as Linux development environment. The controller board has its own libraries to send instructions to the unmanned aerial vehicle and position it at the point of interest; these libraries are used to send the fuzzy controller result to the drone’s flight parameters. The communication protocol used is mavlink wifi, which establishes a stable connection with the PC control medium.

Fig. 6 Structure of the system

142

R. N. Jácome et al.

Fig. 7 Drone 3DR Solo

(a) RGB Image

(b) Object detected

Fig. 8 Image processing

For image processing, OpenCV was used to develop a BGR file with the parameters to highlight the blue object between the ranges (B = 100, G = 65, R = 75) and (B = 130, G = 255, R = 255) placed in a filter layer as in (Fig. 8). In this case, the blue object was used to represent the unauthorized object inside the restricted area, but it could be customized in order to detect different objects, such as vehicles or people, using advanced artificial vision algorithms which are not presented in this paper.

5 Analysis of Results 5.1 Comparison Between Matlab Results and Implemented Code Then, it is analyzed whether the result of the fuzzy control is consistent with the requirements, i.e., whether the velocity value grows and grows according to the position of the detected object. The speed values are then compared with the values obtained through Matlab to check if the implemented code responds correctly.

Fuzzy Logic for Speed Control in Object … Table 5 Comparison between Matlab results and implemented code

Test

143

Speed (m/s) Matlab

Code

1

2.333

0.04375

2

6.88

5.9885

3

8

8

4

12.2

13.49

5

14.25

14.25

Test 1: The object is in one position (1.0) and the speed is 0.04375, tending to zero. Therefore, the result is acceptable as in this position the drone should remain static. Test 2: The object is in a position (−69.43) one and the speed is 5.9885. The two coordinates are at extreme values of the near and normal sets, so it follows that the speed must be within the range of slow and average speeds. The speed value is acceptable. Test 3: The object is in a position (−123.75) one and the speed is 8. The two coordinates are in values belonging to the normal position sets, i.e., the speed of the drone is average. The result is acceptable for this test. Test 4: The object is in one position (−276,166) one and the speed is 13.49. The values of the X, Y coordinates are between the far and far ranges of normal and far, and speed is expected to be between the mid and fast ranges. The speed delivered in this test meets the requirements. Test 5: The object is in one position (−290.199) one and the speed is 14.25. Both coordinate values are in the far range, and the speed is the maximum the algorithm can provide. The result is optimal in this case. Table 5 presents the results where it is compared with the Matlab software where a similarity between the values obtained is evident without this altering the functionality of the implemented controller. As you can see, the results vary slightly, although the growth trend is similar. The probable cause is that Matlab uses a different method to debrister the final result than the one used for code implementation (Discrete centroid method).

5.2 Final Performance Tests Several tests were carried out, three of which are presented at (Fig. 9), which shows an increase in speed as a function of the distance from the drone to the object. The results are shown in Table 6.

144

R. N. Jácome et al.

Fig. 9 Implementation and testing Table 6 Final result in each case

Function

Position

Speed (m/s)

Parameters

A

Near

(−9, 4)

1

Slow

B

Normal

(145, 50)

8

Medium

C

Far

(220, 50)

12.75

Quick

6 Conclusions A fuzzy controller with two inputs and one output (MISO) has been easily implemented on a drone, which provides speed data based on the distance of the object that has been recognized. Real-time images of 640 × 480 pixels have been acquired, resulting in triangular-shaped membership functions defined for the distance in “x” and “y” to be able to associate when an object intended to track is near or far. Fuzzy logic control was implemented in Python programming language because it is a high-level hybrid language, and it is the development native language of the drone. The principal advantage of this fuzzy logic controller is that it could be applied to any type of UAV because the system model of the UAV is not required for control, and it could be developed in any other programming languages depending on the UAV manufacturer. Importance of the presented research work is that the speed control and positioning system can be used and developed within applications mostly in the area of security and defense, such as, for example, non-identified people and objects tracking inside secure areas. Acknowledgements Thanks to the University of the Armed Forces ESPE for the support provided to this research work.

Fuzzy Logic for Speed Control in Object …

145

References 1. Ramírez, N.V., Laguna Estrada, M.: La lógica borrosa: conjuntos borrosos, razonamiento aproximado y control borroso (2012). http://pistaseducativas.itc.mx/wp-content/uploads/2013/01/6RAMIREZ-PE-100–55-65.pdf. Accessed 20 November 2017 2. Camastra, F., Ciaramella, A., Giovannelli, V., Lener, M.: A fuzzy decision system for genetically modified plant environmental risk assessment using Mamdani inference, ScienceDirect, 1710–1716 (2015) 3. Olivares Mendez, M., Mejias, L.: See-and-avoid quadcopter using fuzzy control. In: IEEE World Congress on Computational Intelligence, p. 1 (2012) 4. Gallacher, D.: Drone applications for environmental management in, Science Target, p. 1 (2016) 5. Scott, J.E., Scott, C.H.: Drone delivery models for healthcare. In: 50th Hawaii International Conference on System Sciences, p. 1 (2017) 6. Restas, A.: Drone applications for supporting disaster, SciRes, p. 1 (2015) 7. Saha, A.K., Saha, J., Ray, R.. Sircar, S.: IOT-based drone for improvement of crop quality. IEEE, p. 1 (2018) 8. Hausman, K., Müller, J., Hariharan, A., Ayanian, N., Sukhatme, G.S.: Cooperative control for target tracking with onboard sensing. Exp. Robot. 879–892 (2015) 9. Yang, W.-S., Chun, M.-H., Jang, G.-W.: A study on smart drone using quadcopter and object tracking techniques. IEEE, p. 1 (2018) 10. Gatica, N., Muñoz, C., Sellado, P.: Real fuzzy PID control of the UAV AR.Drone 2.0 for, IEEE (2017) 11. Olivares Mendez, M., Kannan, S., Voos, H.: Vision based fuzzy control autonomous landing with UAVs: From, IEEE, p. 1 (2015) 12. Indrawati, V., Prayitno, A., Utomo, G.: Comparison of two fuzzy logic controller schemes. IEEE (2015) 13. González Morcillo, C.: Lógica Difusa. técnicas de Softcomputing. http://www.esi.uclm.es/ www/cglez/downloads/docencia/2011_Softcomputing/LogicaDifusa.pdf 14. Santos, M., Suescun, E.M.: Aplicación de la lógica difusa en el ambito de las energías renovables (2012). https://dialnet.unirioja.es/descarga/articulo/5085360 15. Ören, A., Koçyı˘gıt, Y.: Landing sequencing modelling with fuzzy logic: opportunistic approach for unmanned aerial systems. In: 2016 International Conference on Unmanned Aircraft Systems (ICUAS), pp. 943–948, (2016) 16. Lovato, A.V., Oliveira, J.C.M.: Airplane level changes using fuzzy control. In: International Conference on Fuzzy Systems (2010)

Part IV

Health Informatics in Military Applications

Micro-controlled EOG Device for Track and Control Military Applications Nayana L. M. Viana, José Ailton L. Barbosa Junior and Francisco A. Brito-Filho

Abstract This paper presents a system with tracking and remote control capabilities for military applications. An electrooculogram and a micro-controlled system with wireless communication were developed as controller based on the eye movement. The system can be used as remote control for military purposes and also for soldier monitoring. A proof of concept to act and be tested as mouse pointer was implemented with reduced circuitry and PIC microcontroller in order to achieve low cost and low profile platform. Experimental results are shown to validate a prototype of the proposed system. Keywords Electrooculogram · Tracking · Remote control

1 Introduction Biopotentials are signals that result from electrophysiological activities that occur inside the cells. Those activities can represent many physical behavior that human being does along the day, for example, to walk, to pick an object and to blink the eyes [1, 2]. Those biopotentials can be measured and applied to track some specific behavior or to treat diseases and to control objects [2]. N. L. M. Viana · J. A. L. Barbosa Junior · F. A. Brito-Filho (B) Federal University of Semiarid Region—UFERSA, Caraubas, RN 59780-000, Brazil e-mail: [email protected] N. L. M. Viana e-mail: [email protected] J. A. L. Barbosa Junior e-mail: [email protected] N. L. M. Viana Federal University of Rio Grande do Norte—UFRN, Natal, RN 59078-900, Brazil J. A. L. Barbosa Junior Federal University of Ceara—UFC, Fortaleza, CE 60020-181, Brazil © Springer Nature Singapore Pte Ltd. 2020 Á. Rocha and R. P. Pereira (eds.), Developments and Advances in Defense and Security, Smart Innovation, Systems and Technologies 152, https://doi.org/10.1007/978-981-13-9155-2_13

149

150

N. L. M. Viana et al.

There are four biosignals covering all activities performed by humans, that are electrocardiogram (ECG), electroencephalogram (EEG), electromyography (EMG) and electrooculogram (EOG), which are the potential arising from muscle contractions of the heart, neural activities (or brain activities), activity from fiber or group of muscle fibers and difference of potential between the cornea and the retina, respectively [1–4]. Rising advances in biopotential acquisition, processing and applications are allowing low profile devices, with more battery life and lower cost, motivating new markets to use them like consumer and defense [4, 5]. Acquisition and control of EOG signals appear in this scenario as an alternative to track soldier behavior, to monitor his health from remote places or to use as a remote control signal for military devices [6–11]. This work presents a device that can be applied for military purposes both in track or monitor field as in control applications. As a proof of concept of device idea, a mouse pointer micro-controlled by eye movement is developed and validated by experimental tests.

2 Proposed System The EOG signals are pulses that vary proportional to the degree of eye movement. As higher the degree of eye movement, higher is the EOG signal amplitude [1, 2]. Figure 1 shows typical EOG signals. To validate the proposed system idea, to track eye movement and to control objects by the same biopotential, a mouse pointer device micro-controlled by the EOG signal was implemented. The EOG system block diagram is shown in Fig. 2.

Fig. 1 Typical EOG signals

Micro-controlled EOG Device for Track and Control …

151

Fig. 2 EOG system block diagram

The implemented system can be divided into two subsystems: the acquisition system, that contains the electrodes, instrumentation amplifier and filters; and the processing and control system that is performed by the PIC18F microcontroller. Also, a wireless transmission system can be added in order to send the control signals to remote devices. Below are presented those subsystems in more detail.

2.1 Acquisition System The acquisition system is basically composed of electrodes, instrumentation amplifier and filters. AgCl Electrodes and Conductive Gel. The biopotential acquisition is done by means of electrodes that perform the transduction from the ionic currents in electrical signals [1–5]. To capture the EOG signal is used silver chloride surface electrodes (AgCl) and to optimize capture of that signal is used a conductive gel. The latter is used to increase the conductivity between the electrode and the patient’s skin. The electrodes are placed as shown in Fig. 3. They are placed below and above the eye to capture vertical movement and at left and at right of the eye to capture horizontal movement. The electrode placed on the forehead is the reference electrode and must be connected to GND (ground) circuit. Instrumentation Amplifier. As seen before, the EOG signals have low magnitudes. So, they need to be amplified in order to be analyzed by measurement instruments and used as input signal to the PIC microcontroller. Is necessary that their amplitudes are larger and therefore the signal must pass through amplifiers. The instrumentation amplifier is the most indicated when need to get a high amplification with minimum noise. Since the EOG biosignal magnitude is on the order of low-frequency noise, the amplifier needs a high Common Mode Rejection Ratio (CMRR), high gain, low drift and high input impedance that is the case of the instrumentation amplifier shown in Fig. 4. In order to get a low profile device was chosen the use of two integrated circuits (IC) to implement the two instrumentation amplifiers (one for acquisition of the horizontal eye movement and other for the vertical eye movement), since each IC has four operational amplifiers. Three operational amplifiers were used to implement the instrumentation amplifier and the other was used to implement the output filter that will be shown in next subsection. Each IC has inputs for signals that come from electrodes and output to PIC microcontroller.

152

N. L. M. Viana et al.

Fig. 3 Electrodes positioning

Rf3

Fig. 4 Instrumentation amplifier schematic

Cf2

Rf2

EL+ Rf3

Vout Cf2

ELRf3

Rf3

Rf2

Filter. In order to get a better pulse acquired from eye movement and suitable to be read in PIC inputs, an active low-pass filter is implemented at the output of the instrumentation amplifier in order to eliminate the high frequency noise. The filter schematic can be seen in Fig. 5. At filter output is a signal ready to be read at PIC microcontroller input and processed.

Micro-controlled EOG Device for Track and Control …

153

Cf2 Rf1

Rf2

Vin Vout

Cf1 Rf3 Rf2

Fig. 5 Filter schematic Start

A/D > 512?

A/D > 512? N

Y X--

N

Y X++

Y--

Y++

Fig. 6 Microcontroller algorithm flowchart

2.2 Processing and Control System The processing and control system can have a transmitter after filter output, that wirelessly transmits the EOG signals to a receiver coupled to the PIC microcontroller, in order to process and control remote devices or to monitor the EOG signals. The use of radiofrequency signals can allow remote control applications for long distances for example in battlefield context. The microcontroller used was the PIC18F. Its function is to acquire the analog signal from the EOG and convert it into a digital signal using one of its analogto-digital converter (A/D) inputs. Then the signal is processed using an algorithm developed exclusively for this application which its flowchart can be seen in Fig. 6. Two PIC A/D inputs are used to acquire EOG signals represented in the flowchart as X and Y inputs, or horizontal and vertical inputs, respectively. Since PIC A/D is 10-bit resolution so the microcontroller can read 1024 different levels representing EOG voltages from 0 to 5 V. Before EOG signal enters in PIC A/D input, it passes through a 2.5 V offset in order to test the average value 512. As can be seen in Fig. 6, the mouse movement is incremented or decremented in two dimensions if the inputs are below or above of the offset.

154

N. L. M. Viana et al.

The post-processed signals are used to control the device. Depending on the device to be controlled, the signals can be sent wirelessly or by serial connection, stored to be monitored in real time by another device.

3 Results In order to validate the proposed system, an EOG micro-controlled mouse pointer was developed and the system was divided into previously described two stages for acquisition and for processing and control purposes. As part of the next tests, wireless transmitter-receiver interface is been provided. Figures 7 and 8 can be seen two prototype boards for the two subsystems designed for the system proof of concept. Since the boards were built independently from each other, it can be used in stand-alone mode and can be stacked in order to get low profile. In order to check the acquired signals, some volunteers were used to test the system. Figures 9 and 10 show the measured EOG signals acquired, respectively, from the horizontal and vertical eye movement.

Fig. 7 Prototype upper view

Micro-controlled EOG Device for Track and Control …

Fig. 8 Prototype side view

Fig. 9 Measured EOG horizontal signal

155

156

N. L. M. Viana et al.

Fig. 10 Measured vertical EOG signal

In Fig. 9, the pulses higher than offset indicate that the volunteer eye movement was to the left side and the pulses less than offset represent the eye movement to the right side. In the same way, Fig. 10 represents the volunteer eye movement down when pulses are higher than offset and up when pulses are less than offset which is in agree with the fluxogram previously presented. Transmission and reception prototypes in the same modular profile are under construction.

4 Conclusions This paper presented a system proposal for track and remote control based on the eye movement and can be used for military purposes both monitor or control applications. In order to validate the system device, a micro-controlled mouse pointer was implemented and tested. The results showed the pulses acquired from the horizontal and vertical movements and the prototype proved a low profile, as low as 5 × 3 cm, and low cost platform to be used. The prototype tested was effective attending the initial proposal, with the functions of track eye movement for each direction: right, left, up and down, and also the blink, that for military control applications could be a confirm signal.

Micro-controlled EOG Device for Track and Control …

157

Some improvements are been done as to adapt acquisition system to special glasses without the necessity of disposable electrodes and cables, to be applied in a more real battlefield context, and also, with receive-transmit capabilities.

References 1. Bronzino, J.D.: The Biomedical Engineering Handboook, 2nd edn. CRC Press, Boca Raton (2000) 2. Malmivuo, J., Plonsey, R.: Bioelectromagnetism: Principles and Applications of Bioelectric and Biomagnetic Fields. Oxford University Press, USA (1995) 3. Yazicioglu, R.F., Hoof, C.V., Puers, R.: Biopotential Readout Circuits for Portable Acquisition Systems. Springer Science & Business Media (2009) 4. Tseng, Y., Ho, Y., Kao, S., Su, C.: A 0.09 µW low power front-end biopotential amplifier for biosignal recording. In: IEEE Transactions on Biomedical Circuits and Systems, pp. 508–516. IEEE (2012) 5. Ha, S., Kim, C., Chi, Y.M., Akinin, A., Maier, C., Ueno, A., Cauwenberghs, G.: Integrated circuits and electrode interfaces for noninvasive physiological monitoring. In: IEEE Transactions on Biomedical Engineering, pp. 1522–1537. IEEE (2014) 6. Patil, N., Iyer, B.: Health monitoring and tracking system for soldiers using internet of things (IoT). In: International Conference on Computing, Communication and Automation (ICCCA2017), pp. 1347–1352, IEEE, Greater Noida, India (2017) 7. Bisio, I., Delfino, A., Lavagetto, F., Sciarrone, A.: Enabling IoT for in-home rehabilitation: accelerometer signals classification methods for activity and movement recognition. IEEE Internet Things J., 1–11. IEEE (2016) 8. Rossi, M., Rizzi, A., Lorenzelli, L., Brunell, D.: Remote rehabilitation monitoring with an IoTenabled embedded system for precise progress tracking. In: IEEE International Conference on Electronics, Circuits and Systems (ICECS), pp. 384–387. IEEE, Monte Carlo, Monaco (2016) 9. Jiang, M., Gia, T.N., Anzanpour, A., Rahmani, A.-M., Westerlund, T., Salantera, S., Liljeberg, P., Tenhunen, H.: IoT-based remote facial expression monitoring system with sEMG signal. In: IEEE Sensors Applications Symposium (SAS). IEEE, Catania, Italy (2016) 10. Reyes, C.R.P., Vaca, H.P., Calderón, M.P., Montoya, L., Aguilar, W.G.: MilNova: an approach to the IoT solution based on model-driven engineering for the military health monitoring. In: CHILEAN Conference on Electrical, Electronics Engineering, Information and Communication Technologies (CHILECON). IEEE, Pucon, Chile (2017) 11. Tavakoli, M., Turicchia, L., Sarpeshkar, R.: An ultra-low-power pulse oximeter implemented with an energy-efficient transimpedance amplifier. IEEE Trans. Biomed. Circuit. Syst. IEEE (2009)

Part V

Leadership and e-Leadership

Multilevel Military Leadership Model: Correlation Between the Levels and Styles of Military Leadership Using MLQ in the Ecuadorian Armed Forces Celio Humberto Puga Narváez, Alex Fernando Jimenez Vélez, Rafael Caballero Fernández and Osvaldo Fosado Téllez Abstract This article presents the advances of a research focused on validating a Multilevel Military Leadership Model, which relates: leadership levels, levels of military strategy, personal and institutional competencies that leaders must develop, and the leadership styles of the Full Range Leadership Model (FRLM). The model applied in a first approximation to the Ecuadorian Armed Forces uses the Multifactor Leadership Questionnaire (MLQ) to determine the predominance of each leadership style within each level of leadership, as well as its influence with variables of performance and satisfaction at work. Keywords Multifactor Leadership Questionnaire (MLQ) · Full Range Leadership Model (FRLM) · Military leadership · Multilevel leadership · Transformational leadership

1 Introduction The increasing complexity and evolution of organizational science means that leadership research should be prepared for a multilevel study approach, supported mainly by the progress in the analysis of its data. This analysis allows the complex field of leadership to develop specific theoretical-methodological advances at each level [1], going from conceiving a type of individual leadership in the 1980s, organizational in the 1990s and multilevel since 2000 [2]. C. H. P. Narváez (B) Programa de Doctorado de Economía y Empresa, Universidad de Málaga, Málaga, Spain e-mail: [email protected] A. F. J. Vélez Centro de Investigación y Desarrollo FAE, Ambato, Ecuador R. C. Fernández Universidad de Málaga, Málaga, Spain O. F. Téllez Universidad Técnica de Manabí, Portoviejo, Ecuador © Springer Nature Singapore Pte Ltd. 2020 Á. Rocha and R. P. Pereira (eds.), Developments and Advances in Defense and Security, Smart Innovation, Systems and Technologies 152, https://doi.org/10.1007/978-981-13-9155-2_14

161

162

C. H. P. Narváez et al.

The multilevel approach appears in 1989 under the name of stratified systems theory, depending on the time and work complexity. Subsequently, Hunt (1991, 1996) adapts the theory to a vertical leadership perspective at the different levels of the organization, called the multi-organizational leadership model. This model establishes that there are critical tasks that must be carried out by the leaders in an organization, according to their level, thus they will have tasks that change its complexity based on their level. The complexity of these tasks can be measured in several ways, one of them will be the time taken to accomplish every single task, activities or projects developed by an individual or team. The tasks assigned to the lower levels will have a shorter time of execution than the tasks performed at higher levels. That is why the multilevel extended model indicates that the variety of tasks complexity leads to a division of three levels; where the lowest level of the organization is called direct or production, the level in the middle is called organizational and the highest level as systems or strategic, being these levels highly related to each other. Leadership at the direct or production level involves administrative or operational procedures and maintenance of individual or collective skills; the organizational comprises the integration of organizational elements focused on the compliance of the objectives and its mission statement; and the leadership of systems implies the development of the mission statement, articulation of objectives, strategies, principles, policies and planning of the organizational systems [3, 4]. In this sense, Reeves-Ellington (2015) presents another classification of leadership: junior, middle and higher. The junior level leadership is in charge of taking care of the lower levels of the organization; the medium has a key role in the company integration, its broader social environment and personal climate have strong links throughout the organizational community; and the superior level with the main responsibility of controlling strategic execution through the supervision of human and budgetary resources, achieving a community and holistic result [5].

1.1 Military Leadership Levels According to the aforementioned, it can be determined that the theory of multilevel leadership is accepted in the organizational field, this multilevel classification being extrapolated to the military leadership, specifically to a vertical institution such as the Armed Forces. Under this organizational structure, military leadership comprises three levels: tactical, operational and strategic leadership; division that is correlated based on the classification of the three levels of military strategy, widely known in the Armed Forces field. The tactical leadership, which is the lower level in the military organization, seeks to develop the ability in the execution of actions in order to achieve the purposes proposed with the available resources. At this level, there is more certainty and less complexity than at higher levels.

Multilevel Military Leadership Model: Correlation …

163

Although the time that the military officer remains at this level depends on the acquired skills, generally spoken, it is made up of military personnel who start their military career until around the 12 years of service, being this reference the limits, and considering that the military profession could last for approximately 35 years; logically, these times vary according to the institution and analyzed country. In mid-level operational leadership, the main objectives and resources are adapted; where a link between the complex decisions, guidelines of the strategic level and the tasks fulfilled by the tactical level is established. At this level, a large amount of useful information is analyzed to advise the superior level; therefore, the people at this level must have a clear understanding of the different systems of the organization, which allows them to decide their correct interaction. The operational level is made up of military professionals whose service time can generally be 12–24 years of service. The strategic leadership is the highest level, where the organizational goals and objectives are established, and it is unfolding in an uncertain, changing, unknown and complex environment, because the strategic leader is the direct link to organizations outside the military environment, with different characteristics. Therefore, the leader at this level has a knowledge focused on the national, regional and global perspective, with a capacity for convincing and negotiating, which allows him to make successful decisions, since the implications will affect the direction and evolution of the institution. The level of strategic leadership is made up of the entire institutional high command, with approximately 24–35 years of service. Hence, leaders are trained to finally achieve their full leadership at a higher level. It is important to point out that during the process, they acquire a series of institutional and personal competencies.

1.2 Institutional Competencies Among the competencies that must be developed by leaders for the growth of their leadership, there are three broad categories: technical, interpersonal and conceptual competencies; at the tactical level, specific skills or technical knowledge is required, which decrease as the level of leadership increases, in contrast to the interpersonal skills that are essential at all levels, as well as the conceptual competencies that grow as the level of leadership increases, allowing to determine important aspects of the organization. In the case of technical competencies, these cover specific knowledge regarding the assigned technical aspects, they are required at all levels, but in a different way, at the tactical level consists generally in using, operating or maintaining a team or system; they focus on solving specific problems, as well as carrying out specific tasks and missions. At the operational level, the personnel can lead to establishing a network of systems, and at the strategic level, they are useful for decision-making, by facilitating the synchronization of information with a system of systems, or to design

164

C. H. P. Narváez et al.

a well-articulated structure between forces or military units, as well as to work on solving major problems that would affect the entire organization. If we talk about interpersonal competencies, they refer to the capacities needed to interact effectively with other people; these competencies also allow the leader to have emotional intelligence, self-awareness, confidence and patience to foster an environment of empowerment, being fundamental at all three levels (tactical, operational and strategic). The personnel at the tactical level can deal with interpersonal relations under the framework of subordination; on the other hand, the personnel at the strategic level have more lateral relationships, where collaboration, integration, persuasion, perception and negotiation skills may be necessary, but generated through mutual trust and respect. Finally, in the case of conceptual competencies, these include the ability to think, mainly related to logical analysis and inference, which is why they are more necessary at the strategic level than at the other levels. At this level, the strategic leaders get information from external sources and internal, which allows them to understand through visualization and to project adequate planning. The ability to search for relevant and priority information is fundamental, as well as the correct decision-making, identifying key situations that visualize potential problems and possible lower risk solutions. Namely, to achieve a prospective vision that allows projection to the organization, with proactive reasoning based on present actions to visualize future objectives [6].

1.3 Personal Competencies The statistics mentioned that no leadership model will have all the elements that make it effective. The effective leader develops skills in three areas: personal, team and organizational. Personal competency is considered as the basis for the growth of the leader, since it refers to leadership over oneself, here self-knowledge and awareness of skills, strengths, weaknesses and values are highlighted, before leading the rest of people; it also works on development, self-growth and continuous learning, which are based on feedback and image as a role model for other followers. In the case of the team competency, it is based on personal leadership, plus the sum of values and competencies of the people, this competency means heading a team that works in a coordinated way that pursues a common goal, and also the leader seeks to lead the entire team toward the same vision, within the scope of their competencies, contributing with collaboration, creativity and innovation, developing relationships of trust and solidarity among team members. Finally, the organizational competency is the one that handles the general context, the macro-tendencies of the different systems, basing its action in the strategic thinking, where the organizational leader must establish the vision of the organizational future, values, objectives, strategies, as well as making decisions and developing

Multilevel Military Leadership Model: Correlation …

165

actions to try to take advantage of the changes that occur in the external environment of their institution [7].

1.4 Leadership Styles (Full Range Leadership Model—FRLM) Leadership has been one of the most important variables in the consideration of the leadership effectiveness perceived by employees and their performance and satisfaction at work. Bass and Avolio, important leadership researchers, based on charismatic leadership and transformational leadership, establish a new leadership model called multifactorial leadership theory in which the types of transformational, transactional and laissez-faire leadership are included. These leadership researchers also raise a questionnaire of multifactorial leadership (MLQ Multifactor Leadership Questionnaire) to evaluate and measure these types of leaders with their characteristics and effects. The transformational leadership is the one that motivates, transforms, inspires employees to overcome personal interests, seeking optimal performance, for organizational benefit. It comprises five factors: idealized influence (attributed), idealized influence (behavior), inspiration or inspirational motivation, intellectual stimulation and individualized consideration (Table 1). The transactional leadership is characterized through the fulfillment of objectives that implies a reward; financial or moral; as well as, negative results due to poor performance have a negative action from the leader. This leadership is not identified by presenting a vision of the future, but rather by focusing on the current organizational situation. It comprises three factors: contingent rewards, management by active exception and management by passive exception (Table 2).

Table 1 Transformational leadership and its components Components

Definition

Idealized influence (attributed)

Leaders serve as a role model for their followers. They are admired, respected and inspire confidence

Idealized influence (behavior)

Besides being models to be imitated, admired, respected and give confidence, the leaders have an idealized behavior

Inspirational motivation

Leaders motivate followers by providing a sense of work and an attractive vision of the organizational future

Intellectual stimulation

Leaders stimulate innovation and creativity. They also try new approaches and seek solutions for themselves

Individualized consideration

Leaders give individualized attention to each follower for their achievement and growth by being a coach or mentor

Source Prepared by the authors, based on [16]

166

C. H. P. Narváez et al.

Table 2 Transactional leadership and components Components

Definition

Contingent rewards

The leaders make an agreement with their follower regarding a reward (material) before a well-executed task. On the contrary, there may also be a punishment

Management by active exception

The leader specifies the rules, limits and punishments. The leader focuses on correcting errors and then corrective measures that should be taken

Management by passive exception

The leader usually leaves things as they are and only intervenes when the problems become serious

Source Prepared by the authors, based on [16]

The laissez-faire leadership is considered a lack or absence of leadership, in which the leader does not decide, does not guide, avoids tasks or making any decision and being involved in important matters.

2 Military Multilevel Leadership Model In the wide range of leadership, there is a topic of permanent relevance and it has been the military leadership. Wong and several authors review the existing literature in this field, highlighting the opportunities and importance of research in this field, bringing up Hunt’s Multilevel Leadership Model, mainly for the various facets involved in that leadership [8]. In addition, we know that this military leadership is classified into three levels, which are related to levels of military strategy, requiring the military to develop different types of competencies. Among the main required developed competencies are the personal and institutional competencies, in order to be properly prepared to be able to ascend to the next level of leadership. That is why institutional competencies, classified in technical, interpersonal and conceptual, are required at all levels of military leadership, but in different ways at each level. In the case of the technical skills, they are very necessary at the tactical level, since the tasks developed at this level are predominantly technical and specific. However, these competencies decrease as the leader rises in the operational level, and his needs are even lower at the strategic level, which is no longer purely technical. On the other hand, interpersonal skills are fundamental at all levels of leadership; however, they are not always the same as they are developed according to the level, because at the tactical level the people deal with military personnel, while at the strategic level is required to have strong networking abilities, coordination and negotiation with other institutions. Finally, highlight that conceptual competencies begin at the tactical level, which are being necessary to understand the importance of specific tasks for the use of

Multilevel Military Leadership Model: Correlation …

167

means, until reaching the strategic level where the global concept that implies leading the organization is essential. However, we not only have institutional competencies, but also personal competencies that exist within the military leadership, which refer to the development of the military’s personal leadership, being these: personal, team and organizational. Personal competency is referred to self-knowledge and self-awareness in the personal formation as a leader, fundamental in the level of tactical leadership at the beginning of the military career, as the time passes by, the military personnel needs less of this competency, as they evolve in the leadership levels, until they become a leader with solid personal training at the strategic level. Regarding the competency of teams and people, it will always be in the evolution of the leader since the composition of the team members varies, for example, at the tactical level teams will be made up of technical personnel, while at a strategic level teams of directors, commanders, leaders of different disciplines must be led. In the case of organizational competencies, these allow the leader to have a vision of their environment, develop strategies to achieve objectives in the unit they lead, which at the tactical level will be a small structure while at the strategic level the whole organization will be. It is important to note that currently the military leaders of countries such as Canada, the USA and Portugal have focused their efforts on developing transformational leadership in depth, since it is due to its own characteristics of motivation, transformation and the vision it transmits superiors to subordinates, the levels of effort and satisfaction with the objectives have been evidenced, when it comes to guiding behaviors in military organizations [9, 10], and also in civil organizations [11, 12]. Therefore, even when the laissez-faire leadership is present in the tactical leader, because he is a leader in the initial stage of their formation; that transactional leadership is found at the operational level in actions, that generate responses and effects that could be whether negative or positive from their superiors, and that transformational leadership is present in the strategic leader, who leads the entire organization; there is no determined interrelation between the levels of leadership and the three styles of leadership, being fundamental to establish this relationship as an initial step to develop a Multilevel Military Leadership Model whose scheme is presented in Fig. 1, and its purpose is focused on promoting the formation of qualified leaders committed to the military organization.

3 Materials and Methods The present study proposes the application of the Multifactorial Leadership Questionnaire (MLQ) of Bass and Avolio [13, 14], whose empirical validity has been recognized and used to obtain data that allow to determine the style of leadership and the relationship with variables of performance and satisfaction at work. The

168

C. H. P. Narváez et al.

Fig. 1 Military Multilevel Leadership Model. Source Prepared by the authors, based on [6, 7]

MLQ has some versions, we will use one adapted to the needs of the Armed Forces of Ecuador [15]. The first stage will be done through surveys, whose questions will be based on measurement instruments determined by the MLQ. This survey will be applied to military officers of the three branches of the Armed Forces of Ecuador (Army, Navy and Air Force), via Google Forms due to the respondents that are in different cities of the country. The officers have been initially considered, since they lead the different levels of the military organization (tactical, operational and strategic) and are the ones that receive the direct influence of their leaders, influencing their subordinates or followers through their leadership, throughout the different charges they occupy. The people surveyed will be stratified, based on the three leadership levels, in order to provide data for each level. For the distribution and application of the survey, the authorization of the corresponding authorities will be requested, establishing preparatory meetings with the focal points determined by each military unit. For the tabulation phase, data analysis and treatment of results, comparisons will be made with similar studies in the military field. In addition, it is intended to determine, based on the surveys and data, the predominant leadership style at each level of leadership, be it transformational, transactional and laissez-faire, as well as its relationship with influence on variables such as performance and satisfaction at work.

Multilevel Military Leadership Model: Correlation …

169

The final results will be needed to proceed to the validation stage using multivariate techniques, determining the modeled relationship between the leadership style and the aforementioned variables. Finally, the results obtained with the organizational authorities will be analyzed and discussed, in order to later move on to a socialization stage and structuring of institutional leadership manuals with regional projection, which seeks to highlight the importance of the influence of the type of leadership of officers in the organizational management variables of the Armed Forces under a multilevel military leadership approach.

4 Conclusion This article proposes, through a Multilevel Military Leadership Model, to establish the correlation between the levels and leadership styles based on the FRLM, emphasizing the importance and topicality of the multilevel analysis in the topic of leadership. In addition, it is intended to determine the relationship between institutional (technical, interpersonal and conceptual) and personal (personal, team and organizational) competencies that are necessary for a greater or lesser degree according to the levels of military strategy. As a result of this study, conducted in first instance to officers of the Armed Forces of Ecuador, it is expected to establish an approach of a Multilevel Military Leadership Model, where the existing correlation between the levels and styles of leadership allows the institutional command of the Armed Forces of Ecuador make better decisions. Besides, they will be able to direct their efforts to guide the preparation of leaders according to the leadership required by the organization, mainly aimed at transformational leadership, for its approach of empowerment, commitment, inspiration and motivation of people. As a future work, this study is planned to be executed in the region, such as Armed Forces officers from other countries, seeking to refine the model and determine improvements in the processes of military leadership training in the region.

References 1. Dionne, S.D., et al.: A 25-year perspective on levels of analysis in leadership research, Leadersh. Q. 25(1), 6–35 (2014) ˇ 2. Batistiˇc, S., Cerne, M., Vogel, B.: Just how multi-level is leadership research? A document co-citation analysis 1980–2013 on leadership constructs and outcomes, pp. 1–18 (2016) 3. Hunt y, J., Ropo, A.: Multi-level leadership: grounded theory and mainstream theory applied to the case of general motors 6(3), 379–412 (1995) 4. Van Wart, M.: Leadership in Public Organizations, 2nd edn. Routledge, New York (2014) 5. Reeves-Ellington, R.: Enviroscapes: a multi-level contextual approach to organizational leadership. Emerald Gr. Publ. Ltd., pp. 337–420 (2015)

170

C. H. P. Narváez et al.

6. 7. 8. 9. 10.

Jacobs, T.O.: Strategic Leadership: The Competitive Edge (2002) Air-University, USAF Leadership Doctrine, vol. II, p. 71 (2015) Wong, L., Bliese, P., McGurk, D.: Military leadership: a context specific review 14(6) (2003) Moreira, M.C.M.: Liderazgo Transformacional y Género en Organizaciones Militares (2010) Ivey, G.W., Kline, T.J.B.: Transformational and active transactional leadership in the Canadian military. Leadersh. Organ. Dev. J. 31(3), 246–262 (2010) Segovia, A.: El liderazgo, la compensación variable, el empowerment psicológico y su impacto en la efectividad del empleado: un enfoque de modelación mediante ecuaciones estructurales, Universidad Autónoma de Nuevo León (2014) Saleem, H.: The impact of leadership styles on job satisfaction and mediating role of perceived organizational politics. Proc. Soc. Behav. Sci. 172, 563–569 (2015) Avolio, B.J., Bass, B.M.: You can drag a horse to water but you can’ t make it drink unless it is thirsty. J. Leadersh. Stud. 5(1), 17 (1998) Avolio, B.J., Bass, B.M., Jung, D.I.: Re-examining the components of transformational and transactional leadership using the multifactor leadership questionnaire. J. Occup. Organ. Psychol. 72, 441–462 (1999) Samanta, I.: Modern leadership types and outcomes: the case of Greek public sector. J. Contemp. Manag. Issues 23(1), 173–191 (2018) Ganga-Contreras, F., Navarrete, E., Alt, C., Alarcón, N.: Percepción de los estilos de liderazgo: el caso de un campus universitario IV (1)(2016), 1–35 (2016)

11.

12. 13. 14.

15. 16.

E-leadership Using WhatsApp, A Challenge for Navy Organizations: An Empirical Study Rolando P. Reyes Ch., Luis Recalde Herrera, Galo Andrade Daza, Victor Gómez Bravo and Hugo Pérez Vaca

Abstract Leadership is a fundamental characteristic of naval–military strategy. However, the development of information and communication technologies (ICTs) has led to the leadership being complemented by an emerging paradigm called e-leadership. Therefore, it is interesting to evaluate this paradigm in a naval–military organization where they are currently using WhatsApp as a tool to issue and follow up on dispositions. Considering the perspectives of technology, communication, and trust proposed by Avolio (Organ. Dyn. 2003, [3]), He et al. (E-leadership strategy in virtual organizations and virtual teams, 2008, [8]) and Liu et al. (Leadersh. Organ. Dev. J. 39(7):826–843, 2018, [10]), a survey is applied to Lieutenant Commanders and Lieutenants of the organization. The results evidence a questionable use of eleadership over WhatsApp. However, e-leadership fails to be effective due to the conflicts that are complicated to mitigate even in the traditional leadership of the organization. Keywords Leadership · E-leadership · Naval · Conflict · WhatsApp

R. P. Reyes Ch. (B) · L. Recalde Herrera · H. Pérez Vaca Departamento de Seguridad y Defensa, Universidad de las Fuerzas Armadas ESPE, Sangolquí, Ecuador e-mail: [email protected] L. Recalde Herrera e-mail: [email protected] H. Pérez Vaca e-mail: [email protected] R. P. Reyes Ch. · G. Andrade Daza · V. Gómez Bravo Centro de Estudios Estratégicos Marítimos CEESMA, Guayaquil, Ecuador e-mail: [email protected] V. Gómez Bravo e-mail: [email protected] © Springer Nature Singapore Pte Ltd. 2020 Á. Rocha and R. P. Pereira (eds.), Developments and Advances in Defense and Security, Smart Innovation, Systems and Technologies 152, https://doi.org/10.1007/978-981-13-9155-2_15

171

172

R. P. Reyes Ch. et al.

1 Introduction Leadership is a topic widely tackled by experts in business administration and management. Its importance has been such that it has also been studied in other areas, such as: social, political, naval–military, among others [6]. However, the great development of information and communication technologies (ICTs), digital communication, and other factors has allowed the leadership to begin to evolve or possibly to complement an emerging paradigm that can act on changing and volatile scenarios that impose the technology. This emergent paradigm is called as e-leadership (or electronic leadership) [3]. In this regard, several authors [2, 3, 10] consider that this type of leadership is an emerging research topic that poses several challenges and contributions, both for business and military organizations, given its main characteristic of allowing interaction and communication between the leader and his followers in a virtual environment via the Internet [1]. Therefore, studying this paradigm not only in business organizations could be interesting. Possibly, in the military–naval field, its study is transcendental due to its hierarchical structure, command authority, and other elements depend on the use of WhatsApp1 as a means to issue and follow up on dispositions. However, to study e-leadership, it is necessary to take into account the proposed perspectives of technology, communication and trust proposed by Avolio [3], He et al. [8] and Liu et al. [10]. With this, it is possible to achieve an empirical evaluation of the existence of possible e-leadership within a naval-military organization in Ecuador. For which, we applied a randomized online survey to 46 Commander Lieutenants and Lieutenants with extensive experience in leading onboard units. The results show the existence of a questionable use of WhatsApp to exercise e-leadership in the naval-military organization. However, there are conflicts that also often appear in his leadership as a naval–military organization, the same ones that impede the effectiveness of e-leadership in a virtual environment. The rest of the article is structured as follows: Sect. 2 presents the background and related studies as well as a brief explanation of the perspectives of e-leadership. Section 3 details the research methodology. The results are explained in Sect. 4. The threats to validity in Sect. 5. Finally, conclusions, discussion, and future work are presented in Sect. 6.

2 Background and Related Studies E-leadership is an emerging research topic and for this reason, several studies [2, 3] have determined its importance for the challenges and benefits it brings to organizations; creating a new way of leading in a virtual environment. In this regard, we find some empirical studies of e-leadership, as shown below: 1 WhatsApp is a messaging application for smartphones, which sends and receives messages via the Internet, complementing instant messaging services, short message service or multimedia messaging system.

E-leadership Using WhatsApp, A Challenge for Navy Organizations: An Empirical …

173

• Liu et al. [10], conducted a qualitative study to study the communicative adoption of the e-leader. A model is used on structural equations proposed by Van Wart et al. [12]. The authors mention that the model adjusts adequately to e-leadership, achieving a greater understanding of the performance of e-leaders. • Fan et al. [7] conducted an experimental study to test the influence of the motivational language of the e-leader in virtual teams. The experiment was conducted on 107 university students in Taiwan. The results showed that the instructions given by digital media have positive effects on creativity and generation of ideas. Although, the positive stimulus depends on the empathic writing of the e-leader. • He et al. [8] conducted a study to establish the difference between leadership and e-leadership, establishing the effectiveness and challenges of e-leadership in the virtual environment, (e.g., technology, communication, cultural differences, trust, and logistics). However, there are limitations such as: lack of confidence, lack of cohesion, level of relationship, among others. The aforementioned empirical works show that empirically evaluating e-leadership in organizations is important. However, the evaluation could be specific to each organization because of the way each one works. Therefore, these authors [3, 8, 10] have proposed strategies based on three perspectives: technology, communication, and trust, as a starting point to evaluate e-leadership, as explained below.

2.1 Perspectives of E-leadership According to [3, 8, 10], everyone considers that technology, communication, and trust are perspectives that allow evaluating e-leadership in an organization. These perspectives are briefly explained below: • Technology. It is an important perspective because e-leadership appears in a virtual environment where it is accessed through technological tools (e.g., WhatsApp, e-mail, and others). That is why it is necessary to establish the competencies, such as: learning, experience, and skills in ICTs of both the leader and members [8]. • Communication. This perspective refers to the communication between members of a virtual group. Normally, the lack of presence of the members of the team allows tension between them or simply that mutual understanding does not come to exist, resulting in misinterpreted communications that limit communication in the virtual environment [10]. • Trust. It is the perspective more complex than the previous ones because its articulation and development are very difficult in a virtual environment. There is no physical presence and other factors [3]. The perspectives mentioned help to determine the situation of e-leadership in an institution. Therefore, through these perspectives, it is possible to evaluate the situation of e-leaders and their virtual teams. Especially in a naval–military organization where its leadership is paramount due it is more rigorous in terms of compliance

174

R. P. Reyes Ch. et al.

and supervision of dispositions, especially when in recent years its members have ventured into the use of WhatsApp as a means to issue and follow up on dispositions. This is the reason why we have posed the following research question (RQ): RQ: What challenges can a naval–military organization find in its e-leadership through electronic messaging (WhatsApp) considering the perspectives of technology, communication, and trust?

3 Research Methodology Our RQ proposes an exploratory research given the limited studies regarding e-leadership in naval–military organizations. Therefore, it is appropriate to conduct a survey as a methodological strategy. We use the survey creation recommendations of Kitchenham and Pfleeger [9], which are summarized in: 1. Design of the Survey. An on-line survey was designed considering the guidelines for creating and conducting surveys proposed by [9, 11]. This allowed, on the one hand, to produce descriptive statistics, and, on the other hand, to obtain important information to answer the research question proposed for this study. 2. Construction of the survey. Based on the perspectives of e-leadership mentioned by [7, 8, 10] in the Sect. 2, the survey is constructed with 18 obligatory questions: 4 general information questions, 6 technology questions, 4 communication questions, and 4 questions of trust. The survey can be answered in an average of 10 min (depending on the feedback provided) which does not fatigue the respondents. The questions are closed so as not to give rise to erroneous interpretations or irrelevant answers. The detail of the obtaining of the questions of the survey can be found in the link: https://bit.ly/2QPLqOV. 3. Evaluation instrument. It was done in two stages. First, a researcher with experience in the subject of leadership reviewed the survey and provided feedback on its readability, understandability, and potential ambiguities. Your feedback allowed to lead to the elimination of 3 mandatory questions. Secondly, piloting was carried out with two members of the naval–military organization (not included as authors), who helped to pilot the survey before being formally applied. 4. Population and sample size. The population is oriented to the members of a naval–military organization of Ecuador in different hierarchical levels: Commanders Lieutenants and Lieutenants of the last years who have proven experience of naval–military leadership. The population is adjusted around 1360 members. To obtain the sample size (η), the following formula was applied (1): η=

(Z 2 ∗ p ∗ q ∗ N ) N ∗ e2 + Z 2 ∗ p ∗ q

(1)

E-leadership Using WhatsApp, A Challenge for Navy Organizations: An Empirical …

175

Using this formula,2 and with a confidence level of 95%, an approximate of 50 members is the sample size obtained. We consider that the sample used is representative, since they are members that have experience in leading their units on board with approximately 18 to 25 years in the institution. 5. Data collection. The data collection was carried out in October 2018. The online survey was anonymous, the same one that was announced by e-mail. In total, 46 46 = 92%, a surveys were successfully received, achieving an effective ratio of 50 ratio that is adequate for this type of study.

4 Results The results obtained from the survey are very interesting and allow us to answer our research question. Next, the characteristics of the respondents are mentioned and then our research question is answered.

4.1 General Characteristics of the Respondents As mentioned above, 46 members of the naval–military organization successfully answered the survey. Of these, 47.06% of respondents possess the rank of Commanders Lieutenant and 52.94 % of those surveyed possess the rank of Lieutenant. All the respondents have between 5 and 7 years in that grade. Of these respondents, 67.65% are weapons specialty, 8.82% are technical specialty, and 23.53% are specialty services and administrative. Of them, 41.18% have been in the naval–military operations areas.

4.2 RQ: What Challenges Can a Naval-Military Organization Find in Its E-leadership Through Electronic Messaging (WhatsApp) Considering the Perspectives of Technology, Communication, and Trust? The results allow us to respond to our RQ. However, they will be explained in the three perspectives of e-leadership, as mentioned below: 2 In

statistics, the sample size (η) is the number of subjects that make up the sample drawn from a population and are necessary for the data obtained to be representative of the population. The variable Z is a constant that depends on the level of confidence for this study is 95%. The variable p is the proportion of individuals who possess the characteristic of study in the population. This data is generally unknown and it is usually assumed that p = q = 0.5, which is the safest option. The variable q is the proportion of individuals that do not possess this characteristic, that is, it is 1-p.

176

R. P. Reyes Ch. et al.

Technology Perspective. According to He et al. [8], to exercise e-leadership, it is necessary to use the right technological tools (e.g., WhatsApp). Therefore, the importance of evaluating the experience and skills in ICTs. Table 1 shows the results, in which it is observed that 55.88% of respondents actually use WhatsApp to issue and make dispositions as part of their e-leadership within the naval–military organization. We believe that this value could increase (to 88.23% of respondents) if the 32.35% of respondents who answered sometimes are included in the future. Also, this 55.88% of respondents mentioned that they have installed WhatsApp in their cell phones 6–9 years ago. As far as the experience and technological ability is insured in the naval-military organization. To establish the effectiveness of the skill and experience using WhatsApp in eleadership, respondents were also asked if they would have created groups to their e-leadership. 91.18% of respondents mentioned having created between 0 and 50 virtual groups since they installed the tool. This percentage is similar to the percentage of respondents (88.23%) who are using WhatsApp to issue their dispositions. In this regard, we also asked about the average number of subordinates participating in their virtual groups, almost the majority of respondents (44.12%) mentioned that their virtual groups are made up of 6–10 subordinates, although sometimes they tend to exceed ten subordinates (29.41%). In summary, almost half of the respondents could be e-leading a small crew virtually. Communication Perspective. According to Liu et al. [10], the communication in e-leadership is usually very complicated due to the virtual environment of ICTs does not allow the communicative fluency that physical presence usually offers. In this regard, Table 2 shows that 50% of respondents confront communication conflicts in e-leadership when using WhatsApp. Conflicts exist because there is no confirmation of receipt of virtual messages by the subordinates or simply, the virtual messages of the e-leader are ignored. Some respondents (29.41%) mentioned that communication conflicts are usually caused by the lack of specific rules of participation at the time of creation of the virtual group or other similar causes. Also, they were asked if they have tried in any way to minimize these conflicts. The 32.35 % and the 35.29% of respondents mentioned that they use an empathic language in their messages and that they try to give immediate feedback to their issued dispositions, respectively. This allows us to deduce that respondents make a communicational effort to avoid conflicts. We also asked respondents if they use other means of communication such as telephone or person-to-person conversation when conflicts have not been mitigated through the virtual environment. The 64.71% of respondents indicated that they have done it that way. Something interesting that we were able to determine was that several of the conflicts were due to the disclosure (received/sent) of exclusive information and high priority of the virtual group by other technological means (e.g., groups WhatsApp, Facebook, among others.). Trust Perspective. Trust according to Avolio [3] is a very important factor in the virtual environment. Without a high degree of trust, relationships in the virtual environment become extremely fragile and e-leadership is condemned to be eroded. In this regard, Table 3 shows that 52.94% of respondents believe that the trust within WhatsApp should be indistinct among the members of the virtual team. Other respon-

E-leadership Using WhatsApp, A Challenge for Navy Organizations: An Empirical … Table 1 Technology perspective in e-leadership Questions Do you use “WhatsApp” to create virtual groups with your subordinates where you mention and follow up on their dispositions?

How many virtual groups with your subordinates have you created since you installed “WhatsApp” for the first time?

What is the number (average) of subordinates that usually tend to conform to their virtual “WhatsApp” groups?

How long have you installed the “WhatsApp” on your cell phone as a form of electronic messaging?

Do you believe that “WhatsApp” is or could be effective as an administrative means to give dispositions?

Are you aware of the different levels of technological competence about the use of whatsapp?

177

Answers

%

Always

55.88

Sometimes Never 0–50 virtual groups

32.35 11.76 91.18

51–100 virtual groups More than 100 virtual groups I do not remember 2–5 subordinates

0.0 0.0 8.82 20.59

6–10 subordinates More than 10 subordinates I do not remember 6 to 9 years

44.12 29.41 5.88 55.88

3 to 5 years 0 to 2 years I do not remember Yes

29.41 8.82 5.88 47.06

Maybe No Yes

14.71 38.24 85.29

Maybe No

14.71 0.0

dents (29.41%) mentioned that there should be trust from subordinates to the e-leader. This allows us to reflect on a possible distrust of the superior; and for this reason, the e-leaders considered that their subordinates could not ethically handle the disposition or information belonging to the virtual group (e.g., forwarded to other groups, etc.). Almost half of respondents (41.18%) mentioned this strange behavior of their subordinates while 44.12% of respondents are not sure if their subordinates would have handled or not ethically the information or dispositions within the virtual group. We also ask respondents if they apply confidence-building strategies. Several of them mentioned writing congratulatory messages or similar within the WhatsApp

178

R. P. Reyes Ch. et al.

Table 2 Perspective of communication in e-leadership Questions Answers During your communication in No confirmation of receipt of messages e-leadership for “WhatsApp”, have you (ignore messages by the subordinate) had any kind of conflict with subordinates due to? Due to technological limitations By the excessive use of messages to give a disposition For the misinterpretation of communications Due to the lack of empathy among members of the virtual team due to the lack of presence Due to the lack of specific participation rules when the virtual group was created (superiors/subordinates) Mistrust between virtual members I have not had conflicts If there were virtual conflicts, did you use Yes any telephone or face-to-face conversation to resolve these conflicts in your virtual team? No Maybe To achieve a more creative performance or An empathic language in the messages to carry out an efficient control and sent to the virtual group follow-up of your dispositions through “WhatsApp”, have you used? Use of an enriched written communication (emoticons) An immediate feedback by messages to avoid doubts regarding the dispositions issued Do not consider any aspect related to improving performance with my subordinates When you receive superior dispositions Only e-leader with exclusive information and high priority through “WhatsApp”, this information arrives To subordinates by other means: Internet, Facebook, other WhatsApp groups, etc.

% 50.0

20.59 20.59 20.59 11.76

29.41

8.82 29.41 64.71

29.41 5.88 32.35

2.94 35.29

29.41

44.12

58.82

E-leadership Using WhatsApp, A Challenge for Navy Organizations: An Empirical … Table 3 Trust perspective in e-leadership Questions What kind of trust do you think may exist within a virtual group with subordinates?

During your leadership in a virtual group, have you noticed that information or dispositions are not handled ethically (forwarded to other groups)?

179

Answers

%

Confidence of the virtual team towards the e-leader Confidence of the e-leader towards the virtual group Confidence among all members of the virtual team There is no distrust Yes

29.41

No Maybe How do you raise confidence and improve Use fluid and proactive communication the performance of your virtual team in in person between members before “WhatsApp”? using WhatsApp Try to minimize individualism and motivate the virtual team “Congratulations” messages or similar within the WhatsApp group I do not apply motivation Within your virtual group in “WhatsApp”, A control model (continuous monitoring have you followed up on its provisions by messages to the virtual group) using some strategy? A method of trust (relies on compliance and only awaits compliance response) None

17.65 52.94 17.65 41.18

14.71 44.12 47.06

20.59 47.06 20.59 55.88

17.65 26.47

group (47.06%) or simply they maintain a fluent and proactive communication before interacting with WhatsApp (47.06%). Likewise, respondents were asked if they follow up on dispositions. The results show that 55.88% of respondents use a control model based on the continuous monitoring of virtual messages while a 17.65% considers leaving the subordinate the compliance.

5 Threats to Validity The possible threats to the validity of this study are discussed according to the aspects mentioned by Creswell [5], as indicated below: • Conclusion Validity. It refers to the degree to which the conclusions obtained are derived from the data [5]. Only possible threat is the lack of representativeness of

180

R. P. Reyes Ch. et al.

the chosen sample. This threat does not operate, since it was chosen respondents who possess a high level of experience and leadership practice within the naval– military organization. • Internal Validity. This type of validity refers to the absence of other influential variables in the study, except those that are being studied [5]. In the online survey conducted, the appearance of other influential variables should only appear due to errors in the formulation or understanding of the survey questions. To mitigate, an introduction to the survey is provided to familiarize the respondent with its purpose. On the other hand, as a way to guarantee the veracity of the data, we have given anonymity to the responders. • Construct Validity. This type of validity refers to the relationship between a theory behind the observation [5]. For this, a study of the previous literature about the e-leadership mentioned in the Sect. 2 was carried out. Likewise, the survey was evaluated by an experienced researcher and piloted by two members (non-authors) of the organization. • External Validity. External validity refers to the generalization of results [5]. The results obtained are valid externally because the selected sample size is very representative in the naval–military organization.

6 Conclusions, Discussion and Future Work The results obtained for our RQ based on the technology, communication, and trust perspectives proposed by [7, 8, 10] show that e-leadership using WhatsApp in a naval–military organization maintains conflicts that are not easy to solve in the virtual environment, it is even possible to say that they are difficult to solve in a traditional leadership. The conflicts we refer to are those that appear between perspectives: (1) technology–communication, (2) communication–trust, and (3) trust–technology. Regarding the first case, conflicts appear when the dispositions sent by virtual messages are ignored. Reason why, the e-leader tries to make communication efforts to mitigate these conflicts such as the telephone or in person. The second case refers to conflicts for distributing exclusive information and high priority of the virtual group by other electronic means (e.g., Facebook, other virtual groups, etc). The latter is very complicated to mitigate by the e-leader given the degree of viralización of information (without restriction) that WhatsApp allows. And regarding the third case, conflicts are generated by the little effective trust that virtual environments allow, since it is shown that subordinates do not usually handle ethically the information and dispositions in their virtual group. Practically, the e-leadership on electronic messaging tools like WhatsApp are still complicated to articulate. Several of these conflicts that we mentioned were detected by their authors [7, 8, 10] but in other virtual environments such as: videoconference or e-mail. Therefore, before a military–naval organization wishes to complement its traditional leadership with e-leadership, it is necessary to develop an attitude framework based on institutional values and principles. Possibly, modeling an ethical

E-leadership Using WhatsApp, A Challenge for Navy Organizations: An Empirical …

181

e-leadership over electronic messaging (WhatsApp) is the answer to the conflicts found; maybe using a proposal similar to the one mentioned by Bass [4] but oriented to virtual environments. Finally, we consider investigating that could be happening in other organizations similar to the study as: land–military and air–military organizations. For this reason, we set out as future work, establish the effect or effects that e-leadership on WhatsApp could have on these types of organizations. Acknowledgements This work was supported by the Academia de Guerra Naval - Armada del Ecuador and Dirección General de Educación y Doctrina and sponsored by Universidad de las Fuerzas Armadas ESPE.

References 1. Aggarwal, A.: E-leadership-a new and modern style of leadership. IJAME (2018) 2. Avolio, B.J., Kahai, S., Dodge, G.E.: E-leadership: implications for theory, research, and practice. Leadersh. Q. 11(4), 615–668 (2000) 3. Avolio, B.J., Kahai, S.S.: Adding the e to e-leadership: how it may impact your leadership. Organ. Dyn. (2003) 4. Bass, L., Israel, M.S.: Modeling ethical leadership: being an ethical leader means modeling principles of self-awareness, reflective practice, transparency, and ethical behavior. In: Developing Ethical Principles for School Leadership, pp. 29–49. Routledge (2018) 5. Creswell, J.W., Clark, V.L.P.: Designing and Conducting Mixed Methods Research. Sage Publications, Thousand Oaks (2017) 6. Esguerra, G.A., Contreras, F.: Liderazgo electrónico, un reto ineludible para las organizaciones de hoy. Estud. Gerenc. 32(140), 262–268 (2016) 7. Fan, K.T., Chen, Y.H., Wang, C.W., Chen, M.: E-leadership effectiveness in virtual teams: motivating language perspective. Ind. Manag. Data Syst. 114(3), 421–437 (2014) 8. He, R., et al.: E-leadership strategy in virtual organizations and virtual teams (2008) 9. Kitchenham, B.A., Pfleeger, S.L.: Personal opinion surveys. In: Guide to Advanced Empirical Software Engineering, pp. 63–92. Springer, Berlin (2008) 10. Liu, C., Ready, D., Roman, A., Van Wart, M., Wang, X., McCarthy, A., Kim, S.: E-leadership: an empirical study of organizational leaders virtual communication adoption. Leadersh. Organ. Dev. J. 39(7), 826–843 (2018) 11. Punter, T., Ciolkowski, M., Freimut, B., John, I.: Conducting on-line surveys in software engineering. In: Proceedings of 2003 International Symposium on Empirical Software Engineering, ISESE 2003, pp. 80–88. IEEE (2003) 12. Van Wart, M., Roman, A., Wang, X., Liu, C.: Integrating ICT adoption issues into (e-) leadership theory. Telemat. Inform. 34(5), 527–537 (2017)

Part VI

Planning, Economy and Logistics Applied to Defense

Career Anchors for the Portuguese Army’s Volunteers and Contract Personnel: Using the Career Orientations Inventory Lúcio Agostinho Barreiros dos Santos and Maria Manuela Martins Saraiva Sarmento Coelho Abstract This study aims to identify and discuss the career anchors of the Portuguese Army’s volunteers and contract personnel, in order to outline a “professional profile” that can help the organization improving its decision-making process, in terms of management practices and working and living conditions. The study addresses the internal dimension of the concept of career using a three-pronged approach—regulatory, theoretical, and empirical—to find out whether career anchors are determinants for the choice to join the military, as well as for the length of time served. The study uses quantitative and qualitative strategies, with emphasis on Schein’s Career Orientations Inventory (COI). The survey had a valid final sample of 1821 members from the Portuguese Army. The study’s findings suggest a professional profile based on a dominant career anchor, “Lifestyle”, which, combined with complementary anchors (e.g. “Security/Stability”), and provide a qualitative explanation for the participants’ choices. Keywords Military volunteers and contract personnel · Career anchors · Career Orientation Inventory · Recruitment · Retention

1 Introduction This study examines the career orientations of the Portuguese Army’s volunteer and contract personnel to discover how this knowledge could help solve and/or mitigate the current constraints to the model’s sustainability. About 15 years after Portugal formally suspended its compulsory military service (CMS) in the wake of it having been declared unconstitutional their requirements L. A. B. dos Santos (B) · M. M. M. S. S. Coelho Research and Development Center, Military University Institute, Lisbon, Portugal e-mail: [email protected] M. M. M. S. S. Coelho e-mail: [email protected] Lusíada University, Lisbon, Portugal © Springer Nature Singapore Pte Ltd. 2020 Á. Rocha and R. P. Pereira (eds.), Developments and Advances in Defense and Security, Smart Innovation, Systems and Technologies 152, https://doi.org/10.1007/978-981-13-9155-2_16

185

186

L. A. B. dos Santos and M. M. M. S. S. Coelho

[1] in 2004, implementing a strictly voluntary service [2–4], the military’s ability to recruit and maintain the personnel needs has been progressively declining. These difficulties are currently on the political agenda and have been addressed by multiple studies conducted by the Ministry of National Defence, by the branches of the Portuguese Armed Forces (AAFF), and by several independent researchers (e.g. [5–12]). In the light of this, the paper attempts to diagnose the causes of the main difficulties and devise ways to mitigate their negative consequences. Despite the studies conducted in Portugal during recent years (e.g. [13, 14]), it was only recently that some authors [9–11, 13] began to examine the AAFF’s management practices and the living and working conditions of citizens, who join the military as volunteers for a maximum period of seven years [2, 3]. However important, this awareness of some intrinsic factors in the organization’s functioning does not seem to be sufficient to guarantee the success of the military professionalization model, as the current situation is critical and calls into question the model’s sustainability. Therefore, it remains necessary to explore, from a theoretical and practical perspective, the reasons that hinder recruitment and retention, especially in terms of “internal career” orientation [15, 16], a concept that will be explained further on. The first phase of the study is focused on the Army branch, and its objective was to identify and discuss the career anchors of the Portuguese Army’s volunteers and contract personnel. In order to outline a “standard professional profile” which the AAFF can use as a guide to select the most appropriate management practices and create suitable working and living conditions, helping members to achieve their “internal career” orientations as well as encouraging citizens to join the military. The research question is: “What are the main ‘internal career orientations’ of the Portuguese Army’s volunteer and contract personnel, and how do they relate to the main reasons that lead citizens to join and remain in the military?”. As for the quantitative dimension of the research, an adapted version of Schein’s Career Orientations Inventory (COI) [15] was used to identify the main “career anchors” of the military volunteers and contractors” of the Portuguese Army. The qualitative dimension of the empirical study can provide a better understanding of members’ professional profile, as well as increase awareness regarding the adoption of management practices that can be used to improve recruitment and retention. The article is divided into five sections. After this introductory section, the second section briefly presents the legal framework of military service and the theoretical framework, examining the concept of career anchors as a determinant for members’ professional choices and the nature and type of link that exists between them and organization. The third section outlines and explains the methodology, and the fourth section presents and discusses the study’s main findings. The fifth and final section presents conclusions based on the main findings and answers the general objective and the research question, also providing suggestions for future research.

Career Anchors for the Portuguese Army’s Volunteers …

187

2 Legal and Theoretical Framework 2.1 Legal Framework The legal framework, an essential part of this research, concerns the service model for volunteer and contract military personnel, its conceptual and legal organization, and its operationalization. The model is defined in the legislation as voluntary service and contract service (VS/CS) and is governed by a set of specific law, which include: the Military Service Law, the Military Service Law regulations, and the regulations of the military Incentive Scheme [2–4]. The concepts of recruitment and retention are especially relevant for this legal framework. Military recruitment can refer to both voluntary and contract schemes, and consists of the set of operations carried out to capture human resources, that is, citizens above the age of 18 who are willing to serve in the AAFF [2]. Retention refers to the need to maintain contract personnel and their expertise in the AAFF, to meet specific organizational goals and obtain an appropriate return on the organization’s investment on education and training [3, 16]. If the “expectation of service completion” is not met, members could have to pay damages proportional to their training period and its associated costs. Currently, the military voluntary and contract service model of the Portuguese AAFF provides only short-term careers: if members fulfil certain requirements after one year of voluntary service, they can become “regular” contract personnel for a period of six years [2, 3]. Under certain conditions, members may also opt for a medium-term career by entering a “special contract scheme” that allows them to extend their contractual link up to 18 years of service. Despite a specific law having been recently approved to allow for this type of career mobility [17], this option has not yet been regulated and is in the process of regulatory expansion. In addition to the possibility of career mobility between regular contract schemes and the possibility to transfer to a longer-term contract scheme, members can also apply to the career staff of the AAFF [2, 18], which provides them a long-term career until retirement age. In 2000, after the enactment of the new Military Service Law [2], an incentive scheme [4] for military personnel serving under VS/CS was approved, which proved instrumental to capture and retain, staff, as well as to support their socio-professional transition to the labour market. Due to fiscal difficulties, this law has been revised and reduced in its scope and possibilities. The increasing difficulties recruiting and retaining staff are compounded by changes in the sociocultural, demographic, and generational context. Therefore, a new law was recently approved expanding the incentive scheme for military volunteers and contract personnel, which now mentions the special military contract scheme [19].

188

L. A. B. dos Santos and M. M. M. S. S. Coelho

2.2 Theoretical Framework The theoretical framework looks at two distinct and complementary perspectives for the concepts of career, external career, and internal career, as suggested by Derr [20] and Schein [15], in order to discuss both the organization’s job offer and members’ responsibilities. A career’s “external” component refers to the organization’s needs, its functioning and culture, its management practices and constraints, and the professional development opportunities it makes available to its employees. A career’s “internal” perspective has an intra-individual nature and is the result of a construction based on values, goals, ambitions, motivations, and individual skills, which are consolidated (by developing self-knowledge) and based on reflection and professional experience. This construction, which Edgar Schein called ‘career anchors’, reveals a rather stable behaviour pattern, when it comes to making important career decisions based on self-knowledge [15, 21]. After an initial classification using five types, Schein divided his career anchors into eight final types, which are associated with certain identity components (e.g. [15, 21–23]): (i) Technical/Functional Competence (TF) refers to specific skills and areas of expertise; (ii) General Management Competence (GM) refers to cross-sectional management skills and a markedly vertical career matrix; (iii) Autonomy/Independence (AU) concerns personal work organization (when and how to work) and is based on flexibility and individual freedom of action; (iv) Security/Stability (SE) emphasizes job and financial security, job stability, and remaining in the same organization and functional area; (v) Entrepreneurial Creativity (EC) reflects an entrepreneurial profile that favours innovation and creativity in the management of objectives and projects; (vi) Service/Dedication to a Cause (SV) is linked to the usefulness of the work and the opportunity to help others and solve societal problems and represents an orientation for the usefulness of the work performed; (vii) Pure Challenge (CH) refers to complex challenges and obstacles that are difficult to overcome and translates into accepting a high level of risk and seeking novelty and variety in the activities performed; and (viii) Lifestyle (LS) represents an appropriate balance and harmony between professional and personal/family requirements, between one’s career and private life. Therefore, the concept of career anchors is a determinant for professional choices and reflects the nature of the link between an individual and his or her organization. Regarding that link, two additional theoretical notions were selected for review and exploration in this study—organizational career and protean career—which are often differentiated by the nature of the link between individual and organization. Organizational careers represent a traditional approach and are conceptually close to a markedly vertical matrix, are defined by law, are largely controlled by the organization, and provide job/employment stability and security. Protean careers have different elements, which are often the opposite of those in organizational careers. They allow for greater fluidity and spontaneity and are therefore less institutionalized since individuals have the power to make decisions and control most of the variables,

Career Anchors for the Portuguese Army’s Volunteers …

189

which are subject to his or her “individual will”, allowing for greater functional and inter-organizational mobility [24–29]. Despite Schein’s [21] suggestion that each person has one anchor that defines their strengths, goals, motivations, and values, several studies have tested the eight types and have obtained different systematizations, which indicates that, while a dominant anchor may exist, individual career orientations are a dynamic process and a professional profile is consolidated over time, resulting from a combination of up to three anchors, according to Feldman and Bolino [30]. On the other hand, the construct’s empirical limitations have led to other studies that attempt to validate and/or refine Schein’s construct by reducing the number of clusters (e.g. [31]) or, in contrast, by adding more clusters (e.g. Security/Stability; Entrepreneurial-Creativity), such as those conducted by Danziger, Rachman-Moore, and Valency [32]. Other studies even suggest additional anchors to meet twentyfirst-century requirements such as “Employability” and “Spiritual Dimension” [25]. Despite these various updates and adaptations, the foundations of the original construct proposed by Schein seem to have remained current due to the instrument’s internal consistency. However, in order to account for sociocultural, economic, generational, and other contextual factors, researchers should use the COI in combination with other instruments to ensure more robust results. This study combines Schein’s eight career anchors with a qualitative dimension. Another concept addressed in this study, which is fully applicable to military contexts, is that of psychological contract, which refers to employee perceptions about the organization’s obligation to reciprocate the work performed [33, 34]. These perceptions are based on explicit or implicit promises and may result from more or less formal agreements with the organization, or simply from the expectations of individuals regarding the organizational context and the interactions with some of its elements [35]. The type of relationship that is established between individual and organization and the degree to which the expectations match reality can either strengthen or weaken the psychological contract and can ultimately lead to its dissolution (breach). This could cause employees to become less involved in activities outside their regular duties, leading to uncertainty and dissatisfaction and increasing employee exits from the organization [36–38]. Some common reasons for a breach of the psychological contract may stem from (1) inadequate management practices, (2) lack of support from the organization and/or management regarding external issues, and (3) lack of equity among employees [35]. A review of the sociocultural and professional characteristics of the Y and Z generations was conducted to investigate the implications of their specific features for recruitment and retention. Despite their relevance, due to space constraints, the results of the review are not included in this article. Based on the regulatory and theoretical review, a chart was prepared to illustrate the guiding structure for the study (Fig. 1).

190

L. A. B. dos Santos and M. M. M. S. S. Coelho

Fig. 1 Guiding structure for the study: summarizes the theoretical framework and the main concepts reviewed

This guiding structure includes the following elements: – The military VS/CS model and its main dimensions—recruitment, retention, and reintegration—were drawn from the review of the legal framework and are a key element of the research framework; – The structuring elements of the study’s theoretical framework are the concepts of internal career (individual perspective), which includes “career anchors” and external career (organizational context), in which opportunities and constraints overlap with individual career and life ambitions and goals; – The link between the perceptions about internal career, external career, and “psychological contract”, which complements the “formal contract”, may help explain the main career trends among the Portuguese Army’s volunteers and contract personnel; – Furthermore, the career profiles may help explain members’ behaviour towards the AAFF: the reasons that lead them to join, the reasons for staying until the end of their contract, and the motivations behind early dropouts; – Similarly, the career anchors and psychological contract are key elements in the definition of the concept of career/professional success, which is also addressed in the theoretical framework and is part of the study’s research object. The organizational aspects correspond to the “external career” concept, in which management policies and practices serve to align organizational and individual objectives, expectations, opportunities, and values (“internal career”). In military

Career Anchors for the Portuguese Army’s Volunteers …

191

contexts, these policies and practices are relatively standardized, although there are some differences according to professional groups and areas.

3 Methodology and Method This section summarizes the methodological procedures and identifies the population, the sample, the gather data instruments, and the analysis techniques.

3.1 Research Methodology The research is based on a case study [39] that investigates the applicability and value of career anchors to the Portuguese Army’s volunteer and contract personnel during 2018. The study used a mixed research strategy (quantitative and qualitative) developed simultaneously in the context of the proof and of the discovery. The findings were obtained by combining the processes of induction and construction of meaning, in which the researchers assumed an ontological and epistemological posture closely related to constructivism and interpretivism [40–42].

3.2 Research Method: The Sample and Data Processing Techniques The study used documentary data based on the legislation that regulates the military VS/CS model, as well as empirical data obtained from a questionnaire that consisted of 40 close-ended questions and four open-ended questions. The target public of the questionnaire was volunteers in contract regime of the Portuguese Army belonging to several professional categories. The questionnaire used in this study (closed-ended questions) is adapted from Dutra and Albuquerque questionnaire [22]. However, this one was also adapted from Schein [15, 21]. In order to apply our questionnaire to the Portuguese military context, we did three procedures: firstly, an adaption to the military population, secondly we did a validation with experts on military human resources, and thirdly, we did a pretest with 11 Portuguese military in regime of contract volunteers. The final version of the questionnaire was online and the link was sent by e-mail to the Army, Navy, and Air Force, being available for filling from March until October 2018. The answers were automatically introduced in an Excel database.

192

L. A. B. dos Santos and M. M. M. S. S. Coelho

The study’s qualitative dimension (the questionnaire’s open-ended questions) was combined with the COI to provide a deeper understanding of the research object [43]. We analyse the respondents’ answers regarding (i) the main reasons that led them to join the military; (ii) the reasons for remaining until the end of their contract; (iii) the motivations that lead to early dropouts; and (iv) the concept of career or professional success. We got 2474 answered surveys, but only 1832 had all the questions answered; however, some were invalid, what correspond to 1821 valid surveys. Table 1 shows the distribution of the final sample by professional category, gender, and type of service. The sample is robust and can be considered representative of the volunteers and contract personnel of the Portuguese Army, which represent 26.78% of the Army’s staff on 31 July 2018 (1821 respondents in a total of 6801 military). This percentage is also close to the values of the population under analysis when compared with the sample segmentation. Crossing the age level with the professional category reveals that most participants in the sergeants and enlisted categories are between 21 and 26 years old (70.56% of sergeants and 70.81% of enlisted personnel), while most participants in the officers category are between 24 and 29 years old (75.00%). The methodological procedure used to process the answers to the closed-ended questions was a descriptive statistical analysis, namely averages, percentages, and frequencies. In the near future, we will perform bivariate and multivariate statistics such as principal components analysis, cluster analysis, and inferential analysis. The qualitative component of the research (the answers to the questionnaire’s open-ended questions) was processed through content analysis using a specialized software, SPSS Text Analysis for Surveys (STAFS), version 4.0.1. The analysis led to the addition of some emerging types to the initial types, which correspond to Schein’s eight career anchors.

Table 1 Sample characterization by professional category, gender, and type of service Professional category Officers

Sum

Contract service

Total male

Volunteer service

Contract service

Total female

Volunteer service

Contract service

Total 120

Value

2

89

91

0

29

29

2

118

%

2.20%

97.80%

100.00%

0.00%

100.00%

100.00%

1.67%

98.33%

6.59%

4

144

148

0

32

32

4

176

180 9.88%

Sergeants Value Enlisted

Volunteer service

%

2.70%

97.30%

1.00%

0.00%

100.00%

100.00%

2.22%

97.78%

Value

125

1248

1373

14

134

148

139

1382

1521

%

9.10%

90.90%

100.00%

9.46%

90.54%

100.00%

9.14%

90.86%

83.56%

Value

131

1481

1612

14

195

209

145

1676

1821

Career Anchors for the Portuguese Army’s Volunteers …

193

4 Data Presentation and Discussion of Results 4.1 Quantitative Research An overall analysis of the results of the quantitative component reveals that the eight career anchors defined by Schein are present in the sample, although in different proportions. The mean values show that all three professional categories (officers, sergeants, and enlisted) and both genders have a dominant career anchor (“Lifestyle”), as well as secondary career anchors (“Stability/Security”, “Technical/Functional”, “Service and Dedication to a Cause”, and “Pure Challenge”), which obtained relatively high values (Tables 2 and 3). This analysis can be fundamental for the definition of a standard professional profile (career orientations) of the citizens currently serving in the Portuguese Army as volunteers or under contract.

Table 2 Mean value of career anchors by military category Types of career anchors Technical/Functional competence (TF)

Sergeants

Enlisted

Value

Officers Order

Value

Order

Value

Order

4.86

2nd

4.74

2nd

4.56

3rd 8th

General management competence (GM)

3.58

8th

3.91

8th

3.63

Autonomy/Independence (AU)

3.79

7th

4.02

7th

3.97

7th

Security/Stability (SE)

4.73

3rd

4.60

5th

4.58

2nd

Entrepreneurial creativity (EC)

3.85

6th

4.02

6th

4.08

6th

Service/Dedication to a cause (SV)

4.58

4th

4.61

4th

4.49

4th

Pure Challenge (CH)

4.52

5th

4.66

3rd

4.41

5th

Lifestyle (LS)

4.99

1st

5.12

1st

4.94

1st

Table 3 Mean value of career anchors by gender Types of career anchors

Male

Female

Value

Order

Value

Technical/Functional competence (TF)

4.60

2nd

4.60

Order 4th

General management competence (GM)

3.66

8th

3.61

8th

Autonomy/Independence (AU)

4.00

7th

3.69

7th

Security/Stability (SE)

4.59

3rd

4.67

2nd

Entrepreneurial creativity (EC)

4.10

6th

3.75

6th

Service/Dedication to a cause (SV)

4.49

4th

4.62

3rd

Pure Challenge (CH)

4.44

5th

4.53

5th

Lifestyle (LS)

4.96

1st

5.02

1st

194

L. A. B. dos Santos and M. M. M. S. S. Coelho

The dominant career anchor obtained much higher values than the second most important anchor, although this difference is less marked in the officer’s category. The relative importance of secondary anchors varies according to professional category and gender: “Technical/Functional Competence” is the second most important career anchor for officers and sergeants, as well as for men, and “Security/Stability” is the second career anchor for enlisted personnel and for women. For the latter, the anchor “Service/Dedication to a Cause” is also important. The least relevant career anchor for all categories and genders is “General Management Competence”, followed by the anchor “Autonomy/Independence”. The career anchors “Service/Dedication to a Cause” and “Pure Challenge”, which are characteristic of the military environment, are moderately important for military volunteers and contract personnel, and are slightly higher for the sergeants category and for female members. It is relevant to conclude that women score high in the main anchors, and on the other hand, they score low in the less important anchors, which can indicate a more developed self-concept in terms of professional orientation.

4.2 Qualitative Research The results obtained from the analysis of the qualitative component of the questionnaire confirm that career anchors are generally present in all phases during the service term in the AAFF—recruitment, retention, and transition to the labour market—although some anchors (General Management Competence, Autonomy/Independence, and Entrepreneurial-Creativity) obtained insignificant values, which confirms the general trend of the results of the quantitative analysis. Figures 2, 3, 4, and 5 summarize the results of the content analysis of the four open-ended questions. Only results that obtained 15 or more occurrences were included in the type networks, regarding the initially defined types (which means that not all types associated with the anchors are shown in the images) and emerging types. The size of the nodes and the thickness of the line segments that link them represent, respectively, the importance of each type within each topic/question and the strength of the relationship between the types.

Fig. 2 Reasons for joining the military

Career Anchors for the Portuguese Army’s Volunteers …

195

Fig. 3 Reasons for remaining in the military

Fig. 4 Reasons for early separation

Fig. 5 Concept of career/professional success

First and Second Open-Ended Question: Reasons for Joining and Remaining in the Military Reasons for joining the military. The types of career anchors Pure Challenge (CH) (505 occurrences) and Service/Dedication to a Cause (SV) (261 occurrences), as well as the relationship between them, obtained more relevant values than the other types, which confirms the results obtained by earlier studies done by AAFF in 2012,

196

L. A. B. dos Santos and M. M. M. S. S. Coelho

2016, and 2017 [7–9]. Security/Stability (SE) is a key type for the process of joining the military, showing a significant correlation with the emerging type Employability. In military context, this last type represents the choice to remain in the organization under VS/CS or to seek further qualifications and/or apply to the career staff. Reasons for joining the military. The type of career anchor Security/Stability (SE) (750 occurrences) is one of the most important reasons given by respondents for remaining in the military and is associated with the types Pure Challenge (CH) (240) and Service/Dedication to Cause (SV) (119), as well as with the type of Employability (138). Work Conditions and VS/CS Incentive Scheme are emerging types closely correlated to the quality of the work environment and the possibility to benefit from the incentive scheme provided by law. Third Open-Ended Question: Reasons for Early Separation The reasons that lead members to separate from the military before completing the minimum length of service defined by law, although they still wish to maintain their “employee” status (Security/Stability) (650 occurrences), stem from dissatisfaction regarding certain individual needs. When faced with Work Conditions (310), which they find inadequate, members look for job and financial stability outside the AAFF, where they believe they can achieve better life conditions (Lifestyle (LS) (270). In most cases, the Employability type (130) refers to the possibility of joining the military for the advantages, when applying to a position in the security forces or to certain positions in the Portuguese civil administration (provided for by law). The VS/CS Incentive Scheme could be especially relevant in these cases. Fourth Open-Ended Question: Concept of Career/Professional “Success” The values obtained in the types of career anchors Security/Stability (271 occurrences) and Lifestyle (103), as well as the correlation between them, show that the respondents’ concept of professional success confirms the overall relevance of the corresponding anchors in the quantitative component of the analysis. Here, the first type correlates to all the others. The emerging type Achievement (161) is the second most important type for the concept of success and correlates with each of the types obtained from the responses to the fourth open-ended question. The other emerging types that explain respondents’ perceptions are Employability (57), Recognition (78), Military Career (101), Career Development (69), and Successful (46).

5 Interconnection Between Quantitative and Qualitative Results This section presents the main conclusions of the methodological procedure and provides an answer to the general objective and research question. The study’s general objective was to identify and discuss the types of career anchors of the Portuguese Army’s volunteers and contract personnel. The purpose is

Career Anchors for the Portuguese Army’s Volunteers …

197

to outline a “standard professional profile” which can be used as a guide to improve the AAFF’s management practices and create suitable working and living conditions, allowing individuals to achieve their “internal career” orientations and encourage them to join the military. The study used a mixed research strategy (quantitative and qualitative) developed simultaneously in the context of the proof and in the context of the discovery. The study used documentary and empirical data obtained from a questionnaire consisting of close- and open-ended questions, which was delivered to the Portuguese Army’s volunteers and contract personnel, resulting in a sample of 1821 respondents. The results of the quantitative component of the study revealed a dominant career anchor “Lifestyle”, as well as several secondary anchors, namely “Stability/Security”, “Technical/Functional”, “Service and Dedication to a cause”, “Pure Challenge”, which vary slightly according to some demographic variables. The results of the qualitative component seem to indicate that anchors are present during all phases of the service term in the AAFF (recruitment, retention, and socioprofessional transition to the external labour market), confirming the general hypothetical perspective presented in the theoretical framework. Crossing the identified anchors with the main reasons for joining and remaining in the AAFF provided a better understanding of the standard professional profile of VS/CS personnel and how it could influence the adoption of management practices that can be used to improve recruitment and retention. The answer to the research question “What are the main internal career orientations of the Portuguese Army’s volunteer and contract personnel, and how do they relate to the main reasons that lead citizens to join and remain in the military?” was obtained from two factors: – The definition of the standard “Professional Profile” of the Portuguese Army’s VS/CS personnel and – The organization’s role in helping its staff to meet their “Internal Career” needs.

6 Final Conclusions This section discusses the study’s contributions, usefulness, and limitations, as well as it makes recommendations and suggestions for future studies. According to the “Professional Profile”, VS/CS Personnel, the Results Suggest – Seek to balance their personal, family, and career/professional needs, which career anchor is the “Lifestyle”. – Prefer to work in specific areas of expertise and/or technical functions, where they can use the qualified skills they acquired previously or developed in a military context, which career anchor is “Technical/Functional Competence”. – Seek stability, work benefits, and financial security through long-term employment in an organization, preferably, and if possible, in a functional and geographic area

198

L. A. B. dos Santos and M. M. M. S. S. Coelho

that allows them to be close to their family/close social group, which career anchor is “Security/Stability”. – Are (moderately) motivated by the symbolic, vocational, and operational aspects of military service, by the utility of the work and its role in helping to solve societal problems, which career anchor are “Pure Challenge, and Service/Dedication to a Cause”. – Find typical management and general management activities and responsibilities, which career anchor is “General Management Competence”, somewhat unappealing. Regarding the Organization’s Role, the Results Suggest – That the AAFF should accept the standard professional profile of its VS/CS personnel and help them achieve their anchors. – It is imperative to implement management practices that can operationalize the professional profile obtained and provide suitable working and overall living conditions for the AAFF’s volunteers and contract personnel. – Due to some of the differences that emerged regarding respondents’ perceptions about secondary career anchors, the skills of the AAFF’s volunteers and contract personnel should be managed in a more flexible and personalized way, in order to meet the organization’s objectives while simultaneously meeting members’ “internal career” needs. The usefulness of this study is based on a deeper knowledge of the career orientations of the Portuguese Army’s volunteers and contract personnel. This will facilitate the adoption of management practices that are better suited to members’ professional profile, aligning organizational and individual objectives. The observed “career anchors” could be also suitable to explain the reasons why someone either decide or not to join the army. Therefore, the suggestions for future research include the expansion the quantitative methodological procedure and a denser qualitative analysis, as well as the replication of the research scope to the other branches of the Portuguese AAFF. It would be very interesting to analyse what are the differences and the similarities among the “career anchors” EU countries, when the current political agenda is about a common Army and a common European defence.

References 1. Law No. 1/97, of 20 September: Approves the Fourth Revision of the Constitutional Law. Assembly of the Republic. Diary of the Republic, Lisbon (1997) 2. Law No. 174/99 of 21 September: Approves the Military Service Law. Assembly of the Republic. Diary of the Republic, Lisbon. https://dre.pt/application/dir/pdf1sdip/1999/09/221A00/ 65416550.pdf (1999) 3. Decree-Law No. 289/2000 of 14 November: Approves the Regulation of the Military Service Law. Portuguese Government. Diary of the Republic, Lisbon. https://dre.pt/application/dir/ pdf1sdip/2000/11/263A00/64256438.pdf (2000)

Career Anchors for the Portuguese Army’s Volunteers …

199

4. Decree-Law No. 320-A/2000 of 15 December: Approves the Regulation of Incentives to Service in RV/RC (amended by Decree-Law No. 118/2004 of 21 May and by Decree-Law No. 320/2007 of 27 September). Portuguese Government. Diary of the Republic, Lisbon. https:// dre.pt/application/dir/pdf1sdip/2000/12/288A01/00020011.pdf (2000) 5. Bragança, N., Santos, L.: Study 2 – Desafios do recrutamento para o futuro em relação aos cidadãos para a prestação de serviço militar nos regimes de voluntariado e de contrato das Forças Armadas portuguesas [Future challenges in the recruitment of volunteer and contract service members for the Portuguese Armed Forces]. In: Santos, L., Sarmento, M., Fachada, C. (Coord.). Prestação de serviço militar em regime de contrato nas Forças Armadas portuguesas. Dificuldades e desafios num contexto em transformação. [Contract service in the Portuguese Armed Forces. Difficulties and challenges of a changing context.] ARES Collection, vol. 22, pp. 53–103. Military University Institute, Lisbon (2018) 6. Branco, A., Santos, N.: Study 3 – Predisposição para ingresso nas Forças Armadas: Um estudo sobre ações de recrutamento dirigido. [Predisposition to join the Armed Forces: A study on directed recruitment.] In: Santos, L., Sarmento, M., Fachada, C. (Coords.). Prestação de serviço militar em Regime de Contrato nas Forças Armadas Portuguesas. Dificuldades e desafios num contexto em transformação. [Contract service in the Portuguese Armed Forces. Difficulties and challenges of a changing context.] ARES Collection, vol. 22, pp. 105–15. Military University Institute, Lisbon (2018) 7. Centre of Applied Psychology of the Army: Summary analysis of Training Dropouts and Personnel Retention. Portuguese Army, Lisbon (2016) 8. General Directorate of Military Personnel and Recruitment: Study on the reasons for VS/CS personnel dropouts in the Armed Forces. Ministry of National Defence, Lisbon (2012) 9. General Directorate of National Defence Resources: Study on the Sociodemographic Characteristics and Organizational Satisfaction of Volunteer and Contract personnel from the three branches of the Armed Forces. Scientific coordinator – Professor Helena Carreiras. Ministry of National Defence, Lisbon (2017). (Restricted document) 10. Rijo, F., Marreiros, J., Mairos, J., Paquete, O.: Employee Retention in the Military. J. Mil. Sci. VI(1), 333–356. https://www.ium.pt/cisdi/index.php/pt/publicacoes/revista-de-cienciasmilitares (2018). Accessed 9 Oct 2018 11. Santos, L., Sarmento, M.: Study 1 – Evolução do modelo de prestação de serviço militar nas forças armadas portuguesas em regime de contrato. Análise crítica e formas de intervenção. [Evolution of the contract service model of the Portuguese armed forces. Critical Analysis and intervention measures.] In: Santos, L., Sarmento, M., Fachada, C. (Coord.). Prestação de serviço militar em regime de contrato nas Forças Armadas portuguesas. Dificuldades e desafios num contexto em transformação. [Contract service in the Portuguese Armed Forces. Difficulties and challenges of a changing context.] ARES Collection, vol. 22, pp. 19–51. Military University Institute, Lisbon (2018) 12. Santos, L.: Reflections Arising from military service under contract in the portuguese armed forces: functional and social and citizenship perspectives. J. Mil. Sci. (online) III(1), 331–362. https://cidium.ium.pt/docs/artigos/Artigo_123.pdf (2015) 13. Batista, M., Ribeiro, R.: Study 7 – Carreira Militar – Motivações e expetativas dos militares do regime de Contrato. [Military Career – Motivations and expectations of military contract personnel.] In: Santos, L., Sarmento, M., Fachada, C. (Coords.). Prestação de serviço militar em Regime de Contrato nas Forças Armadas Portuguesas. Dificuldades e desafios num contexto em transformação. [Contract service in the Portuguese Armed Forces. Difficulties and challenges of a changing context.] ARES Collection, vol. 22, pp. 289–316. Military University Institute, Lisbon (2018) 14. Cunha, F., Costa, P.: Study 3 – Quadro de Incentivos à prestação do serviço militar como estímulo ao recrutamento e à retenção. [The role of the military Incentive Scheme in promoting recruitment and retention.] In: Santos, L., Sarmento, M., Fachada, C. (Coords.). Prestação de serviço militar em Regime de Contrato nas Forças Armadas Portuguesas. Dificuldades e desafios num contexto em transformação. [Contract service in the Portuguese Armed Forces.

200

15. 16.

17.

18.

19.

20. 21. 22. 23.

24. 25. 26. 27. 28. 29. 30. 31.

32.

33. 34. 35. 36. 37.

L. A. B. dos Santos and M. M. M. S. S. Coelho Difficulties and challenges of a changing context.] ARES Collection, vol. 22, pp. 193–238. Military University Institute, Lisbon (2018) Schein, E.: Career Anchors: Discovering Your Real Values. Pfeiffer & Company, University Associates, San Diego (1990) RTO/NATO: Recruiting and Retention of Military Personnel. Technical Report of RTO (Research Task Group), HFM-107, NATO Research & Technology Organization. https://www. nato.int/issues/women_nato/Recruiting%20&%20Retention%20of%20Mil%20Personnel.pdf (2007) Decree-Law No. 75/2018 of 11 October: Second amendment to Decree-Law No. 130/2010 of 14 December, approving the special military contract scheme. Portuguese Government. Diary of the Republic, Lisbon (2018) Decree-Law No. 90/2015 of 29 May: Approves the Statute of the Armed Forces Military Personnel. Portuguese Government. Diary of the Republic, Lisbon http://www.emgfa.pt/documents/ cqw3zjnhvg4s.pdf (2015) Decree-Law No.76/2018 of 11 October: Establishes the incentive scheme for former or active duty AAFF personnel under contract (CS), special contract (SCS) or volunteers (VS). Portuguese Government. Diary of the Republic, Lisbon. (2018) Derr, C.: Five definitions of career success: implications for relationships. Appl. Psychol. 35(3), 415–435 (1986) Schein, E.H.: Identidade Profissional: como ajustar suas inclinações e opções de trabalho. Nobel, São Paulo (1996) Dutra, J.S., Albuquerque, L.: Âncoras de Carreira – In: Career Anchors by Edgar H. Schein. Adapted for Human Resource Management. FEA-USP, São Paulo (2009) Rego, A., Cunha, M., Gomes, J., Cunha, R., Cabral-Cardoso, C., Marques, C.: Manual de Gestão de Pessoas e do Capital Humano, 3rd edn. “Gestão” Collection. Edições Sílabo, Lda, Lisbon (2015) Arthur, M., Rousseau, D.: The Boundaryless Career: A New Employment Principle for a New Organizational Era. University Press, New York, Oxford (1996) Baruch, Y.: Managing Careers: Theory and Practice, pp. 79–80. Prentice Hall, FT, Harlow, England (2004) Chanlat, J.F.: Quais carreiras e para qual sociedade? RAE – Revista de Administração de Empresas. São Paulo 35(6), 67–75 (1995) Dutra, J.: Administração de Carreiras – Uma Proposta para Repensar a Gestão de Pessoas. Atlas, São Paulo (2012) Hall, D.: Careers In and Out of Organization. Sage, London (2002) Martins, H.: Gestão de Carreiras na Era do Conhecimento. Qualitymark, Rio de Janeiro (2010) Feldman, D.C., Bolino, M.C.: Careers within careers: reconceptualising the nature of career anchors and their consequences. Hum. Resour. Manag. Rev. 6(2), 89–112 (1996) Coetzee, M., Schreuder, D.: Using the career orientations inventory (coi) for measuring internal career orientations in the South African organisational context. S. Afr. J. Ind. Psychol. 35(1), 99–111 (2009). https://doi.org/10.4102/sajip.v35i1.806 Danziger, N., Rachman-Moore, D., Valency, R.: The construct validity of Schein’s career anchors orientation inventory. Career Dev. Int. 13(1), 7–9 (2008). https://doi.org/10.1108/ 13620430810849506 Rousseau, D.M.: Psychological and Implied Contracts. Empl. Responsib. Rights J. 2(2), 121–139 (1989) Rousseau, D.: Psychological Contracts in Organizations: Understanding Written and Unwritten Agreements. Sage, Thousand Oaks (1995) Conway, N., Briner, R.: Understanding Psychological Contracts at Work. Oxford University Press, New York (2005) Cable, D.A.J.: The Psychological contract: the development and validation of a managerial measure. Ph.D. Thesis in Phylosophy. Hamilton, Waikato University, New Zeeland (2008) Turnley, W.H., Feldman, D.C.: Re-examining the effects of psychological contract violations: unmet expectations and job dissatisfaction as mediators. J. Organ. Behav. 25–42 (2000)

Career Anchors for the Portuguese Army’s Volunteers …

201

38. Van de Ven, C.: The psychological contract: a big deal! (Chapter 3E). In: Recruiting and Retention of Military Personnel, Literature Reviews for Selected Recruitment and Retention Topic 3-1 Areas, pp. 3E1-19. Research and Technology NATO, Paris. https://www.nato.int/ issues/women_nato/Recruiting%20&%20Retention%20of%20Mil%20Personnel.pdf (2007) 39. Yin, K.R.: Case Study Research – Design and Methods, 5th edn. Sage Publications, Los Angeles (2014) 40. Bryman, A.: Social Research Methods, 4th edn. Oxford University Press, Oxford (2012) 41. Creswell, J.W.: Qualitative Inquiry & Research Design: Choosing Among Five Approaches, 3rd edn. Sage Publications, Los Angeles (2013) 42. Guerra, I.C.: Pesquisa Qualitativa e Análise de Conteúdo: Sentidos e Formas de Uso. Principia, Parede (2006) 43. Denzin, N., Lincoln, Y. (eds.): Collecting and Interpreting Qualitative Materials, 4th edn. Sage Publications, Los Angeles (2013)

Part VII

Simulation and Computer Vision in Military Applications

Implementation of Dubin Curves-Based RRT* Using an Aerial Image for the Determination of Obstacles and Path Planning to Avoid Them During Displacement of the Mobile Robot B. Daniel Tenezaca, Christian Canchignia, Wilbert Aguilar and Dario Mendoza Abstract The application of mobile robots in autonomous navigation has contributed to the development of exploration tasks for the recognition of unknown environments. There are different methodologies for obstacles avoidance implemented in mobile robots; however, this research introduces a novel approach for a path planning of an unmanned ground vehicle (UGV) using the camera of a drone to get an aerial view that allows to recognize ground features through image processing algorithms for detecting obstacles and target them in a determined environment. After aerial recognition, a global planner with Rapidly-exploring Random Tree Star (RRT*) algorithm is executed, Dubins curves are the method used in this case for nonholonomic robots. The study also focuses on determining the compute time which is affected by a growing number of iterations in the RRT*, the value of step size between the tree’s nodes and finally the impact of a number of obstacles placed in the environment. This project is the initial part of a larger research about a Collaborative Aerial-Ground Robotic System. Keywords RRT* · Path planning · Image processing

B. Daniel Tenezaca (B) · C. Canchignia · W. Aguilar · D. Mendoza Universidad de las Fuerzas Armadas ESPE, Sangolqui, Ecuador e-mail: [email protected] C. Canchignia e-mail: [email protected] W. Aguilar e-mail: [email protected] D. Mendoza e-mail: [email protected] © Springer Nature Singapore Pte Ltd. 2020 Á. Rocha and R. P. Pereira (eds.), Developments and Advances in Defense and Security, Smart Innovation, Systems and Technologies 152, https://doi.org/10.1007/978-981-13-9155-2_17

205

206

B. Daniel Tenezaca et al.

1 Introduction Unmanned aerial vehicles (UAVs) and mobile robots are used in different operations according to their performances. UAVs have many limitations, for example, the carrying capacity, flight time, etc. However, UAVs has a more ample range of vision of a workspace and higher velocities [1]. On the contrary, mobile robots present many advantages compared to UAVs. Mobile robots can carry heavier loads, have more power units, and higher capacity for processing [2, 3]. The purpose is to combine them to create a hybrid system taking into consideration the advantages of those two robotic systems. In this proposal, the UAV is able to analyze a broad portion of ground in specific workspace to provide safe displacement directions to the grounded vehicle [4, 5]. The research presents a study to establish a global planner that represents the data obtained from a top view generated by UAV [6]. First, the system develops an image through processing algorithms based on artificial vision in order to determinate the most relevant features inside the picture. In this case, the algorithm estimates the position of the obstacles as well as the mobile vehicle’s destination. LaValle [7] proposed RRT algorithm considered as an optimal technique to construct trajectories in nonconvex high-dimensional spaces and used to solve path planning problems. Commonly, an RRT is deficient to solve a problem related with the planning; therefore, it is necessary to implement a path planning algorithm based on that RRT* to expand the tree in free spaces [8]. The kinematic constraints of the ground vehicle based on Ackerman steering makes RRT* algorithm grows drawing curves easy to adapt the vehicle turning movements. This kind of curves is named Dubins path. Finally, it is important to consider adjusting many parameters to expand the tree; it will allow to know how the algorithm is affected when existing variations of the obstacles, number of iterations and the length of the connection between the tree’s nodes. In Sect. 2, the article presents an image processing algorithm for constructing a global planner. In Sect. 3, it details an analysis to establish a RRT* algorithm using a geometrically method based on constructing tangent lines that connect two points with a shortest curve in a Euclidean plane. Finally, in Sect. 4 results will be presented according to the data obtained in different scenarios and various obstacle configurations.

2 Image Processing The main objective of image processing is to remark certain details on an image. For this reason, processing images from the UAV’s camera is important to obtain a 2D binarized map where obstacles are recognized [9]. All stages are specified in the next flowchart (Fig. 1):

Implementation of Dubin Curves-Based RRT* … Image Acquision

Transformaon from RGB to HSV

Image smoothing

207 Image binarizaon

Morphological operaons

Fig. 1 Flowchart image processing

2.1 Image Acquisition Parrot Bebop 2 was used as UAV. It has a 14-megapixel ‘fisheye’ with a 3-axis image stabilization system that maintains a fixed angle of the view. The drone’s camera has to point to the ground (vertical position), for getting vertical camera’s rotation we have used a software development kit provided by Robot Operating System (ROS). Experimental results show to an altitude of 4 m above ground level our region of interest (ROI) covers an approximately metric area of 6 m × 4 m.

2.2 Changing Color Spaces Object discrimination using red, green, and blue color space (RGB) turns difficult because the object’s color is correlated with the amount of light exposure at the moment. For that reason, hue, saturation, and value color space (HSV) is more relevant to discriminate colors regardless the amount of light at the scene. Change of color spaces from RGB to HSV in this application allows better adapting of the obstacles’ algorithm to light changes, in cases which shadows contribute to false positives [10].

2.3 Image Smoothing Noise reduction is a very important factor for good image processing. In this case, it is applied a Gaussian Blur filter [11]. This means that each pixel was affected for a kernel (neighbor pixels). A kernel of size 5 was used and a Gaussian kernel standard deviation in an X direction of 0, where the target is not to get a blurred image if else to eliminate the small percent of noise.

2.4 Binarization In our application, it was important to separate foreground elements (obstacles) from the background of the image, binarization, or thresholding methods are commonly used for that purpose [12]. The obstacles were considered the most outstanding

208

B. Daniel Tenezaca et al.

element on the ground, therefore, establishing a binarized map where the obstacles have a pixel value of 0 and a pixel value of 1 is a free space. The hue value which represents red color goes from 0º to 30º and 330º to 0º approximately. Another stage is using the binarization to detect the target blue [13].

2.5 Edge Dilatation and Erosion The dilation convolution adds pixels to the boundaries of the obstacles’ structure reducing noise, while erosion removes pixels on the outside boundaries of the obstacles’ structure. The dilatation and erosion combined result in better filter. It fills holes (opening) and removes small objects (closing). The opening operation works with a structuring element a size kernel of 5. The operation affects the pixel and a 5 × 5 matrix size, meanwhile closing operation uses a size kernel of 21 [14].

3 RRT* Algorithm Dubins curves-based RRT* is used as a global planner. It establishes the path by two types of configurations divided into curve—straight—curve (CSC) and curve-curvecurve (CCC), which are the pattern of expansion in the RRT* algorithm. Random sample paths are generated from combination of straight lines and curves.

3.1 Initial Requirements Data input to get the path with the RRT* algorithm serve as initial state and orientation, numbers of iterations, goal state, and step size. To set Dubin curves parameters, it is necessary to know the minimum turning radius of the robot. Finally, another important input data is having the binarized map (workspace).

3.2 RRT* Algorithm According to Noreen et al. [15], RRT* is based in a group of features which allow tree expansion very similar to RRT algorithm. The difference between the systems is that RRT* incorporates two special properties called near neighbor search and rewiring tree operations. The algorithm is represented by a tree denoted as T = (V, E), where V is a set of vertices and E is a set of edges. The initial configuration (qinit ) includes a set of vertices that represents where the tree starts. In each iterations configured (Lines

Implementation of Dubin Curves-Based RRT* …

209

Table 1 Pseudocode RRT* algorithm Algorithm RRT* T = (V, E) RRT* 1: 2: 3: 4: 5: 6: 7: 8: 9: 10: 11: 12:

3–11), the algorithm establishes a random position (qrand ) in free region, (qnear est ) is searched in the tree according to predefined step size from (qrand ) and immediately the algorithm establishes a new configuration (qnew ) with Steer function which guides the system from (qrand ) to (qnear est ). The function Choosepar ent allows to select the best parent node for the new configuration (qnew ) before its insertion in tree considering the closest nodes that lie inside of a circular area, finding (qmin ). Finally, near neighbor operations allow to generate the optimal path repeating the previous process [16] (Table 1).

3.3 Dubin Curves The Dubin curves describe six types of trajectories: RSR, LSR, RSL, LSL, RLR, and LRL. Each configuration comes from an analogy that is denoted by R (right move), S (straight move), and L (left move) [17]. All these configurations use geometrically computing method based on constructing tangent lines between two circles. The first step has two circles C1 and C2 , with their respective radius r1 and r2 , where C1 represents coordinates (x1 , y1 ) and C2 as (x2 , y2 ) (see in Fig. 2).

Fig. 2 Circles initial configuration

210

B. Daniel Tenezaca et al.

Fig. 3 Inner tangents

Inner tangents. Then a line is drawn between two center points C1 and C2 estab− → lishing a vector V1 , magnitude D and the midpoint p3 (point between p1 and p2 ) is calculated, circle C3 is constructed with a radius r3 as we reflected in Fig. 3 D=



(x2 − x1 )2 + (y2 − y1 )2   x1 + x2 y1 + y2 , p3 = 2 2 D r3 = 2

(1) (2) (3)

The next step is to draw another circle C4 located in C1 ’s center, with radius r4 = r1 + r2 , we obtained pt , which is the intersection between C4 and C3 like the one shown in Fig. 3. A triangle is built joining the points p1 , p3 , pt and we can define geometrically that segment pt p1 = r4 and p1 p3 = pt p3 = r3 . The angle γ = ∠ pt p1 p3 is very important to define the coordinates of pt . The next equation − → determinates the amount of rotation about the x-axis, θ for V2 . − → (4) θ = γ + atan2 V1 Using θ , it is possible to obtain pt according to the following equations: xt = x1 + (r1 + r2 ) ∗ cos(θ )

(5)

yt = y1 + (r1 + r2 ) ∗ sin(θ )

(6)

Considering that the inner tangent starts on C1 , it is necessary to normalize a − → − → vector V2 = ( pt − p1 ) and multiply it by r1 , the result will allow to find a vector V3 to pit1 from p1 . It resumes next by:

Implementation of Dubin Curves-Based RRT* …

211

Fig. 4 RRT* algorithm’s visualization

− → V2 − → V3 = ∗ r1 V2 − → pit1 = p1 + V3

(7) (8)

− → Finally, it is possible to draw a vector V4 from pt to p2 , as shown in the figure. − → Using V4 magnitude and the direction, it is possible to find the inner tangent point on C2 . − → V4 = ( p2 − pt )

(9)

− → pit2 = pit1 + V4

(10)

Outer tangents. The process is very similar to the one of inner tangents, having two circles C1 and C2 , and considering r1 ≥ r2 , the procedure is the same as before, C4 is centered at p1 , with a difference the radius r4 = r1 − r2 , after getting pt and − → following all steps performed for the interior tangents, V2 is obtained and the first outer tangent point pot1 . This condition produces that r4 < r1 . To get the second outer tangent pot2 an addition is performed by: − → pot2 = pot1 + V4

(11)

The main difference between calculating outer tangents compared to inner tangents is the construction of circle C4 ; all steps keep the same. In the next figure (see Fig. 4), it can be seen the path establishes using data input.

212

B. Daniel Tenezaca et al.

4 Results The experimental results produced by RRT* path planning based on Dubins curves were performed using a CPU with an i7 third generation processor and 16 Gb RAM memory. For algorithm testing, a scene with a necessary amount of light was chosen to obtain the correct object segmentation. From this point, the test started with no obstacles in the scene. Then, obstacles were gradually added to establish four different configurations, in such way, algorithm functionality and efficiency was tested. The algorithm was tested in five different cases (see Fig. 5), with no obstacles (first case), one obstacle (second case), two obstacles (third case), three obstacles (fourth case), and four obstacles (fifth case). The algorithm satisfies its purpose because the path established (red line) started from car’s position to the final configuration (blue circle) avoiding the obstacles.

Fig. 5 Scene configurations. a No obstacles. b One obstacle. c Two obstacles. d Three obstacles. e Four obstacles

Implementation of Dubin Curves-Based RRT* …

step size

Nº Obstacles vs Input data

150 100

Poly. (nº itera ons)

50

Poly. (step size)

0

150

y = -4.6429x2 + 44.357x + 3 R² = 0.9756

100 50

y = 8.5714x2 - 67.429x + 196 R² = 0.9752 0

1

2

3

4

nº itera ons

40

80

90

105

110

step size

140

90

70

70

70

0

Step size

nº itera ons

213

nº Obstacles Fig. 6 Number of iterations and step size processed for each obstacle configuration

Another aspect to consider was that the goal position should have a tolerance radius of 60 pixels measured from the goal position’s center point. The first measurement of results (see Fig. 6) showed that the cases in which obstacles were added, it was necessary to increase the number of iterations to reach the configuration desired. Forty iterations were the minimum to reach the goal in the first case, compared with the fifth case where it was necessary at least 110 iterations to establish a path for UGV to represent. Based on the data processed, a second-grade polynomial equation was obtained to represent the number of iterations required for a scene where there are more than four obstacles in a scene. y = −4, 6x 2 + 44, 3x + 3

(12)

The ideal value for step size for each case is decreasing if we take into consideration the first test is with no obstacles and the fifth test has four obstacles in the scene. One hundred forty pixels the maximum step size for the optimal performance of the algorithm within the limits of workspace and for UGV to follow the path described. In the last three cases (case 3, case 4, and case 5), the necessary step size is 70 pixels, although the correlation of data is high by number of tests, the best possible solution in case the number of obstacles increase would be tested with the same value of step size. Processing time required by the algorithm compared with the number of obstacles is presented in (Fig. 7). The blue line (time 1) describes the necessary time for expanding RRT* using an optimal data input for each case. The red one (time 2) describes the time required to reach de goal destination if the data is fixed with values of 30 pixels for step size and the maximum number of iterations is 1500.

214

B. Daniel Tenezaca et al.

Fig. 7 Processing time required by algorithm

Time

Nº Obstacles vs Time 100 50 0

0

1

2

3

4

Ɵme 1

2

2

5

5

5

Ɵme 2

6

7

66

70

62

Nº Obstacles

5 Conclusions From the data aforementioned in this paper, we can conclude that exists a direct relationship between several iterations, step size, and quantity of obstacles. In case with few obstacles within the workspace, we can configure the step size in maximum value of 140 pixels allowing to reduce the number of times that mechanical direction of UGV has to rotate. The number of iterations is directly proportional to the number of obstacles because, more obstacles in the scene require to construct more iterations to reach the goal with an optimal path. Establishing constant input data was not the most efficient way to execute the algorithm because if the step size configuration is too low, near 30 pixels, so the UGV can take an excessive time to follow the path and require too many changes in the direction of the robot. This study has determined the importance of configuring initial parameters to construct the RRT* algorithm. Considering the path planning is based in a global planner, it is important to get the parameters in lesser time and optimal manner possible to obtain a better routing of data between UGV and UAV. Also, the study is the first stage of a global research for developing a collaborative robot system; we focused on getting a correct path for a mobile robot to navigate without colliding with objects around it. Finally, this work could be complemented with the development of pure pursuit control for UGV.

References 1. Yulong, D., Bin, X., Jie, C., Hao, F., Yangguang, Z., Guanqiang, G., Lihua, D.: Path planning of messenger UAV in air-ground coordination. IFAC-PapersOnLine 50, 8045–8051 (2017) 2. Sivaneri, V.O., Gross, Y.J.N.: UGV-to-UAV cooperative ranging for robust navigation in GNSSchallenged environments. Aerosp. Sci. Technol. 71, 245–255, 2017 3. Rahimi, R., Abdollahi, F., Naqshi, K.: Time-varying formation control of a collaborative heterogeneous multi agent system. Robot. Auton. Syst. 62, 1799–1805 (2014) 4. Melin, J., Lauri, M., Kolu, A., Koijonen, J., Ritala, R.: Cooperative sensing and path planning in a multi-vehicle environment. IFAC-PapersOnLine 198–203 (2015) 5. Rosa, L., Cognetti, M., Nicastro, A., Alvarez, P., Oriolo, G.: Multi-task cooperative control in a heterogeneous ground-air robot team. IFAC-PapersOnLine 48, 53–58 (2015)

Implementation of Dubin Curves-Based RRT* …

215

6. Ropero, F., Muñoz, P., Moreno, M.D.R.: TERRA: a path planning algorithm for cooperative UGV-UAV exploration. Eng. Appl. Artif. Intell. (2019) 7. LaValle, S.: The RRT Page (1999). [En línea]. http://msl.cs.uiuc.edu/rrt/index.html 8. Abbadi, A., Prenosil, V.: Collided path replanning in dynamic environments using RRT and cell decomposition algorithms. Modelling and Simulation for Autonomous Systems: Second International Workshop, pp. 131–143 (2015) 9. Sieberth, T., Wackrow, R., Chandler, J.H.: Automatic detection of blurred images in UAV image sets. ISPRS J. Photogramm. Remote Sens. 122, 1–16 (2016) 10. Chernov, V., Alander, J., Bochko, V.: Integer-based accurate conversion between RGB and HSV color spaces. Comput. Electr. Eng. vol. 46, pp. 328–337, 2015 11. Huang, J., Feng, H., Xu, Z., Li, Q., Chen, Y.: A robust deblurring algorithm for noisy images with just noticeable blur. Optik 168, 577–589 (2018) 12. Chen, S., Li, D.: Image binarization focusing on objects. Neurocomputing 69, 2411–2415 (2006) 13. Dougherty, E.: Mathematical Morphology in image processing. Marcel Dekker, Nueva York (1993) 14. OpenCV.: OpenCV 2.4.13.7 documentation. [En línea]. https://docs.opencv.org/2.4/modules/ imgproc/doc/filtering.html 15. Noreen, I., Khan, A., Habib, Z.: Comparison of RRT, RTT* and RRT*-smart path planning algorithms. IJCSNS Int. J. Comput. Sci. Netw. Secur. (2016) 16. Karaman, S., Frazzoli, E.: Sampling-based algorithms for optimal. Int. J. Robot. Res. (2011) 17. Yao, W., Qi, N., Zhao, J., Wan, N.: Bounded curvature path planning with expected length for Dubins vehicle entering target manifold. Robot. Auton. Syst. 97, 217–229 (2017)

Machine Learning and Multipath Fingerprints for Emitter Localization in Urban Scenario Marcelo N. de Sousa, Rafael L. Cardoso, Henrique S. Melo, José W. C. Parente Jr. and Reiner S. Thomä

Abstract A hybrid approach is proposed to perform the locate and tracking a noncollaborative radio frequency emitter using ray tracing (RT) simulation tool, channel impulse response (CIR), and machine learning estimation. The technique can enhance communication intelligence (COMINT) systems or even perform the localization using a single sensor in an non-line-of-sight (NLOS) suburban scenario. A multipath fingerprint can identify the target position using the machine learning classification engine to perform the matching. Conventional localization techniques mitigate errors trying to avoid NLOS measurements in processing emitter position, while the multipath fingerprints proposed uses the reflection information to feed the pattern matching engine build on a machine learning classification framework. The method was applied to simulate a tactical scenario, where a navy frigate is in Ipanema and tries to track an RF emitter target in the Rio de Janeiro streets using only one RF sensors fixed in an aerostat in a hypothetical counterinsurgency situation. Keywords Wireless positioning · Multipath exploitation · Hybrid positioning · Machine learning · Ray tracing simulation

1 Introduction The estimation of the source of a signals’ location remains the subject of extensive research for years and is increasingly receiving considerable interest in signal processing scholarly community, including radar, GPS, wireless sensor networks, sonar, animal tracking, multimedia, and mobile communications. In urban and counterinsurgency operations, there is an increasing demand to have an information superiority M. N. de Sousa (B) · R. S. Thomä Technische Universität Ilmenau, 98684 Ilmenau, TH, Germany e-mail: [email protected] URL: http://www.tu-ilmenau.de/it-ems/ R. L. Cardoso · H. S. Melo · J. W. C. Parente Jr. Military Institute of Engineering, Rio de Janeiro 22290-270, Brazil © Springer Nature Singapore Pte Ltd. 2020 Á. Rocha and R. P. Pereira (eds.), Developments and Advances in Defense and Security, Smart Innovation, Systems and Technologies 152, https://doi.org/10.1007/978-981-13-9155-2_18

217

218

M. N. de Sousa et al.

of the tactical scenario, where the communication intelligence systems (COMINT) need to supply the commanders in different levels with the precise information of the target position [1]. Multipath is typically perceived as a limitation, [2], but is possible to exploit interaction effects for localization of the RF emitter using a single sensor. There is always a trade-off between the performance desired and the available resources in deploying the number of sensors needed. It is also worthy to use the scenario features converted by ray tracing simulation to improve the performance of the available sensors even in NLOS condition. There are various approaches [3] tracking and emitter position finding where multipath exploitation is performed to locate the target, again using the image-based approach to identify the sensor position end extract the location with LoS and NLOS calculation [4]. We propose an approach to perform a RF localization in urban scenario binding together a ray tracing propagation tool to extract the multipath fingerprint and a machine learning framework to improve the precision even in NLOS situation. The idea is to reproduce the framework presented in [5] using only one sensor, contrarily to similar approaches, where the NLOS measurements are not used in the position estimation. This approach is relevant to enhance the radio frequency localization performance in an urban environment because it uses NLOS and geographic database as an extra layer of information, that can be fused in the position estimation engine. This paper has the following structure. Section 2 discusses the problem formulation, including the signal data model and the channel impulse response estimation. An experimental setup, showing the application of the proposed localization method and including the results of the simulation setup, is discussed in Sect. 3. Finally, Sect. 4 summarizes the principal aspects of this paper and presents possible research direction using the proposed methodology.

2 Problem Formulation Assuming the typical scenario displayed in Fig. 1, Where a frigate if the navy is placed close to a city cost to support some counterinsurgency operations, the ship has an RF sensor with direction finding capability, and the sensors are installed in an aerostat connected to the frigate flying at 100 meters over the sea level. A schematic view of the sensor, the platform, and the navy frigate is shown in Fig. 2. To map the reflection points in the environment as virtual sensors, assume a “non-collaborative” emitter that has been tracked inside the city, the buildings, and the urban scenario makes it impossible to have a clear line-of-site situation to locate the emitter (Fig. 3). Using the approach introduced by [6], modified to perform the calculation using the sensor S1RT and M scatters points at location Ri = [xi , yi , z i ]T . If the scatters’ position are known, it is possible to locate the RF target P using an adapted TDOA measurements scheme introduced by [7]. The three most significant scatters that affect the signal received S1RT will be considered as three others TDOA sensor, namely as “Virtual Sensors.” The position of

Machine Learning and Multipath Fingerprints for Emitter Localization …

219

Fig. 1 Schematic view using Google Earth of an emitter in a urban scenario (Ipanema-RJ) tracked by a navy vessel with a RF sensor at 100 m

Fig. 2 Sensor on aerostat platform in navy vessel at 100 m

220

M. N. de Sousa et al.

Fig. 3 Location error caused by discretization

the sensors, the scatters, and the emitter presents an error, caused by the discretization of the scenario. As the error also affects the precise reflection points’ information, the virtual nodes modeled as virtual sensors position also presents a grid error. The sensor position will be assumed to be TDOA sensor 1, and the three most significant scatters that affect the signal received S1RT will be considered as three others TDOA sensor, namely as “Virtual Sensors.” The emitter source is in position P = [xs , ys , z s ]T ∈ R3 and the TDOA sensors, Si = [Si x , Si y , Si z ]i=1:4 at known position in 3D coordinate system, as shown in Fig. 4. Θˆ = [τ21 , τ31 , τ41 ]T

(1)

As shown in Fig. 5, the channel impulse response presents a diffracted ray (α1 , τ1 ), and three scatters that where reflected in R2 ,R3 , and R4 , in this situation we have the time of propagation from emitter and the sensor, and the time of propagation from the sensor to each reflection wall, as follows: τ21 = a + b − τ1 ; τ31 = c + d − τ1 ; τ41 = e + f − τ1 .

(2)

Machine Learning and Multipath Fingerprints for Emitter Localization …

221

Fig. 4 Scatters points as virtual TDOA sensors

The Figs. 4 and 5 can illustrate the differences of the RT virtual nodes data model with the normal TDOA system. The sensor position and the reflection points are assumed to be known under certain imprecision. The time differences from the sensor base (S1RT ) to the other TDOA sensor in Fig. 4 or to the reflection points in Fig. 5, a, c, e are also known. So, the central question is how to estimate the time differences and perform the position estimation with this data model of [7]. Assuming that it is possible to perform the CIR estimation and assign to each ray to each reflection point in the scenario, these points can work as virtual TDOA sensors. The error in the reflections points will be assumed as an error in the definition of the station position. The TDOA computation between the signals received by the sensors is computed several; first, the signal is filtered (“whitening”) and oversampled using Farrow filters [8] and generalized cross- correlation function in phase(GCC-PHAT).

3 Proposed Method This section describes the method proposed to enhance a TDOA localization system when the emitter is in NLOS position, that is a typical situation of the outdoor channel characteristics and cannot be avoided. Some methods deal with multipath assuming

222

M. N. de Sousa et al.

Fig. 5 Location using four TDOA sensors

that they are outliers, but this approach is not feasible in a typical suburban scenario. The proposed method is a combination of channel impulse response estimation and ray tracing. The first step is to discover the strong multipath specular components that affect the time measurements, this is performed doing the CIR estimation at sensor position, after that, ones try to identify the main obstacles in the environment, that produces the multipath rays. The multipath parameters, defining the amplitude, time of arrival, and directions of the incoming ray at the sensor position are calculated. The decision to use the reflection points as virtual TDOA sensor, instead of using only the direct range estimation was made because there is no synchronization between the emitter and the receiver. As shown in Fig. 6, the buildings in the scenario cause reflections and diffraction in the incoming signal at the sensor. The method is schematically presented in Fig. 6, first, the ray tracing will perform several simulations of the outdoor scenario, the output of the simulation is a description of the paths from each component of the signal that connects the emitter position with the receiver point. The second step is to estimate the CIR to extract the multipath information (α1 , τi ) from the receiver signal in each sensor position. In the end, a neural-network-based on machine learning engine performs the estimation of the emitter position, enhancing the TDOA performance in the outdoor NLOS situations. Depending on the precision of the obstacles description and the material properties

Machine Learning and Multipath Fingerprints for Emitter Localization …

223

Fig. 6 Channel impulse estimation in RF sensor

Fig. 7 Fingerprints using ray tracing

included in the simulation, the ray tracing gives a reasonable representation of the specular components that characterize the emitter–receiver relationship (Fig. 7). The next step is to take the signal received by each sensor and make the channel impulse response (CIR), where the information (α1 , τi ) extracted from the complex baseband (CBB) received signal is associated with a defined position in the scenario. The matching procedure between the CIR estimated, and the multipath dataset is

224

M. N. de Sousa et al.

performed using a neural-network-based with a machine learning engine. The reflection point used as virtual sensors has some uncertainty not precisely described, as described in [9].

3.1 Ray Tracing Fingerprints When a ray tracing simulation is performed, there is “preprocessing” information, which extracts the emitter–receiver visibility. Depending on the desired number of ray interaction, the image theory allows to identify the reflection points and the virtual nodes, and all possible rays in the simulation domain can be estimated. Then, for each Tx-Rx combination, there is a specifically channel impulse response that defines the time delays for the specular components. Considering that the sensor 1 can extract the angular channel impulse response where the four most robust specular components are identified and measured in angular and time information. With the output file of ray tracing, it is possible to search for a given position where are the main reflectors that the rays bounce before arriving at sensor. It works as if, ones want to extract the “Field of Vision” in optics ray. So, in the case of Fig. 8, after this “searching” phase three buildings were found, and the reflection points are P1, P2, and P3. Now, with the information of the time difference between the sensor position and each reflection point, the time difference between the direct rays and the reflected ones are taken from the CIR estimation. In the end, this configuration looks like a TDOA system where the reflection points

Fig. 8 Wall and edges form the buildings in ray tracing simulation

Machine Learning and Multipath Fingerprints for Emitter Localization …

225

Fig. 9 Single station estimation location

work as virtual sensors and the position estimation engine uses the time differences as a modified TDOA system with only one station. Following the approach of [10], the RT gives the information about the ray path, describing the edges and wall touched by the rays in the emitter–sensor path. The Fig. 8 gives a graphic representation of the visibility matrix calculation. Fig. 9 shows the multipath fingerprints produced by ray trace simulation from sensor in the aerostat.

3.2 Machine Learning in Multipath Fingerprint As described by [11], the machine learning (ML) tool is used to match the multipath information to predict geolocation of the emitter. It is possible to use the regression for prediction because it is a class of supervised algorithm that attempts to establish a continuous relationship between a set of dependent variables (geolocation coordinates) and set of other independent variables (multipath fingerprints—αi , τi ). Figs. 10 and 11 show the machine learning framework proposed. First, we will try to use ML to approximate the position estimation as a target function for optimization, the semi-supervised or supervised ML makes an approximation to find a target function ( f ). This function should be able to map input variables (X ), that are the multipath components (αi , τi ), to an output variable (Y ), that is the emitter position. We used a set of tools as a machine learning framework to develop and test the application, including:

226

M. N. de Sousa et al.

Fig. 10 Machine learning ray tracing dataset refinement

Fig. 11 Supervised machine learning based on RT simulation

– Jupyter Notebook version 5.1.0 (http://jupyter.org); – Tensorflow version 1.4.0 (http://tensorflow.org); – CentOS version 7.4.1708 (https://www.centos.org). The machine learning and kernel methods can work as a supervised, unsupervised, or semi-supervised learning framework. In our approach, the ray tracing gives the initial dataset for the ML and the CIR estimation of the reference or anchor nodes multipath works as a “reinforcement” or supervised learning. The hybrid approach uses real measurements and simulations to perform the dataset, which can be classi-

Machine Learning and Multipath Fingerprints for Emitter Localization …

227

Table 1 Emitter position with single station

1:

2:

3:

4: 5: 6: 7:

Pseudo-code: Emitter Position with Single Station Input: – Sensor Position (X, Y, Z); – Geo database with Buildings Description (Walls and Edges); – Complex Base Band Signal at Sensor 1; – Direction Finding with with 4 Strongest Multipath Info. Perform Ray-Tracing Simulation: – from all points from “AoI” and Sensor 1; – Extract the “Visibility”, Edges and Wall with “mutual vision”; – Map RT simulation to channel estimation to calculate reflection points. From the CBB Signal Received at S1 Perform CIR Estimation; – Assign the First Obstacle in the DoA from 4 strongest Rays; – Identify the Edges and Wall from the Obstacles. Assign the Reflection Points a TDoA Sensor Position (X, Y, Z). With ToA info and Reflection Position: Perfom TDoA Localization. Compare the Localization with Virtual TDoA Sensor Output: Emitter Estimation Position.

fied as semi-supervised learning. The network had ten inputs, i.e., five rays. Neural networks, linear regression, and random forest algorithm were used to achieve the result, and the random forest proved to be the best one. One of the problems in the model proposed was the points that have less than five paths. Then a set of hyper-parameters tuned the linear regression was used, giving better results. In our formulation, we use the five strongest rays that arrive at each point of the simulation domain. The proposed algorithm can be summarized in Table 1.

4 Experiment Setup The algorithm was applied to perform the position estimation in the region of Ipanema, as described in Sect. 2, the target is inside an urban area, and the navy vessel is trying to perform the localization using only one sensor and the ray tracing fingerprint framework presented in Sect. 3 using just the sensor 1 and the ray tracing simulation of the area of interest (AOI). The result was compared with a simulation using the estimation performed by a complete TDOA systems using 4 (four) base station nodes (S1 , S2 , S3 , and S4 ), deployed in the ocean in front of the AOI. First, the complete scenario was reproduced in ray tracing simulation tools to extract the multipath information, then, the most significant reflectors, considering

228

M. N. de Sousa et al.

the TDOA sensor 1 was obtained from the RT output files. The CIR estimation procedure was performed using a signal synthetic generated using a binary pseudo-random sequence with length 15 (PBRS15). Several averaging and filtering operations were performed to reproduce the effects of the urban multipath channel on the received signal S1RT . The CIR estimated, in the time domain, gave the delay and power profile of the known position building. Then the CIR estimated from the target with up to three rays (delay and power) was used to search in the fingerprint database the position of the emitter. The fingerprints of the rays suing the machine learning framework are presented in the results of the application the method are presented in Fig. 12. The Table 2 summarizes the result of the method using three and four multipath components, the error is caused by the scatter position imprecision in the discretization of the scenario. It is also important to point out that the multipath components are first filtered out to taking into account only the first bounce reflections, and the ray association using the real measurements was possible because the emitter stayed in a static position

Table 2 Performance of the single station TDOA location position compared with a complete TDOA systems Method Obs Lat Long Error (m) TDoA 4 sensor

Sol 1 Sol 2

[tau 31 , tau 31 , tau 31 ]

49.591715 11.085821

Qualities [84 57 65]% Duration 20 ms Power [−100 −99 −101 −95] dBm Delays [1.064 1.052 0.471] µs S1 + 3 Paths at R1, R2 and R3 S1 + 2 Paths at R1 and R2

49.592382 11.085037 27 49.592379 11.088463 82

5

Fig. 12 Fingerprints from the first rays amplitude and delay in a urban scenario (Ipanema-RJ)

Machine Learning and Multipath Fingerprints for Emitter Localization …

229

Fig. 13 Fingerprints in 3D from the first rays amplitude and delay in a urban scenario (Ipanema-RJ)

during the measurements done. The TDOA location ellipses are shown in Fig. 13, and the total standard deviation of the position estimator was 89.66 m, and the standard deviation of each component, using the Chan method was: σˆ [σx = 28.40 m, σ y = 33.05 m, σz = 78.36 m].

5 Conclusions The paper proposes a kernel-based machine learning localization scheme based on TDOA fingerprinting. It describes how to use ML for refining the multipath Fingerprint generated by ray tracing simulation. The method proposed showed the ability for enhancing the performance of sensors deployed in NLOS scenario. The simulations performed using the single sensor location with the ray tracing technique could perform the emitter position estimation. Despite the bias and imprecision of the scenario description, the main obstacles that reflect the signal can be used as virtual sensors to improve the TDOA system performance. The effect in localization procedure is not straightforward to analyze, and the ML engine and ray tracing are two tools that only give reasonable results depending on the quality of the representations of the obstacles in a geographic database. The ML can improve the multipath fingerprint generated by ray tracing, but it also depends on the capability of processing the channel impulse estimation and get an appropriate multipath characterization of the calibration point. For the next steps in the research, the authors would like to consider the use of more features either the signal and the scenario, trying to record more multipath information and patterns to build a data fusion engine with the cartographic database

230

M. N. de Sousa et al.

and signal processing. It is necessary to make a mapping procedure to assign the reflected ray from the measurements into the ray tracing. The static scenario and the specific characteristic of the buildings make the multipath components “well defined” with the specular components much bigger them the dense multipath components (DMC).

References 1. Brady, E., Starr, S.: Assessing C3I in support of dismounted operations in complex terrain. Technical Report, Strategic Perspectives Inc., McLean, VA (2002) 2. Phelan, B.R., Lenzing, E.H., Narayanan, R.M.: Source localization using unique characterizations of multipath propagation in an urban environment. In: 2012 IEEE 7th Sensor Array and Multichannel Signal Processing Workshop (SAM), pp. 189–192, June 2012 3. Fertig, L.B., Baden, M.J., Kerce, J.C., Sobota, D.: Localization and tracking with multipath exploitation radar. In: 2012 IEEE Radar Conference, pp. 1014–1018, May 2012 4. Setlur, P., Devroye, N.: Bayesian and Cramer-Rao bounds for single sensor target localization via multipath exploitation. In: 2013 IEEE International Conference on Acoustics, Speech and Signal Processing, pp. 5845–5849, May 2013 5. de Sousa, M.N., Thomä, R.S.: Enhancement of localization systems in NLOS urban scenario with multipath ray tracing fingerprints and machine learning. Sensors 18(11), 4073 (2018) 6. So, H.C.: Handbook of position location: theory, practice and advances. In: Reza Zekavat, R.M.B. (ed.) The Oxford Handbook of Innovation, pp. 23–34. Wiley-IEEE Press, ch. 2 (2011) 7. Ho, K.C., Lu, X., Kovavisaruch, L.: Source localization using TDOA and FDOA measurements in the presence of receiver location errors: analysis and solution. IEEE Trans. Signal Process. 55(2), 684–696 (2007) 8. Kolb, D., Warzügel, S., Schramm, J.: Hardware-reduced system for TDOA-locating of radio frequency emitters, eP Patent App. EP20,130,160,415, 25 Sept 2013. http://www.google.com/ patents/EP2642312A1?cl=en 9. Yang, L., Ho, K.C.: An approximately efficient TDOA localization algorithm in closed-form for locating multiple disjoint sources with erroneous sensor positions. IEEE Trans. Signal Process. 57(12), 4598–4615 (2009) 10. Fuschini, F., El-Sallabi, H., Degli-Esposti, V., Vuokko, L., Guiducci, D., Vainikainen, P.: Analysis of multipath propagation in urban environment through multidimensional measurements and advanced ray tracing simulation. IEEE Trans. Antennas Propag. 56(3), 848–857 (2008) 11. Alsheikh, M.A., Lin, S., Niyato, D., Tan, H.P.: Machine learning in wireless sensor networks: algorithms, strategies, and applications. IEEE Commun. Surv. Tutor. 16(4), 1996–2018 (2014) (Fourthquarter)

Virtual Rehabilitation System Using Electromyographic Sensors for Strengthening Upper Extremities Z. Andrea Sánchez, T. Santiago Alvarez, F. Roberto Segura, C. Tomás Núñez, P. Urrutia-Urrutia, L. Franklin Salazar , S. Altamirano and J. Buele Abstract This work presents a virtual system for the rehabilitation of the upper extremities, using the MYO Smart Band device for the acquisition of electromyographic signals produced by the user. Processing and managing of these signals are done through the SDK provided by the manufacturer of the bracelet which is compatible with the MATLAB software. The virtual environment is developed in the Unity 3D graphics engine, in which three-dimensional objects that were previously designed in the 3ds Max software are implemented. The application presents the user with a virtual scenario set in a natural landscape, in which there is a van that must be driven on a certain path (the complexity is increasing). The videogame is of low complexity, since it seeks to avoid situations of stress while the rehabilitation process takes place. Each task in the application is associated with a hand and forearm movement of the user; it means the patient is given an alternative tool that Z. Andrea Sánchez · P. Urrutia-Urrutia · L. Franklin Salazar · S. Altamirano · J. Buele (B) Universidad Técnica de Ambato, Ambato 180103, Ecuador e-mail: [email protected] Z. Andrea Sánchez e-mail: [email protected] P. Urrutia-Urrutia e-mail: [email protected] L. Franklin Salazar e-mail: [email protected] S. Altamirano e-mail: [email protected] T. Santiago Alvarez · F. Roberto Segura Instituto Tecnológico Superior Guayaquil - Ambato, Ambato 180205, Ecuador e-mail: [email protected] F. Roberto Segura e-mail: [email protected] C. Tomás Núñez CELEC EP, Baños 180250, Ecuador e-mail: [email protected] © Springer Nature Singapore Pte Ltd. 2020 Á. Rocha and R. P. Pereira (eds.), Developments and Advances in Defense and Security, Smart Innovation, Systems and Technologies 152, https://doi.org/10.1007/978-981-13-9155-2_19

231

232

Z. Andrea Sánchez et al.

allows him/her to perform exercises that improve his/her extremity active mobility, mitigating the routine effects of a conventional session. To validate this proposal, it is tested by five retired military personnel in passive state, to whom the using task ease (SEQ) usability test is applied. The result is (58,8 ± 0,27), which shows that this interactive interface has a good acceptance when being in the range between 40 and 65. Keywords Rehabilitation · Virtual reality · Upper extremities · Electromyographic sensor

1 Introduction Military confrontations and wars are considered extreme and challenging situations that a human being must endure [1]. It is subjected to a multitude of psychomotor, sensory, cognitive, and psychological tests that exert great pressure on military personnel [2]. Some get to lose their senses or develop some type of trauma or disability [3]. Among the main intellectual/cognitive disabilities are learning disorders, generalized developmental disorders, attention deficit disorders and behavior [4]. In the sensory field, there are visual, auditory, language, speech, and voice disabilities [2, 3]. Additionally, in the physical aspect, reference is made to motor alterations at osteoarticular level, nervous system, amputations, and disability in the extremities [5]. The most common injuries occur in both the lower and upper extremities [6, 7]. In the upper ones, there is ankylosis, joint stiffness, sequelae of fractures or traumatisms and complete paralysis of fingers or hands. That is why, with the aim of preventing or mitigating any of these illnesses, a physical therapy process must be initiated [8]. Through a cycle of relearning promote and stimulate muscles and joints to regain their functionality [9]. The rehabilitation process must be evaluated by a qualified physiotherapist who provides the patient with the tools to accelerate this process. Previously, routine coordinated physical exercises were developed to produced boredom and fatigue during the sessions [10]. But this reality has changed when merging these procedures with advanced technological tools. A clear example of this is the use of compatible devices with virtual reality (VR) interfaces and auditory accompaniment. Systems involving VR are an advanced form of human–machine interaction (HCI), which allows the user to experience extraordinary situations, without the need to move to another site. These systems achieve a higher level of immersion with the proper use of sensors, devices, and peripherals that enrich the experience. All this stuff, focused on the rehabilitation processes, gives the patient the opportunity to perform exercises in a different way and with greater encouragement [8]. Especially when it comes to military personnel who have been subjected to stress and have certain post-war psychological disorders.

Virtual Rehabilitation System Using Electromyographic Sensors …

233

2 State of Art The implementation of systems that include VR in their architecture for rehabilitation processes has been gaining relevance in recent years and that is why several related researches have been developed. A treatment focused on counteracting the incidence of post-traumatic stress disorder (PTSD) of military personnel using VR is described in the proposal developed by Achanccaray et al. [11]. A virtual environment inspired by Iraq/Afghanistan has been developed and includes the dangers found in these places. This proposal has provided initial reports with positive results, when tested by active staff and veterans. Regarding upper extremity treatment, the development of new sensors and devices allows the development of several options when it comes to rehabilitation. In the prior research carried out by Levin et al. [12], VR is presented as a tool used in rehabilitation processes of upper extremities. Virtual environments that optimize learning and motor skills recovery are managed. Based on the respective experimental tests, the limitations of current technologies are discussed with respect to their effectiveness and feedback in the process of physical and sensory learning. For its part, Liu et al. [13] present a rehabilitation system of upper extremities based on electromyographic sensors (EMG) and portable accelerometers (ACC) for children with cerebral palsy (CP) which is explained. Using the Android platform, games are developed that seek to improve motor function under the guidance of a doctor. The proposal has been validated by three test subjects who have improved their reaction capacity by interacting with this motivating and user-friendly interface. Similarly, López et al. [14] present an interactive system that uses myoelectric sensors to strengthen upper limbs in children. Unity 3D software develops interactive interfaces (games) that promote the realization of therapies by patients. This prototype is tested by 5 users (3 boys and 2 girls) with ages between 6 and 12 years where respective System Usability Scale (SUS) usability test is presented. As can be seen, there are many researches related to this topic, but not focused on the military field as required. In this context, this work proposes a virtual rehabilitation system for upper extremities that interacts intuitively with the patient, in this specific case of military personnel who suffered a combat injury. Data acquisition for the interface configuration as well as the reading of electromyographic signals is done using a low-cost electronic device with which the physical gesture that the person is making can be recognized. Three-dimensional objects have been made in 3ds Max software and imported into the Unity 3D software, where they are given hierarchy and animation. In this context, the aim is to present the patient with an interactive tool that contributes to a comfortable and accelerated rehabilitation process, mitigating the post-traumatic stress disorders that they develop after their exposure to stress situations. This work is composed of five sections, including the introduction in Sect. 1. The works related to the subject are described in Sect. 2 and the methodology used in Sect. 3. In Sect. 4 the tests and the results obtained are presented and finally, in Sect. 5 the conclusions and future work.

234

Z. Andrea Sánchez et al.

3 Methodology This section details the proposed system and the elements that comprise it. The general diagram is described in Fig. 1.

3.1 I/O Device Configuration The MYO bracelet is used as the input device and the HTC VIVE virtual reality glasses and binaural headphones as output devices. Next, the respective configuration of these hardware elements is presented. MYO Smart Band. To perform the digital processing of acquired signals, this device is connected to the MATLAB software through the Myo SDK MATLAB MEX Wrapper library. Through the SDK, information can be accessed provided from the eight electromyographic sensors and the inertial measurement unit (IMU). Among the most relevant data are those obtained from the three-axis gyroscope (angular velocity) and the three-axis accelerometer (linear acceleration). HTC VIVE. The connection of this device (virtual reality glasses) and the environment is generated natively in Unity 3D thanks to the SteamVR plugin. In addition, reorientation and movement scripts have been made due to the fact that HTC controllers are not used. In this device, the teleportation function is activated by default in standard mode, which is replaced by signals from the Myo bracelet. Binaural Headphones. They allow to reproduce the spatial sound that is generated in the virtual scene and in which the user can spatially identify the origin of the sound.

Fig. 1 General diagram of the implemented system

Virtual Rehabilitation System Using Electromyographic Sensors …

235

3.2 Signal Acquisition The signal acquisition process is performed based on the flow diagram described in Fig. 2. The Myo Armband device from Thalmic Labs integrates a set of eight noninvasive sensors, both electromyographic (EMG) and inertial. All sensors collect information that is processed in a central module (sensor number 4), which includes batteries and a communication device based on Bluetooth. Through libraries developed in C++, it has been possible to use it without restrictions letting know the developer certain parameters such as the measurement of each one of the electromyographic sensors, processed and unprocessed information of the inertial sensors, transmission periods, the gesture made by the patient, the values of accelerometers and gyroscopes, properties to eliminate the connected device, transmission rate, among others. Figure 3 shows a capture of the eight sensor’s signals in a 10 s period executing the open hand gesture. The amplitude units of each of the sensor signals are denoted here as maximum factory value (MFV), a dimensionless value used by the developers to represent the minimum and maximum amplitudes (−1 to 1) that the signal can have, given the internal preprocesses and validators (within the SDK).

Fig. 2 Flowchart of the algorithm used for signal acquisition

236

Z. Andrea Sánchez et al.

Fig. 3 EMG signal acquisition in 10 s of the gesture: open hand

3.3 Interface Development The interface design is made based on the flowchart of the interaction of programs, which also includes the configuration in MATLAB and the database presented in Fig. 4. Using Unity 3D software, a virtual environment has been developed, in which objects that simulate an open field with a road across it are placed. The aim of this application is to drive a van-type vehicle and completing a trajectory that is previously defined by itself. Tasks have been carried out in a simple way so as not to produce a stressful situation in patients, but instead they are entertained while performing their rehabilitation exercises. Script programming allows to identify five gestures that patients can perform: (i) fist, (ii) open hand, (iii) palm inward, (iv) palm outward, and (v) join thumb and middle fingers. The three-dimensional objects’ design has been made in 3ds Max software. Once all the necessary objects have been modeled in 3D, their compatibility is checked, the rotation points are located, hierarchies are established, and the axes of each element are oriented. After that, models are imported from 3ds Max using a format with extension *.fbx. Generally, the model of the imported objects is in gray color, or it can have certain properties of previously created materials, but commonly the colors and textures are eliminated. In this way, within the Unity 3D videogame engine, colors and textures are assigned to the 3D model, depending on the material and graphic quality required. In addition, the different tools offered by the palette of this software are used to incorporate the required grass and road, as shown in Fig. 5. Additionally, the respective environmental sounds are coupled, which complement the immersive experience of the patient who can feel how the car drives in the middle of nature.

Virtual Rehabilitation System Using Electromyographic Sensors …

Fig. 4 Flowchart of the interaction of programs

Fig. 5 Design of the interface in Unity 3D software

237

238

Z. Andrea Sánchez et al.

4 Tests and Results 4.1 Test When executing the application, the patient must perform various gestures with his hand and forearm in order to perform the tasks entrusted, as shown in Fig. 6a, b. When the first gesture is made, the vehicle is turned on and you can drive on the road. Once you are in driving mode, moving the palm of the hand outward produces a turn to the right; depending on the intensity of the gesture, the rotation is performed. The movement of the palm of the hand inward indicates that it is going to turn to the left. When you open your hand, it indicates that you have reached the desired point and so the vehicle must stop. In case that a bad maneuver has been made, the reverse option is incorporated by joining the thumb and the middle finger. Patients performed the entrusted exercises in 30 min’ sessions, 2 times a week. This treatment lasted 4 weeks and was carried out by 5 retired soldiers in a passive state (it must be

Fig. 6 a Tests performed by a patient and b virtual environment seen by the patient

Virtual Rehabilitation System Using Electromyographic Sensors …

239

differentiated that 2 of them required rehabilitation in both arms), with an age range between 48 and 63 years. The supervision of the therapist is very important, since it must corroborate that patient performs the exercises correctly so that there is an improvement in the treatment.

4.2 Results At the end of the application, each patient data and the values obtained in the videogame are saved in a text file that is automatically generated with the user’s name. To determine the level of acceptance of this proposal, the usability test SEQ has 14 questions, of which 13 have the values of 1–5 points according to the following scheme: The first seven questions (Q1–Q7) are related to the level of acceptance and immersion after the user experienced the virtual environment. The following four questions (Q8–Q11) are related to the effects and discomfort that the system could cause nausea, disorientation, or ocular discomfort. The next two questions are related to the difficulty encountered when performing the tests. The last question is open, so the user can indicate if there is any discomfort when using the virtual system and its reasons. If the result obtained is in the range of 40–65, the implemented system is considered acceptable. The questions asked to the users about the virtual system and the results of the SEQ usability questionnaire are shown in Table 1. The results of the SEQ test performed by five users after using the virtual system are: (58,8. ± 0,27). The score obtained is greater than 40. Therefore, it is determined that patients feel fully comfortable with the developed interface, so it represents an ideal tool to contribute to the rehabilitation process which requires an extended period of application.

5 Conclusions The experimental results after the implementation of this system show that it is a valid proposal for the rehabilitation of upper extremities. With this, it is pretended to give the patient a new tool to replace the conventional method; immersing his/her in a virtual world (fusing the sense of sight and hearing) a wide range of possibilities for physical rehabilitation is presented. An important fact to note is when dealing with users who were in active service and were subjected to stress situations, the execution of simple tasks and the simulation of an environment with nature help them to mitigate possible side effects of a psychological nature. The application of the usability test SEQ shows that the interface presented to the user is intuitive, easy to use, and does not cause discomfort. It should be pointed out that this research has been evaluated from the point of view of the acceptance generated by this system, but not with a medical approach, given that the results are not conclusive when requiring a longer evaluation period.

240

Z. Andrea Sánchez et al.

Table 1 Results of the SEQ test applied to users Question

Result (N = 5) Mean

SD

1. How much did you enjoy your experience with the system?

4.8

0.43

2. How much did you sense to be in the environment of the system?

4.8

0.43

3. How successful were you in the system?

4.4

0.83

4. To what extent were you able to control the system?

3.2

0.71

5. How real is the virtual environment of the system?

5

0

6. Is the information provided by the system clear?

4.8

0.43

7. Did you feel discomfort during your experience with the system?

4

0.43

8. Did you experience dizziness or nausea during your practice with the system?

4.8

0.43

9. Did you experience eye discomfort during your practice with system?

4.6

0.87

10. Did you feel confused or disoriented during your experience with the system?

5

0

11. Do you think that this system will be helpful for your rehabilitation?

4.2

0.71

12. Did you find the task difficult?

4.8

0.43

13. Did you find the devices of the system difficult to use?

4.4

0.43

Global score (total)

58.8

0.27

In this context, the authors of this research propose as future work the execution of this application in civil patients, in order to make a comparison and determine factors of incidence. In addition, the development of a system that allows the rehabilitation of the lower limbs is proposed, taking advantage of the already established design. Finally, these authors propose the use of a control algorithm to expand the number of gestures presented in this work and thus allow a better recovery process. Acknowledgements To the authorities of Universidad Técnica de Ambato (UTA), Dirección de Investigación y Desarrollo (DIDE), Instituto Tecnológico Superior Guayaquil Ambato and Celec EP, for supporting this work and future research.

References 1. Drakos, N.D., et al.: In good conscience: developing and sustaining military combat trauma expertise. J. Am. Coll. Surg. 227(2), 293–294 (2018) 2. Russell, C.A., Gibbons, S.W., Abraham, P.A., Howe, E.R., Deuster, P., Russell, D.W.: Narrative approach in understanding the drivers for resilience of military combat medics. J. R. Army Med. Corps. 164(3), 155–159 (2018) 3. Wells, T.S., Seelig, A.D., Ryan, M.A., Jones, J.M., Hooper, T.I., Jacobson, I.G., Boyko, E.J.: Hearing loss associated with US military combat deployment. Noise Health 17(74), 34–42 (2015)

Virtual Rehabilitation System Using Electromyographic Sensors …

241

4. Writer, B.W., Meyer, E.G., Schillerstrom, J.E.: Prazosin for military combat-related PTSD nightmares: a critical review. J. Neuropsychiatry Clin. Neurosci. 26(1), 24–33 (2014) 5. Stinner, D.J.: Improving outcomes following extremity trauma: the need for a multidisciplinary approach. Mil. Med. 181(4), 26–29 (2016) 6. Wilken, J.M., Roy, C.W., Shaffer, S.W., Patzkowski, J.C., Blanck, R.V., Owens, J.G., Hsu, J.R.: Physical performance limitations after severe lower extremity trauma in military service members. J. Orthop. Trauma 32(4), 183–189 (2018) 7. Galarza, E.E., et al.: Virtual reality system for children lower limb strengthening with the use of electromyographic sensors. In: International Symposium on Visual Computing, ISVC 2018, pp. 215–225. Springer (2018) 8. Gelman, D., Eisenkraft, A., Chanishvili, N., Nachman, D., Glazer, S.C., Hazan, R.: The history and promising future of phage therapy in the military service. J. Trauma Acute Care Surg. 85(1), 18–26 (2018) 9. Cools, A., Whiteley, R., Kaczmarek, P.K.: Rehabilitation of upper extremity injuries in the handball player. In: Handball Sports Medicine, pp. 433–459. Springer, Germany (2018) 10. Quevedo, W.X., et al.: Assistance system for rehabilitation and valuation of motor skills. In: International Conference on Augmented Reality, Virtual Reality and Computer Graphics, AVR 2017, pp. 166–174. Springer (2017) 11. Achanccaray, D., Acuña, K., Carranza, E., Andreu-Perez, J.: A virtual reality and brain computer interface system for upper limb rehabilitation of post stroke patients. In: 2017 IEEE International Conference on Fuzzy Systems (FUZZ-IEEE), pp. 1–5. IEEE (2017) 12. Levin, M.F., Weiss, P.L., Keshner, E.A.: Emergence of virtual reality as a tool for upper limb rehabilitation: incorporation of motor control and motor learning principles. Phys. Ther. 95(3), 415–425 (2015) 13. Liu, L., Chen, X., Lu, Z., Cao, S., Wu, D., Zhang, X.: Development of an EMG-ACC-based upper limb rehabilitation training system. IEEE Trans. Neural Syst. Rehabil. Eng. 25(3), 244–253 (2017) 14. López, V.M., Zambrano, P.A., Pilatasig, M., Silva, F.M.: Interactive system using myoelectric muscle sensors for the strengthening upper limbs in children. In: International Conference on Augmented Reality, Virtual Reality and Computer Graphics, AVR 2018, pp. 18–29. Springer (2018)

Part VIII

Strategy, Geopolitics and Oceanopolitics

The Portuguese Special Operations Forces as Instrument of Foreign Policy: The Case Study of Afghanistan João Reis and Sofia Menezes

Abstract The aim of this paper is to provide an exploratory overview of the Portuguese special operations forces as instrument of foreign policy. In doing so, we disclose new dynamics that Portugal employing in fragile states and we discuss its implications for the North Atlantic Treaty Organization. The research methodological approach falls into a qualitative case study research, which includes more than one method of data collection for triangulation and corroboration purposes. We have found that Portugal is trying to intensify its international relations within the EU, UN and the NATO, while the Portuguese special operations forces have been playing an important role to strengthen its presence within these international institutions. The new dynamics that Portugal is seeking are based on the democratic values, the rule of law and the human rights, to dissociate itself from the colonial legacy and thereby find new avenues of influence. Further research should focus on the new strategic priorities for the Afghan Government and NATO, as these priorities are changed at a relatively fast pace. Keywords Portugal · Afghanistan · Special operations forces · Foreign policy · Case study research · European Union · United Nations · North Atlantic Treaty Organization

1 Introduction This paper focuses the Portuguese special operations forces as instrument of foreign policy in Afghanistan. We have selected Afghanistan because this country has played a major role with great regional empires of Persia, India and China in late medieval J. Reis (B) Department of Military Science, Military Academy, GOVCOPP, Lisbon, Portugal e-mail: [email protected] S. Menezes Department of Management and Leadership, CINAMIL, Lisbon, Portugal e-mail: [email protected] © Springer Nature Singapore Pte Ltd. 2020 Á. Rocha and R. P. Pereira (eds.), Developments and Advances in Defense and Security, Smart Innovation, Systems and Technologies 152, https://doi.org/10.1007/978-981-13-9155-2_20

245

246

J. Reis and S. Menezes

and early contemporaneous period [1]. The country is known as the graveyard of empires [2] that reputation invokes the British and Soviet defeats in several wars in the nineteenth and twentieth century, to argue that Afghanistan is an unconquerable quagmire that inevitably dooms even the strongest army that venture there [3]. Moreover, Afghanistan was the first major test for state-building in the twenty-first century and embroiled the USA and its allies for almost two decades [4]. Afghanistan required virtually every North Atlantic Treaty Organization (NATO) member’s participation, including Portugal, and its participation is far from certain. Most probably, the economic and political assistance will continue, even after the Afghan forces assumed full security responsibility at the end of 2014—the new, smaller, non-combat mission Resolute Support (January 1st, 2015—until recent days) began to train, advise and assist (TAA) the Afghan security forces and institutions [5]. We have chosen the well-trained special operations forces instrument due to its strategic value, as they were the first to be deployed to Afghanistan, and they will be probably the last to redeploy. The objective of this research is to ascertain the suitability of the Portuguese special operations forces within the NATO Resolute Support Mission (RSM) and its relevance as an instrument of foreign policy for Portugal. We have structured this paper in four sections after the introduction, as follows: we begin by a review of the literature; secondly, we describe the methodological approach; then, we investigate the relevance of the special operations forces to NATO and Afghanistan; in the last place, we provide the implications and suggestions for subsequent research.

2 Theoretical Background 2.1 Allied and the Portuguese Army Special Operations Forces NATO Allied Joint Publication (AJP) defines that special operations as military activities conducted by specially designated, organized, trained and equipped forces, manned with selected personnel, using unconventional tactics, techniques and modes of employment—special operations deliver strategic or operational-level results or are executed where significant political risk exists [6]. On the other hand, the NATO special operations headquarters define special operations as operations that may be conducted across the spectrum of conflict as part of article 5 (collective defence) or non-article 5 (crisis response operations) to fulfil NATO’s three essential core tasks—collective defence, crisis management and cooperative security [7]. The NATO special operations headquarter also describes the special operations missions to include suitable combination or all of the principal tasks of military assistance (MA), special reconnaissance (SR) or direct actions (DA), depending on the circumstances of each operation. While special operations mission may range from

The Portuguese Special Operations Forces …

247

small unilateral actions to large-scale activities of a combined and joint nature, they are tailored to contribute to the accomplishment of the defined political and military/strategic objectives. Each type of special operations missions may be defined as follows: DA is the use of a rapid offensive operations to destroy or capture a target, usually by means of secrecy and surprise [8, 9]; SR involves covert collection of intelligence typically in denied enemy areas [9, 10]; MA involves the provision of training and advising to host nations to improve their skills and capabilities [8, 9]. In Portugal, an order from the Chief of Staff of the army (directive nº 90), October 28th, 2007, determines that special operations are divided in special reconnaissance, direct actions and military assistance. The forces that develop special operations may be employed across the spectrum of operations. In view of its unique characteristics, it constitutes adequate forces to perform tasks within the crisis response operations (CRO), combat operations for search and rescue (CSAR) and counterterrorism operations, consisting in forces especially suitable in counterterrorism [11]. The same directive from the Chief of Staff of the Portuguese army states that: (1) SR objective is to obtain strategic and operationally sensitive information, it can use sophisticated methods, techniques (e.g. human intelligence) and equipment [11]; (2) DA is a precise action, which is usually performed on well-defined targets, high criticism and high value, with strategic or operational significance. The use against tactical objectives is justified only by the action of criticality or by the inadequacy or unavailability of the use of other forces [11]; (3) MA consists of a broad framework of measures in support of friendly or allied forces and populations in peace, crisis or conflict, in situations of need for security and survival, but also development. The MA also fits into all activities related to the actions of military-technical cooperation, support in situations of disaster and conflict mediation [11]. The missions developed by the Portuguese special operations forces are similar to reference countries and NATO members. This section has the objective to define the elementary concepts, which are going to be necessary for the following sections.

2.2 Resolute Support Mission and the Afghan Government The RSM in Afghanistan is focused on train, advise and assist (TAA) the Afghan security forces. Historically, the USA special operations forces (SOF) have had the responsibility to TAA foreign military forces. Even if today, the US National Security Strategies introduced the US military services (Army, Navy, Air Force and Marines) to organize, train and equip themselves to carry out these activities at a larger scale with conventional (non-SOF) forces [12]. Livingston [12] remarks this responsibility in a broader sense of building the capacity of partner states that has been termed security force assistance (SFA). SFA is part of the USA and NATO strategic goal of having Afghanistan responsible for their security. Usually, SFA is directly linked to counterterrorism strategy and is the key to engage underdeveloped and undergoverned nations, often referred to as weak or fragile states.

248

J. Reis and S. Menezes

In Afghanistan, the SFA mission is in part attributed to the NATO special operations component command—Afghanistan. This command brought together NATO and other coalition special operations forces into the same operational headquarters, and it demonstrates both that NATO special operations forces have made major progress in integration and can potentially be a major part of the alliance capability [9]. The created synergies among special operation forces community achieve significant effect due to the combination of different techniques, technologies and expertise. Long [9] reinforces the idea of NATO special operations forces that actively performs three main allied special operations forces missions, i.e. military assistance, special recognizance and direct actions; but particularly, the military assistance mission to develop the Afghan police special units was its core mission. Strandquist [13] also performed a research which suggests the USA and their allies should change their strategy in Afghanistan, since developing military combat operations against the insurgents is not enough and the future will have to go through a focus on the development of local communities to counter-insurgency and that the Afghan central government should win their support. During the war in Africa (1961–1974), the Portuguese armed forces also used the same techniques, in order to obtain public support, as it may be one of the key points to win the war in Afghanistan. This technique was also used in Vietnam and it had been termed “winning hearts and minds” [14]. The term was firstly used by Louis Lyautey, a French army marshal and minister of war. The term included not only military, but also social actions and improvements in the quality of life [15]. Later, Grochowski [16] published an article concerning the Takur Ghar battle, a short, but intensive, military operation that engaged the US special forces and Al Qaeda insurgents in Afghanistan. According to Grochowski [16], the Takur Ghar battle demonstrated the short lead-time and short-duration high-intensity tempo of the special operations missions regarding the war against non-state actors and particularly violent extremist organizations. The course of Afghan conflict and the commitment of the coalition of special operations forces has focused on the tactical objectives, but with time, this path has changed, i.e. the current conflict requires special operations forces highly committed to TAA the Ministries of the Government of the Islamic Republic of Afghanistan (GIRoA), which creates strategic and political objectives for a rapid development of the Afghan Security Forces and its population.

2.3 Brief Overview of the Portuguese Foreign Policy Between April 1974 and January 1986, the Portuguese foreign policy oscillated between two basic orientations, which marked two distinct phases: the transition to democracy and the democratic consolidation [17]. Teixeira [17] also portrays the Portuguese position in the Euro-Atlantic binomial, in the post-colonial relations regarding the new multilateral triangle and based on foreign policy values, which we summarize as follows: (1) The Europe-Atlantic binomial remains the geopolitical equation: Portugal as an EU country seeking to promote and enhance not only the

The Portuguese Special Operations Forces …

249

Atlantic position between Europe and the USA, but also the post-colonial relations (African Portuguese-speaking countries), the key for the strategic guidelines of the foreign policy of democracy [17–19]; (2) Portugal seeks participation in terms of producing international security, such as the military presence in organizations as NATO, EU and UN (the new multilateral triangle), which has provided not only a factor of military modernization, but also the external credibility of the Portuguese State. The armed forces have become a foreign policy tool [17, 19, 20]; (3) The challenge is in the global plan of international security, in an environment marked by terrorism, ethnic and religious conflicts; that is why the framework of security has to be cooperative security [17, 21]; (4) The presence of the Portuguese armed forces in peacekeeping missions under the umbrella of the UN confirms the diplomacy of values that had the independence of Timor as one of its major hits [17]; (5) Lastly, foreign policy always comes from the formulation of interests. However, the contemporaneous democratic model ignores the interests of the states and focuses on the human values. The values for Portugal are democracy, the rule of law and human security [17]. Similarly, Santos [22] argues about the use of the armed forces which is one of the most important instruments available to support its foreign policy, giving it a specific weight within the security scope.

3 Methodology The methodological approach falls into a qualitative case study research. This research therefore uses a contextual rich data setting from the real world to investigate a phenomenon which is understudied. We do use multiple sources of data collection [23], i.e. semi-structured interviews, official documents and direct observations, for data triangulation and corroboration purposes [24]. The data collection method focused on ten interviews from two military branches of the NATO Result Support Mission in Afghanistan; the non-classified documents were collected from open sources; and the direct observations have collected from informal interviews and personal field notes of the observed reality. The field study took place in Kabul, Afghanistan during the following periods: February to July of 2015 and from August 2017 to February 2018. Furthermore, for confidential reasons, all the respondents kept their anonymity and the collected reports came from unclassified sources. The goal of the case study is to gain a holistic view of the phenomenon and to understand the specific reality that is being discussed in this paper [25].

3.1 Data Collection Techniques The interviews were the primary data collection method for this case study. Quivy and Campenhoudt [26] reported the interviews as a method that differs from others

250

J. R eis a n d S. M e n e z es

i n t h e e xt e nt t h at it r ef ers t o t h e f u n d a m e nt al pr o c ess es of c o m m u ni c ati o n a n d h u m a n i nt er a cti o n. T h e s a m e a ut h ors als o m e nti o n e d t h at t his t e c h ni q u e, w h e n pr o p erl y v alu e d, e n a bl es t h e r es e ar c h er t o wit h dr a w i nf or m ati o n a n d v er y ri c h r e fl e cti o n el e m e nts fr o m i nt er vi e ws. H o w e v er, w e k n o w t h at, s o m eti m es, r es p o n d e nts c a n pr o vi d e t h e i nt er vi e w er wit h t h at h e w a nts t o h e ar, r es ulti n g i n p o or a n d i n a c c ur at e r es p o ns es [2 4 ], w hi c h m a y b e miti g at e d t hr o u g h t h e us e of ot h er d at a c oll e cti o n t e c h ni q u es. T h e s a m pl e s el e cti o n c o m pr e h e n d e d t e n i nt er vi e ws, di vi d e d i nt o t w o gr o u ps of fi v e. B ot h gr o u ps s er v e d u n d er t h e c o m m a n d of t h e N A T O s p e ci al o p er ati o ns c o mp o n e nt c o m m a n d — Af g h a nist a n. T h e first fi v e ar e s er vi c e m e n fr o m t h e U S s p e ci al o p er ati o ns a n d t h e s e c o n d gr o u p ar e s er vi c e m e n fr o m t h e P ort u g u es e ( P R T) s p e ci al o p er ati o ns f or c es. T h e U S a n d P R T of fi c ers h a v e i nt er a ct e d wit h t h e m ost s e ni or l e vels of t h e Af g h a n Mi nistr y of I nt eri or a n d dir e ctl y i n fl u e n c e d t h e l o n g-t er m str at e g y t o s ust ai n t h e Af g h a n S p e ci al S e c urit y F or c es ( A S S F). T o g et diff er e nt p ers p e cti v es, w e h a v e als o c oll e ct e d a c o ntri b uti o n fr o m s o m e P ort u g u es e c o m missi o n e d of fi c ers t h at h a v e w or k e d as st aff of fi c ers fr o m t h e S p e ci al O p er ati o ns A d vis or y A n al ysts. T h e c as e st u d y pr ot o c ol w as b as e d o n t h e P ort u g u es e N ati o n al D ef e n c e Str at e g y ( P N D S), w hi c h w as a p pr o v e d b y t h e r es ol uti o n of t h e P ort u g u es e c o u n cil of mi nist ers 9/ 2 0 1 3, A pril 5t h, 2 0 1 3. T h e P N D S w as d e v el o p e d a c c or di n g t o t hr e e m ai n v e ct ors: ( 1) n ati o n al s o v er ei g nt y — n e utr ali z ati o n of t hr e ats a n d ris ks t o n ati o n al s e c urit y, ( 2) a cti o ns t o n ati o n al v ul n er a biliti es a n d ( 3) e n h a n c e m e nt of t h e r es o ur c es a n d n ati o n al o p p ort u niti es. E a c h v e ct or is di vi d e d of s e v er al str at e gi es ( 2 0 str at e gi es i n t ot al), a n d e a c h str at e g y is s u b di vi d e d i nt o li n es of a cti o n ( 1 1 6 li n es of a cti o n i n t ot al) —f or t h at r e as o n, it w as i m p ossi bl e t o dis pl a y all t h e N ati o n al D ef e n c e Str at e g y str at e gi es i n s h ort p a p er. T h e n ati o n al v e ct ors, t h e str at e g y a n d t h e s e v er al li n es of a cti o n c o ntri b ut e d dir e ctl y or i n dir e ctl y t o t h e a c hi e v e m e nt of p er m a n e nt a n d s h ort-t er m n ati o n al g o als i n Af g h a nist a n. T h e c o ntri b uti o n of b ot h gr o u ps w as b as e d o n t h e c h oi c es of e a c h ( 1) v e ct or, t h e c orr es p o n di n g ( 2) str at e g y a n d t h e ( 3) li n es of a cti o n fr o m t h e P N D S ( vi d e Ta bl e 1 ). T a bl e 1 Ve ct ors, str at e gi es a n d li n es of a cti o n of t h e V e ct ors Str at e gi es E x er cis e s o v er ei g nt y, C o ns oli d ati n g n e utr ali z e t hr e ats a n d e xt er n al d ef e n c e ris ks t o n ati o n al r el ati o ns s e c urit y D ef e n di n g P ort u g al’s i nt er n ati o n al p ositi o n

P ort u g u es e n ati o n al d ef e n c e str at e g y Li n es of a cti o n

I nt e nsif yi n g e xt er n al r el ati o ns a n d d ef e n c e c o o p er ati o n wit h t h e U S R e affir m t h e str at e gi c i m p ort a n c e of t h e i n cl usi o n of P ort u g al i n N A T O, as w ell as t h e bil at er al alli a n c e wit h t h e U S

C o ns oli d ati n g e xt er n al d ef e n c e r el ati o ns

P arti ci p at e i n i nt er n ati o n al milit ar y p e a c e missi o ns, p arti c ul arl y i n t h e c o nt e xt of t h e U N, N A T O a n d E U

D ef e n di n g P ort u g al’s i nt er n ati o n al p ositi o n

Str e n gt h e n t h e pr es e n c e of t h e P ort u g u es e i n i nt er n ati o n al i nstit uti o ns

The Portuguese Special Operations Forces …

251

To ensure the data corroboration, we have used other data collection techniques, i.e. secondary sources, such as direct observations and official documents. We did not give priority to these data collection techniques, but they were essential for triangulation and corroboration purposes. The direct observations came from field notes that were taken during the research process and from the author’s own experience, since he has also served as a special operations operator in Afghanistan. Furthermore, we have also collected data from the frequent discussion with NATO servicemen about their judgement regarding the Portuguese special operations forces contribution to NATO and, consequently, to Afghanistan. The informal discussions were registered and once they provided a better understanding of the phenomenon, they were taken into account. We also collected official documents from NATO website that were relevant to provide the Resolute Support Mission perspective and the commander intention about all NATO countries contribution to the overall mission. To analyse the collected data, we defined a set of categories and subcategories to search associations between the various sources of data collection. Then we sought to establish relations between the various categories in order to draw conclusions. The NVIVO10 was used as a qualitative data analysis software to organize, analyse and finds insights into unstructured or qualitative data.

4 Findings This section provides an empirical summary of the data analysis, which includes real-life statements from the respondents. The most valued aspect attributed by the US special operations forces is regarding the knowledge acquisition by the Portuguese forces. The respondents believed the Portuguese special operations forces are looking forward to acquire new experiences and new competencies through the contact with other international military forces, especially in counterinsurgent environments. This exchange of experiences is allowing the Portuguese servicemen to increase their technical skills, which can be boosted by continued experience sharing with contributing NATO nations. These relationships mainly promote the exchange of relevant information that allows technical innovation with European partner. A US special operations serviceman shares this view: I believe that Portugal is participating at this NATO mission (Afghanistan) to intensify its relations with the United States and to better know our modi operandi. This aspect is relevant for us because it allows for a better interoperability within NATO, but also because we can work with our allies (special forces operator, 32 years old, US)

The US special operations operators also consider Portugal as a strategic partner relevant to the USA, as it is a country within the European Union that allows strengthening the transatlantic ties, because of its relative proximity to the USA—being the most western country in Europe. They believe it is increasingly relevant to have a

252

J. Reis and S. Menezes

close cooperation with the Portuguese special operations forces, as well as with all the European countries, as they take into account the European contribution to the international security—cooperative security. The arguments are well evidenced in by a US special forces team leader: Portugal has a very close relationship with the US. To my best knowledge, we have a US base in a Portuguese Island (Azores) that works as a strategic platform to Europe. The relationship we have with Portugal is not only due to its geographical position, since we also have had several special operations exercises, namely cross training in infiltration techniques (special forces team leader, 42 years old, US)

From the second group of respondents, it was possible to ascertain that the most valued aspects were the participation in missions that have contributed to peace and international security, especially the ones that have strengthening the presence of the Portuguese in international organizations. They believe that Portugal has a stable international status as a western democracy, a full member of the EU and NATO, which gives the country the possibility to actively participate in the international panorama. The participation in international missions, namely through elite forces, creates bonds between Portugal and the countries of the alliance and contributes to improve the country´s visibility and credibility. These arguments are well evidenced by the statement of a Portuguese special operations operator: The Portuguese special operations forces were created as a counter-insurgency unity for the Portuguese operations in Africa, during the colonial war period (1961–74). After the revolution of April 25th, 1974, the Portuguese special operations had to change their way of employment. The special operations are now used as an icon of national sovereignty, which acts in the international context as a co-producer of security while working with their international partners (special forces operator, 32 years old, Portugal)

These respondents also believe the use of the Portuguese special forces intensifies the relations and defence cooperation with the US military. A good example they provided is the participation of the Portuguese Commandos in Kabul Capital Division, at the time under the mandate of the International Security Assistance Force (ISAF) and, very recently, with the deployment of special operations forces to which were assigned several roles, e.g. intelligence, ministerial engagement, etc., under the mandate of the Resolute Support Mission (RSM). The Portuguese special forces have long contributed to the develop of the Afghan National Defence and Security Forces capabilities. I personally work with the General Command of Special Police Unit (GCPSU), which is an Afghan police unit responsible for responding to high-profile attacks and to conduct high-risk arrests. The most important for us is to participate in the peacekeeping operations and to strengthen the Portuguese presence within the international institutions i.e. NATO. Yes, the Portuguese special forces are operating in Afghanistan almost since the beginning of the conflict, I am just a small piece this wellorganized and very professional institution (special forces operator and ministerial advisor, 34 years old, Portugal)

During the discussions, we also noticed that Portugal is seeking new dynamics, based on the democratic values, the rule of law and the human rights, and thereby find new avenues of influence. We observed that the Portuguese special operations are

The Portuguese Special Operations Forces …

253

also trying to get more influence within the Afghan Ministry of Interior (rule of law), but also investing on issues such as the gender integration in the Afghan defence and security forces (human rights). Therefore, further research should investigate these topics, i.e. new strategic priorities for GIRoA and the NATO. For instance, it would be interesting to investigate how the Portuguese special operations forces can improve the integration of Afghan women into their special operations force units. The data analysis also demonstrated the Portuguese special operations missions—military assistance, falls under the overall mission of the NATO special operations component command. Regarding the political foreign objectives, the Portuguese special operations forces capacities and the missions performed by the NATO special operations forces are perfectly aligned and have two main goals: (1) influence the long-term strategy and sustainability of the Afghan defence and security forces. I feel that we have an effective influence on the defence and security policy. Our Afghan counterparts are generals and politicians form the Afghan Ministry of Interior. At the same time that we have received the milestones from NATO and Portugal, we can decisively influence the long-term sustainability of the Afghan special forces (informal discussion) and, to develop activities (2) which can strengthen its international present and to stimulate cooperative security relations with their international partners. In recent years there has been an increase of special operations servicemen to be deployed to Afghanistan, which may be a reflection of this policy (field notes)

By using its elite forces in Afghanistan as instrument of foreign policy, Portugal reaches: 1 national vector, 4 strategies and 4 lines of action, from the PNDS. Table 1 summarizes the selected vectors, strategies and actions by the respondents. In light with the above, we view the Portuguese strategy in Afghanistan from two different perspectives. At first, it was interesting to observe that questions related to the scope of the Portuguese mission in Afghanistan, i.e. military assistance, the military discourse was focused on strengthening the Portuguese presence within the international institutions and the military participation in peacekeeping operations, in particular contexts such as the UN, NATO and EU (Table 1—light grey). However, when we evaluated the Portuguese foreign policy, we cross-checked the interviews from both groups of respondents and we verified the existence of a strong relation between the USA and Portugal, with the intention to keep strengthening the transatlantic bonds (Table 1—dark grey). The Portuguese strategies and line of action are consistent within the PNDS, i.e. all the strategies and lines of action falls under the same national vector: to exercise sovereignty, neutralize threats and risks to national security (vide Table 1). The purpose of this vector is to defend national interests and guarantee the security and defence of Portugal, reduce vulnerabilities and execute the international responsibilities of the state and contribute to the realization of the Portuguese vision of the constructive values of the international order [27].

254

J. Reis and S. Menezes

5 Concluding Remarks We have concluded that Portugal is trying to intensify its international relations within the EU, UN and NATO. To this end, the Portuguese armed forces have been playing an important role to strengthen its presence within the international institutions. We have verified that the Portuguese special operations operators are trained well enough to operate in complex environments, as the Afghan mission falls under one of the main special operations missions—the military assistance. In other words, the mission that brings the Afghan Ministerial capabilities to been raised is suitable for the Portuguese special forces. Portugal, as a small European state, has also tried to contribute along with the USA to spread the rule of law. In Afghanistan, once NATO moved from a combat role to a TAA mission, the Portuguese special operations forces have been developing a strategy to build the Afghan Government, as well as the Afghan National Defence and Security Forces. Thus, under the NATO special operations component command, the Portuguese special operations operators have played one of the most demanding missions in the contemporary Portuguese military history. We therefore closely observed some Portuguese servicemen that helped decisively to develop the capabilities of the Ministry of Interior, in order to operate the Afghan police forces and its special units, e.g. GCPSU. What is new is that Portugal is seeking new dynamics—based on the democratic values, the rule of law and the human rights—to dissociate itself from the colonial legacy. Portugal is therefore, trying to get more influence, by using its armed forces, to rebuild fragile states in remote geographical areas, along with their international partners. This paper also presents some limitations, as the identity of the respondents may not be revealed, thus to clarify any questions to address this subject the first author should be contacted. The number of interviews was also limited, once it was restricted to the number of respondents available in a specific period of time. The study is also exploratory in nature and it lacks generalization. The lack of generalization is due to different political goals and different existing types of special forces; thus, the results may differ across NATO countries. Nevertheless, we do believe this study may be useful to other researchers, in the extent that are no similar studies so far; to practitioners, this paper may be relevant too, as it can provide new and fruitful insights into the strategic engagement of special operations forces in fragile states.

References 1. Seddon, D.: Imperial designs. Crit. Asian Stud. 35(2), 175–194 (2003) 2. Jones, S.: In the graveyard of empires: America´s war in Afghanistan. WW Norton & Company, New York (2010) 3. Miller, P.: Graveyard of analogies: the use and abuse of history for the war in Afghanistan. J. Strateg. Stud. 39(3), 446–476 (2016)

The Portuguese Special Operations Forces …

255

4. Nojumi, N.: American State-Building in Afghanistan and its Regional Consequences: Achieving Democratic Stability and Balancing China’s Influence. Rowman & Littlefield (2016) 5. NATO.: NATO and Afghanistan. North Atlantic Treaty Organization (2016). http://www.nato. int. Accessed 1 Dec 2018 6. AJP 3.5.: Allied joint doctrine for special operations. North Atlantic Treaty Organization (2013) 7. NATO SOF.: NATO special operations headquarters: Special operations forces. Military Assistance Handbook. 1st study draft, North Atlantic Treaty Organization (2014) 8. AAP-06.: NATO glossary of terms and definitions. North Atlantic Treaty Organizations. 2-D-7 and 2-S-9 (2013) 9. Long, A.: NATO special operations: promise and problem. Obris 58(4), 540–551 (2014) 10. Dellinger, B.: Special operations command Europe: strengthening partnership for global security. Spec. Warf. (2012) 11. Chefe de Estado-Maior do Exército.: Âmbito de atuação, missões e tarefas das unidades de tropas especiais. Divisão de Planeamento de Forças, diretiva nº9 do Chefe de Estado-Maior do Exército (2007) 12. Livingston, T.: Building the capacity of partner states through security force assistance. Security Force Assistance in the Development of Foreign Forces. Nova Science Publishers (2011) 13. Strandquist, J.: Local defence forces and counterinsurgency in Afghanistan: learning from the CIA´s village defense program in South Vietnam. Small Wars Insur. 26(1), 90–113 (2015) 14. Anderson, D., Ernst, J.: The war that never ends: new perspectives on the Vietnam War. University Press of Kentucky (2014) 15. Tucker, S., Roberts, P.: The encyclopedia of World War II: A political, social, and military history. ABC-clio (2004) 16. Grochowski, G.: Planning for the next Takur Ghar. Def. Secur. Anal. 31(2), 152–158 (2015) 17. Teixeira, N.: Breve ensaio sobre a política externa portuguesa. Relações internacionais, R:I(28), 51–60 (2010) 18. Ruivo, L.: Conceito estratégico de defesa nacional contributos para uma revisão. Boletim Ensino, Instituto de Estudos Superiores Militares (2011) 19. Silva, J.: As forças armadas como instrumento de política externa portuguesa no actual contexto internacional. Trabalho Individual de Longa Duração do Curso de Promoção a Oficial General, Instituto de Estudos Superiores Militares (2006) 20. Azambuja, M.: As Nações Unidas e o conceito de segurança coletiva. Estudos Avançados 9(25), 139–147 (1995) 21. Serronha, M.: A cimeira de Lisboa: Uma NATO para o século XXI. Instituto de Defesa Nacional 126–5ª série, 37–46 (2010) 22. Santos, J.: O conceito estratégico da NATO: Superar contradições, manter a coesão. Nação e Defesa, 37–46 (2010) 23. Barratt, M., Choi, T., Li, M.: Qualitative case studies in operations management: Trends, research outcomes, and future research implications. J. Oper. Manag. 29(4), 329–342 (2011) 24. Yin, R.: Case study research design and methods. Appl. Soc. Res. Methods Ser. 5 (2003) 25. Benbasat, I., Goldstein, D., Mead, M.: The case research strategy in studies of information systems. MIS Q. 369–386 (1987) 26. Quivy, R., Campenhoudt, L.: Manual de investigação em ciências sociais (1998) 27. Governo de Portugal.: Grandes opções do conceito estratégico de defesa nacional (2013). https://www.parlamento.pt/sites/COM/Paginas/DetalheNoticia.aspx?BID=5034. Accessed 1 Dec 2018

The Internal–External Security Nexus: EU Operation Sophia Through the Lens of Securitization Ana Paula Brandão

Abstract The construction of EU security actorness has been accompanied by a narrative on security nexuses associated with the “comprehensive approach”. This trend has been intensified in recent years as demonstrated by the discourse on the “refugee crisis”, Daesh activity, hybrid threats and border security. This paper focuses on the internal–external security nexus analysed through one of its materializations notably the interface between the CSDP and the AFSJ. Based on the EU Naval Force Operation Sophia, it is argued that the comprehensive approach underlying the logic of the nexus is the result of a co-constitutive adequacy: “more security” (appropriation of policies and instruments of a multifunctional actor for security purposes) and “more actorness” (securitization of issues in order to promote the actor and its policies). The main contribution of the paper is to think critically on why and how security narratives, military instruments and securitization dynamics serve convergent processes of gaining political and public space for legitimizing policies and actions. Keywords CSDP · AFSJ · Security nexus · Securitization

1 Introduction In a context in which the European Union (EU) discourse has been fertile in identifying Europe’s challenges in a globalized world, amongst which we find the postWestphalian security challenges, it is paramount to reflect upon the rationale and effects of the evolving EU security actorness. The Union has been innovative in creating a de facto security community that overcame the European interstate conflict, and since the end of the Cold War, it endeavoured both to be a security actor (with global reach and through a comprehensive approach) and to address the multisector and transnational threats of a complex security environment. A. P. Brandão (B) CICP—University of Minho, Braga, Portugal e-mail: [email protected] © Springer Nature Singapore Pte Ltd. 2020 Á. Rocha and R. P. Pereira (eds.), Developments and Advances in Defense and Security, Smart Innovation, Systems and Technologies 152, https://doi.org/10.1007/978-981-13-9155-2_21

257

258

A. P. Brandão

EU security actorness reinforces and is reinforced by the narrative on the security nexuses: internal–external security; public–private security; civilian–military security; security development. This narrative and its practices raise several questionings about the what (means the nexus), the why, the how (to implement it) and the “with what effects”. The paper focuses on a Common Security and Defence Policy (CSDP) operation which demonstrates in the field the nexus between internal and external security (IESN), and more specifically the link between the CSDP and the area of freedom, security and justice (AFSJ). Resorting to the securitization theoretical framework combined with a conceptual matrix on security actorness [1], applied to the analysis of the European Union Naval Force Operation Sophia (a recent operation dealing with complex areas of law enforcement, trafficking of human beings and migration in the neighbourhood), it is argued that it reflects a securitizing move of the European actor explained by the convergence of opportunity (redefinition of security and prioritization of transnational threats in a globalized world), capacity (legal, institutional and operational capacity in the field of security), and (ambition to have a global) presence. The comprehensive approach (CA) underlying the logic of the nexus is the result of a coconstitutive adequacy: “more security”—appropriation of policies and instruments of a multifunctional actor for security purposes (security of the EU and of European citizens); “more actorness”—securitization of issues in order to promote the actor and its policies.

2 The Internal–External Security Nexus The interdependence between the internal and external dimensions is a transverse view to the official EU documents relating to security [2]. What does this interdependence mean? Both in the political and academic contexts, different expressions, not necessarily synonymous, have been used to refer to the phenomenon. This wording cacophony does not facilitate the work of policy-makers and academics. The scientific field has been marked by “empirical ambiguity, theoretical fragmentation and a lack of scholarly dialogue on this issue” [3]. Historically, the study of security, associated with state polity, was based on the separation between “the two arms of the Prince” [4]. The complexity of the phenomenon, associated with the diversification of threats and the multitude of actors, either as providers of security or as a source of threat in the context of intense mobility and communicability worldwide, bucked the traditional paradigmatic, political and institutional separation between the internal and external dimensions of security defined by the realist legacy. The end of the Cold War and the events of 11 September 2001 potentiated the broad conceptualization of security covering four areas: security sectors (multisectoral security beyond political and military sectors); security objects (multiple actors, including individuals and groups beyond the state); security players, either as security providers or as sources of threat; cross-border dynamics (transgovernmental cooperation for security; actions of transnational enti-

The Internal–External Security Nexus: EU Operation Sophia …

259

ties for security purposes; perverse transnational actors). In sum, an evolving trend on internal and external security, from the traditional approach as separated phenomena (Realism) to the perspectives as interdependent (Neoliberalism) or even as merged phenomena (Paris School) [5]. As far as the European Union is concerned, the nexus can be applied to different phenomena which, in short, stem from three dynamics: (a) internalization of external phenomena; (b) externalization of internal phenomena; (c) cross-border phenomena. Underlying the in/out narrative is the idea of “globalization of security” associated with the “predominantly transnational character of postmodern risks” [6]. In this context, a CFSP that is effective in preventing and combating external threats is considered to be a condition to ensure the internal security of the European area. In turn, an effective internal security system is understood as a condition for the former to be an active policy. This “indissoluble link between internal and external aspects of security” [7] is explained by several phenomena, namely: Europe’s vulnerability due to its reliance on an infrastructure interconnected in various areas (transport, energy, information); the external dimension of organized crime; the global nature of terrorism, which has increasing resources, including connection through electronic networks; proximity to troubled areas; regional conflicts that have direct or indirect impact on European interests; climate change that has a “threat multiplier effect” [8]. Thus, in the “era of globalization, distant threats may be as much a concern as those that are near at hand” so “the first line of defence will often be abroad” [7], and it is therefore necessary to “improve the way we reconcile the internal and external dimensions” [8]. The most recent events, particularly in the field of terrorism, have contributed to intensifying the in-out nexus security narrative. In the words of Federica Mogherini, the fight against radicalization and violent extremism must continue to be “a priority, not only for internal and security action, but also for our diplomatic and foreign policy” [9]. The EU agenda on IESN has extended the list of priority issues: transnational organized crime (since the 1990s), terrorism (since 9/11), migrant smuggling and human trafficking (post “refugee crisis”), the link between CSDP-FSJ: “The linkages between internal and external security, including in areas such as irregular migration, trafficking of all sorts, terrorism and hybrid threats are an increasingly important context for the further development of CSDP and when considering any possible new operation/mission” [10]. After the date of entry into force of Lisbon Treaty changes, the CSDP-FSJ link was the subject of a structured reflection contained in the joint report of the commission and the HR, which resulted in an operational roadmap [11] with five areas and 27 lines of action (12 short-time priorities and 15 medium-term priorities). Those initiatives of reflection and operationalization are concerned specifically to the external component (CSDP missions and operations) and the externalization of EU internal security (external dimension of AFSJ). The internalization (of CSDP) component is being approached in a separated vector, in particular associated with the solidarity clause. The civilian missions of CSDP are one of the tangible expressions of the IESN, particularly with regard to security goals, civilian capabilities, planning, situational awareness and strategic analysis [12]: civilian missions of Rule of Law, Security

260

A. P. Brandão

Sector Reform designated to reduce the criminal activity (organized crime and corruption) in the host country, through judicial means, thereby contributing to greater security and stability not only in that country, but also at regional and international levels, and creating the conditions for cooperation with EU and its agencies (EUROPOL, EUROJUST, FRONTEX); missions sustained by internal police and judicial capabilities; crisis management structures which integrate elements from both domains (CSDP and FSJ) in the planning of new missions and revision of ongoing missions; situational reports, risk and threat assessment, early warning and alerting about terrorist attacks and external crisis, which cover both domains (CSDP and FSJ), as demonstrated by the activity of the EU Situation Centre (SiteCen), EUROPOL and FRONTEX. The second manifestation of the CSDP-FSJ link regards to intelligence gathering and sharing whether in Brussels or in the field of the missions/operations, which has been fostering administrative arrangements between the General Secretariat of the Council and FSJ agencies as well as the interaction between these agencies and CSDP civilian missions. The assessment of the first years of implementation of the roadmap (four progress reports) approved in 2011, and the need to adequate it to situational changes led the Council to define three priority areas of action [10]: improving situational awareness and exchange of information within the EU; operationalizing the nexus between internal/external security, especially by improving the cooperation among the EU agencies and between these and the EEAS, the joint training, and the coordination and decision-making process mechanisms; civilian–military convergence and synergies among EU missions and operations. Although the documents value the CSDP civilian missions in the link between this policy and FSJ, we find relevant to focus on a maritime operation which materialize the IESN through military means used to deal with the (so-called) new type threats [13].

3 Operation Sophia: The CSDP-AFSJ Link Through Military Means In order to tackle the so-called refugee crisis, the EU adopted a comprehensive approach (CA) based on short-time (emergency support) and medium-long efforts, multisectoral policies (humanitarian, migration, development, security and external relations), multilevel (European, national, local) and multi-actor action (EU institutions and governments, partner countries, partner organizations, international agencies, NGOs). The use of military means (CSDP operation) and the concentration on the sea (maritime operation) were decided after the tragic sinking and death of hundreds of migrants attempting to cross the Mediterranean [14]. On 20 April 2015, the European Commissioner for Migration, Home Affairs and Citizenship presented a ten-point action plan on migration in response to the crisis in the Mediterranean. The second

The Internal–External Security Nexus: EU Operation Sophia …

261

action of the plan consisted in the capture and destruction of vessels used by the smugglers. This was reiterated by the European Agenda on Migration (May 2015) as an immediate key action. The Crisis Management Concept of EUNAVFOR MED was approved by the Council on 18 May and the operation launched on 22 June 2015. The mandate of the operation “disruption of the business model of human smuggling and trafficking networks in the Southern Central Mediterranean” by “efforts to identify, capture and dispose of vessels and assets used or suspected of being used by smugglers and traffickers” (Article 1, Decision 778/2015). “The objective is less to stop migration flows than to disrupt smuggling routes and capabilities and, hence, reduce the flows originating from the Libyan Coast, which has been (together with the eastern route) the main point of departure of migrants coming to Europe” [13]. The operation is based on the Articles 42(4) and 43(2) of the TEU. The adoption of the Resolution 2240 by the Security Council of the UN, acting under Chapter VII of the UN Charter, on October 2015, reinforced the legitimacy of EU military operation. EUNAVFOR MED Sophia Operation is one of the CA’s elements—a complementary, not a main instrument. Migration had already been implicitly or explicitly mentioned in other EU missions and operations (EUFOR ALTHEA, EUCAP SAHEL, NIGER, EUTM MALI) [15], but for the first time ever an EU military operation is explicitly and specifically associated with migration issues. It “is the first EU Naval Force to operate in the Mediterranean, an area of strategic and economic importance to Europe, and at the centre of many security challenges that affect Europe as a whole” [16]. This operation reinforced the maritime dimension of CSDP being the first EU maritime military operation “aimed to deliver sea-based capacity” [17], with an explicit coercive mandate and the possibility of deploying means on the territory of a sovereign state (authorized by a UNSC Resolution) without its consent: “while the EU had so far adhered to the crisis management principles of consent, limited coercion and relative impartiality for its own CSDP operations, EUNAVFOR MED’s mandate contains the possibility of the EU going beyond these principles and coming close to a peace enforcement situation. In and by itself, this constitutes a qualitative shift in the EU’s security and defence posture” [13]. The operation involves three phases [18]: phase 1 surveillance—“detection and monitoring of migration networks through information gathering and patrolling on the high seas” (June–August 2015); phase 2 operational—search operations “boarding, search, seizure and diversion on the high seas of vessels suspected of being used for human smuggling or trafficking” (2A) in international waters (ongoing since 7 October 2015) or (2B) in the territorial and internal waters of Libya (there are still no legal and political pre-conditions to transit to this sub-phase); phase 3 operational—disposal practices—“measures against a vessel and related assets, including through disposing of them or rendering them inoperable, which are suspected of being used for human smuggling or trafficking”, in the territory Libya (timeline to be determined). In June 2016 and July 2017, the Council approved additional (support) tasks: training of the Libyan Coastguard and Navy and contributing to the implementation of the UN arms embargo on the high seas off the coast of Libya; setting up

262

A. P. Brandão

a monitoring mechanism of the long-term efficiency of the training of the Libyan Coastguard and Navy, conducting new surveillance activities and gather information on illegal trafficking of oil exports from Libya and enhancing the possibility for sharing information on human trafficking with member states law enforcement agencies, FRONTEX and EUROPOL [19]. Despite the focus on the fight against migrant smuggling and human trafficking, there is a clear concern to underline the humanitarian dimension of the operation. The decision of renaming it “Sophia” in September 2015 was justified by Federica Mogherini as a tribute to “the lives of the people we are saving, the lives of people we want to protect” and a way “to pass the message to the world that fighting the smugglers and the criminal networks is a way of protecting human life” [19]. The EU public diplomacy/communication strategy also emphasizes this humanitarian goal as demonstrated by the animated video “EU Global Strategy—The story of Sophia”. According with the Commander of the Operation, “we will remain committed to saving lives at sea upon request by the competent Maritime Rescue Coordination Centre (MRCC) and in accordance with the International Law of the Sea” [20]. According with the UN Secretary-General, the EU should focus on saving lives as a top priority [21]. The Operation Commander commends the “great results” of Sophia Operation and its evolving scope: it “has transformed into a multifaceted maritime security operation carrying out a range of tasks from maritime situational awareness to capacity building and law enforcement while building an extensive network of contacts; all furthering the achievement of the mission objective” [16]. The operation demonstrated EU political will and solidarity, being launched in record time and achieving results both in the security and humanitarian components without collateral damages: arresting smugglers, destructing boats and tackling the smuggling networks; saving lives. As of 31 December 2016, Operation Sophia had apprehended 101 suspected smugglers and traffickers, neutralized or destroyed 372 vessels, conducted 222 rescue missions and rescued at sea 31,899 migrants and carried out 253 arms embargo events. According with Lieutenant General Wolfgang Wosolsobe, the first phase of the operation had two main outcomes in terms of military objectives: “Processes had been ‘developed to gain information from the migrants’, in order ‘to determine not only their particular circumstances, but how and by whom their passage was effected’; identification of a network that exists to traffic women and children with a view to their sexual exploitation”. Both enhanced “understanding of the traffickers’ networks, personnel and tactics” [22]. Despite those tangible results, Sophia Operation has been subject of criticism for its ambitious intentions and objectives [23], ambiguity [24], ineffectiveness [14] and unintended consequences. One of the main critical voices has been the UK House of Lords, whose latest report is entitled “Operation Sophia: a failed mission” [25]: the number of arrests piecemeal is modest in comparison to the scale of smuggling; given the destruction of boats, the smugglers are now using inflatable rubber craft from China that are more unsafe; the operation had apprehended only low-level smugglers (and some of them are in fact co-opted migrants), not the major ones behind the business model; the deterrent effect of migrant flows is not working

The Internal–External Security Nexus: EU Operation Sophia …

263

since the smugglers find new routes, generally more dangerous and expensive for the migrants; the search and rescue missions are feeding hopes of safe passage (“a magnet to migrants”) and having the unintended consequence of “assisting the job of smugglers, who now only need their boats to reach the high seas, rather than EU waters”; “serious abuses of the human rights of migrants by the Libyan coastguard”. In sum, the operation faces “an impossible challenge” [23] and has no meaningful deterrent and disruption results, being “the wrong tool with which to tackle migration in the central Mediterranean” [25, 26]. Operation Sophia is a demonstration of the EU narrative and practice on security nexuses (defence-migration, military–civilian security, internal–external security, border security–human security) that justify the goal of “more security” and “more actorness”. These nexuses are present in the trend to construct the link between the CSDP and the AFSJ [11, 12]. The European Agenda on Migration advanced that migration would “become a specific component of ongoing Common Security and Defence Policy (CSDP) missions already deployed in countries like Niger and Mali, which will be strengthened on border management” [27]. It specifically underlined the role of the operations in the fight against human smuggling: the identification, capture and destruction of vessels used by criminal networks as “a powerful demonstration of the EU’s determination to act” [27]. Those tasks constitute the core business of the operation. The link between maritime border security and migration was initially associated with FRONTEX operations. FRONTEX is in the forefront of the maritime border security related with migration since 2005. The main focus has been the control and surveillance of external borders through EU Joint operations and the forced return of migrants to the departure state [24]. After the adoption of the Sea Border Regulation, on May 2014, the operations also encompass search and rescue: “The objective of Union policy in the field of the Union external borders is to ensure the efficient monitoring of the crossing of external borders including through border surveillance, while contributing to ensuring the protection and saving of lives” [28]. One month later, the Council adopted the EU maritime security strategy (EUMSS). This strategy defines maritime security as “as a state of affairs of the global maritime domain, in which international law and national law are enforced, freedom of navigation is guaranteed and citizens, infrastructure, transport, the environment and marine resources are protected” [29]. It encompasses both the internal and external aspects of EU maritime security. Among others, it identifies the cross-border and organized crime, including trafficking of human beings and smuggling of migrants, organized criminal networks facilitating illegal migration, trafficking of arms as a threat to the security of the EU, its member states and their citizens [29]. EUMSS left out the humanitarian (search and rescue) dimension. Both operations (FRONTEX and Sophia) evolve rescue activities, but while FRONTEX operations are mainly focused on border management; Sophia Operation is dedicated to the disruption of migrant smuggling and human trafficking routes and capabilities [24]. Despite Operation Sophia’s humanitarian component, its rationale is mainly security: Operation Sophia “is also an example of the renewed strategic and political importance of the nexus between internal and external security in the present European security

264

A. P. Brandão

environment. Beyond the EU borders, terrorist organizations and irregular migration flows are profiting from instability and unresolved external conflicts, which eventually have an impact on EU citizens” [10]. In general terms, Rehrl [15] identifies four potential inputs of CSDP to fight crossborder crime in general, and migrant smuggling and human trafficking in particular, and to enhance EU borders security, namely: “border surveillance and prevention of uncontrolled border crossings”; “processing of irregular migrants, in particular by providing training and technical assistance as well as capacity building for the so-called hotspots”; “law enforcement activities against smugglers’ networks by strengthening intelligence sharing”; “security sector reform in countries either of origin or transit”. In sum, surveillance and intelligence gathering, training and SSR, three of which are performed by Operation Sophia. The third axe if the CSDP-FSJ link is multi-actor and multi-policy coordination. As requested in Article 8 of Council Decision (CFSP) 2015/778, “EUNAVFOR MED shall cooperate with the relevant member state authorities and shall establish a coordination mechanism, and as appropriate, conclude arrangements with other Union agencies and bodies, in particular FRONTEX, EUROPOL, EUROJUST, European Asylum Support Office and relevant CSDP missions”. General Mikhail Kostarakos (Chairman of the EUMC) [30] and Enrico Credendino (Sophia Operation Commander) [31] converge in the statement that Operation Sophia has been a success in terms of complex coordination among several EU actors and Italian authorities, and among military and non-military actors.

4 Concluding Remarks EU security actorness has been evolving since the end of the Cold War by a combination of opportunity, capacity and (global) presence (ambition). This construction has been accompanied by the comprehensive approach narrative. This holistic approach is the result of a co-constitutive adequacy: “more security”—appropriation of policies and instruments of a multifunctional actor for security purposes (security of the EU and of European citizens); “more actorness”—securitization of issues in order to promote the actor and its policies. The internal–external security nexus formulated both in terms of threat perception (transnational, cross-border multifaceted threats) and response to threats, justifies and is justified by the comprehensiveness of the approach of a post-Westphalian security actor. Its practices, such as Operation Sophia, demonstrate how security narratives, civilian and military instruments and securitization dynamics serve convergent processes of gaining political and public space for legitimizing policies and actions of the European security actor. Sophia is a pioneering operation in several aspects: the first Naval EU operation in the Mediterranean and aimed to deliver sea-based capacity; the first CSDP operation explicitly and specifically associated with migration issues; the first EU operation with an explicit coercive mandate and the possibility of deploying means on the territory of a sovereign state (authorized by a UNSC Resolution) without

The Internal–External Security Nexus: EU Operation Sophia …

265

its consent. The security rationale underlying the operation results of the merging of several dimensions: border, maritime, military and human security. In the fight against migrant smuggling and human trafficking, EUNAFOR MED operationalizes the CSDP-FSJ by being a law enforcement using military assets, carrying out surveillance, intelligence gathering and training activities involving means and structures of both domains, coordinating actors of both domains (CSDP and FSJ). For this reason, its achievements and failures enlighten the potential limits and effects of IESN operationalization. Acknowledgements This paper has been developed in the framework of the collective research project: “España ante los nuevos retos de la seguridad marítima: Instrumentos y estrategias en el marco internacional, europeo y peninsular” (DER2016-78272-R), funded by Ministerio de Economía y Competitividad (Programa Estatal de I+D+i Orientada a los Retos de la Sociedad 2017–2019). The author also acknowledges the financial support of Fundação para a Ciência e Tecnologia—FCT (Portuguese Science Foundation).

References 1. For the purpose of this paper, the securitisation process was analysed through the lens of the Copenhagen School (Buzan, B., Waever, O., Wilde, J.: Security: A New Framework for Analysis. Lynne Rienner Publishers, Boulder, CO (1998)): the security speech act on Operation Sophia by EU institutions (Council of the EU, European Council, European Commission); the (perceived) threat (human smuggling and trafficking networks in the Southern Central Mediterranean); the securitising instrument (the military operation as an extreme version of securitisation). The analysis of EU actorness was based on the three elements conceptualised by Bretherton & Vogler: opportunity (“denotes factors in the external environment of ideas and events which constrain or enable actorness”), presence (“conceptualizes the ability of the EU, by virtue of its existence, to exert influence beyond its borders”) and capability (“refers to the internal context of EU external action—the availability of policy instruments and understandings about the Union’s ability to utilize these instruments, in response to opportunity and/or to capitalize on presence”)—Bretherton, C., Vogler, J.: The European Union as a Global Actor, p. 24. Routledge, New York (2006). Two security functions were considered in the analysis of the IESN: compulsion (military operations) and protection (internal security)—Kirchner, E., Sperling, J.: EU Security Governance. Manchester University Press, Manchester (2007) 2. See: the European Security Strategy (2003); the Internal Security Strategy for the EU (2010); the European Agenda on Security (2015); EU Global Strategy (2016) 3. Eriksson, J., Rhinard, M.: The internal–external security nexus. Notes on an emerging research agenda. Coop. Confl. 44(3), 243–267 (2009) 4. Pastore, F.: Reconciling the Prince’s Two ‘Arms’. Internal-External Security Policy Coordination in the European Union. Occasional Paper, n. º 30. ISS, Paris (2001) 5. Bigo, D.: Internal and external security(ies): the Möbius ribbon. In: Albert, M., Lapid, Y., Jacobson, D. (eds.) Identities, Borders, Orders, pp. 91–116. University of Minnesota Press, Minneapolis (2001) 6. Rehrl, J., Weisserth, H.-B. (eds.): Handbook on CSDP. Federal Ministry of Defence and Sports of the Republic of Austria, Vienna (2007) 7. European Council: A Secure Europe in a Better World—The European Security Strategy. European Council, Brussels, 12/13 December (2003)

266

A. P. Brandão

8. European Council: 2008. Report on the Implementation of the European Security Strategy—Providing Security in a Changing World. European Council, Brussels (11/12) (2008); European Council: Internal Security Strategy for the European Union: Towards a European Security Model (2010) 9. Council of the EU: Foreign Affairs Council (2015) 10. Council of the EU: CMPD Food for Thought Paper “From Strengthening Ties Between CSDP/FSJ Actors Towards More Security in Europe” (2016) 11. Council of the EU: Strengthening Ties Between CSDP and FSJ—Elements of a Draft Road Map (15562/11) (2011) 12. European Commission and HR: Strengthening Ties Between CSDP and FSJ Actors—Proposals for a Way Ahead (SEC/2011/560) (2011) 13. Tardy, T.: Operation Sophia Tackling the Refugee Crisis with Military Means. EUISS Brief (September). EUISS, Paris (2015) 14. Priconi, M.: EU military Operation Sophia: analysing the shortfalls. Sci. Bull. 2(42), 122–127 (2016); Ventrella, M.: The impact of Operation Sophia on the exercise of criminal jurisdiction against migrant smugglers and human traffickers. Quest. Int. Law 30, 3–18 (2016) 15. Rehrl, J.: Migration and CSDP. In: Handbook on CSDP, vol. 1, 3rd edn, pp. 104–113. Vienna (2017) 16. EEAS: EUNAVFOR MED—Operation SOPHIA—Six Monthly Report: 1 Jan–31 Oct 2016 (EEAS/2016/1587) (2016) 17. EEAS (European External Action Service): Common Security and Defence Policy of the European Union: Missions and Operations Annual Report 2016 (2017) 18. Council of the EU: Council Decision (CFSP) 2015/778 of 18 May 2015 on a European Union military operation in the Southern Central Mediterranean (EUNAVFOR MED). Off. J. Eur. Union L 122, 31–35 (2015) 19. EEAS: EUNAVFOR MED Operation Factsheet (2017) 20. EDA (European Defence Agency): Operation Sophia to stop human trafficking across the Mediterranean. Defence Matters 9, 30–31 (2015) 21. European Parliament: Ban Ki-moon on migration: “Saving lives should be the top priority”. EP News (2015) 22. Wosolsobe in UK (United Kingdom), House of Lords: Operation Sophia, the EU’s naval mission in the Mediterranean: an impossible challenge. European Union Committee—14th Report of Session 2015–16. HL Paper (144) (2016) 23. UK (United Kingdom), House of Lords: Operation Sophia, the EU’s naval mission in the Mediterranean: an impossible challenge. European Union Committee—14th Report of Session 2015–16. HL Paper (144) (2016) 24. Bevilacqua, G.: Exploring the ambiguity of Operation Sophia between military and search and rescue activities. In: Andreone, G. (ed.) The Future of the Law of the Sea Bridging Gaps Between National, Individual and Common Interests, pp. 165–189. Springer, Cham (2017). https://doi.org/10.1007/978-3-319-51274-7_9 25. UK, House of Lords: Operation Sophia: a failed mission. European Union Committee—2nd Report of Session 2017–19. HL Paper (5) (2017) 26. See also critics from: Amnesty International on human rights violations (Amnesty International: European Policies Leaving Migrant Lives Adrift in the Mediterranean (2016)); spokesman of the Libyan Navy, “the first beneficiaries of this operation are not Europe or Libya, but rather the traffickers” (Euronews: The first beneficiaries of Operation Sophia are the traffickers. Insiders (21/12/2016)); Parliament of UK (UK 2016) and Operation Commander on intelligence gathering; Deputy Director of Global Initiative against Transnational Organized Crime on lack of full transparency (in Mathari, A.: “Mission impossible” for Operation Sophia. Global Spotlight (02 June) (2017)) 27. European Commission: Communication from the Commission to the European Parliament, the Council, the European Economic and Social Committee and the Committee of the Regions—A European Agenda on Migration (COM/2015/240). Brussels (2015)

The Internal–External Security Nexus: EU Operation Sophia …

267

28. Council of the EU: European Union Maritime Security Strategy (2014) 29. Council of the EU: Regulation (EU) No. 656/2014. Off. J. EU L (89) (2014) 30. Kostarakos in: EDA: Interview military capabilities: “Europe still lacks strategic enablers”. Defence Matters 11, 36–38 (2016) 31. Credendino in: EEAS: EUNAVFOR MED—Operation SOPHIA—Six Monthly Report: 1 Jan–31 Oct 2016 (EEAS/2016/1587) (2016)

The Evolution of EU’s Maritime Security Thinking João Almeida Silveira

Abstract Maritime security is a fundamental vector in the security of the European Union (EU). Such importance resulted in the adoption of several strategies that contribute to the protection and attainment of maritime interests such as freedom and safety of navigation, maintenance of supply chains, food and energy security, blue growth, assumption of the EU as a global actor, among others. The EU Maritime Security Strategy (EUMSS) is the core strategy guiding the EU in the protection of the seas, yet EU’s thinking extends beyond the EUMSS. This article explores the process that leads to the formulation of the EUMSS and the changing functions that maritime security played in the overall strategic objectives of the Union. It finds that the EU’s maritime security conceptualization was driven by higher levels of EU engagement in maritime socioeconomic development and marine environment protection. Subsequent to perceived successes in its maritime security efforts, the field was instrumentalized by the Union to support and legitimize its security and international actorness. Maritime security is perceived as an integral part of EU’s global strategy, thus, there are attempts to mainstream maritime security in EU policies. Keywords European Union · Maritime security · EU security strategy

1 Introduction The maritime domain is currently characterized as space of growing levels of competition and human activities (e.g., economic, cultural, scientific) and afflicted by a great array of threats and challenges. The number of actors operating at sea has increased (e.g., NGO’s, criminal organizations), adding to the complexity of a domain profoundly interconnected with state interests [1–3]. Furthermore, the good state of oceanic ecosystems is linked with the future and well-being of humanity [4, 5]. J. A. Silveira (B) Department of Political Studies, Faculty of Social Sciences and Humanities—New University of Lisbon (FCSH-UNL), Avenida de Berna 26C 1069-061, Lisbon, Portugal e-mail: [email protected] © Springer Nature Singapore Pte Ltd. 2020 Á. Rocha and R. P. Pereira (eds.), Developments and Advances in Defense and Security, Smart Innovation, Systems and Technologies 152, https://doi.org/10.1007/978-981-13-9155-2_22

269

270

J. A. Silveira

Gradually, the EU started to play a greater role in maritime affairs and in its protection. Since the early 2000s, it adopted over 40 maritime related documents that covered a myriad of aspects, including, among others, safety and security, environment, blue economy, governance, and scientific and technological advancements. Such interest resulted from the perception that the size and complexity of sea matters are better served by a collective European response, rather than through national policies. The Union attempted to potentiate the exploitation of seas for sustainable socioeconomic development, in respect of the environment. As the maritime interests of the EU grew, it became necessary to protect maritime assets [6–9]. This article explores the dynamic process by which the Union became engaged with maritime security. More, it tracks the evolution of the EU’s maritime security strategic thinking, in order to mark strategic shifts and/or enlargements of the concept. It considers as time period the years between 2000 and 2018. It approaches the topic through the rhetorical analysis of EU official communications. The article follows a sequential process. First, it explores the early elements of EU engagement with maritime matters that explain the added importance of the domain to the Union after 2000. Second, it outlines the main security concerns that gave rise to the adoption of a maritime security strategy. Third, it outlines the maritime security thinking of the Union. Fourth, it explains the instrumentalization of maritime security to the achievement of the global objectives of the Union. And fifth, it pays attention to attempts to turn maritime security into a mainstream policy.

2 Theoretical Framework and Methodology The departing question for this article is: How did the EU’s maritime security strategic thinking evolves between 2000 and 2018? To answer this question, this article uses the securitization framework, a constructivist framework, whose objective is to analyze the ascendant or descendant process of topics within political security agendas [10]. The objective is to mark and analyze the several moments of inclusion of maritime security within the EU political speech. To that end, the article makes the rhetorical analysis of official maritime and nonmaritime strategies of the EU. When appropriate, it also makes use of secondary literature to the better frame or explains the content of the official documents. There are two main concepts in this article: maritime security and strategy. Regarding maritime security, there is no consensual definition [11–13]; thus, for the purpose of this article maritime security is understood as a broad concept that includes matters of national security (seapower), marine safety, blue economy, and human security [3]. Regarding strategy, this article pays attention to the grand strategy of the EU, and not just strategy. Thereby, it considers all instruments of possible political use, and not just the link between policy and military power. In doing so, it considers the aspects related to economy, finance, technology, military, and others [14, 15].

The Evolution of EU’s Maritime Security Thinking

271

3 Economic–Environmental–Social Triangle Throughout the 2000s, the strategic conceptualization of the EU toward maritime security increased exponentially. The awakening for a more robust engagement in sea matters resulted from two traumatic events: the sink of the oil tanker Erika in December 1999, in the Bay of Biscay and the sink of the oil tanker Prestige in November 2002, off the northwest coast of Spain. These events caused great damages to the marine and socioeconomic environment and threatened the livelihood of coastal populations. As a result, European populations demanded measures to prevent such occurrences. To respond to public pressures, the EU adopted two maritime safety packages in 2000: the Erika I in March and the Erika II in December. These packages aimed at improving the safety of seaborne trade (specifically oil) and prevent accidental pollution by ships [6, 16]. When two years later the tanker Prestige sank, it became clear that added efforts had to be taken. Henceforth, the EU established the European Maritime Safety Agency (EMSA) to level-up maritime safety in Europe [17]. Adding to the internal efforts, the EU sought to increase global safety standards. To that end, it engaged multilaterally in forums such as the International Maritime Organization (IMO) [18] and bilaterally by including maritime safety and security concerns in agreements with third countries [19]. During the first years of the 2000s, the Union established a link between maritime economic development, social development, and environmental protection (in Europe and elsewhere). These elements were framed in an interdependent and intertwined manner. The Thematic Strategy on the Protection and Conservation of the Marine Environment of 2005 is a good example of how this conceptualization worked. According to the strategy, the erosion of the marine ecological capital jeopardizes employment opportunities and the generation of wealth. A sound ecological state of European seas is deemed as indispensable for a diverse, sustainable, and thriving economic development, which is linked with the well-being of populations (e.g., food security, employment, maintenance of cultural ways) [20].

4 Blue Economic Development As a result of technological developments, it became possible to make a deeper exploitation of marine resources (e.g., blue energy). Consequently, it was important for the EU to ready European agents (public and private) to fully exploit the potentialities of the oceans. To that end, the European Commission formulated in 2007 the Integrated Maritime Policy (IMP) designed to assist EU agents to address challenges such as competitivity, climate change, degradation of the marine environment, maritime safety and security, and sustainability [7].

272

J. A. Silveira

At the core of the IMP are economic development concerns. The IMP is fundamentally an instrument to promote the growth of the maritime industry [21]. As an overarching policy, the IMP provides an holistic and integrated approach to sea matters, by including areas such as maritime transport; maritime research, surveillance, and spatial planning; improvement of national policies; resilience against climate change; reduction of CO2 emissions and ship pollution; maritime labor law; unsustainable fisheries practices; and development of an European network of maritime clusters [7]. Each of these areas was further developed in sectoral and regional strategies. For example, the EU approved strategies to guide an ecosystem-based approach to human activities at sea [22]; to improve marine research, data availability and accessibility, and general knowledge over the maritime domain [23–25]; to prevent, deter, and eliminate illegal, unreported, and unregulated (IUU) fishing [26, 27]; to improve maritime spatial planning [28]; to promote safe and efficient maritime transport [29]; to improve safety standards [30]; and to address the idiosyncrasies of European basins, such as the Atlantic, Arctic, Mediterranean, Baltic, or Black sea [31–35]. Alike the IMP, economic aspects are central to these strategies. In terms of economic activities, the 2012 Blue Growth Strategy clarified that the EU was especially interested in five value chains: (i) blue energy; (ii) aquaculture; (iii) maritime coastal and cruise tourism; (iv) maritime mineral resources; and (v) blue biotechnology. Improvements in these areas were deemed possible due to advancements in the fields of robotics, submersible machinery, and other technologies. Blue economy industries were perceived as a way to generate growth, innovation, and employment [36].

5 Security Concerns The EU anchored its sea power on economic aspects. It was, thus, necessary to engage in maritime security, for there is no economic development without security [2]. Indeed, even if sparse and, at times, in a neglectable matter, several strategies identified risks and threats to EU interest. Because most maritime strategies addressed the European space, which was a safe space, most threats related to the matters of pollution, environmental risks, and IUU fishing [7, 23, 26, 31–35, 37]. Nonetheless, the several regional strategies included a greater array of security concerns, namely: criminality (including trafficking), illegal migrations, marine accidents, need for more effective law enforcement activities, piracy, legal disputes, among others [31–35, 37]. Maritime security concerns found echo also in non-maritime strategies, such as the Counter-Terrorism Strategy of 2005 that called for higher standards on maritime and ports security [38] and the 2010 EU Internal Security Strategy (ISS) that called for improvements on EU maritime borders vis-à-vis illegal trafficking [39]. In strategies that covered imminently international and global sectors such as the strategy for maritime transport, the most pronounced security concerns included

The Evolution of EU’s Maritime Security Thinking

273

terrorism, piracy and armed robbery at sea, trafficking, stowaways, and smuggling. From these, piracy was regarded as “the most urgent priority.” In 2009, the year of approval of the maritime transport policy, the effects of piracy off the Somalia coast were already pronounced, and the strategy recognized the grave danger of piracy to seafarers, fishermen, and passengers [29]. In 2003, the European Security Strategy (ESS) had already included piracy as a potential threat to EU interest. At the time, it was flagged as a threat that could merit further attention in the future [40]. Events off the coasts of Somalia in 2007 proved right the inclusion of piracy in the ESS. An upsurge of piracy from 2007 onward caused harm to Euro-Asian seaborne trade flows and to EU fishing interests in the region. Furthermore, the instability caused by piracy had consequences on land (e.g., impossibility of deliverance of humanitarian assistance). This event demonstrated that sea insecurity had negative effects to the EU, which prompted action [1, 41–44]. Under the avail of United Nations Security Council (UNSC), the EU launched operation Atalanta in the region in order to restore stability [45] and protect EU economic, diplomatic, and humanitarian interests [44]. With operation Atalanta, the first EU maritime operation within the Common Security and Defence Policy (CSDP), it became clear that piracy and maritime instability were real threats to the EU. Consequently, the 2008 report on the implementation of the ESS increased the attention of the EU to maritime security developments. Besides piracy, the report also recognized the possible emergence of disputes over trade routes and maritime territories [46]. The analysis of the above-mentioned strategies suggests that during the 2000s, the maritime security agenda progressively penetrated through EU policy documents. Such introduction was motivated by traumatic events as the Erika and Prestige sunk and by the upsurge of piracy off the Somalia coast. These events forced the EU to think and engage in maritime security in its proximity and in the wider world. Maritime security was framed in a dual capacity of an element necessary for the protection and enhancement of economic, social, and environmental interests. This perception, aligned with the sprout of European activities at and from the sea, leads to the adoption of the EU Maritime Security Strategy (EUMSS) in 2014.

6 European Union Maritime Security Strategy Designed “as a comprehensive framework contributing to a stable and secure global maritime domain,” the EUMSS addressed internal and external security elements. It identified the core maritime interests of the EU, namely: the guarantee of the territorial integrity, sovereignty, and values of member states; the safe development of infrastructures (e.g., ports and port-facilities, seabed cables, off-shore installations); the increase of economic activities at sea; freedom and safety of navigation for ships, cargo, crews, and passengers; the protection of the marine environment; the protection of European cultural heritage (e.g., archaeologic objects) and EU citizens; and the use of the maritime domain for scientific progresses and innovations [8].

274

J. A. Silveira

The strategy consolidated the security risks and threats to EU maritime interest, namely: interstate conflicts; proliferation; sea denial or sea obstruction; transnational criminality; terrorism; environmental threats (natural or man-made events) [8]. To tackle maritime security challenges, the EUMSS formulated a comprehensive and cross-sectoral approach leading to higher levels of internal–external coherence of the EU. EU actions were to be executed in compliance with the law (national, EU, and international), in a cost-efficient manner, and bridging military and civilian authorities and actors,1 local, national and EU agencies, and the maritime industry [8]. While calling for greater autonomy of EU actions, the strategy asserted the full respect for the functional integrity of member states competencies, rights, and interests. Furthermore, the autonomy of action did not equate to isolationism. Instead, the EUMSS directed toward a multilateral approach, anchored on the strengthening and support of regional responses [8]. Autonomously, but next to partners,2 the EU aspired to engage in maritime capacity building, in areas such as maritime governance and the rule of law; development of the port and maritime transport standards; borders management; and combat of IUU fishing [8]. In view to the implementation of the EUMSS, the Union approved in December 2014 the EU Maritime Security Strategy Action Plan (EUMSSAP); a holistic plan to prompt the comprehensive and cross-sectoral approach of the EU. The actions were divided across five workstrands that addressed specific but intertwined aspects: Workstrand 1—External Action. Addressed matters of external coordination between the EU and other international partners. It sought to increase the visibility of the EU; to foster capacity building in third countries, the peaceful settlement of disputes, and the international law; to pursued force development for contingency operations; and to articulate measures to increase the preparedness of EU and members states to engage in maritime security contingency operations [47]. Workstrand 2—Maritime Awareness, Surveillance and Information Sharing. Pursued the implementation of the Common Information Sharing Environment (CISE) to further cross-sectoral cooperation and interoperability at national and EU level. It addressed cross-border cooperation and information exchange to optimize surveillance; and supported the coordination of the several EU surveillance initiatives [47]. Workstrand 3—Capability Development. Fostered technological development, harmonization of instruments to favor interoperability, and standardization and certification in the maritime domain. It was interested in exploring dual-use and multipurpose capabilities (e.g., civil–military capacities, integration of cyber dimension into the maritime domain) and in promoting the sharing of best practices, risks analysis, and threat information [47].

1 E.g., law enforcement, border control, customs and fisheries inspection, environmental authorities,

maritime administration, research and innovation, navies or other maritime forces, coast guards, intelligence agencies [8]. 2 E.g., North Atlantic Treaty Organization (NATO), the United Nations (UN), the African Union (AU) or the Association of Southeast Asian Nations (ASEAN).

The Evolution of EU’s Maritime Security Thinking

275

Workstrand 4—Risk Management, Protection of Critical Maritime Infrastructures and Crisis Response. Promoted EU joint risk analysis; the enhancement of crosssectoral and cross-border cooperation for crisis response and contingency planning; the assessment of the resilience of maritime transport infrastructures against natural or man-made disasters; and promoted mutual understanding and interoperability among maritime security actors [47]. Workstrand 5—Maritime Security Research and Innovation, Education and Training. Promoted the establishment of knowledge and training networks; fostered civil–military maritime research (e.g., though public–private partnerships); and promoted inter-agency joined-up exercises [47]. The EUMSS and its action plan provided a coherent approach to maritime security. The documents balanced internal and external aspects of security and bridged maritime and non-maritime aspects of the Union’s action. Besides directing the EU’s maritime strategy toward the protection and enhancement of strategic interest, the EUMSS seems to work as a supporter of the EU’s international actorness.

7 Instrumentalization of Maritime Security In 2016, the International Ocean Governance Agenda (IOG) continued the pursuit of ensuring EU’s coherence, by bridging several features of maritime security. On the document, the EU presented itself as the global champion of oceans sustainability and protection. To build its case, the EU anchored its thought on internal and external progress and action it had developed thitherto [48]. On the internal front, it deferred to the developments within the IMP that contributed to an overall increase of oceans sustainability. Furthermore, it placed great focus on the environmental aspects of its policies. On the external front, the EU’s contribution to international maritime security was the main argument. It notably called attention to EU’s action in thwarting piracy in the Horn of Africa and its anti-smuggling activities in the Mediterranean Sea [48]. The agenda was built around the idea of the oceans as a common good. It mimicked the EUMSS comprehensive and multilevel approach to ocean governance and sustainability by including several sectors of the maritime economy; by establishing links with nonexclusive maritime aspects of security such as cyber security; by making an effort to consolidate and advance the internal/external nexus of EU; and by pushing cultural components into the agenda [48]. Overall, environmental features were very much embedded into the agenda, and it established a close link between economic and environmental dimensions. It asserted that only a sustainable use of maritime resources could lead to long-term growth. According to the EU, the environmental and economic link is particularly relevant and requires global and multilateral action [48]. This indicates that the ideas formulated in the early 2000s made a school in EU’s maritime conceptualization. Also in 2016, the Global Strategy for the European Union’s Foreign and Security Policy integrated maritime security within the broader EU approach to its foreign

276

J. A. Silveira

and security policy. The document presented the EU as a global maritime security provider. According to it, maritime global growth is linked with EU interests of supporting an open and fair economic system; thus, it is necessary for the EU to take an active role in its protection. The strategy reinforced a multilateral approach that perceived maritime security as a matter of external cooperation (e.g., with NATO) and that required building capacities in Europe and elsewhere [9]. Again, the support for an international order based on the rule of law was present, with the defense of the law of the seas as an important instrument for the peaceful settlement of disputes. Also in conceptualizing its relations with others regions, the Union included maritime security as one of the topics for joint development [9]. In its path toward coherence, the strategy argued for greater coordination between internal and external agencies and instruments, such as CSDP missions and operations, European Border and Coast Guard, and EU specialized agencies to enhance border protection and maritime security, to save lives, fight cross-border crime, and disrupt smuggling networks. To increase the EU’s preparedness and effectiveness, the strategy called for the improvement of maritime security-related capabilities [9]. These two documents seem to portrait maritime security not just as a supporter element of EU’s internal and external actorness, but also as a legitimizer of EU’s action at home and abroad. Consequently, maritime security serves to protect and enhance the economic, environmental, and social interests of the EU and to support and legitimize the security and international actorness of the EU. Maritime security is, thus, not just a strategic component of EU’s maritime actorness, but also an active component of EU’s global strategy.

8 Mainstream Maritime Security The global strategy of the EU of 2016 is the core document guiding the EU toward a more coherent and coordinated action. It argued for a holistic approach of the Union to which all policies and instruments should contribute. The document re-focused EU’s attention to the European continent and to the internal adaptation to cope challenges faced by the EU as, for instance, the Russian behavior that led to the annexation Crimea in 2014 and the effects of the migration crisis in the Mediterranean [9]. In 2017, the Council Conclusion on Global Maritime Security incorporated such re-focusing. Overall, the conclusions reaffirmed the objectives, interests, and approach of the EUMSS, yet it established as areas of particular attention the Mediterranean Sea, the Baltic Sea, the Black Sea, and the Arctic region [49]. Regarding the Mediterranean, the document pushed for increasing levels of coordination between EU instruments such (e.g., EUNAVFORMED Operation Sophia, Frontex, European Fisheries Control Agency, EMSA) and between EU instruments and other actors operating in the Mediterranean in order to increase security levels, namely, to respond to the migration influxes from North Africa [49]. Oddly, however, the effects of the Russian annexation of Crimea to the maritime security of the EU were not addressed, and the document only refers to regional

The Evolution of EU’s Maritime Security Thinking

277

cooperation in its approach to the Baltic and the Black Sea basins [49]. Considering that the annexation of Crimea considerably changed the security spectrum of Europe and that the existent strategies for those basins predate the annexation, it is a considerable gap the absence of further strategic guidance to address the changing regional order. Despite the European re-focus, the strategy also considered relevant to engage in confidence-building measures in the South China Sea, which the EU regards as an area of potential regional conflicts. Furthermore, the document also demonstrates attention to the situation in the Gulf of Guinea. In both cases, the EU’s approach is based on supporting local ownership and regional institutions [49]. In 2018, the action plan for the EUMSS was updated. The principles of crosssectoral approach (civil–civil, civil–military, and military–military cooperation), functional integrity, respect for rules and principles, and maritime multilateralism of the EUMSS were maintained, as well as the defined approach to maritime security [50]. The document presented a total of 90 actions. Those were divided between the horizontal issues (56 actions) and regional and global affairs (34 actions). The section horizontal issues devises the actions to the fulfillment of the EUMSS and is subdivided in six main areas: – Awareness raising. Promotes awareness regarding the EUMSS. – International cooperation. Promotes the mainstream of maritime security in the EU’s foreign action. The objective is to enhance cooperation with other parties (e.g., organization, third parties, private sector), to promote international law, to foster capability building in regions that require, and to bridge internal and external components of the EU in order to increase its the internal–external coherence. – Maritime awareness. Promotes the increase of maritime awareness through the better use and sharing of information produced by several EU agencies, programmes, activities, and policies (e.g., MARSUR, SatCen, CISE, EMSA, ESA). – Capabilities development, research, and innovation. Promotes research and development of capabilities and technologies of both the civilian and military sectors, through synergies and multiple EU-funded grants. The objective is to reduce the fragmentation of EU security demand. There is a great tonic in the development of dual-use technologies for multipurpose use. In this regard, the enhancement of maritime data is deemed crucial. – Risk awareness and management. Promotes a comprehensive approach to risk management and common analysis. The objective is to increase the resilience of maritime infrastructures. – Education and training. Promotes training and education initiatives to increase the readiness of operations in tackling maritime issues. The objective is to standardize procedures within the EU, and if possible, integrate external partners. Regional and global affairs address specific needs of maritime areas in Europe and in the world. From the 34 actions defined only nine address areas outside the European geography: Gulf of Guinea, the Indian Ocean, and Indian and Pacific Ocean [50]. This division appears to confirm the EU’s geographic re-centering to Europe.

278

J. A. Silveira

The actions defined for regional and global maritime affairs pursuit six fundamental objectives: mainstream maritime security in EU sea basins policies (existent and future); increase cooperation and coordination among EU agencies, projects, and instruments, and between these and actors outside the EU; develop capabilities in third countries; develop EU capabilities and encourage research and development projects; increase awareness of maritime security; and support regional development, both in the EU and in the wider world [50]. Overall, the Council Conclusion on Global Maritime Security and the new action plan for the EUMSS do not significantly alter the EU’s maritime security strategy. These documents do re-center the attention to Europe, but maintain an international take, and the objective of asserting the EU at the global stage. Maritime security remains an important piece of EU’s conceptualization, for it “is an essential precondition for development, job creation, research in the maritime and marine environment and global oceans governance” [49]. The more inward approach that these documents seem to convey appears to be a maneuver to strengthen the EU at home, so that the EU can project a stronger and more coherent image abroad.

9 Conclusion Maritime security observed a gradual and incremental evolution within EU strategic conceptualization between 2000 and 2018. At the core of the EU, strategic thinking is the economic–environmental–social triangle that drives and supports deeper levels of EU engagement with maritime security. It was the growing weight of economic and social interest that prompted the EU into a more active role in maritime security in order to protect and enhance its interests. In its engagement, the Union follows a comprehensive approach that links internal and external elements. More are the acts within the international community, and in partnership with third countries. Maritime security is formulated as a task to be executed in a multilateral setting and requires the level-up of maritime capabilities in Europe and elsewhere, as well as added instruments for cooperation and coordination. With perceived successes in its maritime and maritime security policy, the Union instrumentalized maritime security in order to support and legitimize its security and international actorness. Progressively, the EU brought maritime security objectives and instruments to the service of overall EU strategic objectives. There is a notorious attempt to mainstream maritime security into the EU’s strategic speech, in order to increase the coherence of the Union, and consolidate its position in the global stage.

The Evolution of EU’s Maritime Security Thinking

279

References 1. Behr, T., Brattberg, E., Kallio, J. et al.: The Maritime Dimension Of CSDP: Geostrategic Maritime Challenges and Their Implications for the European Union. European Parliament, Directorate-General for External Policies of the Union, Directorate B, Policy Department (2013) 2. Germond, B.: The Maritime Dimension of European Security: seapower and the European Union. Palgrave Macmillan, Houndmills, Basingstoke, Hampshire (2015) 3. Bueger, C., Edmund, T. (2017). Beyond Seablindness: a New Agenda For Maritime Security Studies. Int. Aff. 0, 1–19 4. Noone, K., Sumaıla, R., Díaz, R. J.: Valuing the Ocean. Extended Executive Summary. SEI Stockholm Environment Institute, Stockholm (2012) 5. Figueres, J.M., Lamy, P., Podesta, J.D.: The Fight for Ocean Health. In: Project Syndicate https://www.project-syndicate.org/commentary/un-sdg-14-ocean-regeneration-by-josemaria-figueres-et-al-2017-02(2017). Accessed 28 Feb 2017 6. European Commission: Maritime Safety, Erika I Package (2000) 7. European Commission: An Integrated Maritime Policy For The European Union. Office for Official Publications of the European Communities, Luxembourg (2008) 8. European Commission: European Union Maritime Security Strategy (2014) 9. High Representative: Shared Vision, Common Action: a Stronger Europe – A Global Strategy for the European Union’s Foreign and Security Policy (2016) 10. Buzan, B., Waever, O.: Regions and Powers. The Structure of International Security. Cambridge University Press, Cambridge (2003) 11. Chapsos, I.: Introduction: Strengthening Maritime Security Through Cooperation. NATO Sci. Peace Secur. Ser. E. Hum. Soc. Dyn. pp. 1–7. https://doi.org/10.3233/978-1-61499-536-41(2015) 12. Bueger, C.: What is maritime security? Mar. Policy 53, 159–164 (2015). https://doi.org/10. 1016/j.marpol.2014.12.005 13. Chapsos, I., Malcolm, J.A.: Maritime security In Indonesia: towards a comprehensive agenda? Mar. Policy 76, 178–184 (2017). https://doi.org/10.1016/j.marpol.2016.11.033 14. Gray, C.: Strategy and History. Essays on Theory and Practice. Routledge, New York (2006) 15. Berindan, I.: Not another ‘Grand Strategy’: what prospects for the future European security strategy? Eur. Secur. 22, 395–412 (2013). https://doi.org/10.1080/09662839.2013.808189 16. European Commission: Maritime Safety, Erika II Package (2000) 17. European Parliament, Council of the EU: Directive 2002/84/EC of the European Parliament and of the Council of 5 November 2002 Amending the Directives on maritime Safety and the Prevention of Pollution from Ships (2002) 18. European Council: Copenhagen European Council 12 and 13 December 2002. Presidency Conclusions (2002) 19. European Commission: European Neighborhood Policy - Strategy paper (2004) 20. European Commission: Thematic Strategy on the Protection and Conservation of the Marine Environment (2005) 21. Germond, B.: The EU’s security and the sea: defining a maritime security strategy. Eur. Secur. 20, 563–584 (2011). https://doi.org/10.1080/09662839.2011.635648 22. European Parliament Council of the EU: Directive 2008/56/EC Of The European Parliament and of the Council of 17 June 2008 Establishing A Framework For Community Action in the Field of Marine Environmental Policy (Marine Strategy Framework Directive) (2008) 23. European Commission: A European Strategy for Marine and Maritime Research (2008) 24. European Commission: Marine Knowledge 2020 Marine Data and Observation For Smart and Sustainable Growth (2010) 25. Council of the EU: Council Conclusions on Integration of Maritime Surveillance. Towards The Integration of Maritime Surveillance: a Common Information Sharing Environment for the EU Maritime Domain (2011)

280

J. A. Silveira

26. Council of the EU: Council Regulation (EC) No 1005/2008 of 29 September 2008 Establishing a Community System to Prevent, Deter and Eliminate Illegal, Unreported and Unregulated Fishing, Amending Regulations (EEC) No 2847/93, (EC) No 1936/2001 and (EC) No 601/2004 and repealing Regulations (EC) No 1093/94 and (EC) No 1447/1999 (2008) 27. European Commission: Commission Regulation (EC) No 1010/2009 of 22 October 2009 laying down detailed rules for the implementation of Council Regulation (EC) No 1005/2008 Establishing a Community System to Prevent, Deter and Eliminate Illegal, Unreported and Unregulated Fishing (2009) 28. European Commission: Roadmap for Maritime Spatial Planning: achieving Common Principles in the EU (2008) 29. European Commission: Strategic Goals and Recommendations For The EU’s Maritime Transport Policy Until 2018 (2009) 30. European Parliament, Council of the EU: Directive 2009/18/Ec Of The European Parliament and Of The Council of 23 April 2009 Establishing The Fundamental Principles Governing The Investigation of Accidents in the Maritime Transport Sector and Amending Council Directive 1999/35/EC and Directive 2002/59/EC of the European Parliament and of the Council (2009) 31. European Commission: Black Sea Synergy - A New Regional Cooperation Initiative (2007) 32. European Commission: European Union Strategy for the Baltic Sea Region (2009) 33. European Commission: Towards an Integrated Maritime Policy For Better Governance in the Mediterranean (2009) 34. European Commission: Developing a Maritime Strategy for the Atlantic Ocean Area (2011) 35. European Commission: An Integrated European Union Policy For The Arctic (2016) 36. European Commission: Blue Growth: opportunities For Marine and Maritime Sustainable Growth (2012) 37. European Commission: Action Plan For A Maritime Strategy In The Atlantic Area. Delivering Smart, Sustainable and Inclusive Growth (2013) 38. Council of the EU: The European Union Counter-Terrorism Strategy (2005) 39. European Commission: The EU Internal Security Strategy in Action: Five Steps Towards a more Secure Europe (2010) 40. European Council: A Secure Europe in a Better World. European Security Strategy (2003) 41. Weber, A.: EU Naval Operation in the Gulf of Aden (EU NAVFOR Atalanta): Problem Unsolved, Piracy Increasing, Causes Remain. In: The EU as a Strategic Actor in the Realm of Security and Defence? pp 70–83. SWP, German Institute for International and Security Affairs, Berlim (2009) 42. Larik, J.: Europe’s Fight Against Piracy: from the Barbary Coast to Operation Atalanta. Fondation Pierre du Bois (2013) 43. Silveira, J.: EUNAVFOR Atalanta: O Compromisso da União Europeia com a Segurança Global. Universidade do Minho (2015) 44. Silveira, J.: European Union Security: the Threat of Piracy. In: VIII Congress APCP (Portuguese Association of Political Science). Faculdade de Ciências Sociais – New University of Lisbon, Lisbon, Portugal (2016) 45. Council of the EU: Council Joint action 2008/851/CFSP of 10 November 2008 on a European Union military operation to contribute to the deterrence, prevention and repression of acts of piracy and armed robbery off the Somali coast (2008) 46. European Council: Report on the Implementation of the European Security Strategy. Providing Security in a Changing World (2008) 47. Council of the EU: EU Maritime Security Strategy Action Plan (2014) 48. European Commission: International Ocean Governance: an Agenda For The Future of Our Oceans (2016) 49. Council of the EU: Council Conclusions on Global Maritime Security (2017) 50. Council of the EU: Council Conclusions on the Revision of the European Union Maritime Security Strategy (EUMSS) Action Plan, 26 June 2018

The Obstacles Women Face in Gaining Access to Special Operations Forces João Reis, Rafael Gonçalves, Sofia Menezes and Manuela Kaczynska

Abstract The aim of this article is to provide exploratory insights regarding women integration in the special operations forces. The methodological approach is qualitative and employs a case study research in the Portuguese Army. It includes multiple sources of data collection for corroboration purposes, namely semi-structured interviews, direct observation and official documents. Although it has been considered a hot-topic in the Portuguese armed forces, and as the Portuguese Republic establishes the legal means to integrate women in all professional areas, reality shows that up to the present days no women have been part of the special operations forces. In light with the above, we evidence some obstacles that women are facing in gaining access to the Portuguese special operations forces, and we present Portugal in comparative perspective with two successful case studies. Keywords Women · Special operations forces · Portuguese Army · Military · Gender · Integration · Case study

J. Reis (B) Department of Military Science, Military Academy, GOVCOPP, Lisbon, Portugal e-mail: [email protected] R. Gonçalves Department of Military Science, Military Academy, Lisbon, Portugal e-mail: [email protected] S. Menezes Department of Management and Leadership, CINAMIL, Lisbon, Portugal e-mail: [email protected] M. Kaczynska Tadeusz Ko´sciuszko Land Forces Military University, Polish Land Forces Military University, Wrocław, Poland e-mail: [email protected] © Springer Nature Singapore Pte Ltd. 2020 Á. Rocha and R. P. Pereira (eds.), Developments and Advances in Defense and Security, Smart Innovation, Systems and Technologies 152, https://doi.org/10.1007/978-981-13-9155-2_23

281

282

J. Reis et al.

1 Introduction Special operations forces have become increasingly central to the wars of the twentyfirst century [1]. Every country has military units composed of personnel specially trained and equipped, with higher level of readiness than general military units and typically regarded as elite—these forces are collectively known as special operations forces [2]. Despite consistent downsizing, over the past two decades the armed forces of the industrial democracies have seen extensive growth in special operations forces (SOF). By means of increasing numbers of personnel and of more frequent deployments, SOF units have wielded considerable influence in conflicts around the world [3]. The remarkable growth of SOF is part of the military organizations and has had to dramatically adapt to their diversity and changing external and internal environments [4]. As Shamir and Ben-Ari [4] argue, SOF are unique in that they are specialized generalists, local-level integrators that link between the tactical operational and strategical levels of actions, therefore being hybrid forms of organizational response to environmental pressures. In Portugal, women are not precluded from serving in any military unit by law; however, the reality is that up to the present days no women have been part of the Portuguese SOF. The reasons are yet to be determined, since all the statistics prove the commitment of women to combat—“in approximately 10 years of combat operations in Iraq and Afghanistan, over 283 thousand female members have been deployed, over 800 have been wounded and over 130 have died” [5, p. 1]. Burrelli [5] reinforces that, on numerous occasions, women have been recognized for their heroism, two of which earning US Silver Star medals. We have conducted a research of the literature and we have detected the existence of similar articles. What is the novelty brought by our article? Most gender studies on special operations forces are from the United States (US) Army, mainly after the congress mandated that US SOF adopted gender-neutral occupational performance standards by 2016. After that year, females were no longer barred from applying to SOF rigorous pipeline, and as shown in operations in Afghanistan and Iraq, female enablers made significant, unique contributions to ground combat and SOF missions [6]. Gasca et al. [6] even went further, by arguing that many of these unique contributions, such as the ability to interact with indigenous children and female adults (forbidden by local customs or religious beliefs from interacting with male service members), were critical to mission success. Moreover, women have been employed successively in SOF units worldwide, though some Western countries still subsist to that trend. It is to that extent that our article provides differentiating contribution, as it analyses the reasons that led a Western European Army to not embrace women in their ranks. This article is organized as follows: in the next sections, we review the literature and discuss the gender concept on the military forces. Then, we describe the methodological approach, the study sample and the data analysis.

The Obstacles Women Face in Gaining Access to Special …

283

The following sections discuss the empirical results, by providing statements from the respondents. Finally, we present the academic and practical insights, limitations and guidelines for future research.

2 Literature Review The issue of women in the military has always been, and is likely to remain, one of the most controversial issues in the study of military institutions [7]. Therefore, the issue of women’s empowerment and gender equality is at the top of agendas across the world, as gender inequality is widespread in all cultures [8]. By the beginning of the twenty-first century, most Western democratic countries had admitted and increased the number of women in their armed forces; although restrictions persisted, many were lifted: women were progressively allowed to enter military academies and given access to a wide variety of positions and functions [9]. As the global momentum towards the acceptance of women in ground close combat (GCC) continues to increase, the USA decided, in December 2015, to open all GCC role to women without exception—a controversial and, some say, politically motivated decision [10]. The number of women service in the military and deployed to active duty is unprecedented in the history of developed economies—in the US Armed Services, women constitutes approximately 14.5% of the 1.4 million active component and 18% of the 850,000 reserve component [11]. Yet, little attention has been paid to women integration on military life. Thus, the history of women’s military participation shows that they have systematically been excluded from warfare as warriors or have participated only in exceptional circumstances [7]. In the USA, between 2003 and 2013, women were technically banned from direct assignment to ground combat units; however, the US military deployed all-female counterinsurgent teams in Iraq and Afghanistan [10, 12]. Greenburg [12] also argues that in various forms, these teams searched Iraqi women at checkpoints and in home raids, provided medical assistance to Afghan women and children. Adding to this, women participated in highly combative special operations missions alongside Army Rangers and Green Berets in Afghanistan. Although this empirical evidence has been demonstrated, women are still being marginalized in several military institutions around the world. The gender theme has raised attention all over the world, and it is not just about the USA. In 2016, the Canadian Armed Forces (CAF) launched a recruitment campaign aimed specifically at women, in which the military’s—and society’s—ambivalence towards female participation was palpable—one of the goals of the campaign was to recruit sufficient women so as to change the CAF’s violent, sexist and sexualized culture [13]. The stereotype has made women’s integration in the armed forces challenging and supports a gender binary “that characterizes men as active, women as passive” and is damaging their credibility as combat soldiers [13]. The reality is that many nations still do not allow women to serve in combat roles or in submarines [14]. Still, a greater number of combat-support and possibly service-support roles

284

J. Reis et al.

performed by women will increasingly mean they will find themselves in combat zones [10]. Szayna et al. [15] have recently conducted a survey about women integration in SOF and concluded an opposition to opening SOF specialties to women in both deep and wide, with high levels of opposition across all SOF elements. Those authors’ analysis of responses to open-ended questions suggests that this opposition also is deep-seated and intensely felt. Szayna et al. [15] concluded that principal sources of this opposition are the belief among SOF personnel that women do not have the physical capabilities and other to meet the demands of the specialties; however, the lower level of opposition to women in SOF units than specialties, as well as the fact that about four in ten of our respondents agreed that women might be helpful in conducting sensitive operations and communicating with local populations might present additional opportunities for the participation of women in SOF. Sunde and Kristiansen [16] argue in the same direction, stating that women in certain urban special reconnaissance (SR) roles will increase the SR capacity in small states NATO SOF. As the authors studied the Norwegian Special Operations Forces (NORSOF) case, the found that conducting SR in urban areas often involves indirect or direct human contact. Female SR operators will tear up the stereotypical signature of male operators, and thereby provide operational flexibility and will enable a SOF unit to handle a broader range of SR missions, consequently increasing the strategic and political value of that unit [16]. Meanwhile, Knarr et al. [17] conducted a survey with Retired SOF Senior Leaders who agreed that any chance of success requires no compromise in the assessment and selection standards across the SOF enterprise (e.g. “Do not lower the standards for physical and mental endurance”, “Keep the standard!”), this followed a focus group concerned with the ability of mixed-gender teams to effectively perform direct action missions, perceived to require the highest level of physical capabilities. From a NATO SOF integration, Kristiansen [18] suggests to analyse the different roles and to produce a selection and education programme that suits the standards for these specific roles, instead of forcing women through the SOF selection and education programmes that already exists. In Kristiansen’s [18] view, the physical standards for women NATO SOF need to be adapted, based on potential relevant roles for female operators, such as the level of risk in NATO SOF is expected to be handled in order to accomplish high-risk missions.

3 Methodology This research is qualitative and exploratory, trying to explain a phenomenon for which little or no empirical data exist [19]. To that end, we have conducted a case study to enable the research to acquire in-depth and holistic understanding of multiple aspects of the phenomenon, as well as the interrelationships between different aspects [20]. We have built this study on multiple sources of data collection, for triangulation and corroboration purposes. Those sources of data collection consist on semistructured interviews, direct observations and institutional documents. A case study

The Obstacles Women Face in Gaining Access to Special …

285

that relies on multiple sources prevents the exclusive reliance on a single data collection method, thus aiding to neutralize any bias inherent to a particular data source [21]. The participants for the interviews were selected according to the following characteristics: six servicemen with the army special operations course and six servicewomen from the Portuguese Army. The researchers made use of their personal network in the army to identify the best respondents. The selected respondents were chosen according to their different functional areas and different levels of responsibility in the military organization. Thus, the study was carried out from operational to administrative areas, and the respondents’ ranks ranged from master sergeant to colonel. The direct observations involved systematically seeing and listening [22] and mainly consisted in informal interviews and notes that were collected into a research diary. Those notes contained the respondents’ reactions and expressions to certain type of questions, while the informal interviews had mainly served to deepen the knowledge that otherwise would not possible to ascertain during the formal interviews. The analysis of institution documents served essentially to understand the Portuguese concept of special operations forces, their main missions and to understand into what extent the training/course can be an obstacle to women integration. Similar to other studies, before proceeding, we took the following details in consideration: first, the women in SOF survey dealt with a highly politicized issue in Portugal as respondents might have viewed the survey as an influence to policy, rather than an opportunity to identify the potential challenges and opportunities for smoothing implementation and to change mentalities [15]. Therefore, we believe some respondents might not have answered in an overt way—to mitigate a possible disruption, we have cross-checked these answers with secondary sources of data collection. In order to enhance the reliability and validity of the case study research, we have followed a research protocol (Appendix) and the transcriptions were doublechecked by the participants to avoid misinterpretations. After we collected the data, all the text was analysed according to the technique of content analysis [23]. We had preliminary categorized textual data into categories, to identify consistent patterns and relationships between variables in a way of reducing data and making sense of them [21]. The study was resorted to the use of qualitative data analysis software known as NVivo, which allowed to code in order to build and hierarchize categories and subcategories to identify the main stream and/or new ideas.

4 Findings This section presents the results of the case study research. In addition, we also provide an insightful contribution, by comparing two European realities, Portuguese and Polish, from military point of view of the authors. Finally, we present an inspiring case, witnessed from a military mission in Afghanistan.

286

J. Reis et al.

4.1 The Case Study of the Portuguese Special Operations Forces The special operations forces have a complex organization, a diverse set of capabilities and a broad range of officially assigned missions, all of which can make it difficult to understand exactly who special operation forces are and how they should be used [24]. In Portugal, the army special operation forces are especially assigned to conduct operations of special reconnaissance, direct actions and military assistance. Moreover, the directive nº 90 from the Chief of Staff of the Portuguese Army (CSPA) states that, due to SOF unique characteristics, they act within the spectrum of crisis response operations, combat operations for search and rescue and counterterrorism operations [25, pp. 2–4]. The respondents have supported the ideal of combat-support and service-support roles to be performed by women in the special operations forces. At the same time servicewomen argued that women might be helpful in conducting sensitive operations with local population, they have also claimed that if women gather enough physical and psychological robustness, they should be included in SOF combat missions. In addition, informal interviews have revealed that until very recently the physical component of SOF in Portugal was highly valued; this information is also corroborated with internal documents. However, this paradigm has been changing, while the technical and intellectual component is highly appreciated, e.g. in order to perform complex missions that involves modern military technologies [26]. Although some scholars and SOF servicemen make a distinction about different physical capabilities, there are no scientific evidence concerning technical and intellectual discrepancies between men and women. Furthermore, a survey in Portugal shows that currently the number of women attending higher education in Portugal surpasses that of men [27]. One of the possible solutions to integrate women into the Portuguese special operations forces is what servicewomen call “pioneers”, that is, a group of women being able to successfully undertake a special operations course and combat operations alongside with men. Demystifying the subject, thrust and inspiration may be the formula that women need to undertake a life in SOF units. The interviewed servicewomen are well informed about the progress made until the present days in both US and Afghan SOF. If women are willing to attend and complete a special operations course, they will encourage and inspire others to move forward as they have proven that this opportunity is within the reach of every person. This opportunity is getting closer, because although official documents show that the special operations course is still very demanding from the physical point of view, it is increasingly becoming more technical, whereas the adaptability, the complex technical and tactical capabilities of SOF students are pushed to limit. Nevertheless, direct observations have pointed out that some servicewomen are reluctant about the robustness level that would be imparted to them during the SOF course, when compared with their male counterparts, in order to make them give up. Apparently, the idea of SOF operators is that integration of women into the special operations forces might be identified

The Obstacles Women Face in Gaining Access to Special …

287

as a reduction of the parameters of physical robustness, which in their mind could place the Portuguese SOF in a situation of fragility when compared with other special forces (e.g. Commandos). The Portuguese paratroopers have a different understanding—they are considered a special force, but they also integrate women in their ranks. The Portuguese armed forces included women for the first time in 1961, when Isabela Bandeira de Mello, the first civilian parachutist women in Portugal, influenced the armed forces to create an all-female corps of nurses to be deployed to the colonial war [28]. The first women were then trained as parachutists, with the mission of providing nursing assistance during the war in Portugal’s overseas territories in Africa and served in combat areas in Angola, Mozambique and Guinea-Bissau during the 13-year colonial conflict [29]. On numerous occasions, these female paratroopers had been recognized for their heroism: an example is Maria Justino da Silva, who was wounded in combat in Mozambique, during an evacuation mission to rescue wounded soldiers [30]. Since those days, women from the paratrooper forces had been well accepted and integrated in all types of military operations. We also considered valuable to present two successful cases and thus to present Portugal in comparative perspective. The next cases were selected due to the authors’ affinity with those realities, mainly due to their professional experiences.

4.2 Comparison with European Countries—The Polish Model Currently, Poland employs women in close combat roles [31]. Nowadays, women that complete 5 years of university assume military posts [32]. According to Cawkill et al. [32], from 2003 to 2008 the Polish armed forces made outstanding progresses, as they achieved 1153 servicewomen in the army; moreover, females platoon commanders have been deployed to Polish military contingents in Iraq and Afghanistan. Although the Polish Army is given as an example, several European countries integrate women into combat operations—Denmark, Finland, France, Germany, Netherlands, Norway, Romania, Spain and Sweden [31]. Moreover, recent studies argue that female soldiers’ experiences refute their male colleagues’ assumptions, regarding their ability to serve in combat units [33]. Polish Special Forces comprise five service operation units. The Polish Grupa Reagowania Operacyjino Mobilnego (GROM) is a special operations force unit that has already gained an excellent reputation within the world’s special forces communities, while the GROM training is made as realistic as possible as all the training is mainly carried out using live ammunition [34]. Their training is based on three main areas: black tactics that are focused on counterterrorism operations on the ground, green tactics-special operations in enemy’s rear and activities associated to the releasing of prisoners of war and blue tactics which involves counterterrorism operations on the sea. Polish Jednostka Wojskowa Komandosów (JWK) is specialized in special

288

J. Reis et al.

reconnaissance. Special reconnaissance reinforces national and allied measures and systems to gain information that is relevant to carry out operation in extreme circumstances. The youngest polish special operation force unit is Jednostka Wojskowa AGAT (JW AGAT) that was set up in 2012. This unit involves counterterrorism operation in urban and rural areas. The fourth and last unit is Jednostka Wojskowa NIL (JW NIL); they operate as a support for other special operations force units by providing access to information, commanding and logistic support. According to the constitutional principal of gender equality, both women and men can become professional soldiers; these rules for entering, fulfilling and terminating military service provide equal rights and place the same requirements on both men and women [35]. Currently, this principle is also applied to the special operations forces, as they integrate women at SOF units, especially in logistics areas, but they are not directly involved in tactical operations—this is identified as their main observed obstacle.

4.3 The Inspiring Struggle—The Afghan Case One of the co-authors of this article has served in the NATO Special Operations Component Command-Afghanistan (NSOCC-A). During his duties, as advisor of the Ministry of Interior, he followed the commitment of the North Atlantic Treaty Organization (NATO) gender advisors while accomplishing their duties (train, advise and assist) in order to increase women’s representation at all levels of the Afghan National Defence Security Forces (ANDSF). Women operating in the Afghan special security forces (ASSF) may integrate the General Command of Police Special Units under the Ministry of Interior (MoI) or service with the Afghan National Army Special Operations Command under the Ministry of Defence (MoD). Recruiting women to ASSF is clearly a higher challenge when compared with most of the Western countries. Despite this fact, gender discrimination remains an issue within the ANDSF; our belief is that, being pushed by NATO advisors, the Afghan leaders are progressively changing their minds and accepting females’ integration into police and army special forces. The sacrifices made by the Afghan women in the line of duty might be seen as an inspiration to developed countries that have not yet been able to successfully integrate women in their SOF.

5 Concluding Remarks In this article, it is concluded that although the physical component of the SOF in Portugal is highly valuable, this paradigm has been changing. Due to some changes, motivated by the use of military technologies, the technical and intellectual component is now appreciated in the Portuguese SOF. Although SOF operators defend a distinction concerning different physical capabilities between men and women, there is no scientific evidence concerning the technical and intellectual discrepancies. It is

The Obstacles Women Face in Gaining Access to Special …

289

with the aforementioned arguments that we support the integration of women into the Portuguese SOF; in addition, existing examples have shown not only the feasibility of their integration, but also new acting opportunities in all types of military operations. This article is not free of limitations: due to confidentiality reasons, we have not provided any information about key informants and the respective military unit they serve. It is also limited because of its exploratory nature, but we hope that we can encourage future investigations as it may also be relevant to fill a gap in the literature. However, case studies are difficult to generalize. Thus, the same study in other countries may carry the risk of obtaining contradictory findings. This issue should not be considered a problem itself. To mitigate the generalization aspect, we suggest to adapt our results to each SOF’s and country realities. We instigate academics and practitioners to provide new and insightful contributions to this emergent topic as it represents a fertile opportunity for future research. Future research may focus on the role and impact of women in the Portuguese paratroopers. In addition, we also suggest the role and impact of women in the modern Afghan Armed Forces. Acknowledgements This project was started under the sponsorship of Erasmus+ programme, which allowed students from different European countries to cooperate. We would like to thank the Portuguese and Polish Army, for their outstanding contribution. Lastly, but not least, we would also like to thank the participation of the anonymous respondents, which greatly contributed to the production of the final results.

Appendix Interview protocol—SOF Personnel/Military Women 1. Introduction • • • •

Explain the purpose of the interview Explain the reasons for tape recording Mention the length of the interview Discuss the confidentiality purpose.

2. Generic information • What is your current professional activity? • For how long did you work (have you been working) at the Special Operations Forces? (this question is not applied to women) • For how long have you been working to the Portuguese Army? • Age?

290

J. Reis et al.

3. Generic questions • What do you think about the integration of women in the Special Operations Forces? • If you agree to it, in what branches? Why that branch? • Should women perform tactical/combat operations? • If not, why? 3.1. Specific questions – Do you know any international model of integration of women in the special operations forces? – If so, can you give some examples? – Which model do you consider the most appropriate and why? – As you know, there are no women in the Portuguese special operations. In your opinion what is the reason? – What can be done, so that the integration of women in this type of forces would be successful? – In your opinion, what is the greatest contribution that women can bring to the special operations forces? 4. Closing comments • Is there anything to add which remained unsaid? • Are you available to revise the transcription of the interview? • Thank you for your participation in the interview.

References 1. Long, A.: NATO special operations: promise and problem. Orbis 58(4), 540–551 (2014) 2. Kirby, S., Harrell, M., Sloan, J.: Why don’t minorities join special operations forces? Armed Forces Soc. 26(4), 523–545 (2000) 3. Turnley, J., Michael, K., Ben-Ari, E.: Special Operations Forces in the 21st Century: Perspectives from the Social Sciences. Routledge (2017) 4. Shamir, E., Ben-Ari, E.: The rise of special operations forces: generalized specialization, boundary spanning and military autonomy. J. Strateg. Stud. 41(3), 335–371 (2018) 5. Burrelli, D.: Women in combat: issues for congress. In: Library of Congress Washington DC Congressional Research Service, pp. 1–17 (2013) 6. Gasca, F., Voneida, R., Goedecke, K.: Unique capabilities of women in special operations forces. Spec. Oper. J. 1(2), 105–111 (2015) 7. Carreiras, H.: Gender and The Military: Women in the Armed Forces of Western Democracies. Routledge (2006) 8. Bayeh, E.: The role of empowering women and achieving gender equality to the sustainable development of Ethiopia. Pac. Sci. Rev. B Humanit. Soc. Sci. 2(1), 37–42 (2016) 9. Carreiras, H.: Gender relations in the armed forces: does the draft make a difference? Sicherh. Frieden (S+F)/Secur. Peace, 19–24 (2012) 10. Fitriani, Cooper, R., Matthews, R.: Women in ground close combat. RUSI J. 161(1), 14–24 (2016)

The Obstacles Women Face in Gaining Access to Special …

291

11. Boyd, M., Bradshaw, W., Robinson, M.: Mental health issues of women deployed to Iraq and Afghanistan. Arch. Psychiatr. Nurs. 27(1), 10–22 (2013) 12. Greenburg, J.: New military femininities: humanitarian violence and the gendered work of war among US servicewomen. Gender Place Cult. 24(8), 1107–1126 (2017) 13. Lane, A.: Special men: the gendered militarization of the Canadian armed forces. Int. J. 72(4), 463–483 (2017) 14. Pinch, F., MacIntyre, A., Browne, P., Okros, A.: Challenge and Change in the Military: Gender and Diversity Issues. Wing Publishing Office (2004) 15. Szayna, T., Larson, E., O’Mahony, A., Robson, S., Schaefer, A., Matthews, M., Polich, M., Ayer, L., Eaton, D., Marcellino, W., Miyashiro, L., Posard, M., Syme, J., Winkelman, Z., Wright, C., Zander-Cotugno, M., Welser, W.: Considerations for Integrating Women into Closed Occupations in US Special Operations Forces. Rand Corporation (2016) 16. Sunde, F., Kristiansen, M.: Defying stereotypes: the untapped potential of integrating female reconnaissance operators into small state NATO SOF. Small Wars J. (2018) 17. Knarr, W., Turnley, J., Stewart, D., Rubright, R., Quirin, J.: Special operations forces mixedgender elite teams (SOFMET): examining socio-cultural dynamics of SOFMET. Joint Special Operations University, Centre for Special Operations Studies and Research, 1–117 (2014) 18. Kristiansen, M.: Women: a NATO special operations forces force multiplier. Small Wars J. (2017) 19. Yin, R.: Case Study Research: Design and Methods. Sage, Thousand Oaks (2003) 20. Gummesson, E.: Qualitative Research in Management. Qualitative Methods in Management Research. Sage, London (1991) 21. Given, L.: The SAGE Encyclopaedia of Qualitative Research Methods, vol. 1–2. Sage, Thousand Oaks (2008) 22. Taylor-Powell, E., Steele, E.: Collecting Evaluation Data: Direct Observations. University of Wisconsin, Madison (1996) 23. Mills, A., Durepos, G., Wiebe, E.: Encyclopedia of Case Study Research. Sage, California (2010) 24. Robinson, L.: The future of US special operations forces. Counc. Foreign Relat. 66, 1–39 (2013) 25. Chief of Staff of the Portuguese Army (CSPA): Âmbito de atuação, missões e tarefas das unidades de tropas especiais. Divisão de Planeamento de Forças, diretiva nº 90 do Chefe de Estado-Maior do Exército Português (2007) 26. Knie, J.: Special Forces Skills: Are they Special Enough? U.S. Army War College, Carlisle Barracks PA. (2002) 27. Pordata: Students enrolled in higher education: total and by sex. Statistical database. https:// www.pordata.pt/en/Search/students%20in%20Portugal,%20sex (2018) 28. Torrão, S.: Anjos na Guerra. Oficinas do Livro, 1–168 (2013) 29. Carreiras, H.: Women in the Portuguese armed forces: from visibility to eclipse. Curr. Sociol. 50(5), 687–714 (2002) 30. Grão, L.: Enfermeiras pára-quedistas 1961–2002. Prefácio (2006) 31. Haring, E.: What Women Bring to the Fight. Army War College Carlisle Barracks PA. (2013) 32. Cawkill, P., Rogers, A., Knight, S., Spear, L.: Women in Ground Close Combat Roles: The Experiences of Other Nations and a Review of the Academic Literature. Defense Science and Technology Labroratory, Porton Down (2009) 33. Doan, A., Portillo, S.: Not a woman, but a soldier: exploring identity through translocational positionality. Sex Roles 76(3–4), 236–249 (2017) 34. Ryan, M.: The Operators: Inside the World’s Special Forces. Skyhorse Publishing Inc. (2008) 35. MNDRP, Ministry of National Defence Republic of Poland—Military service. http:// archiwalny.mon.gov.pl/en/strona/333/LG_260_261 (2018). Accessed 31 Oct 2018

The Transformation of the Defense and Security Sector to the New Logistics 4.0: Public–Private Cooperation as a Necessary Catalyst Strategy Manuel A. Fernández-Villacañas Marín

Abstract The logistical needs to develop and maintain both military and police weapon systems and support material are being met to respond to a set of certain risks, which now imply the invasion of the territory as well as other threats, such as international terrorism, cyberattacks, corruption, drug and arms trafficking, global organized crime, or the trafficking of human beings. The strategic support that will guide the new operative concepts and together with the acquisition of new military and police capabilities is based nowadays, and will be increasingly in the future, on technology and logistics capacity. In this sense, the concepts of the new logistics have experienced a great catalyze with the digital transformation, which requires a complete organizational, cultural, and strategic reinvention of both the armed and police forces and the defense and security industry. The qualitative research conducted has detected a gap in the reactive adaptation of the armed and police forces to the new logistics 4.0, in relation to its proactive implementation by the main companies of this sector. The implementation of public–private cooperation models, which in addition to involving advantages in the improvement of military and police efficiency as well as business development for the industrial sector, will involve the permanent and bidirectional transfer of knowledge among the structures implicated, including the adaptation and permanent dynamic adjustment to the great changes that the exponential digital evolution will imply. Keywords Digital transformation · Logistics 4.0 · Defense and security industry 4.0 · Military logistics · Armed and police forces · Public–private cooperation

M. A. Fernández-Villacañas Marín (B) Spanish Air Force, Madrid, Spain e-mail: [email protected] Higher Technical School of Aeronautical and Space Engineering, Technical University of Madrid, Madrid, Spain © Springer Nature Singapore Pte Ltd. 2020 Á. Rocha and R. P. Pereira (eds.), Developments and Advances in Defense and Security, Smart Innovation, Systems and Technologies 152, https://doi.org/10.1007/978-981-13-9155-2_24

293

294

M. A. Fernández-Villacañas Marín

1 Introduction The vast changes that have taken place in this decade have made it necessary and evident for the armed and police forces of the international community, to understand the nature and current factors of the global environment that condition their internal activity. The risks and threats in our days and, therefore, the problems of defense and security are much broader than those of other times as a result of their global impact, the characteristics and possibilities of the new digital technologies applied, as well as of the “intelligent” armament used [6]. Accordingly, the current logistic needs regarding the development and maintenance of weapons systems and their military and police support material are being covered by realistic and adjusted public budgets to give sufficient response to a set of certain risks, that within the framework of the New World Order, moderately implies the usual threat of invasion of the territory, and a significant presence of a new and diverse set of threats, such as international terrorism, cyberattacks, corruption, drug trafficking and trafficking of international arms, global organized crime, or the trafficking of human beings, which are supported by anonymous groups willing to achieve their objectives at the expense of insecurity and international social destabilization [8]. The strategic support that will guide the development of new operational concepts and the acquisition of new military and police capabilities is based on three key premises. Firstly, that the present and future strategic environment is uncertain, complex and conflictive; secondly, that no crisis can be resolved satisfactorily with isolated military and police employment, which must be combined and integrated with other initiatives of a civil, political, economic, humanitarian, or informational nature; and finally, that the military and police forces must have a balanced and an adequate catalog of capabilities both for the conventional conflicts and for irregular and hybrid combat, whose source of “military and police advantage” will be technology and logistics capabilities [10]. All of this has involved, since the end of the previous decade and the beginnings of the present one, a transformation of the defense and security sector, which has been largely driven by the development and implementation of the so-called new logistics in its adaptation to the global world. However, the new technological scenario represented by digital transformation, and in its scope, the implementation of the concepts of Industry 4.0 and logistics 4.0 in this sector [15], both in its public and private sphere, requires a complete organizational, cultural, and strategic reinvention both the armed and police forces and the international industrial sector of defense and security, thus providing the international community and each of the countries with a dynamic redesign capability that can respond to the needs arising from the processes of exponential technological evolution, that this new paradigm carries with it.

The Transformation of the Defense and Security Sector …

295

2 Background and Literature Review 2.1 The New Logistics: Creating Competitiveness and Global Sustainable Development In recent years, logistics has been acquiring a growing importance in the strategy of companies and has become an essential factor for improving their competitiveness in a constantly changing global market. Thus, the application of improvements in the performance of methodologies and technologies in the logistics field, which have developed exponentially with the Internet, implicitly leads to obtaining competitive advantages. These advantages are based not only on increasing efficiency and effectiveness in logistics management, but also on increasing the value provided to customers and global sustainability. The new logistics is based on a conceptual transition from a “push” approach to a “pull” approach, developing new logistic models that are more efficient, sustainable, intelligent, agile, adaptable, scalable, and resilient. The following figure summarizes the main aspects of traditional logistics and its evolution to the new logistics [9] (Fig. 1). On the other hand, the sustainability of global logistics [3–5] and of supply-chain management represents a strategy of transparent integration of the social, environ-

Fig. 1 Traditional logistics versus the new logistics

296

M. A. Fernández-Villacañas Marín

Fig. 2 La prospectiva de evolución hacia el Physical Internet en 2050. Source http://www.etplogistics.eu/wpcontent/uploads/etpalice/Road_map_Alice_PIE.jpg, 17.12.2018

mental, and economic objectives of the organization, within the system of coordination of the main inter-organizational business processes, in order to improve the long-term economic results of each company and its supply chains [1]. The development prospective of the new logistics with a long-term projection has shaped the theory of Physical Internet [14]; a vision that has been adopted both in the USA and in the European Union as the conceptual goal of logistics for 2050. It involves achieving an open global logistic system, based on a physical, digital and operational interconnectivity, through encapsulation, interfaces and protocol design, in order to move, store, perform, provide, and use physical objects throughout the world in an economically, environmentally, and in a socially efficient and sustainable way. The aim is to eliminate inefficiencies in global transport and waste management of logistics networks, in a similar way to what the Internet raised worldwide for information flows [13] (Fig. 2). It will be necessary to create an open market for the transport of goods, with shared, open and adaptable distribution chains, in which the products will be transported in modular, standardized, and intelligent containers, allowing each unit to be followed and controlled. For this, it will be necessary to achieve a full level of global collaboration, redefining the competitive space, taking it out of the logistics processes, and taking it only to the points of sale in which customers define the market share of each product.

The Transformation of the Defense and Security Sector …

297

The viability of the Physical Internet will obviously require the implementation of existing digital-enabling technologies, such as the Internet of things, cyberphysical systems, collaborative robotics, e-commerce, blockchain, virtual reality, augmented reality, additive manufacturing and 3D printing, artificial intelligence, big data, data analytics, predictive maintenance, simulation, collaborative economics, cloud computing, cybersecurity, and machine learning, but above all, the incorporation of other more advanced and complex innovations resulting from the exponential technological evolution that are being developed now and in the future.

2.2 Logistics 4.0: The Transformation of the New Logistics to the Digital Domain Digital transformation can conceptually be defined as the process of organizational, cultural, and strategic reinvention, both of companies and of public entities, necessary for the integral application of the technology that we call digital, which generates, processes, stores, and uses data, information and intelligence, to improve its performance as well as its ability to quickly adapt to disruptive or radical changes generated in the environment. This technological scenario, an authentic new paradigm, is inducing the emergence of shorter and shorter management cycles, in which the environment changes continuously and increasingly faster, with a pace that we assume will continue to accelerate in the future, with an approach in exponential technological evolution as has already been highlighted. The main catalyst for change and the cause of this continuous acceleration is the digital revolution, motivated by the expansion of the Internet, information and communication technologies, as well as the universalization of its use, which is in turn involving the transformation of our lifestyle. In this way, we are today fully imbricated in a process of deployment of disruptive technologies that are profoundly modifying the business reality and the society in general. Although computer science signified a cyclopean advance in the mechanization and automation of the processes, and the later connection of the equipment between them, it also generated the beginning of a formidable process and distribution capacity of the information, the digital factor has exponentially multiplied the connectivity of all public and private actors, among which, logically, citizens are included as protagonists. The inclusion of digital technologies and the transformation induced by them defines what has come to be called the Fourth Industrial Revolution. Effectively globalization, the universalization of the use of the Internet, the full automation of processes and the digitization of information, have led to an authentic Industrial Revolution, in which the hybridization of the physical and virtual world is taking place; products, machines, tools, factories, warehouses, and vehicles are interconnecting each other and work automatically, an interconnection of all the elements of the value chain that is becoming intelligent and that is leading to the creation of

298

M. A. Fernández-Villacañas Marín

authentic centered networks in the clients. In this way, the environment appears to be changing, hybrid, competitive, global, connected, and interdependent. The result of all this will allow us to achieve immediate answers in decisionmaking based on the information captured in real time, through intelligent systems and processes, without variability or errors, with full traceability in the process chains and total sustainability. All of this leads us to a new situation that will affect the way of producing and controlling processes, applying logistic models, and developing marketing and commercialization strategies. In this way and totally linked to the concept of Industry 4.0, the concept of logistics 4.0, which develops and assumes the concepts of the new logistics, transcends and emerges due to its importance. That is that logistics, as a key element of the industrial sector, cannot be kept aside. The Smart Factory implicitly assumes that manufacturing plants, facing a traditional centralization derived from the search for economies of scale by volume, now consider an intelligent relocation by factors of specialization and synergies, giving rise to the creation of different and immense networks distant from interconnected production units, which will imply that the raw materials, components, and semi-finished products require greater mobility, be perfectly synchronized and guaranteed, participating in global supply networks and intra-industry trade. On the other hand, the life cycles of products in international markets tend to be continuously reduced and variations in their demands are increasingly difficult to estimate, so it is crucial to have greater connectivity and integration between the final and initial links of the global supply chains, as well as a greater capacity for the aggregation of data and automatic estimation, rigorous and real-time estimation of the different demands, with expert systems and artificial intelligence that assist the decision-making processes [9]. For all this, the new logistics 4.0 will involve the optimization and full connection of all elements and processes of the supply chain, which should generate improvements in the efficiency and effectiveness of the management of orders and shipments, a production oriented to the individual customer which increasingly demands more personalized offers, the geolocation of customers, the simultaneous multimodal omnichannel, and the optimization of global routes, a full capacity for adaptation, the total international traceability of merchandise, the reduction of stock and space necessary for storage, automation of payments, etc. Smart logistics will not only better meet new demands, but will also more accurately locate and thoroughly know current and potential customers through the processing and interpretation of the data collected and the optimization of processes. Integration, holistic vision, coherence, innovation, and flexibility are the key concepts to support the development of the new logistics 4.0. Finally, it is considered necessary to emphasize that the digital transformation and the emergence of Industry 4.0 and logistics 4.0, beyond the technological developments and their deployment, and the reinvention of strategy, organization and corporate culture, requires, on the one hand, the development of a new holistic style of a humanitarian leadership, but also, on the other, the availability of a permanent digital leadership capacity that achieves the most critical element for the success of the analyzed evolution to be collaborate and be fully involved: This is the human factor.

The Transformation of the Defense and Security Sector …

299

3 The Problem: The Imbalances and Delays Between the Adaptation to the New Logistics 4.0 of the Armed and Police Forces, and the Companies in the Defense and Security Industry In order to know the state of involvement and development regarding the digital transformation in the companies of the defense and security industry in Spain, a qualitative research has been carried out to the main ones that accounted for approximately 85% of sales in 2016 of this sector (according to the Industrial of Defense Report of the Spanish MoD), all of them with European implantation and global projection (Airbus Defense and Space, Airbus Military, Indra Systems, Navantia, ITP Turboprop Industry, Expal Systems, Airbus Helicopters Spain, Cepsa, Iveco Spain, Santa Bárbara Sistemas, Naval Constructions P. Freire, Spanish Company of Aeronautical Systems, ARPA, etc.). The study has been complemented with the vision of the two associations that bring together most of the companies of the sector in Spain: the Spanish Association of Technology Companies of Defense, Aeronautics and Space (TEDAE), and the Association of Companies that contracting with the Public Sector and especially with the Armed and Security Forces (AESMIDE). The results show that almost all companies consider that digital disruption implies opportunities for improvement in the efficiency, quality, and flexibility of processes, which will produce the potential increase in operating profits. The defense and security industry is facing a new global context that poses major strategic and structural challenges, against which practically all have proactively raised lines of action. However, the analysis as to the levels of initiative and implementation of these lines of action, in regard to the armed forces and police forces which are customers of these companies, both nationally and internationally, denotes a reactive attitude. They are aware of the implications that the digital transformation will have on their operation levels and sustainability of their systems, but this has not translated into decided and effective action plans, possibly due to the lack of funding to address them. In short, the probable existence of a gap in the reactive adaptation in the armed and police forces to the new logistics 4.0 is detected in relation to its proactive implementation by the main companies of the defense and security industry, with a delay and desynchronization that predictably will tend to grow due to the exponential technological evolution.

300

M. A. Fernández-Villacañas Marín

4 The Solution: Public–Private Cooperation Models as a Necessary Catalyst Strategy. From Operational Outsourcing to Cooperation Outsourcing, as a strategic process of the transfer of public activities of the armed and police forces to the private sector, has been an aspect in the defense and security sector in continuous development since the beginning of the 1990s in almost all the countries. Since then, their governments have opted for the implementation of outsourcing solutions in response to the growing demand for the in-service support of their weapons systems and progressive scarcity of resources, exploring better solutions in both technical and economic terms. Factors such as the acceleration of the technological evolution of armament and support material, which has made it increasingly difficult for the armed forces and police to maintain the necessary autonomy to ensure proper maintenance of their systems, or the increase in the cost of these equipment and the relative shortage of resources, especially qualified human resources, led to raise in the international arena to operational outsourcing as an effective solution to the new situation generated [12]. The management technique of outsourcing arises in the post-industrial era when competition in global markets begins [11]. Through the externalization of those activities that are not critical, the entity specializes in its core activity, improving its competitiveness and the ability to adapt to rapid changes in the environment. But more than generic outsourcing (outsourcing), as a transfer of certain activities, managed in a localized manner in the short term and with specific objectives, the concept of cooperation appears as a public–private co-management (co-sourcing). It is a strategy of global restructuring to obtain results that allows the best efforts of the organization to be oriented toward its nuclear activities and, with it, toward the fulfillment of its mission and objectives. The cooperative vision is to consider the provider of services as a partner with which a “win to win” relationship is established. To the extent that the organization internalizes that its fate is associated with that of its service provider and vice versa, the need to seek relationships in which there is mutual benefit becomes clearer [2]. Outsourcing, which in principle was proposed as a managerial technique, has potentially moved toward cooperation, partnerships, and strategic alliances. It is not normally a question of formal companies, because the capital and patrimonies are independent and the symbiosis is such that it is not necessary to have a mercantile corporate relationship to be considered business partners. In recent years, in the public sector, the public–private cooperation has become an international reference model that allows effective, efficient, and synergistic integration of entrepreneurship with the management capacity of public services of the administration. Through its mediation, certain collaboration formulas are established through which the public sector interacts with the private sector in the search for greater involvement of the latter in the development of certain complex projects, setting out a scheme of joint objectives, as well as sharing risks, costs, and benefits. It is a new conception of public management through which social needs are met in

The Transformation of the Defense and Security Sector …

301

a different way, through a new frontier of relationship between the public sector and the private sector. The trend toward its use has been generalized in recent years as a management model for large projects, particularly in the innovation and high technology sectors, in internationalization initiatives and, in general, in all those in which it is necessary to guarantee high levels of competitiveness in advance. It is conceptualized as an economic-organizational model of dynamic and plural relationship, which groups a wide range of figures that maintain a link in time between two or more public and private organizations. It signifies a new conception by means of which a transition of a relationship between administration and industry of competitive character which is generated to another one of collaborative character. On the other hand, the application of public–private cooperation in the defense and security sector entails a change of philosophy and mentality in the bodies involved in the public administration that will go on to manage certain essential services through a strategic partnership with private companies, assuming these shared risks and responsibilities with the armed forces and police forces. The advantages for both parties of the application of these models of public–private cooperation are clear. It can mean higher levels of rationalization and efficiency for the armed forces and police forces, as well as important savings via, among other factors, the reduction of military and police personnel and the technical assistance involved, as well as the reduction of costs of investment in facilities and equipment (avoiding duplicated expenses, heavy investments in spare parts stocks that must be maintained during long life cycles of the systems, reducing the costs of continuous training, technical documentation and its corresponding update, etc.). On the other, the defense and security industry can obtain a new field of business development induced by the greater investment capacity derived from public savings, addressing the development of new defense and security systems that will make the strengthening of the industrial base and technology of that sector feasible. Moreover, and perhaps most significantly, the functioning of the public–private comanagement models implies the permanent and bidirectional transfer of knowledge between the public and private sectors, as well as the adaptation and the dynamic and permanent adjustment to major changes, that the exponential digital evolution will imply that “unique entity” formed by the agencies involved in the armed and police forces with the companies participating in the defense and security industry, in a contemporary way. As a development of these models, the Spanish Air Force carried out the FENIX Project at the end of the previous decade. It was a series of applied research studies to improve the operational efficiency and profitability of surplus capacities of its Aerospace Maintenance Centers MRO (Maestranzas Aéreas) through the implementation of public–private cooperation models with the aerospace industry. The aim was to achieve improvements in the level of generation of air support services, with the lowest possible cost and risk. Previously, an international benchmarking study of the solutions applied by the air forces of the USA, the UK, France, Germany, and Italy was carried out [6, 7].

302

M. A. Fernández-Villacañas Marín

5 Conclusions The work carried out in this paper has first of all analyzed the logistical needs of development and maintenance regarding weapon systems and support material for both the military and police. Nowadays, they are based on technology and logistics capacity. In this sense, the new logistics have experienced a great catalyze with the digital transformation, which requires a complete organizational, cultural, and strategic reinvention of both the armed and police forces and the defense and security industry. Secondly, the qualitative research conducted has detected a gap in the reactive adaptation of the armed and police forces to the new logistics 4.0, in relation to its proactive implementation by the main companies of this sector. In order to reduce this gap, the implementation of public–private cooperation models, in addition to involving advantages in the improvement of military and police efficiency, and business development for the industrial sector, will involve the permanent and bidirectional transfer of knowledge among the structures involved, as well as the adaptation and permanent dynamic adjustment to the vast changes that the exponential digital evolution will imply.

References 1. Carter, C.R., Rogers, D.S.: A framework of sustainable chain management: moving toward new theory. Int. J. Phys. Distrib. Logist. Manag. 38(5) (Emerald Group Publishing Limited) (2008) 2. Fernández-Villacañas, M.A.: Externalización del Sostenimiento, Revista de Aeronáutica y Astronáutica. Spanish Air Force, Madrid (2007) 3. Fernández-Villacañas, M.A.: Externalización estratégica Vs. Externalización operativa: Desarrollos potenciales del modelo de Partenariado Público-Privado (PPP) para el sostenimiento de Armamento y Material de las Fuerzas Armadas. Universidad Comercial de Deusto, Bilbao, Spain, 28th and 29th Feb 2008 4. Fernández-Villacañas, M.A.: El outsourcing estratégico como modelo de integración de las cadenas de suministro para una logística sustentable. V Fórum Mundial de Logística, Monterrey, México, 6th and 7th Mar 2008 5. Fernández-Villacañas, M.A.: Reflexiones en torno al logro de una Logística Global Sostenible. V Congreso Internacional ACOLOG - Asociación Colombiana de Logística, Bogotá, Colombia, 28–30 Aug 2008 6. Fernández-Villacañas, M.A.: Modelos de cooperación industrial para el sostenimiento de armamento y material: El Partenariado público-privado y la rentabilización de capacidades excedentes. In: Different authors (eds.) Gobierno en la Industria de Defensa, Aula Abierta y Foro de Estudios de Seguridad y Defensa, Universidad Politécnica de Cartagena, Cartagena, Spain (2009) 7. Fernández-Villacañas, M.A.: Cooperación industrial público-privada para el sostenimiento profundo de las aeronaves del Ejército del Aire y la rentabilización de capacidades excedentes. 5º Congreso Español de Mantenimiento y 16º Congreso Iberoamericano de Mantenimiento, Asociación Española de Mantenimiento, Barcelona, Spain, 15–17 Nov 2011 8. Fernández-Villacañas, M.A.: La Nueva Economía de la Defensa en un Nuevo Orden Mundial. Revista de Aeronáutica y Astronáutica. Spanish Air Force, Madrid, Spain (2017) 9. Fernández-Villacañas, M.A.: Las plataformas logística 4.0 y la mejora del comercio global: Creando ventaja competitiva logística y desarrollo sostenible. VI Simposio Internacional Online

The Transformation of the Defense and Security Sector …

10.

11. 12.

13.

14. 15.

303

de Logística y Competitividad, High Logistics Simposios, Medellín, Colombia, 3rd and 4th Oct 2018 García Arnaiz, J: Discurso de Inauguración del Jefe de Estado Mayor de las Jornadas Aeroespaciales de Economía de la Defensa del Ejército del Aire “La Nueva Economía de la Defensa en un Nuevo Orden Mundial”. ACGEA y UNED, Madrid, Spain, 7–9 Mar 2017 Gidrón, G.: Nuevos modelos de gestión empresarial: El outsourcing de procesos de negocios. Círculo de Empresarios, Madrid, Spain (1998) Más Sabaté, J.: Gestión privada de servicios públicos. La externalización (outsourcing) en la Administración Pública. V Congreso Internacional del CLAD sobre la reforma del Estado y de la Administración Pública, Dominican Republic, 24–27 Oct 2000 Montreuil, B., Meller, R.D., Ballot, E.: Towards a physical internet: the impact on logistics facilities and material handling systems design and innovation. In: Gue, K. (ed.) Progress in Material Handling Research. Material Handling Industry of America, USA (2010) Montreuil, B.: Toward a physical internet: meeting the global logistics sustainability grand challenge. Logist. Res. 3(2–3), 71–87 (2011) Romero Garat, F.P.: Adaptación de la Defensa y las Fuerzas Armadas al concepto industria 4.0 (1 y 2). Infodefensa.com homepage, http://www.infodefensa.com/es/2018/01/29/opinionadaptacion-defensa-fuerzas-armadas-concepto-industria.php, http://www.infodefensa.com/ es/2018/02/01/opinion-adaptacion-defensa-fuerzas-armadas-concepto-industria.php. Last accessed 27 Dec 2018

Part IX

Safety and Maritime Protection

An Autonomous Airship Swarm for Maritime Patrol Constantino G. Ribeiro, Luciano Santos Constantin Raptopoulos and Max Suell Dutra

Abstract The studies of swarm of unmanned aerial vehicles (UAVs) have been increasing, and swarm of UAVs can become part of many daily tasks. But, as matter of fact, even the use of a UAV does not mean the decreasing of operational complexities and, consequently, the costs of performing its tasks, because of high costs of trained operators and remote control facilities to operate state-of-the-art UAVs. So, in order to support the operation of swarm, this work proposes a parallel/distributed framework to control mission of each UAV. These unmanned crafts will less dependent on remote control facilities. Embedding the mission control running in an embedded parallel/distributed computer system will be able to carry on basic mission control tasks. In order to prove this concept, the following items were developed: (i) a prototype of an embedded parallel/distributed computer cluster using low-cost components; (ii) new procedures to resolve navigation and collision evasion issues; and (iii) a parallel/distributed path discover program. The tests carried out in the embedded parallel/distributed computer cluster prototype and with new evasion procedures proved the viability of proposed framework. Keywords Autonomous airship · Swarm of robots · Embedded computer cluster · Parallel/distributed programs

C. G. Ribeiro (B) CEFET-RJ, UnED Itaguaí, Rodovia Mário Covas, lote J2, quadra J, CEP: 23810-000, Distrito Industrial de Itaguaí, Rio de Janeiro, Brazil e-mail: [email protected] L. S. C. Raptopoulos UnED Nova Iguaçu, Estrada de Adrianópolis, 1317, CEP: 26041-271, Nova Iguaçu, Rio de Janeiro, Brazil e-mail: [email protected] M. S. Dutra COPPE - UFRJ, Centro de Tecnologia, bloco G, salas 202/203/204, Universidade Federal do Rio de Janeiro, Cidade Universitária - Ilha do Fundão, Caixa Postal 68.503, CEP: 21945-970, Rio de Janeiro, Brazil e-mail: [email protected] © Springer Nature Singapore Pte Ltd. 2020 Á. Rocha and R. P. Pereira (eds.), Developments and Advances in Defense and Security, Smart Innovation, Systems and Technologies 152, https://doi.org/10.1007/978-981-13-9155-2_25

307

308

C. G. Ribeiro et al.

1 Introduction There is a set of long-range aircrafts tasks, many of them performed over in sensitive geographic places, that demand specialized equipment and well-trained crew. The examples of these tasks are surveillance [1], patrol [2], data collection, search, and rescue. These tasks are being substituted by drones, which eliminate crew fatigue and improve performance. Although there have been many improvements in electronics, robotics, and computing, the costs of using the so-called drones, especially in defense and surveillance, have increased a lot as stated in Table 1. The cost of using drones is high because they need a staff working in remote control station. So, motivated by a scenario of new applications for autonomous unmanned aerial vehicles (UAV), especially using airships, this work proposes a swarm of UAVs (airships) to perform maritime surveillance/patrol. In this propose, a new UAV mission control framework was developed. This framework is composed of a low-cost embedded computer cluster; a new autonomous collision evasion system; a new on fly parallel/distributed mission route generating system; and a set of instruments/sensors for UAV orientation (GPS, electronic compass, accelerometers, among others).

2 Related Work Vehicles path planning is one of the most studied subjects. It is considered a computational complex problem or with exponential time for their resolution. In this section, the most relevant and related works to this subject will be briefly described. Reference [3] proposes an infinity or achievable horizon optimization method to unknown environments. The work is based on a finite optimization time and is dependent on computational performance of the used equipment. It uses a finite state model of the movement of a mini robotic helicopter vehicle. The proposed work uses cost function called tail discharge or path horizon.

Table 1 Operational costs of surveillance crafts Platforms

Costs ($ per hour)

71 m balloon land based (http://www.navsea.navy.mil)

610

MQ-1 Predator (robotic aircraft)—low operational ceiling; long range (http://www.af.mil/information/factsheets/factsheet.asp?fsID=122)

5.000

Grumman E-2C “Hawkeye”—AWACS (airborne warning and control system)—seagoing aircraft (http://www.navy.mil/view_image.asp?id= 140671)

18.000

RQ-4 Global Hawk (robotic aircraft)—high operational ceiling; long range (https://www.americansecurityproject.org/the-us-and-its-uavs-acost-benefit-analysis/#_edn4)

31.000

An Autonomous Airship Swarm for Maritime Patrol

309

The task of controlling a robotic blimp in a strong wind environment is studied in [4]. The work proposes a stability control of the aircraft using following path technic and flying against the wind. The stability control uses a Lyapunov function that follows pre-established rules. An adapted remote control blimp of 12 m of length was used in satisfactory tests. A path optimization that uses finite retreaded horizon is proposed in [5]. The work uses a Lyapunov as a cost-to-go function to construct a viability path graph to be optimized. After that, the optimization problem is solved as sequential finite state control and sequential decision problem and is used as global planner fed by a finite retreaded horizon method. The work compares the performance and computation load of other works and the proposed model. The tests were conducted in real environment using a Blade CX robotic helicopter. A robust and simple navigation system using fuzzy logic is the subject of [6] that uses a Plantaraco robotic blimp. The work uses ultrasonic sensors to get data about the surrounding environment. These data feed a collision evasion fuzzy system and that generates quick turns of 180°. The test results show a well-balanced behavior of the fuzzy controller that can be even trained. The control and surveillance of natural disaster areas by a robotic blimp are proposed in [7]. The blimp uses a path control system with speed field method and an optimal inverted path controller as control strategy. The used strategy was efficient in strong wind environment to avoid complex time corrections in the path. The optimal inverted path controller uses y and x coordinates, a nonlinear looping based in a Hamilton–Jacobi–Bellman equation, and a Lyapunov filter to horizontal vehicle control. The practical tests used a 12.2 m length blimp and payload of 15 kg, with wind sensors and stereo cameras. Reference [8] proposes a path control using a reinforced learning method. The method keeps the blimp high using previous environment conditions and agents (autonomous computational programs). The agent programs get bonus when they execute right actions that maximize a Monte Carlo search method used to search solutions. This method leads a direct learning, with no need of previous data storage by agent programs. A Gaussian function is used to avoid problems with search space size due to learning search of state-action tuples. Ultrasonic sensors capture the state components of these tuples during blimp navigation, and a Kalman filter removes any noise. The tests used an 18 m length blimp and were conducted indoors in a 5-m-high old factory shed. The overall robotic blimp dynamic is described in a mathematical model by [9]. This model is used to develop a predictive and adaptive robust nonlinear control that is used to navigate and guide a robotic blimp over environment disturbs. The blimp data (like longitudinal flight, climb rate, turns, etc.) were collected, and the blimp dynamic mathematical model was generated by MATLAB suite. Reference [10] AURORA project proposes an image capture system that models a set of elementary signals. This set establishes a relationship between the blimp speed and the land targets. It uses a similar model of the pendulum problem but in association with the movement to a zero distance with the blimp. The work takes into account the wind effects in three possible situations: without wind and no environ-

310

C. G. Ribeiro et al.

mental disturbance; with wind and weak environmental disturbance; and with wind and strong environmental disturbance.

3 The Proposed Swarm of UAVs Airships were extensively used during World War II by the US Navy in coastal patrol operations and escort of sea trains [11]. These operations were carried out satisfactorily for escort of sea trains by Goodyear type K airship. Goodyear built 170 type K airships which served on four continents, escorting 89,000 ships. None of the escorted ships were lost in combat [11] (Fig. 1). Today, there is a set of applications that demand long periods of use, stealth ability, low cost, and reliability of the platform to be used. A good example of these tasks is anti-submarine warfare (Fig. 2).

Fig. 1 World War II US Navy airship (http://www. pbs.org/wgbh/roadshow/fts/ raleigh_200902A15_ss. html#1)

Fig. 2 Airships used in anti-submarine warfare (http://www.xconomy.com/ seattle/2008/07/11/boeingand-skyhooks-zeppelincopter-faces-safetychallenges/attachment/ skyhook_arctic/)

An Autonomous Airship Swarm for Maritime Patrol

311

A swarm of robots is a group of robots that work to perform a collective behavior to carry out a task interacting with each other and with the environment [12]. So, this work proposes a parallel/distributed framework to control mission of the whole swarm of robotic airships performing maritime surveillance and search in sensitive and strategic seashore areas of Brazilian coast as offshore petroleum platforms. This is a low cost, less dependable of human interaction, long range, and endurance solution for the task. Next sub-sessions just explain where movement and force equations came from and how they were used.

3.1 UAV Kinematics The degree of freedom (DOF) is a set of independent movements or rotations that can define the position or orientation of a mechanical system. They can be defined by Euler angles and position coordinates (x, y, and z) [13]. The Euler angles and position coordinates can give a total of six DOFs to a rigid body as stated in Table 2. The coordinate vector η totally describes the airship orientation. It uses Euler angles vector θ plus the three-position coordinated vector p shown in (1), (2), and (3) equations. An airship is a sub-actuated vehicle. So, it demands less control data than degrees of freedom (DOFs) [14] and stands for all environmental situations. The minimal representation to describe the control data to aerial vehicle is the roll (), pitch (θ ), and yaw () plus the position vector coordinate p, and they describe the airship behavior. Equations (1)–(9) represent the position, attitude, forces, and moments actuating in the airship. η = [ p, θ ]T

(1)

Table 2 Euler angles Degree of freedom (DOF)

Forces and moments

Linear and angular speeds

Euler angles and position

x-axis movement (surge)

X

u

x

y-axis movement (sway)

Y

v

y

z-axis movement (heave)

Z

w

z

x-axis rotation (roll)

K

p

ϕ

y-axis movement (pitch)

M

q

θ

z-axis movement (yaw)

N

r



312

C. G. Ribeiro et al.

p = [x, y, z]T

(2)

θ = [ϕ, θ, ]T

(3)

v = [V, ]T

(4)

V = [u, v, w]T

(5)

 = [ p, q, r ]T

(6)

τ = [ f, m]T

(7)

f = [X, Y, Z ]T

(8)

m = [K , M, N ]T

(9)

The airship position is described in relation to inertial coordinates and its speed by a body fixed referential. The η vector (1) has the referential coordinates in a specific referential and v vector (4) the speed related to inertial referential. So, the relation among variables of each referential where one referential is passed to each other [10] can be described by: p˙ = R(θ )V

(10)

where R(θ ) is the transformation matrix with rotations and translations to right referential. And after the appropriated transformations, you can use the Jacobean matrix notation (11) and (12) to describe the UAV kinematics.      V p˙ R(θ ) 03×3 η˙ = J (n)v (11) ↔ ˙ = 03×3 T (θ )  θ

(12)

3.2 UAV Dynamic The blimp dynamics, described in [15, 16, 17, 14], takes into account many data about the effects of aerodynamics, structural issues, actuators, and propulsions. Using Newton and Lagrange laws (forces and momentum applied to rigid bodies) [18]: M v˙ + C(v)v + D(v) + g(η) = τ

(13)

⎧ ⎨ η = [x, y, z, ϕ, θ, ]T v = [u, v, w, p, q, r ]T ⎩ τ = [X, Y, Z , K , M, N ]T

(14)

An Autonomous Airship Swarm for Maritime Patrol

313

where M = M RB + M A rigid body system inertia matrix (with added masses). C(v) = C RB + C A (v) coriolis force matrix and centripetal force matrix (with added masses). D(v) aerodynamic dumping matrix. g(η) gravitational forces and momentum and static sustentation vector. τ control data vector. The complete development of proposed airship (with dimensions, mass, and other parameters) can be found in [15], as well the specific control system using computed torque.

4 The New Framework The framework proposed is composed of an embedded parallel/distributed computer, a new collision evasion system, a parallel/distributed mission route generator program, and a set of electronic sensors.

4.1 The Embedded Computer Cluster The Beowulf-based computer cluster [19, 16] is a reliable and easy way to construct a high-performance computer facility. It can provide a lot of benefits as, for example, free parallel/distributed programing environment and high-performance programing tools. This work proposes an embedded computer cluster made from low cost and market (high-performance multicore board used in last generation cellular phones). These boards are connected by a switch, and cluster main node will be connected with Arduino [17] or similar interface to control airship movements and to radio link connect the cluster to the mission supervision center (see basic computer cluster layout in Fig. 3). The speed of route calculations on fly is crucial to make the framework reliable and effective. A brand-new platform is used as cluster nodes: four ODROID-x2 open development platform, based on Exynos 4412 Prime 1.7 GHz ARM Cortex-A9 Quad Core with 2 GB memory and two and 2 e ODROID-C1 [14, 20, 21, 22]. The cluster will supply the airship with enough computer power to generate its routes on fly as it needs, and it will run a new, vivid, and specific obstacle collision detection and avoidance system. The main path generator task must have performed by a graph path discover program. The Traveler Salesman (TS) algorithm was chosen to perform it. A version of TS program was modified to be executed in a parallel/distributed programing environment; using message passing interface (MPI) library, it can be executed in the proposed embedded computer cluster. The parallel/distributed TS

314

C. G. Ribeiro et al.

Fig. 3 Embedded computer cluster of proposed framework

will be performed any time it is needed, because of any disturbance in previously calculated route due to any obstacle avoidance performed procedure and every time an estimated target is reached or any other significant disturbance that affected the airship attitude or flight.

4.2 The New Path Control System The path mission control system (PMCS) uses a set of pre-selected coordinates to be reached by the proposed autonomous managed blimp. The set is used by the parallel/distributed TS module of the PMCS to establish an optimal route to complete the mission. During the execution of planned route, the laser detection and ranging (LIDAR) [23, 24] and normal radar are used to scan possible unpredicted and unexpected obstacles. As soon as an obstacle is detected, it is analyzed and the proper avoidance procedures are taken. Figure 4 shows the flowchart of path mission control system.

An Autonomous Airship Swarm for Maritime Patrol

315

Fig. 4 Flowchart of path mission control system

4.3 New Obstacle Avoidance Systems As a part of PMCS, a new collision avoidance system is proposed. As described, the framework is equipped with a LIDAR and radar to detect obstacles. The radar will make a long-range search and detection of obstacles, while the LIDAR will sort range detection and collect precise data of obstacle like distance, speed, and attitude. The system identifies two classes of obstacles: fixed and mobile ones.

316

C. G. Ribeiro et al.

Fig. 5 Flowchart of new procedure to detect mobile obstacles

The evasion of fixed obstacles uses fallow the wall procedure [25] using LIDAR data to decide which side to turn and contour the fixed obstacle as a wall. A new and innovative procedure to detect mobile obstacles is proposed using data from LIDAR and COLREGS 72 (Convention on the International Regulations for Preventing Collisions at Sea) [26] navigation rules for cross routes to avoid possible collision situations. The reason to use navigation rules is because the airship has similar behavior (dynamic forces) of a ship, and as COLREGS 72 rules are widely used and tested in control ship and their movements and routes, they must work well with the path control of blimps. Figure 5 shows the flowchart of new procedure to detect mobile obstacles.

4.4 UAV Sensors and Navigation Framework Layout A basic set of sensors is necessary to support airship autonomous operation. The basic sensors are electronic compass, speed sensor, wind speed sensor, accelerometer, altimeter, frontal radar, LIDAR with 10 km of range and 90° of aperture, GPS,

An Autonomous Airship Swarm for Maritime Patrol

317

Fig. 6 Prototype of embedded computer cluster

electronic gyroscopic, six ultrasonic close-range sensors, and four digital cameras. These should be basic and always present sensors that will supply data for airship navigation and attitude procedures. Depending on the kind of mission, some extra and specific equipment will be necessary such as night vision high-resolution cameras, electronic surveillance devices for law and enforcement and patrolling tasks.

5 Tests and Results The embedded computer cluster is the core of this propose, so a proof of concept was carried on. A prototype of an embedded computer cluster with six nodes (see Fig. 6) was constructed, and performance and fault tolerance were conducted using prototype parallel/distributed programs.

5.1 Results of Speed Up Execution Tests The basic speed up test was to execute a sequential version of TS program (developed in C language) for: 8, 16, 32, 64, and 128 targets and then to execute a parallel/distributed version of TS program (developed in C language with message passaging interface—MPI—library) for the same set of targets and compare the execution time (see Fig. 7).

5.2 Resilience and Fault Tolerance Tests In these tests, all prototype programs for mission control were executed. During the execution of all mission control programs, the processor parameters (CPU, memory,

318

C. G. Ribeiro et al.

Fig. 7 Speed up test times

Fig. 8 Single processor parameters usage test

Fig. 9 Embedded computer cluster parameters usage test

Fig. 10 Embedded computer cluster fault tolerance test

network usage) of a single processor environment and the same parameters for the embedded computer cluster were compared (see Figs. 8 and 9). Another test was a fault tolerance test where the parameters (CPU, memory, network usage) were compared for the complete cluster (with six nodes—left side of

An Autonomous Airship Swarm for Maritime Patrol

319

the figure) and a damaged cluster with less one node (with three nodes—right side of the figure) (see Fig. 10).

6 Conclusions A prototype of an embedded computer cluster with market computer elements used in mobile phones was constructed and operated as any other computer cluster with the advantage of being small and with very low energy consumption. The basic tests of speed up for embedded distributed/parallel application showed the improvement in the performance of these applications compared with the embedded sequential ones, especially with big set of targets. The usage tests proved that the prototypeembedded computer cluster has smaller consumption of computer resources in each node of the cluster, so other missions’ specific tasks can run into the cluster. The fault tolerance test showed that the embedded computer cluster can operate properly even when there is a damage in one of some of its nodes. All the tasks described above proved the viability of proposed framework.

References 1. Defense Industry Daily staff: JLENS: Coordinating Cruise Missile Defense – And More, 30 December 2012. http://www.defenseindustrydaily.com/jlens-coordinating-cruise-missiledefense-and-more-02921/ (2012) 2. Corporation, A.P.: Aerial Products, 28 December 2012. http://www.aerialproducts.com/ surveillance-systems/aerial-surveillance.html (2008) 3. Dadkhah, N., Korukanti, V.R., Kong, Z., Mettler, B.: Experimental demonstration of an online trajectory optimization scheme using approximate spatial value functions. In: Proceedings of the 48th IEEE Conference on Decision and Control, pp. 2978–2983 (2009) 4. Saiki, H., Fukao, T., Urakubo, T., Khno, T.: Hovering control of outdoor blimp robots based on path following. In: 2010 IEEE International Conference on Control Applications (CCA), pp. 2124–2129 (2010) 5. Mettler, B., Kong, Z.: Receding horizon trajectory optimization with a finite-state value function approximation. In: Proceeding of 2008 American Control Conference, Seattle, Washington, pp. 3810–3816 (2008) 6. González, P., Burgard, W., Sanz, R., Fernadez, J.L.: Developing a low-cost autonomous indoor blimp. J. Phys. Agents 3(1), 43–51 (2009) 7. Fukao, T., et al.: Inverse optimal velocity field control of an outdoor blimp robot. In: Proceedings of the 17th World Congress the International Federation of Automatic Control, Seoul, Korea, pp. 4374–4379 (2008) 8. Rottmann, A., Plagemann, C., Hilgers, P., Burgard, W.: Autonomous blimp control using modelfree reinforcement learning in a continuous state and action space. In: Proceeding of International Conference on Intelligent Robots and Systems, IROS 2007, pp. 1895–1900 (2007) 9. Gammon, S.M., Fryr, M.T., Qian, C.: The mathematical model of the tri-turbofan airship for autonomous formation control research. In: IEEE Region 5 Technical, Professional, and Student Conference, San Antonio, Texas (2006)

320

C. G. Ribeiro et al.

10. Azinheira, J.R., et al.: Visual servo control for the hovering of an outdoor robotic airship. In: Proceedings of the 2002 IEEE International Conference on Robotics & Automation, Washington, DC, pp. 2787–2792 (2002) 11. Machry, C.R.: As Potencialidades de Emprego dos Dirigíveis no Transporte de Carga. Reista da UNIFA (2003) 12. Navarro, I., Matía, F.: An introduction to swarm robotics. ISRN Robot. 2013, Article ID 608164. http://dx.doi.org/10.5402/2013/608164 (2013) 13. Fossen, T.I.: Guidance and Control of Ocean Vehicles. Wiley, West Susses, England (1994) 14. Hardkernel Co., Ltd.: Products, 24 January 2013. http://www.hardkernel.com/renewal_2011/ products/prdt_info.php (2013) 15. Ribeiro, C.G., et al.: A platform for autonomous path control of unmanned airship. J. Braz. Soc. Mech. Sci. Eng. (2017). https://doi.org/10.1007/s40430-017-0891-9. ISSN 1678-5878, Rio de Janeiro, Brazil 16. Becker, D.J., Sterling, T.: Beowulf: a parallel workstation for scientific computation. In: Proceedings of the International Conference on Parallel Processing (1995) 17. ARDUINO: Arduino, 25 March 2013. http://www.arduino.cc/ (2013) 18. Fossen, T.I.: Marine control systems, guidance, navigation, and control of ships, rigs and underwater vehicles. In: Marine Cybernetics, Trondheim, Norway (2002). ISBN 82-9235600-2 19. Reschke, C., Sterling, T., Ridge, D.: A design study of alternative network topologies for the Beowulf parallel workstations. In: Proceedings of the IEEE International Symposium on High Performance Distributed Computing (1996) 20. Ralph, N.: Odroid-x development board brings quad-core Exynos 4 Quad processor to budding Android hackers for $129, 24 January 2013. http://www.theverge.com/2012/7/13/3156032/ odroid-x-development-board-exynos-4412-quad (2012) 21. Larabel, M.: Quad-Core ODROID-X Battles NVIDIA Tegra 3, 24 January 2013. http://www. phoronix.com/scan.php?page=article&item=samsung_odroidx&num=1 (2012) 22. redOrbit Staff & Wire Reports - Your Universe Online: Miniature Quad-Core Computer For Under $130, 24 January 2013. http://www.redorbit.com/news/technology/1112656695/ miniature-quad-core-computer-for-under-130/ (2012) 23. Lidar USA 3D Documentation and Beyond: Lidar USA 3D Documentation and Beyond. http:// www.lidarusa.com/page.php?cat=3, 7 July 2013 (15 de December de 2013) 24. Texas Instruments: LIDAR System Design for Automotive/Industrial/Military Applications, SIGNAL PATH designer SM. Tips, tricks, and techniques from the analog signal-path experts Texas Instruments, Literature Number: SNAA 123 (2013) 25. i Badia, S.B., Pyk, P., Verschure, P.F.: A biologically inspired flight control system for a blimp-based UAV. In: Proceedings of the 2005 IEEE International Conference on Robotics and Automation, pp. 3053–3059 (2005) 26. U.S. Department of Homeland Security: NAVIGATION RULES ONLINE, 24 January 2013. http://www.navcen.uscg.gov/?pageName=navRulesContent (2013)

Assessing the Location of Search and Rescue Stations on the Portuguese Coast Anacleto Correia , Ricardo Moura

and Miguel Fonseca

Abstract The compliance, by a signatory country, with the International Convention on Maritime Search and Rescue (SAR) requires maintaining adequate means of search and rescue at sea. An important component of such infrastructure is the lifesaving stations, distributed along the coast and equipped with human resources and equipment, such as lifeboats. The assessment of the best geographical location of lifesaving stations (LSS) can be carried out considering multiple criteria. Among them, one can mention: the proximity to areas historically associated with the occurrence of accidents; the maritime traffic pattern in the waters under surveillance; the severity of accidents (in terms of loss of human life and environmental impact of accidents); the typology and cause of the recorded accidents, as well as, the technical characteristics of the lifeboats assigned to life-saving stations (e.g., autonomy, speed). This work aimed to assess the adequacy of the current geographic distribution of the resources of the maritime rescue system on the Portuguese coast, and specifically the location of the life-saving stations, considering the history of the number of accidents at sea. For this purpose, spatial analysis tools were used to compare georeferenced information on the location of the LSS, as well as the autonomy of the lifeboats assigned to the stations, with maritime accidents occurred in their proximity in recent years. The aim was to assess the lifeboats’ degree of coverage compared to the location of maritime accidents registered. In addition, the work also aimed at eliciting the functional requirements of a decision support system for maritime search and rescue. Keywords Geographic information systems · Spatial analysis · Study of location · Maritime accidents · Search and rescue

A. Correia (B) · R. Moura CINAV—Alfeite, 2810-001 Almada, Portugal e-mail: [email protected] R. Moura · M. Fonseca CMA, Centro de Matemática e Aplicações, 2825-149 Caparica, Portugal © Springer Nature Singapore Pte Ltd. 2020 Á. Rocha and R. P. Pereira (eds.), Developments and Advances in Defense and Security, Smart Innovation, Systems and Technologies 152, https://doi.org/10.1007/978-981-13-9155-2_26

321

322

A. Correia et al.

1 Introduction The ocean is a development factor that makes possible different traditional and emerging economic activities, such as fishing, shipping transportation, tourism, and recreative sailing, among others. Accidents caused by maritime activities result in heavy losses of human lives every year; hence, the need to effectively prevent and respond to emergencies at sea. Following a maritime accident, response time is critical, as the delay of a rescue is often a factor that makes the difference between life and death [1]. The first edition of the SOLAS Convention, in 1914, established a set of responsibilities related to safety at sea, with an emphasis on safety standards and obligations of shipowners and crews. Later, when many coastal countries already had services for the rescue of castaways, the Convention on the High Seas, in 1958 at Geneva, assigned greater responsibility to coastal countries on safety of ships crossing waters under their jurisdiction, especially near the coastal line, due to the higher propensity to the occurrence of maritime accidents and, not infrequently, given the greater difficulty usually found to give help. Therefore, more maritime safety responsibilities have been assigned to maritime countries, in particular on safeguard human life at sea. Nowadays, more than one hundred countries have ratified the Search and Rescue (SAR) Convention of 1979, in particular, almost all coastal countries, accepted their responsibilities regarding the international maritime community. However, it was the SAR Convention, established since 1985, which structured the international organization for maritime search and rescue. As a result of the convention, the oceans were divided into regions (Search and Rescue Regions—SRR). For the coastal countries, new responsibilities were assigned regarding the coordination of maritime safety actions in their waters (see Fig. 1a). This responsibility entails the existence of emergency service at sea, with immediate readiness (24 h a day), able to respond to all kinds of accidents related to the endanger of human life at sea. The size of the regions was defined according to the capacity of response of coastal countries to distress calls. Portugal is a country on the periphery of Europe with an extensive maritime area, resulting from the continental coastlines and the archipelagos of Madeira and the Azores. The maritime space under sovereignty or jurisdiction of the country includes regions with responsibility for maritime search and rescue (see Fig. 1b), which are crossed by intense maritime, commercial, and recreative traffic (see Fig. 1c). As one of the signatories of the SAR Convention, Portugal has an accident response system at sea to assist ships in distress in the maritime areas of the continent and the North Atlantic. The execution of this mission is carried out by a national maritime search and rescue service, responsible for responding to the emergency from ships or vessels, in the areas of national responsibility. The search and rescue actions are addressed to one of the two centers responsible for coordinating the two Portuguese SRR: (1) the Maritime Rescue Coordination Center (MRCC) Lisbon—supported by the Maritime Rescue Sub-Center (MRSC) Funchal, responsible for SRR Lisbon, which covers the coast of the continent and the archipelago of

Assessing the Location of Search and Rescue Stations …

323

Fig. 1 a SRRs around the world (Source IMO); b Portuguese SRRs superimposed on maritime spaces under Portuguese sovereignty or jurisdiction; and c heat map of maritime traffic density in Portuguese SRRs in the year 2017 (Source Marine Traffic)

Madeira and (2) the MRCC Delgada, which has under its jurisdiction the SRR Santa Maria, in the North Atlantic. Important infrastructures, to support of SAR actions, are life-saving stations (LSS), located along the coast. These means carry out annually numerous actions of maritime search and rescue and relief to castaways, as happened in 2017, the year in which there were recorded 164 accidents with vessels into the two Portuguese SRR. The proper location of these resources of the national maritime search and rescue system is, therefore, an important factor in the effectiveness of SAR missions. The objective of the present work is to assess the adequacy of the configuration of the life-saving stations on the Portuguese coast, given the historical occurrence of maritime accidents in the SRRs under Portuguese responsibility. The article was developed in five sections. In this section, the motivation of study was presented. The second section presents a literature survey on maritime SAR studies. The third section characterizes the profile of the maritime accidents and the maritime rescue stations on the Portuguese coast. The assessment of the location adequacy of the set of LSS, compared to the registry of marine accidents, is carried out in the fourth section, as well as the elicitation of the requirements for a decision support system for maritime SAR activity, to be developed. In the last section, we present the conclusions of the work carried out, as well as the future work.

324

A. Correia et al.

2 Literature Review Plenty of location studies regarding maritime SAR have been carried out to determine or optimize the location of infrastructures or resources of the emergency services. Basdemir conducted a study, in 2004, for the Turkish Air Force [2] with the purpose of distributing new SAR facilities across the Western Mediterranean region and the Aegean Sea. With this purpose, he used the Maximal Coverage Location Problem (MCLP) model to define the locations for the required number of SAR stations, in order to obtain full coverage of the accident zones considering the operational capabilities of SAR units, in this case, helicopters. In another work, Afshartous et al., studied the best location for United States Coast Guard air stations (USCG), in Miami, to respond more adequately to emergency calls [1]. A simulation/optimization methodology was used which, based on real emergency call data, simulated the variation of the location and occurrence of emergency calls, in space and time. Later on, by using the pUncapacitaded Facility Location Problem (p-UFLP), the pMedian Problem (p-MP) and the Uncapacitaded Facility Location Problem (UFLP), they achieved the best location for the stations be able to respond to emergency calls from an uncertain location. However, the model was a simplification that considered only few criteria in the analysis. In the 1990s, a study from Brown et al. resulted in the mixed Integer Linear Programming (ILP) model used for plan the assignment of patrol missions to vessels, of United States Coast Guard (USCG), in several districts [3]. The study was intended to meet the needs of the patrol, considering the weekly maintenance schedule of the vessels, while reducing, at the same time, the related costs. Azofra et al. proposed a tool to objectively distribute the resources assigned to the maritime SAR [4]. To achieve this aim, they developed a methodology based on gravitational models. Firstly, they applied the Zonal Distribution Model (ZDM), to determine the areas with the highest concentration of accidents from the registry of marine accidents in the area. Each zone was represented by a super-accident centroid. Later they defined an Individual Distribution Model (IDM), which were able to provide the distribution of individualized resources to the maritime rescue actions, directing those resources to the locations of the accident, taking in account the distance to the super accident and their range of action. Radovilsky and Koermer built another ILP model to optimize the distribution of small USCG vessels through existing stations in the Pacific area [5]. The main objective of this model was to provide the best distribution of the vessels, minimizing the shortage or overload on stations’ capability, and simultaneously reduce the operational costs. Later, Wagner and Radovilsky improved the previous work by developing the ILP model called the Boat Allocation Tool (BAT) model, considering as the main variable the total of annual hours available for each vessel, i.e., the hours available until each vessel needed maintenance, instead of the number of vessels only [6]. They also introduced risk management, considering a possible shortage of vessels, as well as, the uncertainty in the future needs of the stations. The model also

Assessing the Location of Search and Rescue Stations …

325

considered some logistic restrictions of the USCG, such as the number and type of vessels available per station. Another interesting research, developed by Pelot el al., was a location study for the Canadian Coast Guard Search and Rescue vessels in the Atlantic region in order to ensure maximum coverage of the accidents possible to occur in the area [7]. Based on the Maximal Coverage Location Problem (MCLP), they proposed three extensions: (1) the MCLP model, which assigned weights to different classes of maritime accidents in order to compute the best location of the vessels to provide the maximum coverage for the required areas; (2) the MCLP that computes the best location of the vessels in order to allow maximum coverage by equally distributing the workload through the stations; and (3) the MCLP which considers the uncertainty on vessels availability. The objective of this model was to maximize the coverage considering the probability of boats’ absent, as happens when, for example, they are on maintenance period. More recently, some studies with new perspectives of the localization problem regarding the maritime SAR domain have been presented. They considered simultaneously multiple aspects of the problem. In this context, Razi and Karatas [8] developed the Incident Based-Boat Allocation Model (IB-BAM), a multi-objective model to assign vessels to maritime SAR activities in Turkey [8]. In this model, the Analytical Hierarchy Process (AHP) was used to assign different weights to each type of maritime incident in the area of interest. Subsequently, a ZDM was used to determine highest probability zones for occurrence of maritime incidents. Finally, the Multi-Objective Mixed Integer Program (moMIP) was applied to provide low response times to maritime incidents, while operating costs minimization is imposed, as well as the compatibility between workload and available hours for operations. In 2017, Akbari et al. presented a couple of studies focused on the development of structures to carry out analyzes considering multiple criteria in the location and distribution of resources. The studies targeted the Canadian Coast Guard SAR operations in the Atlantic region, aiming the improvement of quality service and operational costs reduction. In the first study, they presented a multi-criteria analysis on the performance of solutions provided by two well-known models, the MCLP and p-Median Problem (p-MP), for the case of maritime SAR location [9]. The optimal solution of the two models was obtained based on five decision criteria relevant for CCG: (1) access time—criterion that evaluates the average time that the nearest ship takes to arrive at the incident local (time spent in transit); (2) primary—criterion that measures the percentage of incidents that must be covered within a given access time by, at least, one vessel; (3) backup coverage—criterion that expresses the percentage of incidents that are within a given coverage region of at least two vessels; (4) Gini index—criterion that is measured by the level of deviation of access time to all accidents; and (5) Maximum access time. To model the problem, they considered the characteristics of vessels types, the existing SAR stations, and, based on the history of maritime incidents in the area, they forecast the areas where future maritime incidents could occur, in order to anticipate future SAR needs. Subsequently, Akbari et al. developed a Goal Programming Multi-Objective model to optimize the location

326

A. Correia et al.

of CCG vessels in order to achieve high coverage and responsiveness of probable future incidents [10]. The objectives were established on a basis of information and needs provided by the CCG. In order to reach the proposed objectives, they constructed the Mixed Linear Problem model (MILP) where three criteria were used: primary, backup coverage, and mean access time. The model considered different classes and types of vessels with different speeds, capacities, potential locations, and also included operational constraints to simulate possible ship unavailability. As part of the model, it was also considered the spatial distribution of the history of maritime incidents to simulate new occurrences of accidents. In this exploratory work, rather than use algorithmic models, geospatial tools are used to compare the areas where maritime accidents have been registered, with the range of the rescue means assign to stations, in order to assess the degree of coverage of current stations’ configuration regarding previous incidents.

3 Maritime Accidents on the Portuguese Coast By exploring data through statistics collecting and georeferencing of maritime accidents, the aim was to understand the phenomenon and get some clues on how to plan and improve SAR response. The analysis of the history of offshore accidents, revealed their frequency and the most vulnerable areas, to be compared with the arrangement of the overall rescue facilities. To understand the type of maritime accidents in Portuguese waters, the annual trend of the accidents (a total of 791) was analyzed over the period from 2011 to 2016. During that period, the year of 2013 was the one that registered the greatest number of accidents (164), while the year 2015 had the least occurrences (111), with a slight decrease in the number of occurrences at the end of the period (see Fig. 2a). A monthly analysis shows that July is the month with the most accidents (see Fig. 2b), during the summer, in which, although the meteo-oceanographic conditions are favorable to maritime activities, there is an increase in the number of recreational boats at sea. The type of vessels that contributed most for the number of accidents is, precisely, the recreational craft (444 in the total of the period, as depicted in Fig. 2c). However, the number of accidents shows a large variation over the period, with a maximum of 103 accidents recorded in 2013, to a minimum of 50 recorded in 2015. On the other hand, the number of accidents (254) of fishing vessels (the second type of vessel with more registered accidents) is relatively constant throughout the period. Regarding the types of accidents recorded, the most frequent was vessels’ breakdown, followed by sinking and collision. By analyzing the locations where the accidents occurred, the region of Lisbon has the highest number, followed by Funchal in Madeira. With an expressive contribution, there are also the Viana do Castelo, Setubal, and Aveiro regions where fishing activity is significant. Overall, 84.5% of accidents (668) occurred in continental waters, 8.1% (64) in the Madeira region and 7.4% (59) in the Azores region.

Assessing the Location of Search and Rescue Stations …

(a)

327

(b)

(c)

Fig. 2 Distribution of the number of accidents (2011–2016): a per year, b per month, and c by type of vessel

Fig. 3 Distribution of marine accidents between 2011 and 2016 (#791): a overall; b with fishing boats; c with recreational boats; and d SAR actions between 2015 and 2017

328

A. Correia et al.

Fig. 4 Distribution of maritime rescue stations in the Azores, Madeira, and the Continent in 2017

Figure 3a shows the global picture of accidents in Portuguese waters, allowing to perceive the occurrence of accidents in inland waters. Figure 3b and c depict, respectively, the locations of accidents involving fishing and recreational vessels, where can be noted a close proximity of the coastline. As can be seen in Fig. 3d, there is a notorious geographical coincidence between the locations with mapped accidents and the SAR occurrences between the years 2015 and 2017. Regarding the number of rescue stations, there are 28 LSS. Their geographic distribution is depicted in Fig. 4. The LSS is of four types (types-A to D) with each typology having different kinds of infrastructures, staff number, as well as operational and rescue means. Each LSS of type-A has lifeboats with greater autonomy, as well as garrisons capable of operating them. The LSS of type-D, on the other hand, has small rubber boats and water scooters, for use along the seafront.

4 Assessment of the Maritime Rescue Stations Configuration This section describes the process of assessment of current configuration of lifesaving stations, spread by Portuguese coast, compared to the areas in which there have been a greater number of maritime accidents in Portuguese waters. The criteria followed by the maritime rescue authorities to distribute the life-saving stations through the Portuguese coast have been the following: (a) the number and type of previous occurrences of maritime accidents; (b) neighborhood areas of major fishing communities; (c) neighborhood areas of major commercial ports; (d) quick access to areas of greater maritime traffic density; (e) type of the maritime activity developed in the area; and (f) distance between life-saving stations and the range of the lifeboats in each LSS. After the data gathering process, the information was stored in a geographic database (PostgreSQL with PostGIS extension) [11], after that the maritime accidents

Assessing the Location of Search and Rescue Stations …

329

Fig. 5 a Heat map of major density areas of maritime accidents, compared with the location of life-saving stations b Maximum coverage of the lifeboats in each LSS, compared with locations of maritime accidents

as well as the lifeboats coverage were depicted in maps (see Fig. 5), using a geographic information system (QGIS) [12]. The produced maps were built to help perform the spatial analysis in order to assess the current locations and the coverage of the rescue means assigned to the life-saving stations. The analysis of the heat map of Fig. 5a allows to conclude that most of the zones with greater density of accidents have a life-saving station assigned. The exception is the extreme north of the continent (Caminha), where it would be pertinent the implantation of a life-saving station, given the difficult existing traffic in the waters of the Minho’s river mouth. The remaining life-saving stations are located at enough proximity of areas of higher accident density to access them in emergency situations. In Fig. 5b, one can observe the depiction of the sea area covered by the rescue means currently available in the life-saving stations. The depicted buffer from each one of the life-saving station show the limits of intervention (i.e., autonomy) of the means assigned for each LSS, which are enough to ensure the coverage of the distances for the zones of greater density of accidents. This work also sought for the requirements elicitation for future development of a decision support system for maritime rescue and management of the life-saving stations. The overall objectives, considered relevant for the system were: (a) maximization of the quality of service (measured in response time) of rescue means; (b) minimization of operating costs; (c) minimization of the transit time/distance traveled by the rescue means; and (d) minimization of the number of life-saving stations, without compromise the fulfillment of the search and rescue mission. In addition to the criteria currently used for stations’ location assignment, referred at the beginning of this section, it was considered of paramount importance that the decision support system should always prioritize the rescue of people when allocating the means for rescue. Furthermore, were also considered as requirements for the system, the following: (a) classification of historical data on maritime accidents by areas, envisioning future occurrence of accidents, considering assignment of weights based on, inter alia, the type of vessels, the location, and severity of previous acci-

330

A. Correia et al.

dents; (b) counseling on the assignment of existing marine rescue means, given their maximum speed, maximum range with maximum load, as well as operational limits based on the sea turbulence at the rescue area; (c) minimization of the average time to respond to an accident (considering the sum of the time elapsed from the distress call for the life-saving station, the time spent in transit and the time needed for on-the-spot searches); (d) incorporation of meteo-oceanographic conditions at the location and time of the accident and subsequent evolution in the search and rescue zone; (e) real-time monitoring of risky areas for the occurrence of accidents due to the existence of fishing communities and greater density of maritime traffic; (f) consideration of constraints on the availability of life-saving stations means due, for instance, to vacation period of staff, maintenance of the lifeboats or their occupancy in another rescue mission; and (g) incorporation of seasonal factors in the analysis.

5 Conclusion The readiness of the maritime search and rescue system, on the Portuguese coast, is based on a set of life-saving stations of different types located and equipped with rescue means with different capabilities. Using geospatial information, it was assessed the degree of adequacy of the rescue stations locations regarding the actual needs, given the history of accidents in recent years in Portuguese waters. From the work carried out, it was concluded that the current arrangement of life-saving stations is generally adjusted for the location of previous registered maritime accidents. There is, however, one adjustment that must be considered, which involves the setup of an LSS at the north of the continent, where there is a significant fishing and recreational local community and also a significant number of accidents the sea. The work also led to the elicitation of requirements for the development of a decision support system to advise on how to use existing rescue means by applying predictive techniques to the historical data collected, taking into account the technical characteristics of the rescue means, the typology of accidents, as well as, meteooceanographic conditions in the risk zones. Acknowledgements The work was funded by the Portuguese Ministry of Defense and by the Portuguese Navy/CINAV.

References 1. Afshartous, D., Guan, Y., Mehrotra, A.: US Coast Guard air station location with respect to distress calls: a spatial statistics and optimization based methodology. Eur. J. Oper. Res. 196(3), 1086–1096 (2009) 2. Basdemir, M.: Locating search and rescue stations in the Aegean and Western Mediterranean Regions of Turkey. J. Aeronauti. Space Technol. 1(3), 63–76 (2004)

Assessing the Location of Search and Rescue Stations …

331

3. Brown, G.G., Dell, R.F., Farmer, R.A.: Scheduling coast guard district cutters. Interfaces 26(2), 59–72 (1996). https://doi.org/10.1287/inte.26.2.59 4. Azofra, M., Pérez-Labajos, C.A., Blanco, B., Achútegui, J.: Optimum placement of sea rescue resources. Saf. Sci. 45(9), 941–951 (2007) 5. Radovilsky, Z., Koermer, T.: Allocation of US Coast Guard boats utilizing integer programming. J. Academy Business Econ. 7(2), 130–135 (2007) 6. Wagner, M.R., Radovilsky, Z.: Optimizing boat resources at the US Coast Guard: deterministic and stochastic models. Oper. Res. 60(5), 1035–1049 (2012) 7. Pelot, R., Akbari, A., Li, L.: Vessel location modeling for maritime search and rescue. In: Eiselt, H.A. Marianov, V. (eds). Applications of Location Analysis (2015) 8. Razi, N., Karatas, M.: A multi-objective model for locating search and rescue boats. Eur. J. Oper. Res. 254(1), 279–293 (2016) 9. Akbari, A., Eiselt, H.A., Pelot, R.: A maritime search and rescue location analysis considering multiple criteria, with simulated demand. INFOR: Info. Syst. Oper. Res. 56(1), 92–114 (2017). https://doi.org/10.1080/03155986.2017.1334322 10. Akbari, A., Pelot, R., Eiselt, H.A.: A modular capacitated multi-objective model for locating maritime search and rescue vessels. Ann. Oper. Res. 1–26 (2017) 11. PostgreSQL.: https://www.postgresql.org/. Accessed on Dec 2018 12. QGIS.: https://www.qgis.org/. Accessed on Dec 2018

Operational Scenarios Identification and Prioritization in Systems of Systems: A Method and an Experience Report in the Defense Domain Carlos Eduardo de Barros Paes

and Valdemar Vicente Graciano Neto

Abstract Military systems play a crucial role in the defense of a country. They are often interconnected to monitor the frontiers, increase the national sovereignty, and interoperate to offer a unified control of the entire territory. These alliances of systems are currently known as systems of systems (SoS). During the last years, software has been increasingly embedded in those systems to increase their precision. Besides, multiple cyber-physical systems have also been associated to interoperate in the context of SoS, bringing novel challenges related to the conception of those systems, in particular (i) how to elicit systems’ requirements and make them coexist in the same large-scale system, (ii) how to coexist competing requirements, (iii) how to support the distributed development of constituent systems and interoperability links, (iv) how to capture their conceptual operation in a unified document to exploit their functionalities in a fruitful manner, and (v) how to make the SoS cohesive, exploiting their functionalities and creating emergent behaviors to foster the national security. Given those gaps and challenges, this paper reports an experience of a real project carried out in a real Department of Defense, particularly in the navy’s context, besides externalizing a method called OSW to support the ConOps documents in SoS engineering projects. We report the conception of a SoS for defense and maritime surveillance, focusing on the concept of operations (ConOps) of that systems, the challenges faced, and the method established. We claim that this report contributes to the engineering of defense SoS since it provides a panorama of SoS conception, besides presenting solutions that can be replicated in other projects to be carried out worldwide, contributing thus for the advance of SoS research in that domain. Keywords Systems-of-systems engineering · Experience report · National defense and security

C. E. de Barros Paes (B) Pontifical Catholic University of São Paulo, São Paulo, Brazil e-mail: [email protected] V. V. G. Neto Federal University of Goiás, Goiânia, Brazil e-mail: [email protected] © Springer Nature Singapore Pte Ltd. 2020 Á. Rocha and R. P. Pereira (eds.), Developments and Advances in Defense and Security, Smart Innovation, Systems and Technologies 152, https://doi.org/10.1007/978-981-13-9155-2_27

333

334

C. E. de Barros Paes and V. V. G. Neto

1 Introduction Defense systems of systems (D-SoS)1 are a cutting-edge technology. They comprise a set of multiple military constituent systems connected through interoperability links to (i) unify functionalities offered by those systems, (ii) increase the robustness and reliability of the entire military technology set, and (iii) enable the integration of navy, army, and aeronautics forces, obtaining, consequently, a reinforcement of national security and sovereignty. Crucial emergent behaviors have been achieved by them, such as the defense of a country itself. In this sense, SoS engineering (SoSE) has become strategic, guiding how to develop constituent systems and offering insights on how to interoperate them to accomplish complex missions. Despite the existence of D-SoS since the 1960s, many challenges still remain. DSoS have become software-intensive, i.e., systems in which software development is a dominant element that crosscuts the entire system life cycle [1]. Therefore, they have imposed important challenges. Specifically regarding to requirements engineering issues, we can mention the need for (i) dealing with distributed teams of different defense branches (army, navy, and aeronautics) with potentially competing requirements, (ii) establishing and overcoming communication issues due to the distributed nature of such development, (iii) capturing those potentially conflicting and competing requirements in a unified document, and (iv) maintaining them to achieve reasonable quality in the final obtained D-SoS. In this direction, the main goal and contribution of this paper are (i) to present a novel method that was used in a real project to support the elaboration of a concept of operations (ConOps) document [2, 3] and (ii) to share an experience report that captures expertise knowledge that could be shared and reused to aid the conduction of other similar SoSE development projects. Such contributions are of prominent importance, since this knowledge is often restricted to confidential documents that are not broadly available. In this paper, we analyze a real Navy Management SoS project composed of a set of interoperable systems already in use to collect, share, analyze, display operational information, and provide decision support regarding the national security of a given country [4, 5, 6]. Many of these systems are by the Brazilian Navy and others will be developed and integrated during the systems development. The remainder of this paper is organized as follows. Section 2 presents a background. Section 3 presents an overview about the SisGAAz (Blue Amazon Management System). Section 4 outlines the method used for elaboration of ConOps document in the SoS projects. Section 5 presents a case study of operational scenarios elaboration in the SisGAAz. Section 6 discusses the analysis of the results obtained during the identification and prioritization of the operating scenarios. Finally, Sect. 7 presents final remarks.

1 For sake of simplicity, D-SoS and SoS will be herein used interchangeably to express both singular

and plural forms of SoS and to denote military and non-military SoS.

Operational Scenarios Identification and Prioritization …

335

2 Background SoS result from multiple operationally and managerially independent systems (called constituents) that work together to reach common goals. Each constituent performs its individual mission contributing to the success of the global missions [7]. Each different architectural arrangement that results from different combinations of constituent systems is named a coalition [8]. SoS have been distinguished from single systems by a set of characteristics, as follows [9]: (i) operational independence of the constituent systems, since constituent systems share their resources to contribute to the accomplishment of goals at SoS level, but also hold their local activities being performed; (ii) managerial independence of the constituent systems, which are managed by independent organizations and stakeholders (examples include army, navy, and aeronautics managing constituents of a same D-SoS); (iii) evolutionary development of the SoS, since the SoS evolve as a consequence of the individual evolution of its constituents, their functions, and purposes; (iv) distribution, since constituents interoperability is necessarily supported by a communication technology; and (v) emergent behavior, which comprises novel functionalities obtained from the cooperation among constituent systems that could not be offered by a single constituent, and that raises as a result of a coordinated interoperability among them. In particular, missions are also an important concept, especially in D-SoS domain. SoS are often developed under the umbrella of systems engineering (SE), an interdisciplinary approach to enable the realization of successful system under a broad meaning of the word [3]. Several SE processes have been proposed and adapted by different international standard organizations for construction of SoS, including (i) those evolved from the DoD [10], proposed in the 1970s by the US Department of Defense, (ii) other standard systems engineering processes proposed by industry in the 1980s, and (iii) those based on the ISO/IEC 15288 standard, which was adopted by International Council on Systems Engineering (INCOSE) and the IEEE to form the basis of the system-engineering handbook [11, 12]. Most of these standards and systems engineering process are widely used in the technology industry, aerospace and defense departments of several countries. One of the important artifacts developed during the early phases of the systems engineering process is SoS operational concept (ConOps). The ConOps document describes how functionalities will be employed in an operational coalition to achieve the missions established for SoS [13]. It is developed in parallel with the capability objectives. Capability objectives are a statement of top-level objectives for the SoS. They describe the capabilities needed by the user, ideally based on some definitive or authoritative materials. The capability objectives provide a basis for translating operational needs into high-level requirements, assessing performance to capability objectives, and developing an architecture and solution options [13]. As the capability objectives evolve, the ConOps should evolve in detail, as well. ConOps is used to define the SoS requirements space to identify aspects of systems that can affect the SoS architecture and to select performance metrics and test environments [14]. However, SE guides do not recommend which techniques should be used or adopted

336

C. E. de Barros Paes and V. V. G. Neto

to support the ConOps document in the context of SoS projects. Since (i) the scope of SoS can cover the entire territory of a country, (ii) the teams are distributed, and (iii) involve multiple actors, it is necessary to use a technique to identify the operational scenarios of each set of stakeholders of each constituent involved or impacted by the SoS. Section 4 details the technique proposed for SoS context. Next section brings an overview of the SoS engineering project used as input to conceive and apply the method to support the elaboration of the ConOps document for a real military SoS.

3 SisGAAz Overview The Brazilian Navy has developed a system called SisGAAz to meet the guidelines of the National Defense Strategy for managing an important maritime area that surrounds the territory and corresponds to the international areas of responsibility for the search and rescue operations (SAR—search and rescue) and areas of specific interest that go beyond the Brazilian Jurisdictional Waters (BJW) and the SAR area [5, 6]. SisGAAz supports navy’s activities in its constitutional allocations and subsidiaries’ assignments, such as protecting the national frontiers and ensuring the sovereignty. The project involves a diverse team with several professional roles such as telecommunication engineers, systems engineering, software engineers, domain experts, project managers, quality managers, and infrastructure technicians, totalizing around 30 people involved. SisGAAz mission is monitoring and controlling, in an interoperable way, the national maritime area (waters under the Brazil jurisdiction), international areas of responsibility for the search and rescue operations, and areas of specific interest that go beyond the previous ones, in order to contribute to the strategic mobility, represented by the ability to respond promptly to any threat, emergency, aggression, or illegality. SisGAAz’s main purposes are: (i) information sharing and interoperability among the institutions with interest in the sea (national and international, public and private) and (ii) supporting network-centric operations and providing decision support by a shared virtual environment. SisGAAz development was partitioned into two phases: conceptualization and development. The conceptualization phase involved the system specification and design (operational concept and requirements), elaboration of project management plans, and architecture design. This phase was completed in two years and it was developed in the following four tasks: (i) project management planning; (ii) elaboration of concepts of operations (ConOps); (iii) elicitation, analysis, and documentation of system requirements; and (iv) conception and design of SoS architecture. The development of SisGAAz follows an evolutionary approach (iterative and incremental). Initially, a small core of functionalities is already implemented and deployed. Subsequently, the implementation activities are performed in a way that new functionalities are added to constituents according to the local demands of an operational area. For example, one of the navy districts is in State 1 (the operational area) and another one in State 2. State 1 district hosts the first constituents imple-

Operational Scenarios Identification and Prioritization …

337

mented according to the local demands. After that, the implementation will take place in State 2. The new constituents developed there match their specific demands, and after developed, they are joined to the first ones to form the preliminary body of the SoS. Next, this work is iteratively carried out in other regions of the Brazil, creating new constituents, connecting them to the others, realizing the SoS, and expanding the SisGAAz SoS progressively. The proposed SoS has five types of constituents, each one responsible for one of the following assignments: logistics support, cybersecurity, intelligence, command and control (C2), communications, and remote sensing. Logistics constituent aims to provide support to logistics operations regarding navy’s equipment, such as acquisition and transport of material to national oceans supervision. Cybernetics constituent provides capabilities related to SisGAAz safety and security aspects, supporting (i) availability and assurance of confidentiality of information and services, (ii) defense and protection of networks (internal and external), (iii) monitoring the full spectrum of cyber warfare operations, and (iv) safety and protection of services and constituents that form the SisGAAz. The command and control (C2) constituents are military systems that provide features to support planning of command operations and control of constituents in the accomplishment of the mission, playing a type of central authority role. The communications constituents provide the technological infrastructure to integrate the various military organizations and SisGAAz constituents. The intelligence constituent aims to support planning and execution of operations, according to the ConOps. Finally, the remote sensing constituent provides functionality for collecting, processing, and distributing data from a variety of sensors, new or legacy, which make up the SisGAAz and data received through cooperation with other organizations. Each Brazilian geographic area can host a subset of those types of constituents that are linked together to form the national-scale SoS known as SisGAAz. Figure 1 illustrates the scenarios of interoperability among Brazilian Navy’s constituents in the context of SisGAAz.

4 OSW: A Method for Elaboration of ConOps Documents in the SoS Projects Since SE guidelines do not recommend methods to support the ConOps conception for SoS development projects, we have proposed and applied a technique so-named Operational Scenarios Workshops (OSW) to support the elaboration of a ConOps for a SoS project context. Developed in two phases, the first phase was conducted in a three-day workshop exclusively run with Brazilian Navy organizations, focusing on their operational commands and naval districts. The second phase involved workshops with stakeholders that were external to the Brazilian Navy, including agencies and actors directly related to maritime activities. In addition, an approach for the identification of critical scenarios proposed by the Naval War School of the Brazilian Navy was

338

C. E. de Barros Paes and V. V. G. Neto

Fig. 1 SisGAAz operational context and navy constituents

applied to drive the information collection in order to elaborate ConOps [6]. This approach consisted in the identification of threats, critical activities and areas, and a merging of these three sets to define priorities. That was fundamental and helped to design the operational scenarios to which SisGAAz should respond. Based on these operational scenarios, ConOps was specified following the IEEE standard for information technology, the IEEE 1362 [15]. As a result, the Operational Concept Validation Plan was produced, containing instructions to validate the ConOps. The OSW method was structured in the following steps: Step 1. SoS Operational Scenarios Identification and Prioritization: This step consists in the establishment of the interactions between the critical activities of the SoS and the threats considering the areas of action of each constituent. Step 2. Identification of SoS Critical Activities: This step involves the identification of the activities of interest of each constituent and the targets of monitoring and protection fulfilled by the SoS. Step 3. Identification of threats and emergencies: This step comprises the identification of the threats in which critical activities need to be preserved by the SoS. In addition, the main emergencies that may occur and that need to be addressed by SoS are also identified. Step 4. Constituent area identification: In this step, the geographical areas of each SoS constituent are elicited. This information is important for the elaboration of the scenarios, since it allows to identify the critical activities carried out in each area and their respective threats and emergencies that need to be considered. Step 5. SoS critical activities confronting: This activity comprised the confrontation of the identified critical activities with the perceived threats and emer-

Operational Scenarios Identification and Prioritization …

339

gencies. The combination of the elements previously considered elements was used to form the scenarios of interest of each constituent to act in the SoS. Step 6. SoS operational scenario prioritization: In this step, the priorities are defined taking into account each previously identified operational scenario and the meaning in terms of the use of the available capacities for the SoS. A procedure to support this step of prioritizing operational scenarios can be applied. For exemplification purposes, in the next sections, we systematically follow each one of the steps proposed in OSW method, showing how they were applied to conduct a real elaboration of a ConOps for the SisGAAz.

5 Case Study: Elaboration of Operational Scenarios in the SisGAAz Context Operational scenarios were established to optimize the use of monitoring and protection resources available in the Brazilian Navy. They were elaborated by interactions between critical activities and threats in each maritime area of interest. We present how the OSW method steps were used to support the elaboration of the ConOps of the SisGAAz, as follows: Step 1. Operational scenarios identification and prioritization. During the workshop sessions in the SisGAAz, the SE team used an approach to identify the operational scenarios for each naval district (ND) area. The activities proposed by the approach allowed to delimit the operational scenarios of each naval command. As a result, the compatible response was properly delineated in the SisGAAz to meet the operational processes of each SoS operational area. This approach consisted of the following four steps: (i) identification of SoS critical activities; (ii) identification of threats and emergencies; (iii) constituent area identification; (iv) SoS critical activities confronting; and (v) SoS operational scenario prioritization. The combination of the previously considered elements was used to form the scenarios of interest of each ND. Step 2. Identification of SoS critical activities (CA). This step consisted of enumerating the critical activities, i.e., activities of interest in the area of jurisdiction of the 1st ND that need monitoring and protection: (i) exploration and exploitation of the oil basins (CA1); (ii) exploitation of living resources (fishing areas) (CA2); (iii) LCM protection and security (CA3); (iv) preservation of the environment (CA4); (v) safeguarding human life at sea (CA5); (vi) protection of genetic heritage (CA6); (vii) protection of scientific knowledge (CA7); (viii) protection of the submerged assets, including the protection and inspection of the research, removal, demolition, or exploitation of buried goods belonging to third parties or to the

340

C. E. de Barros Paes and V. V. G. Neto

Table 1 Maritime areas of interest in the 1st ND jurisdiction Occurrence

Physical areas

Acronym

Oil basins

Basin of Espírito Santo Basin of Campos Basin of Santos

BES BCA BSA

MCL navigation focal points

Surroundings of Cabo Frio

SCF

Focal points for access to the main ports and maritime terminals

Port of Vitória and Tubarão Port of Santos Port of Rio de Janeiro Terminal of TEBIG and Ilha da Madeira (Ilha Grande Bay) Terminal of Ponta do Ubu (ES) Terminal of São Sebastião (SP)

PVT PSA PRJ TBM TUB TSB

Reserves or incidences of living and non-living resources (excluding oil and natural gas)

1st ND jurisdiction area

1DN

Maritime environmental protection sites

1st ND jurisdiction area

1DN

Brazil, in addition to activities in archaeological sites incorporated to Brazil domain (CA8); (ix) exploration and exploitation of non-living resources (except oil and natural gas) (CA9); and (x) protection of strategic areas of interest (CA10). Step 3. Identification of threats and emergencies. This step consists of enumerating possible threats or possible perceived emergencies. In the 1st ND area were identified the following threats and emergencies: (i) maritime terrorism; (ii) illegal fishing; (iii) transnational illicit acts; (iv) environmental crimes; (v) national illicit acts; and (vi) SAR emergency. Step 4. Constituent area identification. This step consisted of identifying the existence of maritime areas of interest for each SoS constituent, geographically delimited. In the 1st ND area were identified the following maritime areas: (i) oil basins; (ii) strategic interests; (iii) MCL navigation focal points; (iv) focal points for access to the main ports and maritime terminals; (v) reserves or incidences of living and non-living resources (excluding oil and natural gas); (vi) maritime environmental protection sites; (vii) archaeological sites or other areas containing submerged goods; and (viii) objectives or areas defined by the 1st ND commander. Table 1 organizes the 1st ND maritime areas with respective physical area and acronym. Step 5. SoS critical activities confronting. This step consisted of confronting the critical activities identified for the SoS with the envisaged threats and emergencies, indicating the marine areas of interest evaluated as capable of interacting between critical activities and threats and emergencies. Table 2 shows

Operational Scenarios Identification and Prioritization …

341

Table 2 Intersection of physical areas with threats and critical activities Maritime terrorism

CA1

CA2

CA3

CA4

CA5

CA6

CA7

CA8

CA9

BES BCA BSA

X

PCF PVT PSA PRJ TBM TUB TSB

1DN

X

X

X

X

1stND

the confrontation of critical activities with the threat of maritime terrorism for 1st ND. Step 6. SoS operational scenario prioritization. At the end of the aforementioned steps, we consolidated the threats, emergencies, critical activities, and areas of interest for the SoS. After the completion of the last step, the scenarios were prioritized. The need to prioritize the scenarios arises from the fact that there is hardly any naval, air, and aircraft resource in enough quantity to simultaneously and uniformly meet all the operational scenarios. Therefore, prioritization is not only desirable, but also essential if the available resources are to be effectively used, maximizing the system’s prevention and response capabilities. These steps could be used as guidelines to be adopted in other SoS development projects, not only in other military domains (aeronautics and war), but also in non-military domains, such as public governments of smart cities. We then finished that iterations of the SisGAAz and concluded that the use of OSW method enabled us to obtain a more precise specification of operations in the resulting ConOps.

6 Analysis of Results and Discussion The conception of SoS in the defense area is a major technological challenge, not only because of the complexity of the constituents and equipment that will be part of it, but also because of the complexity of the environment in which it will operate. During the systems engineering of the SisGAAz, the identification and prioritization of the operational scenarios were of extreme importance. The task of SoS conception that can deal with a diverse reality and its many constituents would not be possible without first identifying the scenarios in which these constituents would be inserted. All operational scenarios identified during the OSW were also registered in the Operational Scenarios Report (OSR). The final version of OSR has approximately 100 pages and considers the scenarios from nine naval districts. During the design of the SisGAAz, only the operational scenarios of the internal actors of the Brazilian Navy were considered. These scenarios were analyzed during the elaboration of the operational concept and helped to identify the necessary capabilities for the SisGAAz.

342

C. E. de Barros Paes and V. V. G. Neto

The operational scenarios served to establish the situations in which the SisGAAz will act to fulfill the established mission in the operating concept document of the system. The SisGAAz center of gravity resides in areas of concern where critical activities occur that must be protected from threats or where there may be emergencies. During the OSW in the SisGAAz was observed that most of the information obtained becomes repetitive, with only a few important variations, which make up the nuances that distinguish one operating command from another. The critical activities identification evidenced differences between the regions involved, which were adherent with the economic vocations of each of them. However, despite this differentiation, a high degree of convergence was observed regarding the understanding of the concepts and definitions used. Regarding threats, it can be observed that most of them were identified in a common way in all the places visited, demonstrating a convergence of understandings on this subject. In contrast to the information about the activities and threats, which showed a convergence between the collected data, the comparison between the information on the areas of interest obtained from the various operational commands was more dispersed. This heterogeneity between the commands can be attributed to the differences between the nature and intensity of the maritime activities developed in each region. Each operational command, with its regional characteristics, has a more intense group of activities, concentrated in a given area, justifying differences in the data collected. In addition to the scenario definition information, the OSW was also used to obtain additional information needed by the SisGAAz. This information consisted of four groups of data: the stakeholders, the SoS requirements, the publications and legal standards employed, and the SoS needs and characteristics identified by the users. Finally, it is important to point out that the operational scenarios of a SoS represent a specific conjuncture of a defined moment, that is, the operational scenarios raised refer to a given period. If there is a very long time between scenario-gathering and the deployment of SoS, the reality will have undergone changes that, if relevant, could invalidate SoS operational concepts, claiming for a new application of OSW method. Next section brings final remarks.

7 Final Remarks This paper presented OSW, a method specially tailored to support the ConOps documents for SoS projects. Among several challenges brought by SoS to the SE projects, the ConOps is one of them, given the distributed nature of the teams, the conflicting requirements, the multiple constituents, institutions, and stakeholders involved, and the operational and managerial independence of the constituents. OSW method was applied to a real project in the context of the navy of a given country. The contribution of this paper is twofold: One contribution comprises the OSW method itself, which can be replicated in any SoS project developed by distributed teams, certainly adoptable for defense domain, but potentially extensible for application in SE projects for SoS of other domains; the another contribution consists of the experience report

Operational Scenarios Identification and Prioritization …

343

presented according to the steps prescribed by OSW method, which provides a proof of concept for the method (since it has been applied to a real project), and a report that can be valuable for other teams that also intend to adopt the method. Future work includes the application of OSW method in other SoS development projects and its adaptation for reevaluation of SisGAAz. We expect that OSW method can broaden the state of the art and practice in SoS engineering, consolidating the SoS body of knowledge, enriching the collection of available techniques to be systematically applied in the development of SoS, increasing the quality of the delivered product, and ensuring the correct operation of the system, resulting in a guarantee of national security and sovereignty as a result of emergent behaviors obtained from the SoS operation in the defense domain. We also hope these techniques can be evaluated and incorporated as novel practices and approaches to SoS engineering in forthcoming SoS development projects around the world.

References 1. ISO/IEC/IEEE Systems and Software Engineering—Architecture Description: ISO/IEC/IEEE 42010:2011(E) (Revision of ISO/IEC 42010:2007 and IEEE Std 1471-2000), pp. 1–46 (2011) 2. Department of Defense (DoD): Systems Engineering Guide for Systems of Systems, V 1.0. (2008) 3. Walden, D.D., Roedler, G.J., Forsberg, K., Hamelin, R.D., Shortell, T.M.: International Council on Systems Engineering: Systems Engineering Handbook: A Guide for System Life Cycle Processes and Activities. Wiley, Hoboken, NJ 4. Paes, C.E.B., Graciano Neto, V.V., Oquendo, F., Nakagawa, E.Y.: Experience report and challenges for systems-of-systems engineering: a real case in the Brazilian defense domain. In: Proceedings of Workshop on Distributed Development, Software Ecosystems and Systems of Systems, pp. 41–50 (2016) 5. de Barros Paes, C.E., Neto, V.V.G., Moreira, T., Nakagawa, E.Y.: Conceptualization of a systemof-systems in the defense domain: an experience report in the Brazilian Scenario. IEEE Syst. J., 1–10 (2018) 6. de A. Chaves, S.F.: Sistema de Gerenciamento da Amazônia Azul (SisGAAz): o passo inicial para o efetivo controle da área marítima brasileira. Monografia apresentada ao Departamento de Estudos da Escola Superior de Guerra como requisito à obtenção do diploma do Curso de Altos Estudos de Política e Estratégia (CAEPE), Rio de Janeiro (2013) 7. Silva, E., Cavalcante, E., Batista, T., Oquendo, F., Delicato, F.C., Pires, P.F.: On the characterization of missions of systems-of-systems. In: Proceedings of the 2014 European Conference on Software Architecture Workshops—ECSAW’14, pp. 1–8. ACM Press, New York, NY, USA (2007) 8. Oquendo, F.: Architecturally describing the emergent behavior of software-intensive systemof-systems with SosADL. In: 2017 12th System of Systems Engineering Conference (SoSE), pp. 1–6. IEEE (2017) 9. Maier, M.W.: Architecting principles for systems-of-systems. Syst. Engin. 1, 267–284 (1998) 10. Department of Defense (DoD): Software Development and Documentation—Military Standard MIL-STD-498 (1994) 11. Hamelin, R.D., Walden, D.C., Krueger, M.E.: INCOSE Systems Engineering Handbook v3. 2: Improving the Process for SE Practitioners. INCOSE International 11 (2010) 12. ISO:IEC 15288: 2008 Systems and software engineering-System life cycle processes. In: 2nd International Organization for Standardization/International Electrotechnical Commission, 18th March. (2008)

344

C. E. de Barros Paes and V. V. G. Neto

13. Dahmann, J., Rebovich, G., Lane, J.A., Lowry, R.: System engineering artifacts for SoS. In: 2010 IEEE International Systems Conference, pp. 13–17. IEEE (2010) 14. Dahmann, J.S., Baldwin, K.J.: Understanding the current state of US defense systems of systems and the implications for systems engineering. In: 2008 2nd Annual IEEE Systems Conference, pp. 1–7. IEEE (2008) 15. IEEE Computer Society. Software Engineering Standards Committee., Institute of Electrical and Electronics Engineers., IEEE-SA Standards Board.: IEEE guide for information technology– : system definition– concept of operations (ConOps) document. Institute of Electrical and Electronics Engineers (1998)

Science Diplomacy: Strategic Initiative to Create a Buffer Zone in the Caribbean Colombian Marine Protected Area Seaflower Fabián Ramírez-Cabrales

and Sergio Iván Rueda Forero

Abstract It recognized the science diplomacy as a path to address national and common interest related to marine spaces declared as human heritage areas. Science diplomacy is an instrument that allows scientific cooperation between states and international organizations, and stakeholders interested in to obtain scientific evidence to the decision-making process at the political level in all senses. The Specially Protected Areas And Wildlife to the Convention for the Protection and Development of the Marine Environment of the Wider Caribbean Region (SPAW) Protocol is the main instrument to evaluate the possible course of action that allows us to propose the implementation of this initiative. To do so, it is necessary to consider aspects as sovereignty and maritime jurisdiction of state parties and non-state parties. Besides, are considered threats related to maritime transport. It analyzed the geographical conditions and the managerial institutions and stakeholders that participate in the management. Also, the uses of this strategic marine space within the Colombian Caribbean Sea are reviewed. The aim is to address the possibility to create a buffer zone in the boundary of the Marine Protected Area (MPA) Seaflower considering the borders of the Seaflower biosphere reserve. Keywords Science diplomacy · Buffer zone · Caribbean · Colombian · Marine protected area · Seaflower

1 Introduction Science diplomacy (hereafter SD) is an international, interdisciplinary, and inclusive process that allows us to understand and give a better solution, both to global problems and to national strategic interests. It also incorporates the international dimension F. Ramírez-Cabrales (B) · S. I. R. Forero University Escuela Naval Almirante Padilla, Cartagena D.T.y C, Colombia e-mail: [email protected] S. I. R. Forero e-mail: [email protected] © Springer Nature Singapore Pte Ltd. 2020 Á. Rocha and R. P. Pereira (eds.), Developments and Advances in Defense and Security, Smart Innovation, Systems and Technologies 152, https://doi.org/10.1007/978-981-13-9155-2_28

345

346

F. Ramírez-Cabrales and S. I. R. Forero

of science, technology, and innovation policies in the foreign policy of the states. It links the participation of scientific advisors and considers the value of scientific evidence in international relations for informed decision-making by the state. This study is an exploratory investigation originated from the analytical review of the available literature on Marine Protected Areas (MPAs), and a proposal for some governance and spatial management mechanisms in the Colombian Caribbean Sea (2016), particularly in the Marine Protected Area (hereafter MPA) Seaflower. Currently, this proposal is linked to the scientific diplomacy research project for the protection of the marine environment as a contribution to the development strategy for the archipelago of San Andrés and Providencia led by the Republic of Colombia Navy. The purpose is to develop a strategic initiative to create a buffer zone in the boundary of the MPA Seaflower considering the borders of the Seaflower biosphere reserve. The Seaflower MPA is recognized as possessing one of the highest levels of marine biodiversity in the Caribbean region. Actually, over 72% of all-regional Central and South American transshipments move through the southwestern Caribbean area—wherein the Seaflower is located, a high amount of cargo, being containers and petroleum products, the principal commodities moved through the area. These factors are threatening the surrounding ecosystems by ship-source pollution. In consequence, the Seaflower MPA requires a buffer zone to protect one of the highest levels of marine biodiversity in the Caribbean region. This study proposes as a research question, how science diplomacy can contribute to create a buffer zone in the Caribbean Colombian MPA Seaflower? To answer this question, the article is structured into five parts. The first part gives a brief introduction; the second part defines the science diplomacy and identifies its dimensions; the third part describes international and regional scenarios that promote the sustainable development and care of the MPAs within the framework of the United Nations Sustainable Development Programme (UNEP) and the regional programs implemented in the Wider Caribbean. As a fourth part, we describe the geographical aspects of the MPA Seaflower highlighting the importance that this MPA represents the Wider Caribbean Region and the threats that loom over it. The fifth part describes the three dimensions of the SC and its possible contribution to the strategic initiative to create a buffer zone in the Caribbean Colombian Marine Protected Area Seaflower.

2 Science Diplomacy 2.1 What Is Science Diplomacy? According to Langenhove [1], diplomacy and science represent, in principle, two fields of human activity that have little in common. However, scientific networks and their constant activity play an increasingly determining role in diplomatic efforts

Science Diplomacy: Strategic Initiative to Create a Buffer …

347

initiated by state and non-state actors. The result of this conjugation of efforts results in a practice that can be called “science diplomacy” (SD) [1]. The SD is a fundamental tool for international relations; it promotes science, technology, and innovative knowledge, which includes the deployment of scientific and career diplomatic personnel, and the insertion of these topics in the development of the foreign policy and the new global agenda. The SC demands new contributions and primary expertise in international relations associated with science and scientific and technological development. This discipline is a response to the new global challenges imposed by the effects of climate change, energy efficiency, biodiversity loss, food security, the impacts of pollution, and the conservation of water resources, among others [2]. Considering this scenario, the exercise of the foreign policy requires tools, techniques, and tactics adaptable to a world that is increasingly complex in technical and scientific terms. This has created a new role for science in the field of policy formulation and diplomacy. But nevertheless, how science can contribute to foreign policy objectives? This question had a wide debate during a two-day meeting on “New Frontiers in Science Diplomacy,” which was hosted by the Royal Society from June 1 to 2, 2009, in partnership with the American Association for the Advancement of Science (AAAS). The result identified three dimensions in which science diplomacy can contribute to policy objectives as well [3]: • Science in diplomacy: Science providing advice to inform and support foreign policy objectives. • Diplomacy for science: Diplomacy facilitating international scientific cooperation. • Science for diplomacy: Scientific cooperation improving international relations. These three dimensions can contribute to the strategic initiative to create a buffer zone for the MPA Seaflower. The structure and programs to implement these actions are based on the following regional structure.

3 United Nations Environment Programme (UNEP) At the regional level, the UN system recognizes the Caribbean Sea as one of the most unique and complex areas of the world. Its extension is over 2.5 million km2 [4]. It is geographically and politically diverse as it embraces 44 states and territories with over 100 maritime boundaries [5–7]. For that reason, to manage this vast area, several initiatives have been developed. There are only two biggest environmental protection programs concerning the entire area that today will be discussed. The first one is the Caribbean Environment Programme (CEP), and the other is Caribbean Large Marine Environment (CLME) project. For the purposes of this initiative, the following section will focus on the CEP.

348

F. Ramírez-Cabrales and S. I. R. Forero

3.1 United Nations Environment Programme-Caribbean Environment Programme—(UNEP-CEP) The CEP was the second UNEP established in 1982 with all its institutional components (convention, action plan, funds, and coordination unit), building one of the most valuable programs of the UN system. In addition to this, the CEP is recognized by the countries in the region for its high commitment to the development of environmental management initiatives [8]. The UNEP’s main goal is to reduce and tackle the degradation of environments by marine pollution [8]. In that sense, the Caribbean is considered the second most complex marine route worldwide. This is a consequence of the traffic through the Panama Canal, and the increase of seaborne trade in many nearby ports in the area due to its economic attractiveness. Ship-source pollution is the factor that is threatening the surrounding ecosystems, including the issue of invasive species [9]. Likewise, tourism and fisheries are the main activities in the zone and are also sources of marine pollution. Furthermore, the Caribbean faces several political conflicts and presents diverse economies in terms of development, with significant influence of developed countries exerting colonial powers and hosting many different languages [8–10]. To address all the above issues and challenges, in 1982, countries unanimously decided to work for the adoption of a cooperative and integrated approach through a convention for the protection and sustainable development of the Caribbean region [6].

3.2 Convention for the Protection and Development of the Marine Environment in the WCR and Its Protocols The Wider Caribbean Action Plan was the first step taken by CEP and countries to start developing a binding agreement. As a result and through the commitment of the countries involved, it was possible to move forward to the adoption of the convention for the Wider Caribbean, which was agreed in Cartagena de Indias, Colombia, in 1983 (hereafter the Cartagena Convention). This regional treaty entered into force in 1986. To date, 25 states have ratified it [6, 11]. The Cartagena Convention is the current binding regional environmental agreement serving as an essential reference for both managers and policy-makers when protection measures and management development of coastal and marine resources are in place, either individually or jointly. Nevertheless, due to developing economies surrounding the area, many issues are facing the Caribbean region. For that reason, one of the main targets is the establishment of MPAs as a mechanism to ensure the conservation of biodiversity, pollution reduction, education, and capacity building. Following the adoption of the convention in 1983, two legal instruments were subsequently adopted. The first one was the Protocol Concerning Cooperation in

Science Diplomacy: Strategic Initiative to Create a Buffer …

349

Combating Oil Spills in the Wider Caribbean Region (Oil Spills Protocol). This protocol and the convention itself were convened simultaneously. The second is the Protocol Concerning Specially Protected Areas and Wildlife (SPAW Protocol), which was adopted in the early 1990s and entered into force in 2000. The latter was the third program driven by UNEP with the aim of preserving and improving the marine ecosystem in the region [5, 6, 11]. Currently, the SPAW Protocol has been ratified by only 16 states, including Colombia. The aim is “to take necessary measures to protect, preserve and manage, in a sustainable way, zones that require protection to safeguard their particular value and that threaten or endanger species of flora and fauna” (Art. 3) [6, pág. 40]. To date, nine coastal states in the Wider Caribbean encompass over 32 MPAs under the SPAW Protocol and the CEP. The marine area covered is approximately 100,000 km2 corresponding to only 4% of the all-regional sea (CAR-SPAW-RAC 2018). Although the number of MPAs is relatively small, in 2014, the Caribbean Regional Programme (CRP), under the umbrella of the SPAW Protocol, established the Caribbean Challenge Project [5]. The objective is to enlarge regional MPAs up to 20% by 2020 through a sustainable governance mechanism based on cooperation between countries for the protection of the ecosystem in the larger Caribbean [7]. Colombia is the depository country for the convention and one of the first to ratify it. Therefore, it is making efforts to fulfill the agreement. Currently, around nine percent of its maritime jurisdiction is protected through MPAs [12]; one of 30 the largest in the region is the Seaflower MPA [5, 13].

4 The Seaflower MPA 4.1 Background Colombia’s constitutional mandate considers the preservation and conservation of its unique ecosystems and areas of ecological importance, and is respectful of the obligations prescribed in the international environmental treaties to which it is party [14]. The government, in 1998, proposed the designation of the archipelago of San Andres and Providencia and Santa Catalina, and the surrounded marine area, as a biosphere reserve (Fig. 2) to be known as the Seaflower Biosphere Reserve [15]. Later in 2000, UNESCO included it within the MAB World Biosphere Network [13] (Fig. 1). Moreover, in January 2005, the Ministry of Environment and Sustainable Development (MADS, Spanish acronym) through Resolution 107 declared an MPA, within the biosphere reserve [15], the largest in the Caribbean Sea, and the first of its category in Colombia [13, 16, 17, 18].

350

F. Ramírez-Cabrales and S. I. R. Forero

Fig. 1 Seaflower biosphere reserve

4.2 Geographical Position The Seaflower MPA is located in the southwestern Caribbean region, surrounding the San Andres Archipelago (i.e., three small inhabited oceanic islands and eight unsettled cays and atolls) (Fig. 2). The largest island, San Andres, is 800 km (480 NM) northwest of the Colombian continental territory [17, 19]. The MPA covers a maritime area of 65,000 km2 , with only 1% of the terrestrial surface (650 km2 ) [17]. The population of over 100,000 base their livelihood on artisanal fisheries and tourism. The MPA is part of the Caribbean coral reef hotspot, which is among the richest areas in marine species diversity, but also one of the most threatened [13]. For instance, the area is considered to be of regional and national ecological significance. The Seaflower MPA is recognized as possessing one of the highest levels of marine biodiversity in the Caribbean region [19]. It representatively defines Colombia’s six tropical marine ecosystems: coral reefs, seagrass beds, mangroves, rocky littorals, sandy beaches, and soft bottoms, which influence the high productivity of the area [13]. The coral reef is the key ecosystem for conservation in the MPA because as one of the largest in the Caribbean, it represents about 14% of the world’s coral reefs [20], the second in the western hemisphere, and most productive in the region [17]. The reef extension covers approximately 2000 km2 [18] and is composed of two barrier reefs, five atolls, reef lagoons, and coral banks [13], which is 78% of the national total [16].

Science Diplomacy: Strategic Initiative to Create a Buffer …

351

Fig. 2 Seaflower marine protected area

Colombia set, through the National Law 99/93 [21], overall targets for the MPAs in the territory, based on the global IUCN objectives [11, 22, 23]. Moreover, the Seaflower MPA has defined its objectives with the participation of islander communities and stakeholders based on the concept of community-based management ([13]; De Pourcq et al. 2015) to ensure effective governance and protection of the ecosystems in the zone. Moreover, to succeed in achieving the activities within the MPA, five core objectives were defined [24]: 1. Preservation, recovery, and long-term maintenance of species, biodiversity, ecosystems, and other natural values including special habitats; 2. Promotion of sound management practices to ensure long-term sustainable use of coastal and marine resources; 3. Equitable distribution of economic and social benefits to enhance local development; 4. Protection of rights concerning historical use; and 5. Education to promote stewardship and community involvement in planning, and management.

Based on these, and using the EBM approach [7, 25], both the community and authorities designed the zones within the MPA to ensure the protection of ecologically relevant areas. The zoning criteria took into consideration the ecological criteria (Roberts et al. 2003), seeking an easy demarcation based on representativeness and connectivity of key habitats (i.e., coral reefs, seagrass beds, algal beds, and mangroves).

352

4.2.1

F. Ramírez-Cabrales and S. I. R. Forero

MPA Zoning

The MPA is divided into three administrative sections, established through Accords 021 and 025 [24], seeking a better implementation of the biosphere reserve. Therefore, the areas are: northern (37,522 km2 ), central (12,716 km2 ), and southern (14,780 km2 ) (Figs. 3 and 4). Moreover, within the administrative divisions, there are five zone types for in situ conservation and sustainable use [13, 18]: 1. No-entry (116 km2 ), preservation zones restricted only for research and monitoring activities; 2. No-take (2,214 km2 ), conservation zones that incorporate and allowing non extractive uses; 3. Recovering and sustainable use of marine resources (2,015 km2 ), allows traditional artisanal fishing activities, and artisanal sport fishing and all the activities allowed in the special use, no-entry and no-take zones; 4. Special use (68 km2 ), shipping related, leisure, and waters sports; 5. General use (60,587 km2 ), minimal restrictions apply, seeking and maintaining the MPA objectives to promote marine conservation.

4.2.2

Threats to the MPA in the Region

Vallega [26], UNEP [7, 20], Lopez and Krauss [27], Fanning et al. [25], Biggs [28], and Morris [29] have described the threats and issues concerning the marine ecosystems in the Caribbean region, and thus, they are not different for this MPA. Moreover, today, new challenges concerning shipping activities have emerged as

Fig. 3 Seaflower marine protected area, administrative division

Science Diplomacy: Strategic Initiative to Create a Buffer …

353

Fig. 4 Caribbean shipping density

a consequence of ongoing and short-term developments in the region. About 90% of global trade is moved by seagoing vessels [30], and keeping pace with global economic development, shipping has grown by 30% over a period of four to five years [31], increasing routes and connectivity. For instance, according to UNCTAD [32], Latin America and the Caribbean have the highest liner shipping connectivity with Panama. At least, 21 vessel operators connect with the USA directly through the Panama Canal, which is the crossroad between east–west and north–south routes, followed by Mexico, Jamaica, and Colombia. Thus, the Panama Canal expansion allows regional expansion of ports, and therefore, larger vessels transit through the Caribbean (Fig. 4). The opening, on June 26, 2016, of the expanded new Panama Canal [33] has enabled the transit of approximately 4750 additional ships per year, handling over five percent of global goods and about eight percent of all transshipments worldwide [34]. Furthermore, over 72% of all-regional Central and South American transshipments move through the southwestern Caribbean area—wherein the Seaflower is located, a high amount of cargo, being containers and petroleum products, the principal commodities moved through the area [31]. Approximately, 60 voyages per week (14,000 annually) to over 100 maritime routes pass through the Caribbean Sea [35, 36]. Thus, the risk of an incident occurring during transit through the routes in the Caribbean is much greater. Further, it will cause an increase of ship-source pollution such as CO2 and greenhouse gas (GHG) emissions [32], which contribute to ocean

354

F. Ramírez-Cabrales and S. I. R. Forero

acidification [37, 38], affecting the coral reefs in the Wider Caribbean Area, and those within the Seaflower MPA.

5 Strategic Initiative 5.1 Science for Diplomacy The SPAW Protocol establishes in its Article 8 that state parties may, as necessary, “strengthen the protection of a protected area by establishing, within areas in which it exercises sovereignty, or sovereign rights or jurisdiction, one or more buffer zones.” For this, it sets the following guidelines: 1. If a Party intends to establish a protected area or a buffer zone contiguous to the frontier or to the limits of the zone of national jurisdiction of another Party, the two Parties shall consult each other with a view to reaching agreement on the measures to be taken and shall, inter alia, examine the possibility of the establishment by the other Party of a corresponding contiguous protected area or buffer zone or the adoption by it of any other appropriate measures including co-operative management programmes. 2. If a Party intends to establish a protected area or a buffer zone contiguous to the frontier or to the limits of the zone of national jurisdiction of a State that is not a Party to this Protocol, the Party shall endeavour to work together with the competent authorities of that State with a view to holding the consultations referred to in paragraph 1.

In this regard, countries such as Costa Rica, Jamaica, Haiti, Honduras, and Nicaragua share the border with the Republic of Colombia, but none of them is a state party to the SPAW Protocol. Besides, in late 2001, Nicaragua instituted proceedings against Colombia at the International Court of Justice (ICJ), claiming rights to the islands and maritime waters [39]. Also, Nicaragua requested to the Seabed Authority an extension of their continental shelf with the pretension of granted in concession the area claimed for exploration and exploitation of oil and gas, jeopardizing the marine biodiversity, in particular, the coral reef areas. The ICJ ruled, based on customary law, that Colombia has the sovereignty over the islands in question. However, the ICJ decision is a disaster for the pristine marine ecosystems, dividing the integrity of the Seaflower MPA and the biosphere reserve declared by UNESCO in early 2000 [13] and backed by many international environmental organizations such as UNEP and GEF [40]. In consequence, political constraints and economic interest are affecting and jeopardizing the protection of the marine ecosystems within the Seaflower Biosphere Reserve and thus the MPA. In that sense, to establish a buffer zone contiguous to the frontier of the zone of national jurisdiction of a state that is not a party, it will be necessary to apply science for diplomacy. This dimension can be driven by the scientific community. The scientific community focuses on common problems beyond national borders. Likewise, this facet is flexible and promotes the emergence of new forms of diplomacy that require non-traditional alliances between states, sectors, and non-governmental

Science Diplomacy: Strategic Initiative to Create a Buffer …

355

organizations. “These scientific exchange channels can contribute to the formation of coalitions and the resolution of conflicts as long as they are aligned with the broader objectives of foreign policy” [3]. Likewise, these actions can be beneficial to build a better relationship with Costa Rica, Jamaica, Haiti, Honduras, and Panama, the latter state party of the SPAW Protocol and pioneer in the practice of SD in Latin America. These actions could contribute to the protection of a common interest such as the Biosphere Reserve Seaflower.

5.2 Diplomacy for Science The SPAW Protocol establishes a Scientific and Technical Advisory Committee (STAC) to assist with the implementation of the protocol. The STAC is a Consultative Body, Technical Adviser to the Meeting of the Parties, subordinate to the Contracting Parties, which acts by express mandate of the Meeting of the Parties, and whose function will be to advise the Meeting of the Parties on matters of a technical nature. Within its structure, it includes four committees: a. Subcommittee on Protected Areas; b. Subcommittee on Species Protection; c. Education and Information for the Public; and d. Legal and Regulatory Matters. The STAC promotes national projects for conservation and identifies ways to help national governments to incorporate biodiversity conservation into their resource management activities. It also encourages the coordination of regional activities related to protected areas. Particularly, the role of the Secretariat for the development of projects for the regional implementation as support for the objectives of the protocol is highlighted. The STAC can develop cooperation with other international, regional, and national organizations, under the direction of the parties, in order to generate experiences and knowledge to promote partnerships for cooperation and collaboration, as appropriate. In that sense, “multi-country and regional mechanisms can be effective for building relationships between the scientific and the policy communities, as well as between countries” [41]. Under this framework, the STAC is considered a strategic regional actor to implement the strategic initiative for the creation of a buffer zone through the diplomacy for science.

5.2.1

Science in Diplomacy

Currently, the Republic of Colombia has been planning and executing scientific research cruises in the MPA Seaflower for almost four decades since before its declaration as a Biosphere Reserve (BR). These actions are part of the integral strategy for the exercise of sovereignty in the archipelago of San Andrés and Providencia and Santa Catalina, led from the presidency of the Republic of Colombia to strengthen the management and conservation of the Seaflower Biosphere Reserve through knowledge management and scientific evidence (contributions to Seaflower knowledge).

356

F. Ramírez-Cabrales and S. I. R. Forero

During the last four years, Colombia has made four scientific expeditions in the BR Seaflower: Cayos de Roncador Island 2015; Cayos de Serrana Island 2016; Cayos de Serranilla Island 2017; and Cayo Alburquerque Island. The objective is to specify a 2015–2023 monitoring plan and standardize sampling protocols. In addition, the objective of scientific expeditions to the BR Seaflower is to demonstrate on a scientific basis the importance of the Seaflower Biosphere Reserve for Colombia and as a common heritage of humanity. This national initiative has had an impact on the epistemic community, however, and taking into account [42] it is necessary that the foreign policy of the Colombian state “…should place greater emphasis on science within their strategies, and draw more extensively on scientific advice in the formation and delivery of policy objectives.” Particularly issues related with the RB Sea-flower.

6 Conclusions The MPA Seaflower is facing several threats mostly due to shipping activities as well as political conflicts. The zoning of this large area is not enough to protect these pristine ecosystems located there. Due to its geographical location in the southwestern Caribbean region, surrounding the San Andres Archipelago and within the borders of the BR Seaflower, the outer limit of this MPA links maritime borders with countries such as Honduras, Nicaragua, Costa Rica, and Jamaica. Nevertheless, none of these countries is a state party of the SPAW Protocol. Thus, using the SPAW Protocol as a tool to engage these countries to protect and to preserve the marine ecosystems within the MPA it allows the interested party to initiate a broad process of consultations with the authorities of the adjacent countries to the maritime borders to continue protecting this large area considered as a human heritage site. To this end, Colombia can use and promote the three dimensions of science diplomacy. Each of these dimensions has to be institutionalized as a guideline for the foreign policy. This action will allow a stronger regional projection supported by the instruments and mechanisms of the SPAW Protocol. One of the main targets is the Scientific and Technical Advisory Committee (STAC). This is a key actor to assist the implementation of the SPAW Protocol. Through STAC, Colombia can influence and provide advice to the parties on the scientific and technical aspects of MPA Seaflower. Also, Colombia can promote this kind of initiatives for the conservation of marine biodiversity and lead the coordination at the regional level. To achieve this goal, Colombia must establish a multidisciplinary national scientific committee to integrate each of the STAC Subcommittees, particularly the Subcommittee on Regulatory and Legal Matters for the development of laws and/or model guides for the implementation of SPAW obligations. In terms of strategic alliance, Panama can play an important role given its status as a state party because its political and economic influence on shipping represents an important neighborhood to support this initiative. Particularly, because Panama

Science Diplomacy: Strategic Initiative to Create a Buffer …

357

is one of the pioneer countries in Latin America that promotes Science Diplomacy as a tool for the development of its foreign policy. It is concluded that Colombian foreign policy requires more science to integrate the value of scientific evidence in international relations that will allow to implement this strategic initiative through SD.

References 1. Langenhove, L.V.: Global science diplomacy for multilaterlism 2.0. Sci. Diplomacy (2016) 2. Gorgone, H.R.: La Diplomacia Científico Tecnológica Como Herramienta De Internacionalización. Retrieved June 2017, from Repósitorio Institucional UFSC (2013, November 29): https://repositorio.ufsc.br/handle/123456789/116162 3. The Royal Society.: New Frontiers in Science Diplomacy. Retrieved June 2018, from royalsociety.org (2010): https://royalsociety.org/topics-policy/publications/ 4. “The Caribbean Sea”: The Columbia Encyclopedia, 6th edn. Retrieved 11 Feb 2016, from Encyclopedia.com: (2015) http://www.encyclopedia.com/doc/1E1-Caribbea.html 5. CEP.: The Caribbean Environment Programme—about Cartagena Convention. Retrieved 11 Feb 2016, from The Caribbean Environment Programme (2015): http://www.cep.unep.org/ 6. UNEP: Convention for the Protection and Development of the Marine Environment of the Wider Caribbean Region and its Protocols, 2nd edn. UNEP, Kingston, Jamaica (2012) 7. UNEP.: Setting a Course for Regional Seas. Retrieved 13 Mar 2016, from UNEP—Regional Seas Programme (2014a): http://apps.unep.org/publications/index.php?option=com_ pub&task=download&file=011736_en 8. UNEP.: Measuring Success: Indicators for the Regional Seas Convention and Action Plans. Retrieved 13 Mar 2016, from UNEP (2014b): http://apps.unep.org/publications/index.php? option=com_pub&task=download&file=011400_en 9. Patrick, S., Storm, B.: Council on foreign relations. In: Brown, K., Thaler, F.F. (eds.) Producers, & Media Strom. Retrieved 23 Jan 2016, from Global Governance Monitor/oceans: http://www.cfr.org/global-governance/global-governance-monitor/p18985#!/ oceans?gclid=CN_QxJjLwMoCFSTecgodjDoHAw#Issue-Brief 10. Van Tatenhove, J.P.: How to turn the tide: developing legitimate marine governance arrangements at the level of the regional seas. Ocean and Coastal Manage. J 71, 296–304. Retrieved 23 Jan 2016, from (2013, January) http://www.sciencedirect.com/science/article/ pii/S0964569112003171: http://doi.org/10.1016/j.ocecoaman.2012.11.004 11. Lausche, B.: Guidelines for Protected Areas Legislation, vol. 26. IUCN, Gland, Switzerland (2011) 12. PNNC-RUNAP.: Unique National Registry of Protected Areas—RUNAP. Retrieved 05 June 2016, from National Natural Parks of Colombia (2016): http://www.parquesnacionales.gov. co/portal/es/sistema-nacional-de-areas-protegidas-sinap/registro-unico-nacional-de-areasprotegias/ 13. Howard, M.W.: Evaluation Report Seaflower Biosphere Reserve Implementation: The First Five Years 2000–2005. Retrieved 2 Nov 2015, from UNESCO: http://www.unesco.org/csi/ smis/siv/Caribbean/San_actEnvEd_Seaflower2000-2005%20.pdf 14. Minambiente.: National Policy for the Integral Management of Biodiversity and its Ecosystemic services (NPIMBES). (J. Weiskopf, Trans.) Bogotá, Colombia. Retrieved 10 Apr 2016, from cbd.int (2011): https://www.cbd.int/doc/world/co/co-nbsap-v2-en.pdf

358

F. Ramírez-Cabrales and S. I. R. Forero

15. Minambiente.: Resolución número 107 (27 de Enero de 2005) Por el cual se declara un área marina protegida y se dictan otras disposiciones. Retrieved 22 Nov 2015, from Ministerio de Medio Ambiente y Desarrollo Sostenible (2005, January 27): http://www.coralina.gov. co/coralina/rendiciondecuentas/monitoreo-plan-accion/normatividad-1/normatividad-portemas/areas-protegidas/238-resolucion-107-de-2005-por-la-cual-se-declara-un-area-marinaprotegida 16. Coralina-Invemar.: Atlas de la Reserva de Biósfera Seaflower. Archipiélago de San Andrés, Providencia y Santa Catalina, vol. 28. In: Gómez-López, D., Segura-Quintero, C., SierraCorrea P., Garay-Tinoco, J. (eds.) Santa Marta, Colombia: Instituto de Investigaciones Marinas y Costeras “José Benito Vives De Andréis”- INVEMAR- y Corporación para el Desarrollo Sostenible del Archipiélago de San Andrés, Providencia y Santa Catalina –CORALINA (2012) 17. UNEP.: Seaflower Marine Protected Area—Colombia. Proposed areas for inclusion in the SPAW list. Retrieved 11 Feb 2016, from Unite Nations Environment Program—Caribbean Environment Program (2010, October 5): http://www.spaw-palisting.org/uploads/reports/ Report_31.pdf 18. Taylor, E., Baine, M., Killmer, A., Howard, M.: Seaflower marine protected area: governance for sustainable development. Marine Policy 41, 57–64 (2013, January 23) 19. Murillo, I., Ortiz, R.: Analisis de los metadatos sobre proyectos de investigación científica marina desarrollada en el area de la reserva de biosfera Seaflower. Secretaria Ejecutiva de la Comision Colombiana del Océano- Area Asuntos Marinos y Costeros, Bogotá D.C (2013) 20. UNEP.: Wider Caribbean Region Profile. Retrieved 11 Feb 2016, from UNEP (2005): http:// www.unep.org/regionalseas/programmes/unpro/caribbean/instruments/r_profile_car.pdf 21. Alonso, D., Barbosa, H., Duque, M., Gil, I., Morales, M., Navarrete, S., … Vasquez, J.: Conceptualización del Subsistema de Áreas Marinas Protegidas en Colombia. Documento de Trabajo (Versión 1.0). Proyecto COL75241 Diseño e implementación de un Subsistema Nacional de Áreas Marinas Protegidas (SAMP) en Colombia. Retrieved 06 June 2016, from INVEMAR (2015): http://www.invemar.org.co/documents/10182/14479/cartilla-conceptualizacionfinal-web.pdf 22. Jones, P.J.: Governing Marine Protected Areas: Resilience through Diversity. Taylor and Francis, New York (2014) 23. Speed, J.R., Levine, A.: What makes a “successful” marine protected area? The unique context of Hawaii’s fish replenishment areas (Elsevier, Ed.) Mar. Policy 44, 196–203 (2014, September 12) 24. Coralina: Acuerdos 021 y 025 de 2005 Por medio del cual se delimita internamente el Área Marina Protegida de la Reserva de la Biosfera Seaflower y se dictan otras disposiciones. Retrieved 18 Feb 2016, from Registro Unico Nacional de Areas Protegidas (2005): http:// runap.parquesnacionales.gov.co/reportes/generardocadminzip/aid/838 25. Fanning, L., Mahon, R., McConney, P., Angulo, J., Burrows, F., Chakalall, B., … Toro, C.: A large marine ecosystem governance framework. Marine Policy, 31(34), 434–443. Retrieved 19 Mar 2016, from Marine Policy (2007, July): http://www.sciencedirect.com/science/article/pii/ S0308597X07000097 26. Vallega, A.: The regional approach to the ocean, the ocean regions, and ocean regionalisation—a post-modern dilemma. Ocean Coast. Manag. 45, 721–760 (2002) 27. Lopez, V., Krauss, U.: National and Regional Capacities and Experiences on Marine Invasive Species, Including Ballast Waters, Management Programmes in the Wider Caribbean Region—A Compilation of Current Information. United Nations Environment Programme. CAB International Caribbean and Latin America Regional Centre (CLARC), Trinidad & Tobago (2006) 28. Biggs, C.: Indo-Pacific Red Lionfish: Pterois Volitans, Invasion of the Western Atlantic and Caribbean Oceans. Retrieved 23 Aug 2015, from http://www.washington. edu (2009): http://depts.washington.edu/oldenlab/wordpress/wp-content/uploads/2013/03/ Pterois-Volitans_Biggs.pdf

Science Diplomacy: Strategic Initiative to Create a Buffer …

359

29. Morris, J.J.: Invasive Lionfish: A Guide to Control and Management, vol. 1. Gulf and Caribbean Fisheries Institute Special Publication. Marathon, Florida, USA (2012) 30. IMO: International shipping facts and figures—information resources on trade, safety, security, environment. Retrieved 10 July 2016, from IMO-Knowledge Centre section: Ships and Shipping—Facts and Figures (2012 March 5): http://www.imo.org/en/KnowledgeCentre/ ShipsAndShippingFactsAndFigures/TheRoleandImportanceofInternationalShipping/ Documents/International%20Shipping%20-%20Facts%20and%20Figures.pdf 31. AGCS: Panama Canal 100 Shipping safety and future risk. Retrieved 17 Jan 2016, from Allianz Global Corporate & Specialty (2014): http://www.agcs.allianz.com/assets/PDFs/risk% 20bulletins/Panama-Canal-100_low-res.pdf 32. UNCTAD.: Review of Maritime Transport 2015. Retrieved 10 July 2016, from UNCTAD/RMT (2015): http://unctad.org/en/PublicationsLibrary/rmt2015_en.pdf 33. The World Bank.: Panama Canal expansion: A smart route for boosting infrastructure in Latin America. Retrieved 30 July 2016, from The World Bank(2016, May 07): http://blogs.worldbank.org/voices/panama-canal-expansion-smart-route-boostinginfrastructure-latin-america 34. Rodrigue, J.-P., Ashar, A.: Transshipment hubs in the New Panamax Era: The role of the Caribbean. J. Transp. Geogr. 10 (2015 October 2) 35. US Department of Transportation.: Panama Canal Expansion Study_Phase I Report. Retrieved 17 Jan 2015, from U.S. DEPARTMENT OF TRANSPORTATION MARITIME ADMINISTRATION (2013, November 20): http://www.marad.dot.gov/wp-content/uploads/ pdf/Panama_Canal_Phase_I_Report_-_20Nov2013.pdf 36. Webster, M.: Redrawing Global Shipping Routes: The Panama Canal Gets an Upgrade. Retrieved 17 Jan 2016, from Brown Brothers Harriman (2015, December): https://www. bbh.com/blob/10958/f7d1d1c2fd63c146c469fb0d384b1c75/redrawing-global-shippingroutes–the-panama-canal-gets-an-upgrade-pdf-data.pdf 37. Harrould-Kolieb, E.R., Herr, D.: Ocean acidification and climate change: synergies and challenges of addressing both under the UNFCCC. Clim. Policy J. Retrieved 22 Oct 2015 (2012, January): http://doi.org/10.1080/14693062.2012.620788 38. Hassellöv, I.-M., Turner, D.R., Lauer, A., Corbett, J.J.: Shipping contributes to ocean acidification. Geophys. Res. Lett. Retrieved 22 Nov 2015, (2013, June 6): http://onlinelibrary.wiley. com/doi/10.1002/grl.50521/epdf 39. Bekker, P.: The World Court Awards Sovereignty Over Several Islands in the Caribbean Sea to Colombia and Fixes a Single Maritime Boundary between Colombia and Nicaragua. Retrieved from The American Society of International Law (2013, Jan 15): http://www. peacepalacelibrary.nl/ebooks/files/354062581.pdf 40. Gorricho, J.: Global Pacebuilding II: High Time for Environmental Diplomacy in the Caribbean. Retrieved 7 July 2016, from Governance and Development (2012, November 28): http://www. governanceanddevelopment.com/2012/11/global-peacebuilding-ii-high-time-for.html 41. Gual, S.M., Wang, T.C., Robinson, C.R.: Science Diplomacy. Retrieved June 2018, from science diplomacy.org (2017, March): http://www.sciencediplomacy.org/article/2017/internationalcollaboration-in-connecting-scientists-policy 42. Gluckman, P.D., Turekian, V., Grimes, R.W., Kishi, T.: Science Diplomacy: A Pragmatic Perspective from the Inside. Retrieved 20 June 2018, from Science & Diplomacy (2017, December): www.sciencediplomacy.org/article/2018/pragmatic-perspective

Part X

Wearable Technology and Assistance Devices

Quantum Well Infrared Photodetector for the SWIR Range Pedro Pereira, Lesslie Guerra, G. M. Penello, M. P. Pires, L. D. Pinto, R. Jakomin, R. T. Mourão, M. H. Degani, M. Z. Maialle and P. L. Souza

Abstract An InGaAs/InAlAs superlattice infrared photodetector is developed to reach the forbidden gap in the SWIR range for arsenides, between 1.7 and 2.5 µm, appropriate for surveillance imaging. The figures of merit of the device are determined to be 120 K for the BLIP temperature and 2.1 mA/W and 3 × 106 Jones for the best responsivity and detectivity, respectively, obtained at 120 K under +4 V bias. Possible approaches to improve the device performance are addressed. Keywords QWIPs · SWIR · III-V semiconductors

1 Introduction Infrared photodetection was for a long time dominated by the Hgx Cd1 − x Te detectors, which can be easily tuned to reach any operation wavelength in the mid-infrared range (2–20 µm), by simply varying the alloy composition [1]. Since these devices are based on band-to-band transitions, varying the Hg content in the alloy leads to different band gap energies and, consequently, different threshold detection waveP. Pereira · L. Guerra · L. D. Pinto · P. L. Souza (B) LabSem/CETUC, Pontifícia Universidade Católica, Rio de Janeiro 22451-900, Brazil e-mail: [email protected] P. Pereira · L. Guerra · G. M. Penello · M. P. Pires · L. D. Pinto · R. Jakomin · R. T. Mourão · M. H. Degani · M. Z. Maialle · P. L. Souza DISSE-Instituto Nacional de Ciência e Tecnologia de Nanodispositivos Semicondutores, PUC-Rio, Rio de Janeiro 22451-900, Brazil G. M. Penello · M. P. Pires · R. T. Mourão Instituto de Física, Universidade Federal do Rio de Janeiro, Rio de Janeiro 21945-970, Brazil R. Jakomin Campus de Xerém, Universidade Federal do Rio de Janeiro, Duque de Caxias 25245-390, Brazil M. H. Degani · M. Z. Maialle Faculdade de Ciências Aplicadas, Universidade Estadual de Campinas, Limeira, São Paulo 13484-350, Brazil © Springer Nature Singapore Pte Ltd. 2020 Á. Rocha and R. P. Pereira (eds.), Developments and Advances in Defense and Security, Smart Innovation, Systems and Technologies 152, https://doi.org/10.1007/978-981-13-9155-2_29

363

364

P. Pereira et al.

lengths. However, due to difficulties in producing a homogeneous material over a large area, these devices are not adequate for focal plane arrays, in which application each pixel in a matrix of devices should have exactly the same performance [2–4]. In the past decade, the most investigated alternative has been the quantum well infrared photodetector (QWIP) fabricated with III-V semiconductors, which are based on intraband transitions [5, 6]. Intraband transitions are, in principle, intrinsically limited to energies below the heterostructure conduction band offset. In the case of III-V semiconductors, this means that energies above 500 meV cannot be detected, which corresponds to wavelengths shorter than 2.5 µm. As a consequence of lattice mismatch to the substrate, band-to-band photodetectors using III-V semiconductors can only be fabricated for wavelengths shorter than 1.7 µm. So, there is a sort of forbidden gap for these materials between 1.7 and 2.5 µm. This spectral interval is, though, of major importance for satellite imaging and surveillance [7]. In 2016, an InGaAs/InAlAs superlattice structure with a central quantum well with a different thickness from the otherwise periodic structure has been proposed to reach this forbidden gap [8]. Absorption and photocurrent have been demonstrated at 2.1 µm at 77 K, as well as, the possibility of tuning this operation wavelength within the SWIR range by slightly changing the thicknesses of the quantum wells and barriers. In this work, the behavior of this device, which from now on will be called SLIP, as a function of temperature is addressed, the different figures of merit are determined and operation up to 160 K is demonstrated.

2 Experimental Details The InGaAs/InAlAs SLIP structures have been grown by metalorganic vapor-phase epitaxy lattice matched to the InP substrate at 100 mbar and 650 °C. Trimethylindium, trimethylgallium, trimethylaluminum, arsine and phosphine were used as sources for In, Ga, Al, As and P. After the deposition of an InP buffer layer, the QWIP structure, which consists of a twenty-period SL with a wider central quantum well (CQW), is grown. This modified SL is itself repeated twenty times. An InAlAs spacer of 30 nm is deposited between each SL. A scheme of the structure is depicted in Fig. 1. The CQW is n-doped with Si using a SiH4 flux to reach 2 × 1018 electrons per cm3 . The whole active layer is sandwiched between two n-doped contact layers with the same doping level as the CQW. The thicknesses of the CQW, barrier (B) and superlattice quantum well (SLW) for the main investigated SLIP are nominally 7, 3 and 1.5 nm, respectively. Absorption and photocurrent measurements were carried out with Fourier transform infrared spectroscopy (FTIR) in a waveguide mode, since perpendicular incident radiation does not couple to intraband transitions in QWIPs. For this purpose, the grown samples were polished to produce a 45° coupling facet on which light is focused at normal incidence. Additionally, this waveguide geometry allows multiple reflections of the beam inside the sample increasing the interaction of radiation

Quantum Well Infrared Photodetector for the SWIR Range

365

Fig. 1 Scheme of the photodetector structure. Adapted from [8]

with the quantum wells (QWs) and therefore the absorption itself. The responsivity measurements were carried out using a calibrated blackbody source. Calculations using the split-operator method were employed to design appropriate SLIP structures. In the simulations, the Schrödinger equation is solved for the desired potential obtaining the eigenstates and eigenvalues. As the energies involved are comparable to the bandgap of the material, a nonparabolic approximation is used. The correction of these effects is considered in the effective mass of the electron, which becomes dependent on the electron energy, rendering the calculation selfconsistent.

3 Results and Discussion The calculated band structure of the previously designed SLIP structure is depicted on the left panel of Fig. 2 [8]. In blue are the probability densities of the even parity states, while in green are the odd parity ones. In gray are the probability densities for the miniband states. On the right are the absorption spectra for transitions starting at the ground (blue) and first excited (green) states, showing the possibility of operation within the forbidden gap with such structures, since absorption from the ground state to the third excited state gives rise to a peak at 578 meV (2.14 µm). Photocurrent spectra for different bias voltages have been detected at 77 K for such structure at this energy, as reproduced in Fig. 3 [8], demonstrating that, in fact, a SLIP can detect radiation in the band 7 of the SWIR (2.105–2.155 µm). A second photocurrent peak around 300 meV (4 µm), attributed to the optical transition between the ground state and the miniband states, in which case the current is produced by tunneling from one period to the subsequent one, is also observed. Even though it has already been demonstrated that SLIPs based on III-V semiconductors can reach this important wavelength range [8], it is crucial to determine up to

366

P. Pereira et al.

Fig. 2 Left panel: probability densities for the first four localized states (blue for even and green for odd parity states) and for the miniband states (gray) superimposed on the conduction band profile on the growth direction. Right panel: absorption spectra for transitions with the ground state (blue) and the first excited state (green) as the initial states for the best-fit structure thicknesses Adapted from [8] Wavelength (m)

Photocurrent (arb. u.)

108 6 100

4

2

80

60

40

20 0 0.2

0.4

0.6

0.8

1.0

1.2

Energy (eV)

Fig. 3 Photocurrent as a function of bias for the investigated SLIP structure Adapted from [8]

Quantum Well Infrared Photodetector for the SWIR Range

367

Fig. 4 Photocurrent as a function of temperature under a +2 V bias

which temperature the device can work and its figures of merit such as responsivity, detectivity and BLIP temperature, temperature at which the current generated due to the background radiation equals the dark current. Figure 4 shows the photocurrent spectra for temperatures between 80 and 160 K at +2 V bias. It is clear that the peak at 587 meV is detected up to 160 K with an evident raise in noise as the temperature increases. The lower energy peak around 300 meV is only observed up to 120 K most likely because carriers excited to the miniband have the additional possibility of being thermally excited to states out of the miniband, which do not allow current propagation, reducing the tunneling current. Figure 5 shows the dark current versus voltage curves (I × V ) for temperatures between 80 and 300 K in black using a heat shield with steps of 20 K. At null bias, the dark current is of the order of pA and for +2 V bias is below nA at 80 K, increasing to mA at room temperature. In green is the current versus voltage curve generated by incident background radiation. From this plot, we determine the BLIP temperature to be 120 K. The activation energy of the dark current generation mechanism can be determined from the I × V data, building an Arrhenius plot for each different bias. The determined activation energies are shown in Fig. 6 for biases between + 5 and −5 V. The activation energy for null bias is 295 meV, which agrees with the energy difference between the ground state and the first miniband, indicating that the tunneling current through the minibands is the main mechanism for dark current generation. In order to reduce the undesirable dark current, thicker barriers between the different superlattice periods should be implemented.

368

P. Pereira et al.

Fig. 5 Dark current versus voltage for temperatures between 80 and 300 K in steps of 20 K. In green, the current versus voltage curve generated by incident background radiation, showing a BLIP temperature of 120 K

Fig. 6 Activation energy as a function of bias obtained from the data of Fig. 5

The responsivity of the SLIP, current generated per incident radiation power, as a function of bias is depicted in Fig. 7 for temperatures between 80 and 160 K. The highest responsivity is obtained with a +4 V bias at 120 K and is equal to 2.1 mA/W. At the highest operating temperature, 160 K, the device provides a 2.0 mA/W responsivity when the applied bias is +2 V. The relatively low responsivity is somewhat expected, since the oscillator strength for the optical transition between ground state and third excited state is intrinsically low, as shown in the right panel of Fig. 2. There one can clearly see that the absorption is more than two orders of magnitude lower than the one from the ground state to the first excited state. Figure 8 shows the detectivity as a function of bias of the SLIP, essentially responsivity divided by the noise, for the same temperatures. The highest detectivity of 3 × 106 Jones is achieved under three different conditions, namely temperatures 120, 140 and 160 K with +4, +3 and +2 V biases, respectively. This occurs because as the bias voltage increases not only the photocurrent increases, but also the dark current and, consequently, the noise.

Quantum Well Infrared Photodetector for the SWIR Range

369

Fig. 7 Responsivity as a function of bias for temperatures between 80 and 160 K

Fig. 8 Detectivity as a function of bias for temperatures between 80 and 160 K

4 Conclusion An InGaAs/InAlAs superlattice infrared photodetector, here called SLIP, which can reach band 7 of the SWIR (around 2.0 µm operation wavelength) has been investigated in terms of its figures of merit. Even though this previously forbidden wavelength has been reached, the performance of the device still falls short from being competitive. The optical transition involved to absorb 2.1 µm is from ground state to third excited state, which has a relative small probability of taking place. An alternative would be to seek different designs, using genetic algorithms, for instance, to be able to have the final state of the optical transition of lower order, which would strengthen the absorption’s oscillator strength. Another important issue is reducing the dark current and, therefore, the noise. It has been shown that the main mechanism for generation of the dark current is the tunneling current through the superlattice miniband states. Using a wider InAlAs barrier between each of the 20 periods should significantly reduce the dark current, since it decays exponentially with the layer thickness.

370

P. Pereira et al.

Acknowledgements This work was partially supported by FAPERJ, CNPq, FINEP, CAPES and FAPESP.

References 1. Rogalski, A.: HgCdTe infrared detector material: history, status and outlook. Rep. Prog. Phys. 68(10), 2267 (2005) 2. Shen, S.: Comparison and competition between MCT and QW structure material for use in IR detectors. Microelectron. J. 25, 713–739 (1994) 3. Goldberg, A.C., Kennerly, S.W., Little, J.W., Shafer, T.A., Mears, C.L., Schaake, H.F., Winn, M.L., Taylor, M., Uppal, P.N.: Comparison of HgCdTe and quantum-well infrared photodetector dual-band focal plane arrays. Opt. Eng. 42, 30–47 (2003) 4. Downs, C., Vandervelde, T.: Progress in infrared photodetectors since 2000. Sensors 13, 5054–5098 (2013); Author, F., Author, S., Author, T.: Book title, 2nd edn. Publisher, Location (1999) 5. Liu, H.C.: Quantum dot infrared photodetector. Optoelectron. Rev. 1, 1–6 (2003) 6. Rogalski, A.: Infrared detectors: an overview. Infrared Phys. Technol. 43, 187–210 (2002) 7. See www.landsat.usgs.gov or www.landsat.gsfc.nasa.gov 8. Guerra, L., Penello, G., Pires, M., Pinto, L., Jakomin, R., Mourão, R., Degani, M., Maialle, M., Souza, P.: Detecting infrared radiation beyond the bandoffset with intersubband transitions. IEEE Photonics Technol. Lett. 28, 1641–1644 (2016)

Independent Feeding of People Affected with Osteoarthritis Through a Didactic Robot and Visual Control Arturo Jiménez, Katherine Aroca, Vicente Hallo, Nancy Velasco and Darío Mendoza

Abstract This chapter presents a work that will help in the consumption of the food of people with osteoarthritis. In this way, the patient will avoid discomfort in the joints of their hands. The system consists of an artificial vision algorithm and independent feeding part. The vision algorithm allows the detection of the face, localization, and tracking the person’s mouth. The independent feeding part consists of a didactic robotic arm. The robotic arm takes the food from the dish. The vision algorithm detects and tracks the face, then locates the position of the mouth. Finally, the robotic arm delivers the food to the user. The program was developed in Python using OpenCV and Dlib libraries. The face alignment method has an average of 94% of effectiveness. Keywords Visual control · Recognition and monitoring · Robot feeding people · Osteoarthritis

1 Introduction Osteoarthritis is characterized by progressive degenerative changes primarily involving the hips, knees, spine, hands, and feet. Osteoarthritis causes stiffness, inflammation, pain, deformity, and weakness. The person with osteoarthritis cannot hold objects such as a spoon. Military troops are more likely to develop osteoarthritis than civilians. Robots are one strategy to help with osteoarthritis. Robots are directly relationship with computer control. Computer vision offers the robot information about its work A. Jiménez · K. Aroca · V. Hallo (B) · D. Mendoza Departamento de Energía y Mecánica, Universidad de las Fuerzas Armadas ESPE, Sangolquí, Ecuador e-mail: [email protected] N. Velasco Departamento de Ciencias Exactas, Universidad de las Fuerzas Armadas ESPE, Sangolquí, Ecuador © Springer Nature Singapore Pte Ltd. 2020 Á. Rocha and R. P. Pereira (eds.), Developments and Advances in Defense and Security, Smart Innovation, Systems and Technologies 152, https://doi.org/10.1007/978-981-13-9155-2_30

371

372

A. Jiménez et al.

environment so that this information can be used for its own control. An eyeinhand system is a robot equipped with an autofocus camera. There are many designs and constructions of self-feeding robots like Handy1, Winsford feeder, Neater Eater, My Spoon, Meal Buddy, and Mealtime Partner Dining System [1]. Meal Buddy [2] has three degrees of freedom in robotic arm and three bowls. After the system scoops the food, the robotic arm scrapes the surplus food off the spoon with the rod on the bowls. Handy1 [3] consists of a five degree of freedom robot, a tray unit and a gripper. Handy1 allows a user to select food from any part of the tray. Some robots are for specific food like the design of feeding robot for Korean food [4]. This robotic system consists of two robotic arms: One is a food transferring robotic arm and one is a food grasping robotic arm. Moreover, some students build robots to help with osteoarthritis in [15]. They utilized their robot models to educate the public on osteoarthritis. We propose a low-cost system to facilitate feeding to patients with osteoarthritis. For this, the hardware part is represented by a robotic arm. The software part is implemented in Python and image processing libraries.

2 Development 2.1 Description of Tools The Raspberry Pi 3 is a reduced-board computer. One of the great advantages is its free use both educationally and privately. Its operating system offers an adapted version of Debian, known as Rasbian [5]. A low-weight high-resolution camera is required for the recognition and monitoring of the user’s mouth, so that the robotic arm has no problems when it comes to mobilizing and being compatible with the Raspberry Pi 3. The camera that meets the requirements of the system is the FACECAM 1000X for its weight of 50 g, size (60 × 20 × 22 mm), resolution (1 MP, 1280 × 720, 640 × 480 pixels), and the sensor (CMOS HD 720p pixels) that make this alternative the most convenient for this project.

2.2 System Flow The project is composed of the following processes (Fig. 1).

Independent Feeding of People Affected with Osteoarthritis …

373

Fig. 1 System flow diagram

2.3 Face Detection Algorithm The Haar Cascade function of the OpenCV library is used in face detection, which uses pre-trained classifiers to identify faces. The front face classifier function is used. The video input is loaded and it is transformed to grayscale mode for a better processing of the image. Then the detection of the face is done by the detectMultiScale function, the same as with a copy in scale of grays finds all the faces in each frame [6].

2.4 Face Tracking Algorithm It is not necessary to apply face detection in each frame in real-time videos. Instead, tracking algorithms can be used because it is easier to track the face during the video. When the person makes a movement, the detection algorithm has to return to zero in order to detect the new position of the face; so a tracking algorithm is needed to complement the detection algorithm. This not only improves the tracking speed eliminating the need to redetect the face in each frame. The system also becomes more robust for two reasons. First, it retains the facial characteristics after detecting the face and second, it becomes more resistant to rotation and photometric changes [7, 8]. This project is designed to track only one face (one user) because tracking more than one face can generate errors in the control of the didactic robotic arm. Dlib allows to track the position of an object as it moves frame by frame in a video sequence [9].

374

A. Jiménez et al.

Fig. 2 Face detection using Haar Cascade

The Dlib correlation tracker is used to track the face. It extracts the characteristics of the detected face, to subsequently track it. After that, a rectangle is drawn in the area where the face has been detected (Fig. 2). It is necessary to calculate the area of the rectangle that contains the face to know if Haar Cascade has detected the face. If it is greater than zero, the detector works well and initiates the tracker.

2.5 Mouth Detection Dlib allows to track the position of an object as it moves frame by frame in a video sequence. This correlation tracker has a bounding box to follow the object in the video frame. Correlation trackers, as the name implies, work by correlating a set of pixels from one frame to another [8].

2.6 Tracking the Mouth The correlation tracker Dlib is used for face tracking. The same one that extracts the characteristics of the detected face, to later track through the video sequence. After face detection and tracking, an identifier of the geometric structure of the face of the Dlib library is used. It is called shape_predictor_68_face_landmarks. It is a standardized library to deliver the reference points of the face by through the alignment in one millisecond [10]. The library delivers 68 facial points (Fig. 3), which define the facial regions, such as the corners of the mouth, eyebrows, eyes, nose, and mouth, using parts of a simple matrix in Python [11]. The facial alignment is performed by a set of regression trees, which consists of taking an image in the form of a numpy matrix and returning a 68 × 2 matrix, each row is conformed with the x, y coordinates of a common point of the facial feature in

Independent Feeding of People Affected with Osteoarthritis …

375

Fig. 3 Points of facial coordinates

the input image, in other words, it contains coordinates for a particular facial feature. In this project, points 49 through 68 are used because they contain all the information in the mouth [12].

2.7 Reverse Kinematics of the Robotic Arm The inverse kinematics is necessary to perform a correct positioning of the robotic wrist. The geometric transformation method is used in order to determine the angle to which each of the servomotors must turn [13]. In this project, the calculations will be made first on the YZ plane of the arm to determine the angles of each joint that allow the didactic robotic arm to take the required position according to the coordinate because in this way the calculations will be made in the plane for greater ease. The arm has been divided into three vectors (Fig. 4) to facilitate the calculation. The vectors are arm, forearm, and wrist. When the analysis is carried out, the angle δ is calculated which is imposed by the user. The established spatial coordinate of the armrests in this angle, and it will serve as the basis for calculating the rest of the angles. Next, we decompose the wrist vector: wrist Y = wrist ∗ cos(δ)

(1)

wrist Z = wrist ∗ sin(δ)

(2)

Calculate Y  and Z  in order to find the diagonal AC and the angle α:

376

A. Jiménez et al.

Fig. 4 Geometrical components of inverse kinematics

Y  = Y  − wrist Y

(3)

Z  = wrist Z + Z 

(4)

AC =

√  2 ((Y ) + (Z  )2 )

(5)

tan(α) = Z  /Y 

(6)

  α = tan−1 Z  /Y 

(7)

Angles β and γ are calculated using the law of cosines:   cos(β) = arm2 − forearm2 + AC2 /(2∗arm∗AC).

(8)

  cos(γ ) = arm2 + forearm2 − AC 2 /(2∗arm∗forearm).

(9)

After we found β and γ , it is possible to determine the angles that the forearm can turn based on Fig. 4. arm angle = 90◦ + β + α.

(10)

forearm angle = γ .

(11)

Independent Feeding of People Affected with Osteoarthritis …

377

The triangles ABC, CDE, and ACE are located to calculate ε. After that, ε is determinate based on the complementary angles: ε = ε + ε + ε

(12)

In triangle ABC, the following equation is obtained ε = 180◦ − β − γ

(13)

In the triangle ACE, we have the following equation ε = 180◦ − 90◦ − α

(14)

ε = 90◦ − α

(15)

In the triangle CDE, we use the following equation ε = 180◦ − 90◦ − δ

(16)

ε = 90◦ − δ

(17)

ε = 180◦ − β − γ + 90◦ − α + 90◦ − δ

(18)

Then ε is:

The angle of rotation of the base of the robot that allows the manipulator to locate in the X, Y, Z plane is determined, and the angle between X and Y must be known to obtain its spatial position [14]. Angle of rotation = tan(−1) (x/y)

(19)

The first image is the starting point to know the position of the mouth. A copy of the original frame is made to guarantee a greater acceleration in the processing of the image and further, resize with values of 320 × 240 pixels. The selected resolution makes the system more robust and tracking more efficient compared to larger dimensions. Basically, there is a delimiting area of the mouth; if the user moves outside the area of the center of the image as shown in Fig. 5, then the robotic arm begins the tracking process to be located within the center of the area. For example, if the user is located in a sector with coordinates P(60, −100), through the serial communication already established, the respective location characters will be sent to position the robotic arm at the center of the image.

378

A. Jiménez et al.

Fig. 5 Central area of the image

Fig. 6 Screening and monitoring of the mouth using face alignment method and Haar Cascade

3 Experimental Results 3.1 Experiments Figure 6 shows the robotic arm tests tracking user’s mouth in face alignment method and Haar Cascade’s. The precision and speed of the system are considered to determine the best method for the recognition and tracking of the mouth. If the recognition of the mouth is not exact, the robotic arm would produce failures when feeding the user.

Independent Feeding of People Affected with Osteoarthritis …

379

3.2 Results The tests were performed on five users, inside a house, with different light settings and in different distances. The test distance is from 5 to 50 cm. It is based on the reach of the robotic arm. Once the user’s face is located, a green rectangle is drawn around him and his mouth (Fig. 7). Figure 8 shows the percentage of effectiveness of each method at different distances up to 50 cm. The alignment method of the face of Dlib has a higher percentage of effectiveness than the Haar Cascade method of OpenCV. The face alignment method has a maximum percentage of 98% and a minimum of 90, 94% in average. Haar Cascade has a maximum percentage of 70% and a minimum of 40%. These percentages were obtained from the average effectiveness of tracking of each user.

Fig. 7 Left: face alignment with Dlib. Right: face recognition image with Haar Cascade

Fig. 8 Accuracy of tracking of the mouth according to the tracking distance

380

A. Jiménez et al.

The Haar Cascade method for mouth detection is less precise and its parameters much more difficult to adjust, while alignment of the face is more precise with simpler adjust parameters. Both methods are executed in a speedy manner and do not require a very powerful processor for their execution since can run from a Raspberry Pi 3 without problems. Face alignment method can be adapted to any environment for detection and tracking of the mouth. Haar Cascade needs to reset of its parameters every time the environment changes.

4 Conclusions This project was carried out in free software to reduce costs. The Dlib library is used to efficiently implement the mouth tracking algorithm. Dlib allows the recording of the detected region frame by frame, the reduction of errors and the improvement of the tracking speed. The visual servo control provides an adequate response for the control of the movements of the manipulator. The face alignment method has an average of 94% of effectiveness. In the other hand, Haar Cascade has an average of 55% of effectiveness. Therefore, face alignment method has the best results in this application. The calculation of the inverse kinematics allows to position the end effector of the arm exactly. For this in space (X, Y, Z), the angles of your joints are calculated. Assistive technology allows to considerably improve the quality of life of a war veteran affected with osteoarthritis. Acknowledgements We are grateful to “Universidad de las Fuerzas Armadas ESPE” to carry out the tireless duty of teaching.

References 1. Song, W.-K., Kim, J.: Novel assistive robot for self-feeding. En Robotic Systems-Applications, Control and Programming. InTech (2012) 2. Sammons Preston http://www.sammonspreston.com/ 3. Topping, M.: Handy 1, a robotic aid to independence for severely disabled people. In: 7th International Conference on Rehabilitation Robotics, pp. 142–147 (2001) 4. Song, W.-K, et al.: Design of novel feeding robot for Korean food. In: International Conference on Smart Homes and Health Telematics. Springer, Berlin, Heidelberg, pp. 152–159 (2010) 5. Aranda, D.: Electrónica: plataformas Arduino y Raspberry Pi. Dalaga, Buenos Aires (2014) 6. Lee, S., Lee, C.: Illumination normalization and skin color validation for robust. de IS&T International Symposium on Electronic Imaging 2016, Seoul (2016) 7. Castrillón, M., Déniz, O., Hernández, D.: A comparison of face and facial feature detectors based. Mach. Vision Appl. 22, 481–494 (2011) 8. Dollár, P., Appel, R., Belongie, S., Perona, P.: Fast feature pyramids for object detection. IEEE Trans. Pattern Anal. Mach. Intell. 36(8), 1532–1545 (2014)

Independent Feeding of People Affected with Osteoarthritis …

381

9. Klare, B.F., Klein, B., Taborsky, E., Blanton, A., Cheney, J., Allen, K., Grother, P., Mah, A. Jain, A.K.: Pushing the frontiers of unconstrained face detection and recognition: IARPA Janus Benchmark A. In: de IEEE Conference on Computer Vision and Pattern Recognition (CVPR), Boston (2015) 10. Liao, S., Jain, A.K., Li, A.Z.: A fast and accurate unconstrained. IEEE Trans. Pattern Anal. Mach. Intell. 38(2), 211–223 (2015) 11. Cheney, J., Klein, B.: Unconstrained face detection: state of the art baseline and challenges. In: International Conference on Biometrics (ICB), Phuket (2015) 12. Kazemi, V., Sullivan, J.: One millisecond face alignment with an ensemble of regression trees. Estocolmo (2017) 13. Baki Koyuncu, A.M.G.: Software development for the kinematic. World Acad. Sci. Eng. Technol. 1(6), 1575–1580 (2007) 14. Cruz, A.B.: Cinemática inversa, de Fundamentos de robótica. McGraw-Hill, Madrid (2013)

Adjustable Structure in Height for Assisted Feeding in Patients with Osteoarthritis Katherine Aroca, Arturo Jiménez, Nancy Velasco, Vicente Hallo and Darío Mendoza

Abstract This article presents the design and building of an adjustable height structure for the assisted feeding of people with osteoarthritis. The project will improve the quality of life of these patients. The design is governed by the INEN (Ecuadorian Standardization Service) norm. The fundamental criteria in the document are oriented to create and size a mechanical structure matching with the anthropometric measurements of the users. The parameters include the feeding area, mechanism materials, and the correctly defining a height adjustment. Keywords Ergonomics · Osteoarthritis · Feeding · Mechatronic design

1 Introduction The osteoarthritis is the most common joint disease. According to international studies, more than 70% of adults older than 50 years of age present radiological signs of osteoarthritis in any joint; women are the most appropriate. The osteoarthritis affects joints like hands, hip, or knee [1]. Besides, osteoarthritis is one of the main causes of disability in the retired militaries. It affects a greater proportion of retired militaries than in comparison with civilians. The most frequent reasons are stress and injuries caused by repetitive basic training and combat risks such as bomb explosions, which directly affect the cartilage cells and soft tissues [2]. Currently, there are some structures to feed people with physical disabilities, such as Obi, which is a simple minimalist structure that works with a robotic arm, two push buttons and a plate divided into four compartments for different types of food K. Aroca · A. Jiménez · V. Hallo (B) · D. Mendoza Deptartamento de Energía y Mecánica, Universidad de las Fuerzas Armadas ESPE, Sangolquí, Ecuador e-mail: [email protected] N. Velasco Deptartamento de Ciencias Exactas, Universidad de las Fuerzas Armadas ESPE, Sangolquí, Ecuador © Springer Nature Singapore Pte Ltd. 2020 Á. Rocha and R. P. Pereira (eds.), Developments and Advances in Defense and Security, Smart Innovation, Systems and Technologies 152, https://doi.org/10.1007/978-981-13-9155-2_31

383

384 Table 1 Data for sizing the surface of the table height

K. Aroca et al.

Dimension

Reference

Measure (mm)

A

Popliteal height

470

E

Lumbar support height

180

[3]. There are many designs and constructions of self-feeding robot like Handy1, Winsford feeder, Neater Eater, My Spoon, Meal Buddy and Mealtime Partner Dining System [4]. Handy1 [5] consists of a five degree of freedom robot, a tray unit, and a gripper. Handy1 allows a user to select food from any part of the tray. Meal Buddy [6] has three degrees of freedom in robotic arm and three bowls. After the system scoops the food, the robotic arm scrapes the surplus food off the spoon with the rod on the bowls. The study presented below emphasizes the design of a height adjustable table for patients with osteoarthritis. For this, it is necessary to determine technical criteria to implement a structure with good space of operation according to the physical characteristics of the user.

2 Elements of the Structure The structure will have a space for: a robotic assistance arm with a range of at least 520 mm and lightweight, food plates (soup, entrée, dessert), a drinking glass with a straw, a spoon with its spoon holder, electronic and drive elements, a screen, and a height adjustment mechanism.

2.1 Feeding Area The maximum feeding area is determined by the maximum reach of the robotic arm, in this case 1040 mm2 . Within this space, the user can be fed, and Table 1 shows the elements comprising it. Figure 1 shows the elements of the maximum feeding area. Number one represents the prototype of robotic arm. Number two represents the area of work or feeding area and number three represents the maximum robotic arm reach.

2.2 Support Table The INEN norm 1641:1988 [7] determines nominal values of the free space for the accommodation of legs as shown in Fig. 2. The table width (T ) has a minimum value of 580 mm; it should be defined such as requirement for the implementation of the

Adjustable Structure in Height for Assisted Feeding in Patients …

385

Fig. 1 Elements of the maximum feeding area

Fig. 2 Free space for legs accommodation according to NTE INEN 1641:1988 standard

elements on the table with a value of 810 mm. In addition, the depth at the feet level is determined by a minimum value (V ) of 600 mm. After adding a structure for driving mechanism operated by patient’s feet on the back of the mechanical structure, its height will be increased to 670 mm for user comfort. The dimensions of the table are 810 mm × 670 mm for width and depth, respectively. The Pythagorean theorem is used to calculate the exact location radius of food utensils. The right triangle is formed by the oscillation arm and the union of the arm with the final wrist or effector (Fig. 3). If bo is measurement of the oscillation arm, bm is arm measurement + wrist measurement, and r T is measurement of the radius of location of food utensils, thus: bm2 = bo2 + r T2 rT =

√

(bm2 − bo2 ) =

√

(3172 − 1022 ) mm = 300 mm

(1) (2)

The optimum total radius of location of the food utensils is 300 mm with respect to the initial point of location of the robotic arm. There are three food dishes in total,

386

K. Aroca et al.

Fig. 3 Triangle with real measurements used to calculate the radius of location of food utensils with respect to the center of the robotic arm Fig. 4 Location of food utensils

and the location of each plate is required to calculate the separation length between them in first place. The arc length formula is used assuming an angle of 36°, as shown in Fig. 4. If ∅ is the angle formed between radius, c is the length of separation between each plate, thus, c = 2r T2 sin(∅)

(3)

c = 2(300) sin(36◦ /2) = 185 mm

(4)

The separation between plates is delimited by a value of 185 mm and angle of 36°. The glass of drink is at 300 mm with an angle of 90° with respect to the second plate of food. The total mechanism is present in Fig. 5.

Adjustable Structure in Height for Assisted Feeding in Patients …

387

Fig. 5 Parts of the mechanical structure

The transmission of circular to continuous linear motion is used with the rack and pinion elements in the mechanism of regulation of the height range.

2.3 Adjustable Height Range of the Table One of the parameters of ergonomics is to reduce user’s risks and discomforts in certain positions [8]; therefore, the regulation in height fulfills the need to adapt it to different heights. For this purpose, the following measurements have been carried out to obtain a good range of regulation and offer more ergonomics and comfort. The scheme is presented in Fig. 6. In the first place, the INEN 1641: 1988 NORM [9] is used. The minimum height (hcodo) to which the support table should be fixed depends on popliteal height in the sitting position (A) and the distance of the space from the chair to the elbow known as lumbar support height (E). hcodo = A + E

(5)

The data for sizing the surface of the table height is shown in Table 1. The minimum height of adjustment of the work surface is given by the height hcodo: hcodo = (470 + 80) mm = 650 mm

(6)

388

K. Aroca et al.

Fig. 6 Seated workplace

Fig. 7 Sitting position of women from 60 to 90 years Table 2 Data for sizing the surface of the table height

Dimension

Reference

Measure (mm)

30

Popliteal height

360

37

Lumbar support height

190

The sitting position of adult women is taken as the minimum height of adjustment of the work surface. In the same way, the popliteal height in the sitting position (30) and the distance of the space from the chair to the elbow (37) are added, in order to obtain the minimum height at which the support table should be fixed [10]. The details are in Fig. 7. The data for sizing the surface of the table height is shown in Table 2. The minimum height of adjustment of the work surface is given by the height hcodo, where:

Adjustable Structure in Height for Assisted Feeding in Patients …

hcodo = (360 + 190) mm = 550 mm

389

(7)

In another study based on the INEN 1641: 2015 standard [7], a minimum height value of 650 mm is established for fully adjustable tables, up to a height of 850 mm. In this case, according to previous studies, a height of 550 mm is established, up to a maximum of 670 mm, in other words, an interval of 120 mm, to satisfy the need to regulate the height of all the elderly.

2.4 Pinion and Rack Design Due to the low cost and improvements in the mechanical properties, the plastic materials have increased their utility within the industry. In the market, there is a great variety of pinions with standard zips, mostly with pressure angle of 20° and made of nylon [11], for applications of uniform shock, safe operation, low friction, corrosion resistance, and good properties in how much to wear. The rack and pinion elements must comply with certain design requirements to avoid possible failures during the movement of the mechanism, mainly for plastic gears are supporting the impact of the tip of the tooth and the bottomland or chamfer of the tooth of the rack, which produces fracture in the material due as the failure by fracture between plastic gears due to the impact of the tip of the tooth of the pinion with the bottomland or chamfer of the tooth of the rack. This defines a minimum number of teeth of the gear, due to the established pressure angle, not less than 18 teeth are used [12]. The material used is nylon. The allowable bending stress S at is 41 MPa. The minimum reference space for pinion location DReference is 92 mm. The length of the rectangular zip l is 120 mm. The number of teeth selected from the pinion zp is 21. The geometry of the rack and pinion is defined by [13]. Thus, the pinion and rack module is: m = DReference /z p = 92/21 mm = 4 mm

(8)

The diameter of the step DP is equal to the multiplication of pinion or rack module m and number of teeth selected from the pinion zp : D P = mz p = (4) (21) mm = 84 mm

(9)

The diameter step pd is equal to the multiplication of pi (π ) and pinion or rack module m: pd = π m = π (4) = 12.57 mm

(10)

390

K. Aroca et al.

The number of steps carried out (n°steps) is equal to a length of the rectangular zip divided by diametral step of the pinion or rack: n ◦ steps = 1/ pd = 120 mm/12.57 mm = 9.56 steps

(11)

The number of turns of the pinion n°turnsp is equal to the number of steps carried out (n°steps) divided by the number of teeth selected from the pinion zp : n ◦ turns p = (n ◦ steps)/z p = 9.56/21 = 0.45 Turns

(12)

The revolution per minute of the pinion np is equal to the number of turns of the pinion divided by the thickness of the tooth t: ◦ turns

p

n p = n ◦ turns p /t = (0.47Turns/1.8s) (2π rev/1Turns) (60s/1min) = 95 RPM

(13) The zipper speed V ac in mm/s is equal to the multiplication of diametral step of the pinion or rack pd , the number of teeth selected from the pinion zp , and revolutions per minute of the pinion np : Vac = pd Z p n p = 12.57 mm × 21 × 1.59REVS = 419 mm/s

(14)

The dimensions of head a, root b, and spacing c are 4, 5, and 1 mm, respectively. The external diameter Do is equal to a primitive diameter of the pinion Dp plus double the head a: Do = D p + 2a = 84 + 2 × 4 = 92 mm

(15)

The diameter of the root DRP is: DRP = D P − 2b = 84 − 2 × 5 = 74 mm

(16)

The total depth ht is: h t = a + b = 4 + 5 = 9 mm

(17)

The depth of work hk is: h k = a + a = 1 + 1 = 2 mm

(18)

The tooth thickness t is: t = pd /2 = 12, 57/2 = 6.285 mm

(19)

Adjustable Structure in Height for Assisted Feeding in Patients …

391

The face width F is: F = 12(m) = 12(4) = 48 mm ≈ 50 mm

(20)

The critical load on a tooth will be generated when the load is at the highest point of contact in a single tooth, so the force will have used the value of the mass that will be lifted, this one, a tangential force will generate a tension momentum or maximum bending at the base of the teeth of the rack and pinion. To solve this requirement was used a design software, it is estimated an approximate value of the total mass to be lifted of 36.16 kg which consists of the robotic assistance arm, the food plates, food utensils, to screen, the table, the support profiles of the table, and the approximate mass of the user’s arms. The tangential force W t :   Wt = mass(gravity) = 36.16 Kg 9.81 m/s2 = 354.73 N

(21)

2.5 Bending Stress on the Teeth of the Pinion and Rack These calculations are based on AGMA Standards. If K a is an application factor for the flexural strength, K s is the size factor for the flexural strength, K m is the load distribution factor for the flexural strength, K B is the factor thickness of the crown, K v is the dynamic factor for the flexural strength, K v dynamic factor for the flexural strength, and then, the tension or bending stress on the teeth σ t is: σt = (Wt /F m J )((K a K s K m K B )/K V )

(22)

The effort for maximum tension in the pinion is: σtp = (354.73/(50 × 4 × 0.33))((1 × 1 × 1.126 × 1)/0.85) = 7.12 MPa

(23)

Therefore, it is true that the maximum bending stress on the pinion σ tp is less than the permissible bending stress S at : σtp < Sat → 7.12 MPa < 41 MPa

(24)

If J p is geometry factor for the pinion and J r is geometry factor for the rack, the factors in the flexural stress equation are the same for the rack except for the J value of geometry so the effort in the rack can be calculated as follow: σtr = σtp (Jp /Jr ) = 7.12 (0.33/0.35) = 6.71 MPa

(25)

392

K. Aroca et al.

Fig. 8 von Mises stress analysis of the pinion

Also, the maximum bending stress on the rack is less than the allowable bending stress: σtr < Sat → 6.71 MPa < 41 MPa

(26)

2.6 Maximum Bending Stress of the Pinion and Rack It is the maximum bending at the base of one of the teeth of the pinion giving a value of 8.99 MPa using the von Mises deformation method, therefore: σtp < Sat → 8.99 Mpa < 41 Mpa

(27)

The von Mises effort of the pinion is less than the allowable bending stress. Figure 8 shows the stress through design software. In the same way, the maximum bending at the base of one of the teeth of the rack gives a value of 6.84 MPa, thus: σtr < Sat → 6.84 Mpa < 41 Mpa

(28)

The von Mises effort of the rack is less than the allowable bending stress. Figure 9 shows the analysis of efforts through a design software.

Adjustable Structure in Height for Assisted Feeding in Patients …

393

Fig. 9 von Mises stress analysis of the rack Table 3 Minimum height Study type

Minimum support table adjusting height (mm)

INEN 1641:1988 Norm

650

Study of the anthropometric registry in females between 60- and 90-year-old in Mexico

550

INEN 1641:2015 Norm

650

3 Experimental Results The feeding area has the dimensions 1040 × 1040 for width and depth, respectively. The support table has the dimensions 810 × 670 for width and depth, respectively. Then, the dimensions of the support table have lower values compared to the dimensions of the feeding area allowing to take these dimensions as part of the design of the mechanical structure. Table 3 shows the minimum height obtained through a study of the INEN norms and an anthropometric registry study conducted in Mexico [10]. Based on Table 3, the minimum height required to adjust the adjustable support table will be 550 mm, based on the anthropometric registry study of women aged 60–90 years in Mexico. Since these users are those of shorter stature, and the maximum height will be 670 mm, to allow the regulation of the height of the mechanical structure for the other beneficiaries. The results obtained from the analysis of bending stresses are in Table 4. The stress measurements obtained by the AGMA standards and computer-aided engineering are very similar; therefore, it guarantees the reliability of these calcu-

394 Table 4 Results obtained from the analysis of bending stresses

K. Aroca et al.

Study type

Pinion

Rack

Bending stress according to AGMA (MPa)

7.12

6.71

Bending stress according to von Mises (MPa)

8.99

6.84

Permissible nylon bending stress (MPa)

41

41

lation methods. Also, these values are lower than the permissible bending stress; therefore, the material selected for the rack and pinion is nylon.

4 Conclusions The variation in height in each elderly affected with osteoarthritis allows that the design of the table adjustable in height offers greater ergonomics, comfort, and effectiveness when feeding and also reduces risks or discomfort that the user may have in certain positions, improving their quality of life. Among the technical criteria implemented in the research is the area of power the one that limits the dimensions of width and depth of the mechanical structure in order to avoid shocks and discomfort with the elements that make up protecting the user from being hurt. The correct distribution of the spaces for the food utensils and the robotic arm allow the patient to position themselves in a suitable place to taste their food. The height adjustment mechanism, formed by means of the rack and pinion elements, allows the vertical translation of the mechanical structure. The range of height regulation is delimited by the studies of the INEN standards and an anthropometric study carried out in Mexico. AGMA standards and computer-aided engineering are proven to be effective and reliable tools to select nylon as a material for the implementation of the elements and thereby comply with the economic and design parameters necessary for the selection of materials.

References 1. Solis, U., García y A. de Armas, V.: Rasgos demográficos en la osteoartritis de manos. Revista Cubana de Reumatología 16(3), 275 (2014) 2. Cuddigan Law, [En línea]. Available: https://www.cuddiganlaw.com/library/making-theservice-connection-for-arthritis.cfm?fbclid=IwAR2vnmcOQHx9nETuGUp_bWKDo9Zr9se_ pnASD0rHiyfjOPaJTFH4HuVRQvs [Último acceso: 22.01.2018] 3. Decker, J.: [En línea]. Available: https://meetobi.com/ [Último acceso: 2019 Enero 6] (2019) 4. Song, W.-K., Kim, J.: Novel assistive robot for self-feeding. En Robotic Systems-Applications, Control and Programming. InTech (2012) 5. Topping, M.: Handy 1, a robotic aid to independence for severely disabled people. In: 7th International Conference on Rehabilitation Robotics, pp. 142–147 (2001) 6. Sammons Preston. http://www.sammonspreston.com/

Adjustable Structure in Height for Assisted Feeding in Patients … 7. 8. 9. 10.

395

N. I. 1641:2015, Muebles de oficina, escritorios y mesas. Requisitos, Quito (2015) Herrera Saray, P.: Ergonomía y el hábitat para la tercera edad. Páginas 87, 35–89 (2010) N. I. 1641:1988, Muebles de oficina, escritorios y mesas. Requisitos, Quito (1988) De Lavalle Herrera, Y.: Diseño y ergonomía para la tercera edad, 1st ed., México: Universidad Nacional de México, p. 304 (2014) 11. Roydisa, Roydisa [En línea], (2012). Available: https://www.roydisa.es/wp-content/uploads/ 2012/12/engranajes.pdf [Último acceso: 24 Junio 2017] 12. Carro Suárez, J., Flores Salazar, F., Flores Nava, I.: Análisis de esfuerzo de contacto en engranes rectos aplicando el criterio de Hertz con ingeniería asistida por computadora. Científica 16(1), 25–32 (2012) 13. Mott, R.: Diseño de Elementos de Máquinas. In: Guerrero, P. (ed.) Pearson Education, México (2006)

Voice-Controlled Assistance Device for Victims of Gender-Based Violence Miguel A. Domínguez , David Palomeque , Juan M. Carrillo , José Mª Valverde , Juan F. Duque , Bruno Pérez and Raquel Pérez-Aloe

Abstract One of the biggest problems that society is currently facing is violence against women. In recent years, tangible progress in protecting and saving the lives of female victims of intimate partner/family-related homicide has not been made, so targeted responses are clearly needed. In this work, an electronic device to help victims of gender-based violence who live with their aggressor has been designed. The system is built on Bluetooth Low Energy technology allowing a wireless communication between device and mobile phone with a low power consumption. The device is controlled by three different commands and is capable of sending messages through a mobile phone to a Control Center. Depending on the nature of the received messages, the Control Center will take the appropriate measures to assist the victim. The design has been made paying special attention to a reduced size so that the device can easily be camouflaged in any accessory of the victim’s jewelry, thus going unperceived to the possible aggressor. Keywords Gender-based violence victims · Assistance device · Voice-controlled M. A. Domínguez · J. M. Carrillo · J. M. Valverde · J. F. Duque · B. Pérez · R. Pérez-Aloe (B) University of Extremadura, 06006 Badajoz, Spain e-mail: [email protected] M. A. Domínguez e-mail: [email protected] J. M. Carrillo e-mail: [email protected] J. M. Valverde e-mail: [email protected] J. F. Duque e-mail: [email protected] B. Pérez e-mail: [email protected] D. Palomeque Institute of Microelectronics (US-CSIC), 41092 Seville, Spain e-mail: [email protected] © Springer Nature Singapore Pte Ltd. 2020 Á. Rocha and R. P. Pereira (eds.), Developments and Advances in Defense and Security, Smart Innovation, Systems and Technologies 152, https://doi.org/10.1007/978-981-13-9155-2_32

397

398

M. A. Domínguez et al.

1 Introduction Nowadays, one of the problems of great social impact is the violence against women. According to a new research published by the United Nations Office on Drugs and Crime (UNODC), released for the International Day for the Elimination of Violence against Women, a total of 87,000 women were intentionally killed in 2017. Around a third (30,000), were killed by their current or former intimate partner, someone they would normally expect to trust [1]. The main health consequences of this violence are: death and injury, depression, problems with alcohol consumption, sexually transmitted diseases, unwanted pregnancies and abortions, babies with low birth weight, among others [2]. Since the 1980s, Spain has focused its attention on the problem of violence suffered by women, reflecting his concern in several legal and regulatory programs [3–12]. It is worth highlighting, the agreement signed between several Spanish Ministries for the implementation of the protocol of action of the monitoring system by telematics means of the fulfillment of the restraining orders in matters of genderbased violence. The monitoring system consists of a radiofrequency transmitter that emits a radio frequency (RF) signal that is received by a track unit. Both are devices for the accused. The track unit is a GPS tracking device that incorporates the basic functionalities of a mobile phone in addition to the reception of the RF signal emitted by the RF transmitter. The device for women is a track unit, practically equal to the defendant’s and incorporates an external radio frequency antenna that allows to detect the radio frequency signal of the RF transmitter of the accused. Figure 1 shows the devices for the accused and the corresponding for the woman. When an alarm occurs, the Control Center performs the communications provided in the protocol of intervention for each type of alarm. Despite these actions, gender-based violence in Spain still constitutes one of the main social problems. Although there are multiple mechanisms that currently have the victims of gender violence to protect themselves and ask for help, the reality is that the number of fatalities does not seem to decrease and remains at a very high number. It should

Fig. 1 Devices of the monitoring system by telematics means: for the accused (a) and (b), (c) is the device for women

Voice-Controlled Assistance Device for Victims of Gender-Based …

399

be noted that, only in the last year, 60 women were killed, of which 78% had never processed a complaint against their aggressor [13]. Therefore, the current conception of surveillance of the aggressor is insufficient for a real prevention of situations of gender violence and targeted responses are clearly needed. In order to provide a possible solution to this problem, in this paper we present a device for asking help, designed for people suffering from gender-based violence, including those who have not reported their aggressor and continue to live with him. The purpose is to develop a system of personalized help, through electronic devices easily camouflaged in the accessories of the victims, such as watches, necklaces, bracelets, brooches, as well as a set of actions and strategies, aimed at returning to a normal life this group at risk of social exclusion. The present paper has been organized as follows: Sect. 2 describes the system proposed. The voice capture system and its characteristics are detailed in Sect. 3. In Sect. 4, the design of the Bluetooth application is mentioned. The conclusions are set out in Sect. 5. Finally, Sect. 6 is dedicated to the acknowledgments.

2 Proposed System The main goal has been to develop a device of small size that the victim or potential victim of gender-based violence would carry at all times. The device is able to detect a limited number of voice commands through a relatively simple hardware, which respects the conception of low consumption device. The system will be built on Bluetooth Low Energy (BLE) technology, which will allow to establish a wireless communication between device and mobile phone with a low power consumption thanks to the transmission of small data packets instead of a continuous flow of information. The smartphone will perform alert functions via calls and SMS, as well as GPS location, thus allowing the device designed to be small in size and then easily camouflaged. It, also, will incorporate a microphone that will allow the smartphone to perform GPS locator functions and launch calls to the authorities, depending on the detected voice command. On the other hand, the smartphone will fulfill the functions imposed under an Android application. The use of some capabilities that, a priori, the majority of low-mid-range smartphones have (GPS and Bluetooth 4.0) is essential to reduce the cost and its size to the minimum possible. The entire system described is outlined in Fig. 2. Regarding the products focused on the prototyping and development of systems based on BLE technology, RFduino has been the selected for this work. Specifically, the RFD22301 chip includes a microcontroller based on an ARM Cortex M0 architecture (nRF51822) that implements BLE technology. In addition, this platform is fully compatible with the Arduino IDE environment and with the libraries developed for Arduino. RFD22301 incorporates a 10 bits analog-to-digital converter, serial port, a microcontroller operating at 16 MHz, 128 kB of Flash memory and 8 kB of RAM [14]. To have an easy access to the chip’s ports, the module RFD22102

400

M. A. Domínguez et al.

Fig. 2 Block diagram of the voice command recognition system

“RFduino DIP” has been used [15] along with the RFD22121 “USB Shield.” This device provides serial conversion-USB necessary to load compiled code on the chip, thanks to the built-in FTDI [16]. Regarding the development environment, RFduino is programmable through Arduino IDE in a language that mixes functions of C and C++ [17]. Arduino IDE incorporates code editor, compiler and serial monitor, being the only software debugging element available. With respect to the smartphone, Android, which is based on the Linux kernel, has been the operating system selected. Today it is present in smartphones, tablets, smart watches, televisions, and automobiles, among others, being already the most popular operating system in the world [18]. There are several environments that enable the development of Android applications in Web languages (HTML5, PHP, CSS3); however, it has been selected Android Studio (IDE facilitated by Google) because it offers greater possibilities for optimization and use of the hardware, as well as the use of APIs that facilitate the implementation of services such as Bluetooth or location in a relatively simple way. IDE Android Studio incorporates various utilities for programming: code editing, compilation in real time, debugging, possibility of simulation of mobile devices, and others. As an Android device, we have used a Motorola Moto G, with Android Lollipop 5.0. In order to implement the libraries related to BLE technology in an Android application, it is necessary that the device has API Level 18 or higher, that is, Android 4.3 or higher. Regarding the BLE protocol just highlight that it is built on three fundamental layers: application, host, and controller. This work mainly deals with the highest-level layer, that is, the application layer.

3 Voice Capture System The voice capture system is shown in Fig. 3. As can be seen, consists of a microphone, a sound amplifier, and an analog-to-digital converter. Since the voltage variations will be very small, it will be necessary to pass the electrical signal through an amplifier

Voice-Controlled Assistance Device for Victims of Gender-Based …

401

Fig. 3 Functional block diagram of the audio capture system

Fig. 4 Audio amplifier with low-pass type characteristic

so that, finally, it is digitized, processed and the result sent through the Bluetooth module. The amplified acoustic signal will be a signal with a continuous frequency spectrum between 150 Hz and 4000 Hz, which is the range of frequencies on which the male and female voice extends [19]. For an adequate sampling of the audio by the ADC, it is necessary to ensure the Nyquist Theorem. The amplification structure selected is based on a non-inverting configuration with Operational Amplifier (OpAmp). In order to comply with the Nyquist criterion and avoid the possible spectrum overlapping, a capacitor has been included to reduce the bandwidth of the input signal. The amplifier is shown in Fig. 4, along with the

402

M. A. Domínguez et al.

capacitor Cf added to provide the amplifier with low-pass characteristics. Its value will be selected following an empirical procedure. The OpAmp used has been the LM324. It is a single-supply low power OpAmp. The resistors Rcm1 and Rcm2 are responsible for providing the input signal with a common mode voltage, necessary due to the 3.3 V single-supply operation of the OpAmp. On the other hand, resistors R1 and R2 set the gain of the circuit according to the expression (1) with R1 implemented with a potentiometer for a manual adjustment of the gain. Gain = 1 +

R1 R2

(1)

In addition, the C 1 condenser decouples the input signal and the common mode voltage of the circuit. Finally, the C 2 capacitor allows that the common mode feedback will be complete. The values of the discrete components are the following: R1 = 218 k, R2 = 2.2 k, Rcm1 = 22 k, Rcm2 = 10 k, C 1 = 4.7 µF, C 2 = 4.7 µF so the maximum value of the gain will be approximately 100 V/V, i.e., 40 dBs. The total harmonic distortion of the amplifier has been obtained for a 10 mV input signals at 1 kHz frequency obtaining a value of 0.3%, more than acceptable considering that the maximum input signal will be around 2 mV (voltage level for conversations of great intensity).

4 Design of the Bluetooth Application The effective sampling frequency of RFduino was determined resulting to be about 13 kHz, which, in the first instance, allows the development of an audio application, since the Nyquist Theorem would be respected for signals within the human conversation frequency spectrum. Regarding the roles in the BLE network, the smartphone would take a role of observer whereas for the implemented device the role will be broadcaster.

4.1 Audio Processing The digital processing of the audio signal through RFduino has been implemented through the μspeech library [20]. Thanks to this open-source software library, it is easy implement phoneme recognition functions in RFduino. It is written in C++ and works by taking a number of audio samples that evaluates and associates with a phoneme, distinguishing between English phonemes associated with the characters ‘e’, ‘h’, ‘v’, ‘f ’, ‘s’, and ‘I’. This library will be modified to adapt it to the needs of this work.

Voice-Controlled Assistance Device for Victims of Gender-Based …

403

4.2 Operation and Modification of the µspeech Library It is worthwhile to mention that it is not intended to perform a complex digital audio processing, since this requires advanced hardware (a DSP would be necessary) and a higher power consumption. The valuable of this method is the possibility of distinguishing a few voice commands from a simple hardware with a minimum power consumption allowing the device to operate with a coin cell battery. The μspeech library works by calculating the complexity of the sampled signal. To do this, take a number of samples (32 by default) that are stored in an array of integers. Afterward, discrete derivation operations in absolute value and the discrete integral of the absolute value of the samples are performed. These operations are reflected in Eq. (2).  |array[i] − array[i − 1]|  (2) complexity = |array[i]| In a first approximation, it can be deduced that signals with a waveform with important high-frequency contents will have a higher value of complexity, thus being able to distinguish some phonemes from others. In addition, with the aim of filtering noise, the library performs a buffer of a 7 size in which a new value of complexity is stored in each sample collection. Finally, an amplification factor adjusts the weight of the derivative action against the integral action of the algorithm. Once the complexity of the sample vector has been calculated, the obtained value is compared with the tabulated coefficients for each of the phonemes, selecting the corresponding one. All phonemes are recognized by this algorithm excepting the phoneme /f/, which is obtained by comparing a new parameter with a tabulated coefficient. Finally, in each execution it is possible to calibrate the level of the ambient noise present through a method that takes 10,000 samples and performs the average, setting an ambient sound level that is subtracted from each of the measures that are subsequently taken. μspeech can distinguish 6 phonemes in the English language, which should not necessarily correspond to phonemes of the Spanish language. Thanks to a developed script and empirical tests, it was determined that the Spanish phonemes that this library allows to distinguish were: /i/, /a/, /g/, /s/, and /f/ [21].

4.3 Development of a Script for Voice Command Detection Considering the environment in which the designed device is going to be used, a program has been developed that distinguishes three different Spanish voice commands: “socorro,” “informacion,” and “GPS.” These words contain phonemes that can be distinguished one from the other. The “socorro” command is distinguished

404

M. A. Domínguez et al.

Table 1 Results after a voice command acquisition test Cf capacitor (pF)

Command

Hits

15

Socorro

18

2

90

GPS

20

0

100

100

150

Errors

Hit rate (%)

Información 20

0

100

Socorro

16

4

80

GPS

20

0

100

Información 17

3

85

16

20

20

0

100

Información 13

7

65

Socorro GPS

4

Average rate (%) 96.67

88.33

61.67

by the /s/phoneme, whereas the “informacion” command will be discerned by the /f/phoneme. The “GPS” command is recognized thanks to the /g/phoneme. To implement this script, the syllable class—defined in the library μspeech—has been used. It acts as a “container” of detected phonemes, and an array of characters that allows us to know the history of phonemes detected in chronological order.

4.4 Frequency Response of the Amplifier As already mentioned, the choice of the value of capacitor Cf has been made following a completely empirical procedure. This has consisted of a test in which the three commands to be detected were repeated 20 times. The results obtained are shown in Table 1. It is concluded that the best response is obtained using a 15 pF capacitor in the feedback path. These tests have been carried out in ideal conditions of loudness (without ambient noise). The results were very similar with certain levels of ambient noise.

4.5 Design of a Prototype For the design of a prototype, the shield of the RFduino RFD22127 has been used [22]. It allows the circuit to be powered through a AAA 1.5 V battery since it implements a DC/DC boost converter that sets a supply voltage of 3.3 V. Three LEDs have been also included so they will light up when any of the voice commands were detected. The estimated current consumption is around 15 mA. In the design of the printed circuit board (PCB) special attention has been paid to the layout in order to size the board so that have the dimensions and shape similar with those of the Shields used.

Voice-Controlled Assistance Device for Victims of Gender-Based …

405

Fig. 5 Photograph of the PCB with components and shields

Fig. 6 Miniaturized circuit and necklace where it can be placed a circuit top view, b bottom view and c top and bottom part of the necklace

All traces have been arranged in the lower layer. In addition, three capacitors have been included in the path from the power supply to ground to filter noises. Figure 5 shows the photograph of the PCB with all components and Shields. As already mentioned, the final goal of this work is that the device can be carried by a victim of gender-based violence camouflaged in some jewelry accessory. To achieve this, it is necessary to minimize the size of the designed prototype. Figure 6a, b shows the top and bottom part of the new prototype implemented, respectively. The dimensions are 28.6 × 25.5 mm2 which allow it to be placed in a necklace as shown in Fig. 6c.

4.6 Development of the Android Application The application has a graphical user interface (GUI) which shows how the phone scans for advertising packets during a limited period of time, then going to sleep. If advertising packets are detected, the information will be read to discern if it comes from the implemented device (through the name) and what is the detected voice

406

M. A. Domínguez et al.

Fig. 7 Screenshots of the Android application a main screen, b “socorro” (SOS) screen, c “información” (info) screen and d “GPS” screen

command (through the user data included in the package). The functions, depending on the detected command, are: – Voice command “socorro”: GPS location via satellite and telephony antennas, ability to send SMS alerts with this information and to launch a telephone call. – Voice command “informacion”: collecting the battery level of the device and sending SMS with this information. – Voice command “GPS”: GPS location and sending SMS with information regarding the coordinates (latitude and longitude). Figure 7 shows the different screenshots of the Android application depending on the functions carried out. Once the SMS is sent to the Control Center, the appropriate steps will be taken in order to protect the victim from its possible aggressor or simply store the received information.

5 Conclusions The prototype of an electronic device to help victims of gender-based violence has been designed. The device is controlled by voice commands and is capable of sending messages through a mobile phone to a Control Center. Depending on the nature of the messages, the Control Center will take the appropriate measures to assist the victim or simply store the information received. The design has been done paying special attention to a minimum power consumption so a coin cell battery can be used and a reduced size so that the device can easily be camouflaged in any accessory of the victim’s jewelry. In this way, the device can go unnoticed to the aggressor with which the victim could even to be living together.

Voice-Controlled Assistance Device for Victims of Gender-Based …

407

Acknowledgements The authors are grateful to the Junta de Extremadura and FEDER (Fondo Europeo de Desarrollo Regional “Una manera de hacer Europa”), for financial help by project IB16211.

References 1. United Nations Office on Drugs and Crime, UNODC: Global Study on Homicide, Vienna (2018) 2. World Health Organization, Department of Reproductive Health and Research, London School of Hygiene and Tropical Medicine, South African Medical Research Council: Global and regional estimates of violence against women. Prevalence and health effects of intimate partner violence and non-partner sexual violence. World Health Organization, Switzerland (2013) 3. Organic Law 1/2004, of December 28, on comprehensive Protection Measures against Genderbased Violence 4. Organic Law 3/1989, of June 21, updating the Criminal Code, which introduces the criminal activity of habitual violence in article 425 5. Organic Law 11/2003, of September 29, on Measures on Citizen Security, Domestic Violence and Social Integration that brings habitual violence to Article 173 of the Criminal Code and introduces for the first time the “criminal activity of occasional mistreatment” in his article 153 6. Law 27/2003, of July 31, regulating the Order for the Protection of Victims of Domestic Violence 7. Organic Law 5/2010, of June 22, which modifies the Organic Law 10/1995, of November 23, of the Penal Code 8. Organic Law 2/2009, of December 11, reform of the Organic Law 4/2000, of January 11, on rights and freedoms of foreigners in Spain and their social integration 9. Royal Decree-Law 3/2013, of February 22, which modifies the regime of fees in the area of the Administration of Justice and the system of free legal assistance 10. B.O.E of June 6, 2014, Instrument of ratification of the Convention of the Council of Euro-pa on prevention and fight against violence against women and domestic violence, done in Istanbul on May 11, 2011 11. Organic Law 3/2007, of March 22, for the effective equality of women and men 12. Guide to criteria for judicial intervention against gender-based violence, 2013 13. Ministry of Health, Social Services and Equality of the Government of Spain on Genderbased Violence Homepage, http://estadisticasviolenciagenero.msssi.gob.es/. Last accessed 2018/12/19 14. RFduino (RFD22301) Homepage, http://www.rfduino.com/product/rfd22301-rfduino-blesmt/index.html. Last accessed 2015/06/01 15. RFduino (RFD22102) Homepage, http://www.rfduino.com/product/rfd22102-rfduino-dip/ index.html. Last accessed 2015/06/01 16. RFduino (RFD22121) Homepage, http://www.rfduino.com/product/rfd22121-usb-shield-forrfduino/index.html. Last accessed 2015/06/01 17. Arduino programming language Homepage, http://www.arduino.cc/en/pmwiki.php?n= Reference/HomePage. Last accessed 2015/06/01 18. Android Homepage, https://www.android.com/. Last accessed 2018/012/26 19. Nortel Networks, Voice Fundamentals, Reference Guide (2014) 20. GitHub directory of the μspeech library, https://github.com/arjo129/uspeech/releases. Last accessed 2015/06/13 21. Association of phonemes and letters of Spanish, Homepage, http://www.livingspanish.com/ correspondencia-fonetica-grafia.htm. Last accessed 2015/06/13 22. RFduino (RFD22127) Homepage, http://www.rfduino.com/product/rfd22127-single-aaabattery-shield-for-rfduino/l. Last accessed 2015/06/13

Part XI

Analysis and Signal Processing

Analysis and Evaluation of the Positioning of Autonomous Underwater Vehicles Using Acoustic Signals Enrique V. Carrera and Manolo Paredes

Abstract Autonomous underwater vehicles (AUVs) present as much potential as unmanned aerial vehicles, or drones, for being used in military missions. Major defense-oriented applications of AUVs will be intelligence, surveillance, and reconnaissance tasks, besides payload delivery and communication support. Nevertheless, AUVs have several constraints that make it difficult to rely completely on their autonomous functions. Particularly, AUVs do not have access to accurate positioning mechanisms like GPS. Thus, expensive inertial navigation techniques have been developed for these vehicles in the last decades. However, small and low-cost AUVs require newer and simpler positioning techniques that can be adapted to specific military missions. Based on that, this work analyzes and evaluates common trilateration techniques based on the propagation of acoustic signals in the water for a precise underwater positioning. Although the results show that the method known as time difference of arrival (TDoA) presents good accuracy, its application in military missions is still limited due to the large number of noise sources required; therefore, several future works in this direction are additionally proposed. Keywords Autonomous underwater vehicles · Underwater positioning · Trilateration techniques · Underwater sound propagation

E. V. Carrera (B) · M. Paredes Departamento de Eléctrica, Electrónica y Telecomunicaciones, Universidad de las Fuerzas Armadas ESPE, Sangolquí 171103, Ecuador e-mail: [email protected] M. Paredes e-mail: [email protected] M. Paredes Centro de Investigación de Aplicaciones Militares CICTE, Universidad de las Fuerzas Armadas ESPE, Sangolquí 171103, Ecuador © Springer Nature Singapore Pte Ltd. 2020 Á. Rocha and R. P. Pereira (eds.), Developments and Advances in Defense and Security, Smart Innovation, Systems and Technologies 152, https://doi.org/10.1007/978-981-13-9155-2_33

411

412

E. V. Carrera and M. Paredes

1 Introduction Unmanned aerial vehicles, or drones, have rapidly grown in popularity during the past years. They have broken through rigid barriers in traditional industries which otherwise seemed impenetrable by similar technological innovations. Today, drones are key elements in activities like aerial photography, search and rescue operations, precision agriculture, and express shipping and delivery. In fact, even though the military use of drones has a large history [7], military applications are currently the main service of this type of vehicles [3, 16]. It should also be noted that unmanned vehicles have been an important catapult for the battlefield industry and security operations, improving the fight against organized and transnational crime, and other threats that attempt to violate or alter citizen security. However, illegal activities are also being developed in coastal profiles and offshore, and drones are not able to fully cover the needs existing in underwater environments. For instance, security operations require support to combat drug trafficking in submersible vehicles, guerrillas next to zones covered by water, criminal activities through rivers and estuaries, etc. It is then essential to use the benefits and capabilities of unmanned vehicles for these operations in order to support military activities in those areas. Thus, the need of developing the so-called unmanned underwater vehicles arises. These vehicles are divided into remotely operated vehicles (ROVs), which are controlled by a remote human operator, and autonomous underwater vehicles (AUVs), which operate with no direct human intervention. AUVs are the most complex and expensive vehicles since they completely rely on autonomous functions, and they have gradually evolved offering capabilities in some missions and applications that no other platforms can offer [1]. As a matter of fact, due to the technological innovations in the past years, small and inexpensive AUVs are becoming feasible and profitable. Although AUV field is still an emerging sector, AUVs will be part of many academic, commercial, and military applications [5]. In fact, military usage of AUVs is a sector of increasing demand, due to their ability of conducting missions that no other equipment can perform [8]. However, current existing alternatives face several challenges like position uncertainty, limited and noisy communication, bounded time autonomy, overall reliability, among others. In addition, the usage of AUVs into military operations deserves a deeper analysis, since it is necessary to exploit the principles of the war and carry out military missions effectively. Hence, it has been promoted the creation, strengthening, and improvement of R+D+i processes to generate technological solutions that energize and optimize the use of all kind of resources in military operations. Following this line of reasoning, this work seeks to solve a main AUV problem: underwater positioning. This is a challenging task since the GPS network is not available to submersed devices, and considering that radio signals do not propagate very far in water, AUV positioning and navigation are still areas of active development [14]. When georeferenced navigation is necessary, current AUVs use internal navigation systems, such as inertial guidance systems that measure and integrate

Analysis and Evaluation of the Positioning of Autonomous Underwater …

413

acceleration or a Doppler velocity log system. A controller then integrates velocity to dead-reckoned positions, which are generally updated via GPS. GPS update represents a risk to underwater missions since the AUV must emerge jeopardizing its covertness. Accurate navigation without GPS has been also proposed using bottomterrain maps, which must be developed in advance. In test conditions, AUVs might even navigate using a set of acoustic beacons, forming an LBL network [12]. Based on the above, this paper analyzes and evaluates common trilateration techniques used for Cartesian positioning, but adapted to underwater scenarios through the usage of acoustic signals [6, 15]. Specifically, a simulator of an AUV was implemented to compare received signal strength indication (RSSI), time of arrival (ToA), and time difference of arrival (TDoA) techniques for the positioning of a vehicle inside a large volume of water. The results show that TDoA presents the best accuracy, but the number of noise sources is relatively high for a 3D positioning. Thus, some future works are also proposed for the positioning of AUVs, especially using microphone arrays (i.e., hydrophone arrays). The rest of this paper is organized as follows. Section 2 introduces AUVs, sound propagation in water, and trilateration techniques. Section 3 presents the simulation of three common trilateration techniques for AUV positioning. The main results and their implications are analyzed in Sect. 4. Finally, Sect. 5 concludes this paper and discusses some future works.

2 AUV Positioning An important step for the development of powerful AUVs is to know the vehicle’s position adequately, using any available tool or technology. With this requirement in mind, the basics of AUVs, the propagation of acoustic signals in water, and common trilateration techniques are discussed below.

2.1 AUV Basics Many different AUVs have been designed in the past five decades, ranging in size from portable AUVs to very large vehicles. Large AUVs have advantages in terms of endurance and sensor payload capacity, but smaller AUVs benefit significantly from lower logistics. Today, the AUV market is evolving, and new designs are following commercial requirements rather than being purely experimental [2]. AUVs must basically control themselves while accomplishing a predefined task [18]. For that, the following list represents the main technologies that have been addressed over the last years [1, 5]:

414

E. V. Carrera and M. Paredes

– Autonomy. Topics such as mission planning, perception and situation assessment and intelligent systems have been studied. All these are hard problems, and the issue of autonomy still remains unsolved. – Energy. Endurance of AUVs has increased from a few hours to tens of hours. Some systems now contemplate missions of days or months. However, extended endurance is at the expense of sensing capability and very limited transit speeds. Better battery-powered systems are been still researched. – Navigation. Early AUVs relied on dead reckoning for their navigation. Inertial navigation systems were available for expensive AUVs, while acoustic transponder navigation provided good accuracy but at a significant logistics cost. In recent years, AUVs have taken advantage of GPS when the vehicle surfaces. However, there is strong interest in being able to navigate relative to the environment within which the system exists. Without a doubt, an effective and robust positioning system will provide a significant increment in AUV capabilities. – Sensors/3D imaging. The paradigm of sensor radio integration has changed the way AUVs work today. New sensors are smarter, lower power, highly reliable, smaller in size, etc. In fact, much work continues now on the development of imaging systems with higher resolutions and artificial vision support. – Communications. In the past years, there have been significant advances in acoustic communications such that relatively low error rate communications are possible over ranges of kilometers at bit rates of a few kbps. In summary, there are still technological problems to be solved. The economic viability of the technology also has to be demonstrated, and AUVs must be proven in an operational regime in order to be finally applied in military missions.

2.2 Sound Propagation in Water In the water, acoustic signals propagate faster than in the air and their absorption at low frequencies is weak, making the water a good conductor of acoustic waves [10]. However, speed propagation of acoustic signals in the water depends on several parameters like salinity, depth, and temperature. Del Grosso’s equation, an up-to-date alternative to the UNESCO algorithm, determines the existing relationship among all these parameters [19]. The base velocity considered by Del Grosso’s equation is 1402.4 m/s, and instead of depending on the depth, this equation uses pressure as an input parameter. Obviously, the pressure depends on the depth and geographical latitude. Del Grosso’s equation was derived from fitting it to experimental data, and it has an associated uncertainty in its prediction of sound speed. There are also other proposed equations and algorithms, but the choice of a particular model depends on the accuracy and precision acceptable for the application in which it is being employed. In any case, the range of validity for Del Grosso’s equation is temperature

Analysis and Evaluation of the Positioning of Autonomous Underwater …

415

between 0 and 30 ◦ C, salinity between 30 and 40 parts per thousand, and pressure between 0 and 1000 kg/cm2 . The absorption of acoustic signals in the water also depends on some properties such as temperature, salinity, and acidity of the water, as well as the frequency of the acoustic signal [9]. This absorption generates some losses in the transmitted energy from the source to the receiver. However, the major contribution to transmission losses is the spreading of the acoustic wave as it propagates away from the source. It is well known that sound spreads such that its intensity decreases exponentially with distance.

2.3 Trilateration Techniques Trilateration is the process of determining absolute or relative locations of points by the measurement of distances to known locations [17]. When it is possible to establish the intensity of some transmitted signals, the most common trilateration techniques are RSSI, ToA, and TDoA. However, if it is also possible to establish the angle of arrival of the transmitted signals, the trilateration technique knows as angle of arrival (AoA) might be employed. Any of these techniques can be implemented by the vehicle (i.e., client-based positioning) or the surrounding infrastructure (i.e., infrastructure-based positioning) [4]. In the case of a client-based positioning, the surrounding infrastructure includes N acoustic or radio-frequency (RF) sources transmitting some specific signals, and the vehicle listens to those transmissions. In the case of an infrastructure-based positioning, the vehicle emits an acoustic or RF signal and N transducers in the surrounding infrastructure listen to that transmission. AUVs will prefer client-based positioning techniques, since the autonomy of these vehicles is increased and less energy is wasted listening rather than transmitting acoustic signals. Because of that, the main trilateration techniques are described as a client-based implementation, although an infrastructure-based positioning is also possible for each of the variants. RSSI. This technique measures the strength of the signals received from the surrounding acoustic sources at known locations. Since the intensity of a signal is attenuated by the travelled distance, the distance of each source can be modeled as a function of its intensity. However, this modeling is hard to achieve accurately due to multipath interference, path loss, Doppler distortion, and other fading effects. An important advantage of this technique is that no time synchronization is required between sources and receivers. ToA. In this technique, the distance is computed as a function of the time required by the signal to travel from the source to the receiver. Since the acoustic signals maintain almost constant speed, the accuracy of this trilateration technique is very good. However, the main problem here is to keep synchronized sources and receivers in order to determine the time of flight of each transmitted signal. Geometrically, both

416

E. V. Carrera and M. Paredes

Fig. 1 Geometry employed by ToA/RSSI and TDoA using three sources

RSSI and ToA describe circumferences centered in the sources and the position of the vehicle is determined by the intersection of those circumferences as explained in Fig. 1. TDoA. This method determines the vehicle position measuring the time differences between every pair of received signals. Geometrically, the differences in time generate hyperbolas that intersect in the position of the vehicle as shown in Fig. 1. This technique requires that only the acoustic sources are synchronized among them, being clearly advantageous an infrastructure-based TDoA positioning technique as there is only one source.

3 Underwater Positioning Simulation In order to analyze and evaluate different positioning techniques for AUVs, an underwater vehicle simulator has been developed. The technical details of the simulator and its default configuration are presented in the following subsections.

3.1 Simulation Platform The simulator was implemented in MATLAB® and includes a very complete model of the sound propagation in water. The acoustic sources can be configured in space and number with specific directivity, while the hydrophones are always omnidirectional. The speed propagation is modeled according to Del Grosso’s equation, and the absorption of the sound is modeled by the Ainslie and McColm algorithm. After determining the existing response to the discrete unit impulse between any source

Analysis and Evaluation of the Positioning of Autonomous Underwater …

417

and the receivers, the response to a discrete pulse is computed. Obviously, multipath interference and path loss are considered in these calculations in order to obtain a realistic pulse response. Using the corresponding responses to the pulses emitted by the acoustic sources, the simulator is able to compare RSSI, ToA, and TDoA algorithms using both client-based and infrastructure-based trilateration techniques. The computed positions through these methods can be compared for any specific AUV position, or the average error for a fraction of the whole volume can be determined. The fraction of volume used in this last case is the centered volume limited by a scaling factor between 0 and 1. Figure 2 allows us to understand better the volume corresponding to a specific r factor. In addition, the simulator implements an intuitive graphical interface shown in Fig. 3. Through the interface, the final user can configure the size of the water container, its latitude, the water temperature and salinity, and the reflection coefficients of every wall in the container. Moreover, the sampling frequency for digital processing and the duration of the pulses emitted by the sources are also configurable.

Fig. 2 Volume determined by a scaling factor r

Fig. 3 Graphical interface of the implemented simulator

418

E. V. Carrera and M. Paredes

3.2 Default Configuration The default configuration of the simulator considers the parameters found in a lake at the Andean region of Ecuador. Basically, we use a latitude of 0.216667◦ , a temperature of 4 ◦ C, and a salinity of 10 ppm. The reflection coefficients are 1.0 for the surface and 0.46 for the other walls of the water container. The default size of the water container is 5 × 5 × 3 m, and the average errors are obtained examining points separated 0.2 m in each direction along the volume scaled by a factor r (see Fig. 2). Lastly, the sampling frequency used in the computation of the unit impulse response is 160 kHz and the duration of the pulses is 500 ms.

4 Results The main results obtained by the simulator are presented below. A parameter space study is also included before discussing the main implications of these values.

4.1 Accuracy Results The default configurations for comparing the three main trilateration techniques are: five noise sources, temperature of 10 ◦ C, and a water container of 5 × 5 × 3 m. The results shown in Table 1 confirm that TDoA is the most accurate technique. Although ToA has similar results, ToA requires additional time synchronization among acoustic sources and receivers. Remember that TDoA only requires that the acoustic sources are synchronized. Other import observation from Table 1 is the importance of keeping away the vehicle from the acoustic sources. When the operational zone of the AUV is contained inside an r factor of 0.8 (i.e., approximately half of the total volume), the accuracy is lower than 10 cm. Furthermore, if the operational zone is inside an r factor of 0.6, the accuracy could be as low as 1 cm. Results for an infrastructure-based positioning instead of a client-based positioning also show the same trend.

Table 1 Average position accuracy (in cm) for different trilateration techniques Technique Scaling factor r (volume) 1.0 (100%) 0.8 (51%) 0.6 (22%) RSSI ToA TDoA

326.2 163.0 162.8

29.0 9.6 9.6

3.4 1.1 1.1

Analysis and Evaluation of the Positioning of Autonomous Underwater …

419

Table 2 Average position accuracy (in cm) for TDoA using different configurations Temperature (◦ C) Scaling factor r (volume) 1.0 (100%) 0.8 (51%) 0.6 (22%) 4 10 20 Number of sources

130.8 5.6 162.8 9.6 169.5 12.1 Scaling factor r (volume) 1.0 (100%) 0.8 (51%)

4 5 6 Size of the container (m)

114.0 58.3 162.8 9.6 25.7 3.2 Scaling factor r (volume) 1.0 (100%) 0.8 (51%)

0.6 (22%)

5×5×3 10 × 10 × 6 20 × 20 × 12

162.8 177.0 107.6

1.1 1.4 1.4

9.6 6.0 3.3

1.0 1.1 1.1 0.6 (22%) 25.7 1.1 0.9

4.2 Parameter Space Study In order to determine how some parameters included in the simulator affect the accuracy of the TDoA technique, Table 2 presents accuracy values for different temperatures, number of noise sources, and size of the water container. We can see that low temperatures slightly improve accuracy, even for high values of r . On the other hand, the number of acoustic sources significantly affects the accuracy of TDoA. Basically, five acoustic sources are the minimum number of allowed sources for getting an acceptable accuracy. Working with six acoustic sources, instead of five, still causes a noticeable improvement. In addition, although the initial configuration of the water container might seem a bit small, we can see that accuracy values are similar for larger configurations of the container. In the case of a container of 20 × 20 × 12 m, the positioning accuracy is lower than 5 cm for an r factor of 0.8.

4.3 Discussion These results definitively place TDoA as the best trilateration technique without using angles of arrival. Results are pretty good, especially for low temperatures, using at least five noise sources, and big containers of water. In summary, an accuracy of a few centimeters can be easily achieved by TDoA.

420

E. V. Carrera and M. Paredes

The main problem of TDoA is the high number of noise sources required in the surrounding infrastructure. This problem is enlarged if military applications are considered, since enemy forces can easily be aware of the presence of such devices in the battlefield. Despite that, we have already started the hardware implementation of the TDoA technique to validate previous results in the field. In order to detect the starting point of the acoustic pulses, a variation of Goertzel’s algorithm [11] has been implemented in a Raspberry Pi motherboard. At this moment, time difference errors lower than 0.1 ms are achieved with this hardware platform.

5 Conclusions This work demonstrates that trilateration techniques are good enough to provide a positioning accuracy below 10 cm in a broad range of configurations. The implemented simulator considers a complex model for the sound propagation in the water, providing acceptable trends in the comparison of trilateration techniques based on signal intensities. We expect that this type of technology can be used soon in military application after minimizing the number of noise sources. For the reduction of noise sources, as a future work, we plan to extend the simulator for comparing the AoA technique to the other trilateration techniques. In fact, we would like to study microphone/hydrophone arrays [13] to combine AoA and TDoA techniques in order to reduce significantly the number of noise sources. Finally, the hardware implementation of the TDoA technique is almost finished and we expect to make tests in the field using a commercial ROV in a lake of Ecuador to validate the results presented along with this document. Acknowledgements This work was partially supported by the Universidad de las Fuerzas Armadas ESPE under Research Grant PREDU-2016-005.

References 1. Allard, Y., Shahbazian, E.: Unmanned underwater vehicle (UUV) information study. Technical report, DRDC-RDDC-2014-C290, OODA Technologies Inc., Montreal, Canada (2014) 2. Antonelli, G.: Underwater Robots, 4th edn. Springer, Berlin (2018) 3. Baek, H., Lim, J.: Design of future UAV-relay tactical data link for reliable UAV control and situational awareness. IEEE Commun. Magaz. 56(10), 144–150 (2018). https://doi.org/10. 1109/MCOM.2018.1700259 4. Bensky, A.: Wireless positioning technologies and applications. Artech House (2016) 5. Button, R.W., Kamp, J., Curtin, T.B., Dryden, J.: A survey of missions for unmanned undersea vehicles. RAND National Defense Research Institute, Santa Monica, CA (2009) 6. Carrera, E.V., Perez, M.S.: Event localization in wireless sensor networks. In: 2014 IEEE Central America and Panama Convention (CONCAPAN XXXIV), pp. 1–6. IEEE, Panama, Panama (11 2014)

Analysis and Evaluation of the Positioning of Autonomous Underwater …

421

7. Cook, K.L.B.: The silent force multiplier: The history and role of UAVs in warfare. In: 2007 IEEE Aerospace Conference. pp. 1–7 (March, 2007) 8. Damian, R.G., Jula, N., Paturca, S.V.: Autonomous underwater vehicles - achievements and current trends. Sci. Bull. Nav. Acad. 21(1), 85–89 (2018) 9. Doonan, I.J., Coombs, R.F., McClatchie, S.: The absorption of sound in seawater in relation to the estimation of deep-water fish biomass. ICES J. Mar. Sci. 60(5), 1047–1055 (2003) 10. Etter, P.C.: Underwater Acoustic Modeling and Simulation. CRC Press (2018) 11. Jiang, J., Brewer, R., Jakubowski, R., Tan, L.: Development of a piano frequency detecting system using the goertzel algorithm. In: 2018 IEEE International Conference on Electro/Information Technology (EIT), pp. 0346–0349. IEEE (2018) 12. Khan, R., Ercan, M.F., Metarsit, L., Le, A.N.S., Lim, K.V., Tan, W.T., Ang, J.L.: Underwater navigation using maneuverable beacons for localization. In: OCEANS 2016 MTS/IEEE, pp. 1– 5. IEEE, Monterey, CA (2016) 13. Paredes, D., Apolinario, J.: Shooter localization using microphone arrays on elevated platforms. In: 2014 IEEE Central America and Panama Convention (CONCAPAN XXXIV), pp. 1–6. IEEE, Panama, Panama (2014) 14. Paull, L., Saeedi, S., Seto, M., Li, H.: AUV navigation and localization: A review. IEEE J. Ocean. Eng. 39(1), 131–149 (2014) 15. Perez, M.S., Carrera, E.V.: Acoustic event localization on an arduino-based wireless sensor network. In: 2014 IEEE Latin-America Conference on Communications (LATINCOM), pp. 1– 6. IEEE, Cartagena, Colombia (2014) 16. Solodov, A., Williams, A., Al Hanaei, S., Goddard, B.: Analyzing the threat of unmanned aerial vehicles (UAV) to nuclear facilities. Secur. J. 31(1), 305–324 (2018) 17. Stojanovi´c, D., Stojanovi´c, N.: Indoor localization and tracking: methods, technologies and research challenges. Facta Univ. Ser. Autom. Control. Robot. 13(1), 57–72 (2014) 18. Wadoo, S., Kachroo, P.: Autonomous underwater vehicles: modeling, control design and simulation. CRC Press (2016) 19. Wong, G.S., Zhu, S.m.: Speed of sound in seawater as a function of salinity, temperature, and pressure. J. Acoust. Soc. Am. 97(3), 1732–1736 (1995)

Part XII

Chemical, Biological and Nuclear Defense

Evaluation of 3D Printing Parameters on the Electrochemical Performance of Conductive Polymeric Components for Chemical Warfare Agent Sensing Joseane R. Barbosa, Pedro H. O. Amorim, Mariana C. de O. Gonçalves, Rafael M. Dornellas, Robson P. Pereira and Felipe S. Semaan

Abstract Aiming at the use of 3D printing as a tool in the design of electronic components and sensors, the present work has developed a preliminary evaluation of the influence of printing conditions on the electrochemical behavior of PLA and commercial carbon conductor-based composites. For this, characterization experiments were carried out by means of infrared spectroscopy and thermal analysis, pointing to a clear variability of behavior of different types of PLA, as well as the influence of this variation on the thermal behavior of the composites, which consequently influences the processes of 3D printing. Experiments allowed a speculation of the conductor load present in the conductive filament (about 20.5%), although it was not possible to define its category. The print modes, in turn, proved to be a decisive factor for the electroanalytical behavior in front of a redox probe, being responsible for improving or worsening the resolution, or favoring the exchange in a given direction. In all cases, the relations between currents (cathodic and anodic) were acceptable, although the loss of resolution, translated in the form of spacing between peaks, has greatly contaminated the calculation of electroactive areas in some cases. Clearly, an optimization of the printing process, as well as a better control in the particle size, will allow obtaining high electroactive areas, with adequate resolution, even with low driver loads, which shows a promising line currently in development in our Group. Keywords Conductive 3D printed component · Surface effects · Electrochemical sensing

J. R. Barbosa · P. H. O. Amorim · M. C. de O. Gonçalves · R. M. Dornellas · F. S. Semaan (B) Analytical Chemistry Department, Fluminense Federal University, Niteroi RJ, Brazil e-mail: [email protected] R. P. Pereira · F. S. Semaan Military Institute of Engineering, Rio de Janeiro RJ, Brazil © Springer Nature Singapore Pte Ltd. 2020 Á. Rocha and R. P. Pereira (eds.), Developments and Advances in Defense and Security, Smart Innovation, Systems and Technologies 152, https://doi.org/10.1007/978-981-13-9155-2_34

425

426

J. R. Barbosa et al.

1 Introduction Among the current trends in electroanalytical chemistry, the development of customized sensors for lab-free applications has received special attention. Aiming to combine sensitivity, selectivity and velocity with small sample amounts, such sensors are highlighted in the literature as a promising technology among other detection systems for defense proposals, due to reduction of the sample-operator contact [1, 2]. In this regard, some reports summarize current efforts performed in order to optimize analytical capabilities of in locu analysis. Various examples of electrodes have been described in the literature for defense-related applications, and the next section lists some of these works. Ismail and Matsuda [3] have developed a procedure for organophosphate chemical warfare agent determination using a 3-cyanopropyltriethoxysilane film by cyclic voltammetry, reaching linearity in the range of 0.5–10 pg mL−1 , with a detection limit of 0.25 pg mL−1 . Mahyari [4] performed amperometric determinations of thiodiglycol using electrodes based on a mixture of nano-sheets of graphene, silver nanoparticles and ionic liquid (octylpyridinium hexafluorophosphate), detecting a pair of well-defined redox peaks in a range of 10–3700 µmol L−1 , with detection limit of 6 µmol L−1 . Sikarwar et al. [5] analyzed thiodiglycol stimulant using the xanthine oxidase as electrode modifier onto a gold substrate. The linear response in cyclic voltammetry was obtained in the range between 1.94 and 9.70 mmol L−1 . Elliot et al. [6] developed a square wave voltammetric procedure for the quantification of pyocyanin using a transparent carbon microelectrode (bare or modified by gold nanoparticles). Hondred et al. [7] prepared printed graphene electrodes via non-inkjet lithography. These electrodes were modified by platinum nanoparticles to improve electrical conductivity. The biosensor was able to rapidly measure an insecticide with 3 nmol L−1 detection limit. Wang et al. [8] initiated a different approach while coupling a detection platform to popular smartphones by a plug-able to perform the electrochemical measurement for the determination of low nitrate concentrations in water, being a very reliable proof of concept for lab-free applications. Shim et al. [9] also used a portable potentiostat, performing square wave voltammetry measurements with electrochemical sensors, exchanging all data via Bluetooth to a smartphone. Analytical signals for the observed peaks were compared to those obtained on a lab-based commercial potentiostat, without significative difference. Fan et al. [10] tested nanocomposites synthesized by graphene and gold nanoparticles as customized agent for the determination of some specific enolases using a microfluid analytical paper-based device controlled by smartphone. Such system was used under differential pulse voltammetric conditions, allowing a good linear relationship ranging from 1 to 500 ng mL−1 with a detection limit of 10 pg mL−1 .

Evaluation of 3D Printing Parameters …

427

Kanchi et al. [11] and Ainla et al. [12] described different characterization projects for a universal wireless electrochemical system in a wide range of voltammetric techniques, with compatible results from all assessed systems. Colozza et al. [13] reported a new paper-based electrochemical sensor for detection of sulfur mustards. The working electrode was modified with a nanocomposite consisting of Prussian blue and carbon black nanoparticles. From these examples, it is easy to conclude that use of smartphones and other controlling systems is no longer a challenging task but, on the other hand, customizing sensors is still a decisive step for a reliable application. It becomes important to note that many reports describe the use of carbon-based substrates, which is not only cheap and easy to manipulate, but also versatile in terms of chemical modification. Carbon composites have numerous advantages over conventional substrates such as easy surface renewal (allowing the removal of adsorbates), fabrication in different shapes and sizes, application in a wide range of pH and working potential, possibility of chemical modification on the surface of the electrode or in the bulk itself, low production cost, among others [14–20]. Among many available binders for 3D printing, poly(lactic acid) (PLA) stands for a promising substrate, being a biodegradable aliphatic polyester with melting point between 130 and 180 °C, derived from renewable resources such as corn, rice, cassava roots and sugarcane. Among its characteristics, PLA has a high mechanical resistance that promoted the study of the same in the use of several materials [14, 21–23]. Various forms of carbon-based particles have been used as conductive phase in composite materials suitable for electrochemical sensors [20]. Among these, graphite [15], vitreous carbon particles [24], graphenes [25], nanotubes [26] and fullerenes [27], among others, have shown promising properties. Graphene consists of a monoatomic layer formed by carbon atoms with sp2 hybridization, with a surface area that is almost twice as large as that of single-walled carbon nanotubes [28], and with a long p-conjugated system, the electrons are confined in two dimensions, which gives the graphene remarkable properties such as high electrical and thermal conductivity, good transparency, good mechanical strength, inherent flexibility and huge specific surface area [29–31]. 3D printing is a technology that allows the fabrication of solid 3D objects, with high precision, and with a great deal of flexibility over the adopted design, from a digital file. The material used for printing is considered low cost, and the production process generates minimal waste. There is a commercial drawback, related to fabrication of conductive parts by 3D printing, using carbon black, graphene and nanotubes, which reflects the higher cost compared to pure non-conductive filaments. One important point is the lack of information about the load (%) and nature of the conductive phase employed, since a significant part of the producers and suppliers do not declare such information, offering the product on the market as a “conductive filament” without further technical details. Guo et al. [32] presented in their article electrodes fabricated from PLA nanocomposite and multilayer nanotubes by 3D printing with a helical structure. In addition to high electrical conductivity, this system has good rigidity, excellent sensitivity and selectivity.

428

J. R. Barbosa et al.

Vernadou et al. [33], in his paper on 3D-printed graphene electrodes, deal with the use of graphene printed with pyramidal geometry using a double-extruded 3D FDM type printer for application on lithium-ion batteries. Palenzuela et al. [34] used graphene-PLA composite electrodes imprinted in ring and disk form in a 3D process of melting and deposition molding and, after characterization by cyclic voltammetry and scanning microscopy, applied in the detection of picric acid and ascorbic acid. In spite of the clear application potential and the variety of commercial filaments available, it is interesting to consider that few previous works take into account the printing conditions or describe them in detail, although the interfacial properties obtained are directly affected by the relationship between nature, properties, charge and dimensions of the dispersed conductive phase, extrusion and printing conditions. In such context, the present work aims to exploit the influence of printing conditions over the electrochemical performance of carbon-PLA sensors as prototypes for customized sensor strips for lab-free use.

2 Materials and Methods 2.1 3D Printing of Prototypes Prototypes were prepared by direct printing, using a Voolt 3D GI3 3D printer, with Slic3r software. These test bodies/work electrodes with dimensions (2.8 × 1.5 × 0.1 cm) (length, width and thickness) were printed in horizontal and vertical modes and with Linear type interfacial geometries, Helbert, Archiemedean, Octagram and concentric, with filler density of 10 and 100%. Filaments of PLA (FPLAP), and conductive PLA (FPLAC), supplied by 3D Lab® (Brazil), Proto-Pasta® (USA), respectively, both with a diameter of 1.75 mm, were used as feeder filaments. Table and nozzle temperatures were adjusted following properties described by the manufacturers and modes suggested by the printer default (nozzle temperature between 195 and 225 °C, bed temperature circa 60 °C).

2.2 Thermal Characterization of the Material The thermal behavior was evaluated by thermogravimetric analysis (TGA and DTG), using a Shimadzu thermogravimetric analyzer model TGA-60, sample mass about 20 mg of pre-sampled sample in alumina samples and under heating at 10 °C min−1 , at room temperature at 900 °C, under a dynamic atmosphere N2 inlet at a flow rate of 40 mL min−1 . Further additional information was obtained by means of differential scanning calorimetry (DSC), using a Shimadzu differential scanning calorimeter model DSC-60, between room temperature and the thermal stability

Evaluation of 3D Printing Parameters …

429

limit temperature of each evaluated material, under a heating rate of 10 °C min−1 in heating-cooling-heating mode, under N2 atmosphere.

2.3 Electroactive Surface Response and Geometries Voltammetric cells printed directly on ABS with diameter 2.2 and 2 cm in height were used, with total volume of 5 mL in electrolyte solution. The working electrode was positioned at the bottom, in a window allowing an exposure of 0.385 cm2 in the geometric area of the sensor, and the sealing was performed with the aid of O’ring and acetic curing silicone glue. The electroactive areas were measured by a system of three electrodes (Ag|AgClsat ) as reference, surgical steel needle as against electrode and the printed working electrodes to be tested) carefully inserted into the voltammetric cell already described, using a potentiostat, Ivium® CompactStat model (Ivium Technologies, Netherlands). Cyclic voltammetry (CV) in potassium hexacyanoferrate (5 mmol L−1 ) and 0.5 mol L−1 potassium chloride was used, and the Randles–Sevcik ratio, previously presented and defined, was used for calculations. Three successive scans were performed in the potential range from −1.0 to 1.5 V, under different scan rates (10, 20, 30, 50, 75 and 100 mV s−1 ), as usual.

3 Results and Discussions 3.1 3D Printing of Prototypes Conductive FPLAC blades were obtained in the dimensions previously described by the FDM process employing different conditions. Such blades were suitably positioned to the bottom of BIA cell in ABS and sealed with silicone glue, having a fixed and calculated exposed area. Due to printing parameters, the surface is more roughened or regular, which must be evaluated, additionally, based on the dimensions of the conductive particles.

3.2 Thermal Characterization of the Material Considering the sample as an electroactive composite, it is possible to infer that the homogeneity of charge dispersion (%) of the conductive phase, as well as particle size and thermal conductivity, will directly influence the heat conduction process and consequently in the relevant thermal events observable. In general, information

430

J. R. Barbosa et al.

about the nature and loading of the conductive phase in these filaments are not provided, which points to a clear need for prior characterization by thermal techniques, aiming to evaluate not only the thermal decomposition profile, but possible faults in the final components obtained by the formation of conductor clumps or regions of heterogeneity in the parts as a consequence of large differences in thermal behavior between insulating (PLA) and conductive (undeclared) phases. Both materials have different thermal stabilities, the FPLAC being more thermally resistant, since the thermal decomposition starts at 312.12 °C (Tonset ), about 100 °C higher compared to FPLAP. Another interesting aspect is the third mass loss in FPLAC (25.4%), compared to the event in the same temperature range for FPLAP (4.9%), indicating a percentage of conductive phase of about 20.4% and in agreement with the observed temperatures for carbon-based phases in several reports [21, 35]. Certainly, such an approach is semi-quantitative, since the content is not stated, as well as the nature of the driver, besides the fact that these are different PLA structures. This behavior suggests that the conductor filler dispersed in PLA matrix may increase the thermal stability of the material as a function of the colligative properties, making it more resistant. It was also possible to note in the DTG two small mass loss processes of FPLAP, between 84.94 and 109.47 °C and between 217.80 and 248.69 °C, the latter shortly before thermal decomposition, which may suggest a minimum presence of moisture and additives used in the manufacture of the filaments, respectively. Thermal decomposition occurs between 248.69 and 338.51 °C, where there is a large mass loss, followed by another small loss between 559.2 and 900.0 °C, with no residue in the crucible at the end of the test, as reported in the literature [36]. Differential scanning calorimetry (DSC) for the FPLAP and FPLAC samples was carried out by heating-cooling-heating mode [37] between room temperatures ~25 and 220 °C, based on previous results from TG curves. The percentage of crystallinity was obtained through Eq. 1, where Hc is the enthalpy of crystallization in (J/g), Hm is the enthalpy of fusion, being obtained by the area of the peak in the DSC curve, and Hm 100 is the melting enthalpy of 100% crystalline PLA, which according to the literature is given by 93.7 J/g [38, 39]. Xc (%) =

|Hm − Hc | × 100 H 100%

(1)

The PLA has a characteristic heating profile, with three thermal events—a glass transition around 60 °C, crystallization around 100 °C and melting around 160 °C—and can vary between 130 and 180 °C. Comparing the first and second heating cycles, it was observed that the FPLAC had its DSC profile more affected by the cooling than the FPLAP; however, the glass transition temperatures (Tg ) did not change significantly. In both cases, the glass transition is followed by a small endothermic peak, which suggests some degree of physical aging of the materials, since the polymers have undergone previous processing when the filament is made. The crystallization process characterized by the exothermic peak at 101.96 °C in the first heating for the FPLAC does not occur after the cooling process in addition

Evaluation of 3D Printing Parameters …

431

to a substantial decrease of the melt peak at 150.96 °C, suggesting that the solid upon cooling did not form a more organized or crystalline structure due to the presence of the conductive filler. It was observed that FPLAP presented higher crystallinity (13.97%) than FPLAC (10.21%), which was improved with cooling since the crystallization peaks (first and second) besides the melting became better defined, pointing to a better reorganization of the polymer chains. The melting temperatures Tm for the two samples show a significant difference, which is attributed mainly to the structural difference between the two materials, possibly due to different proportions between the l-lactic acid/d-lactic acid type monomers along the polymer, resulting from different synthesis processes, in which the chains can aggregate in points with a greater proportion of l or d, forming clusters that make the material more crystalline, as described in Kumar et al. [40] and Ceregatti et al. [41]. On the other hand, the presence of conductive filler dispersed in the PLA chains promotes a better thermal stability; however, it makes it difficult to reorganize the chains, making the material less crystalline, after successive cycles of fusionsolidification, unlike that presented by Ceregatti et al. [41], in which the material presented higher crystallinity in the second heating with CNT in concentrations up to 5%.

3.3 Electroactive Surface Response and Geometries A batch-injection cell was used to determine the electrode area and subsequent evaluation of the best printing conditions (direction, fill type and surface), which was filled by electrolyte-holder for later activation and measurements. Subsequent addition of adequate mass of the probe allowed, under homogenization, to reach the desired concentration of the same and then the measurements of the electroactive areas. The determination of the electroactive area was performed using cyclic voltammetry between −1 and +1.5 V versus Ag|AgClsat , at different scanning rates and three consecutive cycles for each condition. A preliminary activation step in the same window was performed at a speed of 100 mV s−1 . Parameters such as peak shift, and the relationship between anodic and cathodic currents, as well as the linear relations between current and velocity values (Ip × v1/2 ) were used to evaluate the best conditions of application of the material as an electrode surface. In addition, the equation of Randles and Sevcik was used to determine the electroactive area. Different electrode prototypes (with dimensions 2.8 × 1.5 × 0.1 cm) obtained in the horizontal and vertical modes and with Linear, Heli, Archiemedean, Octagram and concentric interfacial geometries, with fills of 10 and 100%, were adapted to the described BIA cell, the seal being obtained by the use of acetic curing silicone glue, so that the exposed geometric window remained constant throughout all the tests. Based on the voltammetric profile, as well as the evaluation of the aspects calculated, it was possible to notice that in this impression condition, the electroactive area (0.070 cm2 , 18.18% of the exposed geometric area −0.385 cm2 ) is almost

432

J. R. Barbosa et al.

corresponding to the percentage of conductive phase (20.45%), safeguarding the approximations and errors. Comparing 100% with 10% of fill, following the same printing pattern, a peak resolution loss of 10% is observed, according to the observed peak distances, which has a direct impact on electro-area measurements, as calculated (0.158 cm2 ). However, in both fills, behaviors in cathodic and anodic regions are compatible with each other and between the different test bodies. In this way, it was decided by the evaluation, at this moment, of the vertical direction of impression with 10% of filling. Based on the results and aspects of the printing, it is noticed that using vertical printing, the results are also contaminated by resolution loss, as expected, since other characteristics such as resolution and filling become less important, in this way, the subsequent evaluations in horizontal printing. Helbert geometry now shows a lower spatial spike and compatible behaviors in the anodic and cathodic scanning regions, and a calculated electroactive area of 0.375 cm2 , 97.40% of the exposure window (percentage well higher than expected given the determined percentage of conductive phase). This was followed by an exploration of Archiemedean geometry, resulting in profiles not only better resolved (with spikes between 0.325 V peaks), but also as better Ipa /Ipc ratios, suggesting higher reliability in the determination of electroactive area, which in this case was 0.224 cm2 . Following the optimization, in order to explore all allowed modes, impressions were made in the Octagram and concentric modes. Under printing on Octagram, it was possible to observe the worst resolution (1.05 V between peaks), culminating in a terribly contaminated electroactive area, calculated at 0.409 cm2 ; (0.664 V), the area was the lowest, demonstrating loss of load exchange capacity. All the results for electroactive areas were calculated through the direct application of the Randles–Sevcik equation.

4 Final Remarks The present work developed a preliminary evaluation of the influence of printing conditions on the electrochemical behavior of PLA and carbon-based composites. It is clear the behavior variability of different types of PLA, as well as the influence of this variation on the thermal behavior of the composites, which consequently influences the 3D printing processes. Experiments allowed a speculation of the conductor load present in the conductive filament (about 20.5%), although it was not possible to define its category. Such nature as well as particle size are vital aspects for a better modeling of heat transfer phenomena by the composite, which is clearly related to the printing mode. The print modes, in turn, proved to be a decisive factor, being responsible for improving or worsening the resolution, or favoring the exchange in each direction. In all cases, the relations between currents (cathodic and anodic) were acceptable, although the loss of resolution, translated in the form of spacing between peaks, has greatly contaminated the calculation of electroactive areas in some cases. With this, it can be seen that the best orientation among the evaluated geometries was Helbert.

Evaluation of 3D Printing Parameters …

433

Notably, an optimization of the printing process, as well as a better control in the particle size, will provide higher electroactive areas, with adequate resolution, even with low driver loads, which shows a promising line currently in development in our Group, line is dedicated to the preparation of different filaments for future direct printing of disposable conductive components for lab-free applications.

References 1. Cinti, S.: Electroanalysis: towards a “laboratory-free” approach. SM Anal. Bioanal. Tech. 1(1), 1004 (2016) 2. Wang, J.: Analytical Electrochemistry. 3ª edição. Weinheim: Ed. Wiley-VCH (2006) 3. Ismail, W.N.M., Matsuda, A.: Electrochemical imprinted sol-gel films detection of organophosphate chemical warfare agent. Def. S T Tech. Bull. 8(1), 9–16 (2015) 4. Mahyari, M.: Electrochemical sensing of thiodiglycol, the main hydrolysis product of sulfur mustard, by using an electrode composed of an ionic liquid, graphene nanosheets and silver nanoparticles. Microchim. Acta 183(8), 2395–2401 (2016) 5. Sikarwar, B., Shama, P.K., Singh, V.V., Boopathi, M., Singh, B., Jaiswal, Y.K.: Xanthine oxidase immobilized electrode for electrochemical sensing of chemical warfare agent simulant. Sens. Lett. 15(3), 227–235 (2017) 6. Elliot, J., Simoska, O., Karasik, S., Shear, J.B., Stevenson, K.J.: Transparent carbon ultramicroelectrode arrays for the electrochemical detection of a bacterial warfare toxin, pyocyanin. Anal. Chem. 89(12), 6285–6289 (2017) 7. Hondred, J.A., Breger, J.C., Aalves, N.J., Trammel, S.A., Walper, S.A., Medintz, I.L., Claussen, J.C.: Printed graphene electrochemical biosensors fabricated by inkjet maskless lithography for rapid and sensitive detection of organophosphates. ACS Appl. Mater. Interfaces 10(13), 11125–11134 (2018) 8. Wang, X., Gartia, M.R., Jiang, J., Chang, T.-W., Qian, J., Liu, Y., Liu, X., Liu, G.L.: Audio jack based miniaturized mobile phone electrochemical sensing platform. Sens. Actuators, B: Chem. 209, 677–685 (2015) 9. Shim, W., Han, J.-H., Kim, S., Kwon, H.J., Pak, J.: A portable potentiostat with Bluetooth communication for square wave voltammetry measurement. Trans. Korean Inst. Electr. Eng. 65(4), 622–627 (2016) 10. Fan, Y., Liu, J., Wang, Y., Luo, J., Xu, H., Xu, S., Cai, X.: A wireless point-of-care testing system for the detection of neuron-specific enolase with microfluidic paper-based analytical devices. Biosens. Bioelectron. 95, 60–66 (2017) 11. Kanchi, S., Sabela, M.I., Mdluli, P.S., Inamuddin, I., Bisetty, K.: Smartphone based bioanalytical and diagnosis applications: a review. Biosens. Bioelectron. 102, 136–149 (2018) 12. Ainla, A., Mousavi, M.P.S., Tsaloglou, M.-N., Redston, J., Bell, J.G., Fernández-Abedul, M.T., Whitesides, G.M.: Open-source potentiostat for wireless electrochemical detection with smartphones. Anal. Chem. 90(10), 6240–6246 (2018) 13. Colozza, N., Kehe, K., Popp, T., Steinritz, D., Moscone, D., Arduini, F.: Paper-based electrochemical sensor for on-site detection of the Sulphur mustard. Environ. Sci. Pollut. Res. 1–12 (2018) 14. Buzzetti, P.H.M., Oliveira, G.C., Azevedo, A.L.M., Michel, R.C., Ponzio, E.A., Semaan, F.S.: Conductive composites as versatile substrates for modeling of interfaces and sensors. In: Wythers, M.C. (ed.) Advances in Materials Science Research, vol. 21, 1st edn, pp. 1–24. Nova Science Publisher, New York (2015) 15. Fonseca, C.A., Dos Santos, T.A.D., Resende, A.C., Semaan, F.S.: Paraffin-graphite based composite electrodes as substrate for the electrodeposition of conductive films with analytical purposes. Revista Virtual de Química 5(4), 538–540 (2013)

434

J. R. Barbosa et al.

16. Amorim, R.B.S., Rabelo, A.C., Bottecchia, O.L., Muñoz, A.A., Richter, E.M.: Desenvolvimento, caracterização e aplicação eletroanalítica de um compósito fluido de adesivo epóxi, grafite e ciclo-hexanona. Quim. Nova 33(6), 1398–1402 (2010) 17. Vasiliev, V.V., Morozov, E.V.: Mechanics and Analysis of Composite Materials, 1st edn. Elsevier Science Ltd. (2001) 18. Tallman, D.E., Petersen, S.L.: Composite electrodes for electroanalysis: principles and applications. Electroanalysis 2, 499–510 (1990) 19. McCreery, R.L.: Advanced carbon electrode materials for molecular electrochemistry. Chem. Rev. 108, 2646–2687 (2008) 20. Corb, I., Manea, F., Radovan, C., Pop, A., Burtica, G., Malchev, P., Picken, S., Schoonman, J.: Carbon-based composite electrodes: preparation, characterization and application in electroanalysis. Sensors 7, 2626–2635 (2007) 21. Silva, A.L., Corrêa, M.M., Oliveira, G.C., Michel, R.C., Semaan, F.S., Ponzio, E.A.: Development and application of a routine robust graphite/poly(lactic acid) composite electrode for the fast simultaneous determination of Pb2+ and Cd2+ in jewelry by square wave anodic stripping voltammetry. New J. Chem. 42, 19537–19547 (2018) 22. Pereira, R.B., Morales, A.R.: Estudo do Comportamento Térmico e Mecânico do PLA Modificado com Aditivo Nucleante e Modificador de Impacto. Polímeros 24(2), 198–202 (2014) 23. Milde, J., Hrušecký, R., Zaujec, R., Moroviˇc, L., Görög, A.: Research of ABS and PLA materials in the process of fused deposition modeling method. DAAAM International, Vienna (2017) 24. Kalcher, K., Svancara, I., Buzuk, M., Vytrias, K., Walcarius, A.: Electrochemical sensors and biosensors based on heterogeneous carbon materials. Monatsh. Chem. 140, 861–889 (2009) 25. Geim, A.K., Novoselov, K.S.: The rise of graphene. Nat. Mater. 6, 183–191 (2007) 26. Rivas, G.A., Rubianes, M.D., Pedano, M.L., Ferreyra, N.F., Luque, G.L., Rodrigues, M.C., Miscoria, S.A.: Carbon nanotubes paste electrodes: a new alternative for the development of electrochemical sensors. Electroanalysis 19(7–8), 823–831 (2007) 27. Miranda-Hernandez, M., González, R.: Characterization of carbon–fullerene–silicone oil composite paste electrodes. Carbon 43(9), 1961–1967 (2005) 28. Scida, K., Stege, P.W., Haby, G., Messina, G.A., García, C.D.: Recent applications of carbonbased nanomaterials in analytical chemistry: critical review. Anal. Chim. Acta 691(1–2), 6–17 (2011) 29. Vieira, J.E.D., Vilar, E.O.: Grafeno: Uma revisão sobre propriedades, mecanismos de produção e potenciais aplicações em sistemas energéticos. Revista Eletrônica de Materiais e Processos 11(2), 54–57 (2016) 30. Zarbin, A.J.G., Oliveira, M.M.: Nanoestruturas de carbono (nanotubos, grafeno): quo vadis? Quim. Nova 36(10), 1533–1539 (2013) 31. Castro, M.O.: Síntese de grafeno pelo método CVD. 84 f. Dissertação de Mestrado em Física. Departamento de Física, Universidade Federal do Ceará, Ceará (2011) 32. Guo, S., Yang, X., Heuzey, M., Therriault, D.: 3D printing of multifunctional nanocomposite helical liquid sensor. J. Name 00, 1–3 (2012) 33. Vernadou, D., Vasilopoulos, K.C., Kenanakis, G.: 3D printed graphene-based electrodes with high electrochemical performance. Appl. Phys. A 123, 623 (2017) 34. Palenzuela, L.C.M., Novotný, F., Krupicka, P., Sofer, Z., Pumera, M.: 3D-printed graphene/polylactic acid electrodes promise high sensitivity in electroanalysis. Anal. Chem. (2018) 35. Azevedo, A.L.M., Oliveira, R.S., Ponzio, E.A., Semaan, F.S.: Sensor development exploiting graphite-epoxy composite as electrode material. In: IOP Conference Series: Materials Science and Engineering, vol. 97 (2015) 36. Auras, R., Harte, B., Selke, S.: An overview of polylactides as packaging materials. Macromol. Biosci. 4, 835 (2004) 37. Vidovic, E., Faraguna, F., Jukic, A.: Influence of inorganic fillers on PLA crystallinity and thermal properties. J. Therm. Anal. Calorim. (2016) 38. Du, Y., Wu, T., Yan, N., Kortschot, M.T., Farnood, R.: Fabrication and characterization of fully biodegradable natural fiber-reinforced poly (lactic acid) composites. Compos. Part B Eng. 56, 717–723 (2014)

Evaluation of 3D Printing Parameters …

435

39. Van de Velde, K., Kiekens, P.: Biopolymers: overview of several properties and consequences on their applications. Polym. Test. 21, 433–442 (2002) 40. Kumar, S., Bhatnagar, N., Ghosh, A.K.: Effect of enantiomeric monomeric unit ratio on thermal and mechanical properties of poly (lactide). Polym. Bull. 73, 2087–2104 (2016) 41. Ceregatti, T., Pecharki, P., Pachekoski, W.M., Becker, D., Dalmolin, C.: Electrical and thermal properties of PLA/CNT composite films. Revista Matéria 22(3) (2017)

Part XIII

Information and Communication Technology in Education

Interactive System to Improve the Skills of Children with Dyslexia: A Preliminary Study Jorge Buele , Victoria M. López , L. Franklin Salazar , Jordan-H. Edisson, Cristina Reinoso, Sandra Carrillo, Angel Soria , Raúl Andrango and Pilar Urrutia-Urrutia Abstract This paper describes a virtual system for the strengthening of linguistic abilities of children with dyslexia. To achieve this objective, an intuitive interface has been developed. The interface consists of three games (each with three levels of difficulty), all of them being part of a rehabilitation program. These applications combine visual and auditory messages that complement each other, in order to provide an immersive experience and to train more than one of the five senses at a time. The virtual environment has been developed in Unity 3D software, and the diffusion of sounds is performed through binaural hearing aids. To evaluate the performance of the presented proposal, a sample of eight infants (three girls and five boys) with J. Buele (B) · L. Franklin Salazar · J.-H. Edisson · C. Reinoso · S. Carrillo · P. Urrutia-Urrutia Universidad Técnica de Ambato, Ambato 180103, Ecuador e-mail: [email protected] L. Franklin Salazar e-mail: [email protected] J.-H. Edisson e-mail: [email protected] C. Reinoso e-mail: [email protected] S. Carrillo e-mail: [email protected] P. Urrutia-Urrutia e-mail: [email protected] J. Buele · V. M. López Universidad de Las Fuerzas Armadas ESPE, Latacunga 050104, Ecuador e-mail: [email protected] A. Soria Purdue University, Lafayette 47907, USA e-mail: [email protected] R. Andrango Universidad Técnica de Cotopaxi, 050102 Latacunga, Ecuador e-mail: [email protected] © Springer Nature Singapore Pte Ltd. 2020 Á. Rocha and R. P. Pereira (eds.), Developments and Advances in Defense and Security, Smart Innovation, Systems and Technologies 152, https://doi.org/10.1007/978-981-13-9155-2_35

439

440

J. Buele et al.

ages between 8 and 12 years was chosen. As an inclusion criterion, it is determined that the user must have an age greater than 7 and less than 13 years, and as a criterion of exclusion the presence of some visual impairment and/or an auditory disorder or other disorder that affect the educational environment. It must be taken into account that this is a preliminary study and that the experimental results in the patients will be seen in the long term. Therefore, to evaluate the acceptance of the system, and to make future corrections a SUS usability test has been applied, with the following result (82.5 ± 0.52). At the end of the execution of the applications, a report is issued in which the specialist can control how the patient has progressed during sessions. Keywords Human–computer interaction · Dyslexia · Open educational resources

1 Introduction Children school performance depends on their physical, emotional sensory, and psychological well-being [1, 2]. When the child suffers from a disorder or syndrome, their performance drops markedly, as well as its ability to relate to their environment [3–5]. Among the main learning disorders include dyscalculia, non-verbal learning disorder (NVT), disortography, dysgraphia, dysphasia, and dyslexia [6, 7]. The latter affects the development of the infant, preventing him from acquiring efficient reading and spelling skills despite adequate teaching and the absence of manifest sensorial and/or neural deficiencies [8]. This language learning disability affects one in five people in the world, and it can produce anxiety due to the subject imagining scenarios of mockery or social rejection [9, 10]. They also have difficulty remembering words, and problems distinguishing certain sounds, so they pronounce words with similar sounds, and this increases confusion [11]. Determining the origin of this syndrome coupled with the visual and/or auditory disjunction that it produces has led to the development of several related investigations. As shown in [8], several prominent sensory theories of dyslexia are evaluated. Particularly, the effects of experiencing scarce reading in the brain of affected children are presented as a priority cause. In [12], a study is presented, which aims to discover the nature of this disorder, and to achieve this complete brain regions are examined using the non-invasive auditory P300 paradigm in children with dyslexia and neurotypical controls. Cases of dyslexia occur in all known languages, although it differs according to the spelling, which triggers labor problems as well. That is why identifying it at an early age and carrying out an adequate therapeutic process is a high priority. In [13], we describe a cognitive behavioral therapy based on mindfulness called Mindfulness Based Rehabilitation of Reading, Attention & Memory (MBR-RAM ©). It uses techniques of visual meditation to improve attention focusing on formats assisted by therapists, and practice at home to rehabilitate reading deficits, lack of visual attention, lack of visual motor coordination, and visual memory. The proposal was tested in three children between 8 and 10 years of age, obtaining significant clinical and statistical improvements after a trial period of 6 months. In [14], several

Interactive System to Improve the Skills of Children …

441

technological proposals that seek to remedy the reading difficulties of children who have this disability are described. “Friendly” features are incorporated for the user, such as altering the size and format of the text and converting text to speech. As can be seen in [15], the use of technological tools contributes to the rehabilitation of users. A book that incorporates a set of mental games, mazes, and coloring pages, ideal for creative children with dyslexia, ADHD, Asperger syndrome, and autism (volume 3), is presented. The latter being an interesting proposal that incorporates 180 pages and that is distributed at an affordable cost to parents. The exponential development of technology has enabled greater and better benefits to be implemented, in this case focused on medical applications. The main tools currently available are virtual reality (VR) and interactive systems [16, 17]. Therefore, in [18] a virtual psychometric tool for the rehabilitation of patients with dyslexia is presented. Such tool is based on the Nintendo Wii video game system. The results show that it is a tool that improves attention, but it has no immediate effect on reading performance, suggesting a longer protocol as future work. The standard rehabilitation of this learning disorder involves a classic format of paper and pencil training, meaning that it presents boring and routine exercises to the subject. In this context and based on the previously cited studies, this paper proposes the implementation of a virtual system for the rehabilitation of children suffering from dyslexia. For this, an application has been developed in Unity software that incorporates three games with three levels of difficulty each (easy, medium, and hard), which must be executed in order for a better recovery process. In the first, word games are presented, allowing the user to develop spelling and reading skills. In the remaining games, the sense of hearing is added, and through this the identification words, and if their syllables are in the proper order. In order to corroborate the performance of this proposal, the respective experimental tests were carried out. This work is composed of four sections: Sect. 1 describes the introduction and the works related to the topic. Section 2 shows the description and design of the implemented system. Section 3 presents the tests and results, and finally in Sect. 4 the conclusions and future work are presented.

2 Methodology 2.1 Materials Hardware The implemented system has a commercial brand laptop, where the keyboard and mouse are used as input devices and the screen is the visual output to the patient. Binaural hearing aids are used to provide complementary therapy with sound outputs, when it is required. The general diagram of the project is presented in Fig. 1. Software The software used for the development of the interactive interface is Unity 3D and its complementary packages. This software has tools that allow the imple-

442

J. Buele et al.

Fig. 1 General diagram of the implemented system

mentation of virtual environments simulating real scenarios. You can develop environments in 2D and 3D; in this case, the application is set to work in two dimensions, since the child’s eye development is not yet complete, and taking into account the ergonomics. Video games must be executed in order, since the training program has been carried out with a pre-established sequence.

2.2 Methods Sample It is made up of eight children (three female and five male), and their demographic characteristics are described in Table 1. The criteria for the selection of the sample are: age (between 8 and 12 years) and the diagnosis of dyslexia (with/without dysgraphia). Exclusion criteria include: presence of other disorders in the educational setting and other psychological or neurological diseases. Also, there are some visual deficiency and/or an auditory disorder. Development of scripts In this section, the development of each one of the implemented scripts are explained, for example, how these flat text files manage the information entered by the user and allow him or her to interact with the environment. Based on the flow diagram presented in Fig. 2, scripts are programmed to perform the necessary tasks in the proposed rehabilitation program.

Interactive System to Improve the Skills of Children …

443

Table 1 Demographic characteristics of the sample Item

Gender

1

F

8

4

2

F

11

6

3

F

11

6

4

M

9

4

5

M

9

4

6

M

10

6

7

M

12

7

8

M

12

6

Fig. 2 Flowchart of the application developed

Age (year)

Education (years)

444

J. Buele et al.

Video games designed to encourage the enrichment and compensation of language skills and linguistic functions of children with dyslexia. The main interface requires the child to enter their name and age, using the computer keyboard and thus start. In the selection menu, there are three games (with two levels of difficulty each) that allow the user to train their language skills, while entertaining. In addition, an option is presented to obtain and save the data of each user, since this can be used to evaluate their evolution. Figure 3 shows the main screen of the application, and the screen for entering a user’s data is in Fig. 4. Procedure In this subsection, the developed games and a brief description are shown.

Fig. 3 Main screen of the application

Fig. 4 Screen for user registration

Interactive System to Improve the Skills of Children …

445

Fig. 5 Interface of the first game presented to the user

(i) In the first game, an incomplete word is presented on the screen and a series of syllables are placed, and the user has to choose the complement to said word and thus make sense of it. Figure 5 presents the developed interface. (ii) In the second game, the child is presented with a word composed of two syllables, and in parallel, an auditory message is reproduced. It must be determined if the sound corresponds to the word that is shown on the screen, meaning that if the order of the syllables is the right one or is the other way around; according to that, “yes” or “no” is chosen. Figure 6 shows the interface. (iii) In the third application, a word is played through the headphones and on the screen two words of two or three syllables (depending on the level of difficulty) each one are presented. The user has to choose between these two options, which is the word that has the correct order of their syllables. Figure 7 shows the developed interface.

Fig. 6 Interface of the second game running

446

J. Buele et al.

Fig. 7 Interface of the third game that the user uses

3 Tests and Results 3.1 Tests The participants were subjected to sessions of 30 min, two times per week, for a period of 5 weeks. Figure 8 shows how the tests were developed in two of the eight users. For the execution of tasks, it is recommended that a psychologist or educational therapist conducts the session and supervises each video game. Although it has been designed intuitively, an adult that is not specialized can direct it. The exercises were held in private sessions, and outside school hours with the intention of avoiding some kind of social pressure, and the indications were repeated until the child understands them.

3.2 Results At the end of the execution of the games, the interface has an option that exports the data to a file with a.txt extension. This report contains information detailing the name of the patient and his score obtained, so that the specialist can verify if there has been an improvement. In addition, the SUS evaluation test has been applied to users, also known as the system usability scale created by John Brooke in 1986. The system usability scale (SUS) provides a reliable and fast tool for the measurement of usability, thus allowing the evaluation of a wide variety of products and services, including hardware, software, mobile devices, Web sites, and applications. The SUS test can be used in small samples with very reliable results and is valid since it can effectively differentiate between usable and non-usable systems. For this reason, in the developed application the scores have been normalized in order to obtain a

Interactive System to Improve the Skills of Children …

447

percentile classification, which is (82.5 ± 0.52), and as the obtained result is superior to 68, it means that the developed system has a good degree of acceptance.

4 Conclusions and Future Work Through this research, patients with dyslexia have been given a virtual tool to strengthen their linguistic abilities. It is a valid technological proposal because it has visual and audible feedback to improve concentration and the learning process. In this way, conventional exercises have been set aside, with the intention of providing greater understanding, confidence, and better experiences to the student. It should be clarified that the present work is a preliminary study that shows the development and application of video games with the use of Unity 3D software in a mixed set of children. Being an approved system, there are different types of experimental results in each patient, and it depends on the reactions each of them presents. As a metric to determine the validity of this prototype, the SUS usability test was implemented, which has demonstrated that the interface presented to the user is easy to use, intuitive, and friendly since it is focused in evaluating infants. As an important remark, this research has been evaluated from the point of view of the acceptance generated by this system, but not by evaluating the approach of educational therapy, which is part of a broader investigation scope. Despite obtaining positive results, the authors of this research are making modifications to the system based on the feedback obtained from the comments of the users. In addition, what is proposed, as future work is to present a more personalized interface that can be applied to a larger sample, also enabling the assessment of what is the impact it has on adolescents, and even on adults who have this syn-

Fig. 8 Patients runs the games as part of their rehabilitation sessions

448

J. Buele et al.

Table 2 SUS results Question

Result (N = 8) Mean

SD

1. I think I would like to use this virtual reality system frequently

4

0.63

2. I found the virtual reality system to be unnecessarily complex

2.2

0.4

3. I thought that the virtual reality system was easy to use

4.4

0.8

4. I think I would need the support of a technical person to be able to use this virtual reality

2.8

0.4

5. I found that the various functions in this virtual reality system were well integrated

4.2

0.75

6. I thought there was too much inconsistency in this virtual reality system

1.8

0.4

7. I would imagine that most people would learn to use this virtual reality system very quickly

4.8

0.4

8. I found the virtual reality system to be very cumbersome to use

1.4

0.48

9. I felt very confident using the virtual reality system

4.8

0.4

10. I needed to learn a lot of things before I could get going with this virtual reality system

2.6

0.48

Global score (total)

82.5

0.52

drome. It should be emphasized that for the test users, 3D games could already be implemented, which will provide another type of experimental results (Table 2). Acknowledgements This work was financed in part by Universidad Técnica de Ambato (UTA) and Dirección de Investigación y Desarrollo (DIDE) under project PFISEI 26.

References 1. Cameron, C.E., Cottone, E.A., Murrah, W.M., Grissmer, D.W.: How are motor skills linked to children’s school performance and academic achievement? Child. Dev. Perspect. 10(2), 93–98 (2016) 2. Baroutsis, A., Lingard, B.: Counting and comparing school performance: an analysis of media coverage of PISA in Australia, 2000–2014. J. Educ. Policy 32(4), 432–449 (2017) 3. Björkenstam, E., Dalman, C., Vinnerljung, B., Weitoft, G.R., Walder, D.J., Burström, B.: Childhood household dysfunction, school performance and psychiatric care utilisation in young adults: a register study of 96 399 individuals in Stockholm County. J. Epidemiol. Community Health 70(5), 473–480 (2016) 4. Skerfving, S., Löfmark, L., Lundh, T., Mikoczy, Z., Strömberg, U.: Late effects of low blood lead concentrations in children on school performance and cognitive functions. Neurotoxicology 49, 114–120 (2015) 5. Ranning, A., Laursen, T., Agerbo, E., Thorup, A., Hjorthøj, C., Jepsen, J.R.M., Nordentoft, M.: School performance from primary education in the adolescent offspring of parents with schizophrenia and bipolar disorder—a national, register-based study. Psychol. Med. 48(12), 1993–2000 (2018)

Interactive System to Improve the Skills of Children …

449

6. Mayes, S.D., Frye, S.S., Breaux, R.P., Calhoun, S.L.: Diagnostic, demographic, and neurocognitive correlates of dysgraphia in students with ADHD, autism, learning disabilities, and neurotypical development. J. Dev. Phys. Disabil. 30, 1–19 (2018) 7. Berninger, V.W., Richards, T.L., Abbott, R.D.: Differential diagnosis of dysgraphia, dyslexia, and OWL LD: behavioral and neuroimaging evidence. Read. Writ. 28(8), 1119–1153 (2015) 8. Goswami, U.: Sensory theories of developmental dyslexia: three challenges for research. Nat. Rev. Neurosci. 16(1), 43–54 (2015) 9. Eden, G.F., Olulade, O.A., Evans, T.M., Krafnick, A.J., Alkire, D.R.: Developmental dyslexia. Neurobiol. Lang. 815–826 (2016) 10. Knoop-van Campen, C.A., Segers, E., Verhoeven, L.: The modality and redundancy effects in multimedia learning in children with dyslexia. Dyslexia 24(2), 140–155 (2018) 11. Diamanti, V., Goulandris, N., Stuart, M., Campbell, R., Protopapas, A.: Tracking the effects of dyslexia in reading and spelling development: a longitudinal study of Greek readers. Dyslexia 24(2), 170–189 (2018) 12. Papagiannopoulou, E.A., Lagopoulos, J.: P300 event-related potentials in children with dyslexia. Ann. Dyslexia 67(1), 99–108 (2017) 13. Pradhan, B., Parikh, T., Sahoo, M., Selznick, R., Goodman, M.: Current understanding of dyslexia and pilot data on efficacy of a mindfulness based psychotherapy (MBR-RAM) model. Adolesc. Psychiatry 7(1), 44–55 (2017) 14. Caute, A., Cruice, M., Marshall, J., Monnelly, K., Wilson, S., Woolf, C.: Assistive technology approaches to reading therapy for people with acquired dyslexia. Aphasiology 32(sup1), 40–42 (2018) 15. Brown, S.J., Games, D.: Brain Games, Mazes & Coloring Pages-Homeschooling with Minecraft: Dyslexia Games Presents an Activity Book-Great for Creative Kids with Dyslexia, ADHD, Asperger’s Syndrome and Autism, vol. 3. ACM, USA (2018) 16. Ali, M., et al.: iCBLS: an interactive case-based learning system for medical education. Int. J. Med. Informatics 109, 55–69 (2018) 17. Renganathan, S.M., Stewart, C., Perez, A., Rao, R., Braaten, B.: Preliminary results on an interactive learning tool for early algebra education. In: 2017 IEEE Frontiers in Education Conference (FIE), pp. 1–5. IEEE (2017) 18. Pedroli, E., Padula, P., Guala, A., Meardi, M.T., Riva, G., Albani, G.: A psychometric tool for a virtual reality rehabilitation approach for dyslexia. Comput. Math. Methods Med. 2017, 1–6 (2017)

Use of Information and Communication Technologies in the Classroom: An Exploratory Study in Professional Military Education Nuno Alberto Rodrigues Santos Loureiro and Lúcio Agostinho Barreiros dos Santos Abstract The study focuses on the use of information and communication technologies in professional military education, delimited to the present moment and to the current professors and students of the Portuguese Armed Forces Military University Institute. This study aims to analyse the relationship between ICT and teaching–learning methodologies in the context of professional military education, with emphasis on the concept of innovation in the classroom, on teaching–learning styles and on the role of ICTs in learning. The study is based on a mixed research strategy combining deductive and inductive approaches, materialized in a case study, with data collected through questionnaires, interviews and documentary analysis. This article focuses on the conceptual framework and its interconnection with the methodological strategy for the empirical study. The results presented refer to the pilot validation test of the data collection instruments and indicate that, in general, students and teachers share the same teaching–learning styles, both students and teachers show a good propensity to use ICT in the classroom and an almost complete substitution of traditional lessons, lectures and conferences by distance learning with a face-to-face approach in the discussion of case studies is possible. Keywords Professional military education · Teaching–learning methodologies · Information and communication technologies

1 Introduction Military Higher Education in Portugal consubstantiates itself in the accomplishment of courses and cycles of studies, within the military sciences, preparing highly qualified cadres, developing command, direction and leadership skills. Within the N. A. R. S. Loureiro (B) · L. A. B. dos Santos Military University Institute Research and Development Centre (CIDIUM), Military University Institute, Lisbon, Portugal e-mail: [email protected] L. A. B. dos Santos e-mail: [email protected] © Springer Nature Singapore Pte Ltd. 2020 Á. Rocha and R. P. Pereira (eds.), Developments and Advances in Defense and Security, Smart Innovation, Systems and Technologies 152, https://doi.org/10.1007/978-981-13-9155-2_36

451

452

N. A. R. S. Loureiro and L. A. B. dos Santos

Portuguese Armed Forces, this responsibility is assigned to the Military University Institute (MUI), formally created in October 2015. In today’s changing environment, new generations (e.g. Millennials and Zeds) learn differently. The increasing use of information and communication technologies (ICT) in the context of military education makes it easier to achieve training objectives by concentrating on the selection and integration of technological tools with the highest educational returns, adequate for the teaching of the main military competences for the twenty-first century. The study object of this research is the role of ICT in the teaching–learning process in the context of military professional education (PME), and is delimited at three levels: • Temporally, to the academic year of 2018/2019; • Spatially, at the MUI covering the students who attend and the teachers teaching in the Senior Officer Course (SOC), Joint Staff Course (JSC) and Flag Officer Course (FOC); • Conceptually, in the analysis of the relationships between students ‘learning styles and teachers’ teaching styles, the attitude of students and teachers towards the introduction of ICT in the classroom and the adaptation of new teaching–learning methodologies to curricular areas. The general objective of the research, starting from the concept of innovation in the “Classroom”1 (in its physical and virtual dimensions), is to analyse the teaching–learning styles of teachers and students of the Post-Graduated Studies Department of the MUI, evaluating the role of ICT and their suitability for PME. It is important to analyse the relationships established between the key variables of the study, which are expressed from some specific objectives (SO). SO1 Identify and relate the students’ learning styles and teachers’ teaching styles in order to understand their impact on the teaching–learning process; SO2 Identify and analyse the attitudes of students and teachers, towards the increased use of ICT in the “Classroom”, in order to understand their impact on the teaching–learning process; SO3 Evaluate the appropriateness of the new teaching–learning methodology to the curricular areas, resulting from the use of ICT in the “Classroom”, in order to identify opportunities for their introduction into the teaching–learning process. The study seeks to answer the following research questions: 1. What are the learning styles preferred by the students and what are the teaching styles adopted by teachers, what is the relationship between them, and how do they contribute to the teaching–learning process? 1 The

traditional classroom concept presents it as a space, usually a room of variable dimensions, with environmental, dispositional, work and study conditions, where the formal teaching–learning process takes place, whose development dates back to the century with the democratization of access to education [1]. The “Classroom” concept of the future was created in 2011 under the iTEC Project under the coordination of the European Schoolnet, with the purpose of designing a new proposal of organization of classroom space and practices, by harmonious conjugation of three key elements: space, technology and pedagogy [2].

Use of Information and Communication Technologies …

453

2. What are the attitudes of students and teachers towards the use of ICT in the “classroom”, what is the relationship between them, and how do they contribute to the teaching–learning process? 3. Which, among the new ICT teaching–learning methodologies, are the most appropriate to the curricular contents of the PME? In the present article we intend, essentially, to develop the conceptual framework and its interconnection with the methodological strategy, that is, the definition of the instruments of data collection for the complete empirical study to be developed. In this way, the results presented now refer to the validation pilot test. To this end, the article is structured in three chapters, besides the introduction and conclusion. Initially, a review of the literature is done, highlighting the PME new challenges facing the VUCA2 context of today, the importance of technology in the development of new teaching–learning methodologies, the emergence of new learning environments, the teaching–learning styles and the acceptance of the technology. Secondly, a detailed description of how the study is being conducted is given, including an explanation of the concepts and the operationalization of the variables under study. Next, the preliminary data collected are summarized, characterizing the sample and analysing the relevance of the data collected.

2 Literature Review The military profession, analysed from a vocational or professional perspective [3], requires a broad set of very comprehensive competencies to fulfil its duties and obligations, within its demanding functions of command, direction, leadership, staff and execution, among others. The education and training given should be in line with what the military must do to fulfil the mission. Thus, systemic approach should include four steps: analysis, curriculum design, development and implementation. In the construction of course references, functional analysis is essential and takes into account the work context, organizational systems, functional relations and their future application [4]. Another challenge facing the armed forces is related to the comprehensiveness and demands of knowledge, since it is increasingly necessary to develop more complex knowledge, skills and attitudes, such as critical knowledge and understanding of military operations; different joint military capabilities; deep understanding of the operational planning process, rules of engagement and risk management; understand and deal with the complexity and diversity of tasks; be very adaptable to different positions and functions, in contexts of change and new challenges; demonstrate authority, autonomy, professional integrity, sustained commitment to the development of processes at the forefront of work and innovation. 2 Concept

introduced by US Army War College to describe the multilateral world resulting from the end of the Cold War: Volatile, Uncertain, Complex and Ambiguous.

454

N. A. R. S. Loureiro and L. A. B. dos Santos

In the classroom, technology can span all types of tools. The latest technologies allow us to experiment classroom methodologies, both physical and virtual, that were not previously possible, depending fundamentally on what we want to accomplish. Teaching with technology can deepen student learning by supporting training objectives. However, it can be a huge challenge to select the “best” technology tools without losing sight of the learning objectives. Once identified, the integration of these tools can also be another challenge, although it can become an experience with high educational returns.

2.1 Technology and New Teaching–Learning Methodologies It is common to find some reluctance of higher education institutions regarding the adoption of new procedures, processes and models of teaching–learning, being sometimes more common to exist a conservative position on the part of the military institution [5]. But the recovery of the backwardness that may exist in higher education institutions, whether civilian or military, may be based on a strategy, not to go through all unfulfilled development stages, but to neglect the intermediate and try as much as possible to keep abreast of the current stage of scientific and technological development by adopting new technologies to support the teaching–learning process. It may, however, be necessary to evaluate the implications of “burning” some steps. Another perspective on the role of technology in the teaching–learning process is brought to us by Peter J. Denning and Susan L. Higgins that report a marked increase in the discussion of technological advances in learning environments, in particular the massive open online courses (MOOC) Internet-based platforms, thus being an automated learning environment (ALE). In a military context, senior leaders are asked to go beyond basic skills and are required to be at least at the “proficient” level, which cannot be achieved by recourse to a mere ALE, as those leaders work in environments where rules are constantly changing, so the challenge in PME is to go beyond technologies when trying to develop higher levels of leadership [6]. According to Neal [7], although there has been a growing use of simulators in training programs, however, care should be taken to evaluate in advance the contexts of training in which this use allows effectively gains in effectiveness and those in which this does not happen. As an example, this author points out that topics related to military doctrine, politics or history are typically treated through lectures, discussion and debate, and should not be addressed by technological products that would be downright inappropriate. As Schatz et al. [8] argue, the military context requires a permanent development of skills in its staff, and this process is also based on the teaching and training given to them: “[…] military personnel require an expanded set of competencies, higher levels of nuanced skills such as critical thinking and emotional intelligence, and more efficient and agile pathways to expertise, and that achieved at least in part depends on revising the military learning enterprise”.

Use of Information and Communication Technologies …

455

2.2 Emerging Learning Environments Referring to WEB 2.0 technology, Foon Hew and Sum Cheung [9] affirm that this is an interactive tool that facilitates bidirectional interaction, requiring the students to contribute to the placement of information on the site available, interacting and sharing their ideas and knowledge in a collaborative way. According to the authors, the technological tools of WEB 2.0 allow teachers to apply a constructivist approach to teaching–learning. There are three possible ways to use WEB 2.0 tools within the scope of teaching: Blogger; Active Worlds; and Teachertube [9, 10 cited by 11]. Pinel [12] states that students are required to acquire twenty-first century ICT skills because they are related to the real world, develop as students construct knowledge, to contribute to society. An Australian publication on educational technology edited by Education Technology Solutions (ETS), recently published an online article entitled Technology That Will Shape Education In 2017, in which it present a guide to the main trends in educational technology that will have a relevant impact on way of learning and teaching. The way in which education professionals learn, teach and collaborate is changing significantly with the emergence of new technologies, such as virtual reality, augmented reality, learn from anywhere, teaching from anywhere mobile devices, collaboration technology, gamification, coding, evolving learning spaces and styles and the maker movement [13].

2.3 Learning Styles The assimilation and processing of information and knowledge are different from individual to individual. They can do this by seeing or hearing, reflecting or acting, reasoning in a logical or intuitive way and analysing or visualizing. Students have different strengths and preferences in how they receive and process information, which means that they have different learning styles [11, 14]. Teaching methodologies also vary, with some teachers preferring lectures and others favouring demonstration or leading students to self-discovery; some focus on principles and others on applications; some emphasize memory and others emphasize understanding. Difficulties arise when there are incompatibilities between the student’s learning style and the teacher’s teaching style; students may become bored and inattentive in class, poor test performance, and lose motivation regarding courses or curriculum [15–17]. Empirical studies have confirmed the hypothesis that teachers who are able to adapt their teaching style to include the two poles of each dimension, ideal for most (if not all) students in a class, are more effective [14].

456

N. A. R. S. Loureiro and L. A. B. dos Santos

The Felder and Silverman model, in the 2002 update [18], ranks a student’s learning style from one category or another in each of the four dimensions through the answers given to four questions: 1. What kind of information the student preferentially perceives: sensory (visions, sounds, physical sensations, concrete, practical, fact-oriented and intuitive thinking) or intuitive (possibilities, perceptions, hunches, abstract thinking, innovative, oriented to theories and meanings underlying)? 2. Through which sensitive channel is the most effectively perceived external information: visual (prefer visual representations of displayed material such as figures, diagrams, flowcharts, graphs and demonstrations) or verbal (prefer written and spoken explanations)? 3. How does the student prefer to process information: actively (through involvement in physical activity or discussion, experiencing things and working in groups) or reflexively (through introspection, prefer to work alone or with a single family partner)? 4. How does the student progress towards understanding: sequentially (linear thinking process and learning in small incremental steps in continuous steps) or globally (holistic thinking process and learning in big leaps)? The same model states that teaching style can also be defined, in one or another category of each of the four dimensions, in terms of answers to four questions: 1. What kind of information is emphasized by the teacher: concrete (factual) or abstract (conceptual, theoretical)? 2. Which form of presentation is preferred by the teacher: visual (visual images, diagrams, films and demonstrations) or verbal (verbal talks, readings and discussions)? 3. What mode of student participation is facilitated by the teacher’s presentation: active (do students speak, move and reflect) or passive (do students watch and listen)? 4. What kind of perspective is provided on the information presented: sequential (step-by-step, showing “the trees”) or global (relevance (presenting “the forest”)? Table 1 (see Appendix) summarizes learning and teaching styles in their different dimensions and categories, as proposed by Felder and Silverman [18]. The Index of Learning Styles is an online research tool used to evaluate preferences in four dimensions (active/reflexive, sensory/intuitive, visual/verbal and sequential/global), built from the model of learning styles formulated by Felder and Silverman [11, 18]. The scale was developed and validated by Felder and Soloman [15–17]. In their original version, respondents answer online to 44 questions, with two alternative answers, “a” and “b”, and submit responses online, and receive a report on their preferences regarding the four dimensions under review.

Use of Information and Communication Technologies …

457

2.4 Acceptance of Technology in the Classroom Acceptance of technology results from users’ beliefs and attitudes about their use, which was measured with the perceived utility and perceived ease of use in the technology acceptance model (TAM) [19, 20]. The Technology Acceptance Barriers questionnaire is an instrument that intends to measure the extent and nature of the barriers to the integration of ICT in the perspective of technology acceptance, compiled by Gu, Zhu, and Guo [21] into four constructs often referred to as predictors of ICT acceptance: expectation of results, adjustment of technology to task, social factor and personal factor. Outcome expectancy (OE) results from internal beliefs and attitudes about the use of ICT, and it is measured as perceived utility and perceived ease of use, being considered the most important predictor of the use of technology [20, 22, 23]. Tasktechnology fit (TTF) is the degree to which technology helps an individual to accomplish his or her tasks [24]. It is based on the assumption that users accept the technology because of its potential benefits, such as performance improvement, regardless of their attitude. Social factor (SF) is a type of social norm defined as “perceived social pressure to carry out behaviour or not” [25, 26]. Studies suggest that perceived social influence has a significant positive influence on individual beliefs about the usefulness of technology and positively and significantly affects the use of ICTs [22, 27, 28]. Personal factors (PF) construct includes computer self-efficacy and personal innovation through technology, widely recognized as explanatory factors that influence the use of IT by end-users [27, 28].

3 Methodology The study is based on a mixed research strategy [29] combining hypotheticaldeductive [30] and inductive [31] approaches, materialized in a cross-sectional study [32], with data collected at a given moment, through questionnaires, interviews and documentary analysis. Due to the temporal, spatial and conceptual boundaries defined for this research, the target population of the present study is composed of students and professors of the SOC, JSC and FOC of the Military University Institute, in the academic year 2018/2019. The preliminary results presented are based only on the pilot test, submitted to a convenience sample composed of the students and professors of the SOC, JSC and FOC of the Military University Institute, in the academic year 2017/2018.

458

N. A. R. S. Loureiro and L. A. B. dos Santos

3.1 Questionnaire For the empirical study, a data collection instrument was prepared based on the following scales: (1) Index of Learning Styles (ILS), proposed by Soloman and Felder [17]; and (2) Technology Acceptance Barriers (TAB), proposed by Gu, Zhu, and Guo [21]. It was also complemented by ad hoc items designed to collect biographical and professional data on the one hand and, on the other hand, to verify the traditional teaching methodologies currently used in curricular areas and to identify the most appropriate ICT-based teaching methodologies to each of the specific curricular areas. In the present case, the scales were adapted to the Portuguese language and to the military context. In order to develop a comparative analysis between preferences of learning styles and teaching styles between students and teachers, the results of the ILS questionnaire were transformed into a seven-point Likert scale, according to Table 6 (see Appendix—Table 6). All items of TAB questionnaire were also measured using a seven-point Likert scale ranging from “strongly agree” to “strongly disagree”. Regarding the ad hoc items introduced in the questionnaire, some asked the respondent to choose the three most traditional teaching methodologies3 currently used in each curricular area; others asked the respondent to choose the three ICTbased teaching methodologies4 best suited to each of the specific curricular areas. The questionnaire, necessarily different between students and teachers, after a brief introduction, explaining the objectives of the study and guaranteeing the treatment of the data in an anonymous and confidential way, and its exclusive use for the mentioned effects, was thus composed of the four sections shown in Table 2 (see Appendix). In order to adapt the questionnaire formulation to the Portuguese language and the military context, three sequential stages were followed: (1) translation and retroversion, (2) pretest and (3) pilot test. As the items were initially expressed in the English language, the retroversion was applied to guarantee a precise translation into Portuguese. Subsequently, a pretest was applied, using a small number of respondents (focus group) to test the adequacy of the questions and their comprehension. This focal group, consisting of 10 subjects from each course (30 participants) suggested some modifications in the items of the questionnaire, in order to be better understood by the target population. After the pretest, a pilot test was performed and submitted to all individuals of the SOC, JSC and FOC of the 2017/2018 school year, whose objective was to test the reliability of the items (consistency). 3 Lesson/Lesson-Debate; directed discussion/learning discussion/topic discussion; individual appli-

cation work/application work group; CPX/CAX computer assisted; field exercises; field trip; seminar/panel/symposium; workshop; lecture/ conference/cycle of conferences; presentation and defence of work. 4 Digital Book; ongoing training online; gamification; project-based learning (PBL); case study; team-based learning (TBL); flipped classroom; coding; virtual reality; computer-based training (CBT); augmented reality; learn from anywhere/teach from anywhere mobile devices.

Use of Information and Communication Technologies …

459

3.2 Pilot Test Sample and Preliminary Results The pilot test universe consisted of 76 teachers and 225 students. The questionnaire was answered by 46 teachers and 150 students. After validation of the answers given, there were some errors and incongruities, which led to the elimination of some of them. The final number of validated respondents was 24 teachers and 124 students, corresponding, respectively, to 31.6 and 56.4% of the target population. Their demographic and professional characteristics can be seen in Table 3 (see Appendix). As preliminary results from a pilot test, they should be interpreted as exploratory data for the preparation of the questionnaire that will be submitted to the target population of the study, SOC, JSC and FOC students of the 2018/2019 school year. Reliability was estimated by the Cronbach’s alpha coefficient, and the items of the study achieved a reliable result (ILS > 0.60, use of ICT > 0.97). The answers to the questionnaire were, therefore, submitted to a first analysis, using the statistical package for the social sciences (SPSS V.23) with regard to descriptive statistics and comparative analysis of means, centred on two aspects: (1) analysis of learning style profiles and (2) analysis of the barriers to the use of ICT in the “Classroom”.

3.3 Learning Style Profiles Analysis Table 4 (see Appendix—Table 4) summarizes the learning style profiles reported by students and teachers of each of the courses under analysis (SOC, JSC and FOC), and reads as follows, for example, of the 101 SOC students who completed ILS, 91% were classified as having a “sensory” profile (implying that 9% were classified as having an “intuitive” profile) 88% prefer “visual” input of content (so 12% prefer to “verbal” content input) and so on. It is verified that an expressive majority of the students of the SOC and JSC show preference for the “Sensory”, “Visual”, “Active” and “Sequential” style. Regarding the FOC students, an expressive majority shows a preference for the “Visual” style, but only a relative majority shows a preference for the “Sensory” and “Active” style, even reversing the preference in the field of “Understanding” with a minority to opt the “Sequential” style. Regarding how teachers perceive the learning styles of the students, the majority follow the preferences expressed by students, however, with regard to the dimension “Perspective”, a significant majority of teachers presents a “Sequential” style, contrary to the majority of FOC students. The analysis of ILS responses also allows us to identify (see Appendix—Table 5) that a large percentage of students and teachers have a mild preference for a particular learning style. This suggests that both students and teachers may be able to switch between styles, adapting more easily to the “classroom” environment. However, in the case of the students of all the courses, there is a moderate or strong preference for

460

N. A. R. S. Loureiro and L. A. B. dos Santos

the “Visual” style in the “Input” dimension, and also, in the case of the SOC and JSC students, a moderate or strong style preference of “Sensory” perception, not being followed by the teachers in this preference. Particular attention should be given to these situations. Therefore, with regard to global learning and teaching styles (see Appendix—Table 7), and by gender (see Appendix—Table 8), the results indicate no significant differences in the preferences and attitudes of students and teachers. According to the branch of the armed forces to which the respondents belong (see Appendix—Table 9), the results indicate significant differences in the air force branch in the dimensions of “Perception/Content”, “Input/Presentation” and “Understanding/Perspective”. According to the age group to which the respondents belong (see Appendix—Table 10), the results indicate significant differences in the 51–55 group in the four dimensions. According to the course the respondents attend/teach (see Appendix—Table 11), the results indicate no significant differences in the preferences and attitudes of students and teachers, except in the dimension “Understanding/Perspective” within the FOC students and teachers. According to the effective service time to which the respondents belong (see Appendix—Table 12), the results indicate significant differences in the 31–35 group in the four dimensions.

3.4 Analysis of Barriers to the Use of ICT in the “Classroom” Given the results of the data collected, both pupils and teachers indicate a good outcome expectation with the use of ICT in the “Classroom”, consider ICT reasonably adjusted for use in the context of professional military training, consider themselves capable of to use and consequently accept the introduction of technology into teaching–learning processes. The deepening of this study will allow us to see how concretely it can be done.

3.5 Analysis of Adequacy of Teaching–Learning Methodologies to Curricular Areas Taking into account the preliminary results achieved (see Appendix—Tables 13, 14 and 15), whereas in the SOC, the most used traditional teaching–learning methodologies are the “Lesson/Lesson-debate”, followed by “Directed Discussion/Learning Discussion/Topic Discussion” and “Lecture/Conference/Lecture Series”, in JSC preference is given to “Directed Discussion/Learning Discussion/Topic Discussion”, followed by “Lesson/Lesson-debate” and “Lecture/Conference/Lecture Series”. This

Use of Information and Communication Technologies …

461

preference is emphasized in FOC, with the primacy given to “Directed Discussion/ Learning Discussion/Topic Discussion” followed by “Lecture/Conference/Lecture Series” and “Lesson/Lesson-debate”. As regards the teaching–learning methodologies based on ICT, considered more appropriate, the results point to a preference for the “Digital Book”, the “Continuing education online” and the “Case Study”, although the “Team Based Learning” and “Gamification” were also referred. These results seem to indicate an almost complete substitution of traditional lessons, lectures and conferences by distance learning with a face-to-face approach in the discussion of case studies.

4 Conclusions Being part of a broader research on teaching–learning methodologies in PME, through an approach to the introduction of ICT in the classroom, this paper has essentially sought to develop the conceptual framework and their interconnection with the methodological strategy, that is, the definition of the instruments of data collection for the complete empirical study to be developed. The results presented refer to the pilot validation test of these same data collection instruments. The first stage of preparation of data collection instruments is in the process of being finalized, a preliminary analysis indicates that: • The adapted ILS scale has an acceptable internal consistency (alpha > 0.60) and the adapted TAB scale has a very good internal consistency (alpha > 0.97); • In general, students and teachers share the same teaching–learning styles; however, when analysed according to control variables, there are significant differences; • Both students and teachers show a good propensity to use ICT in the classroom; • An almost complete substitution of traditional lessons, lectures and conferences by distance learning with a face-to-face approach in the discussion of case studies is possible. In the light of these preliminary results, the data already collected will be further analysed in the continuation of the study, and it will be sought to gather more information and complete the objectives of the study, which will be a major contribution to improving the quality of PME in the Portuguese Armed Forces.

Appendix See Tables 1, 2, 3, 4, 5, 6, 7, 8, 9, 10, 11, 12, 13, 14 and 15.

462

N. A. R. S. Loureiro and L. A. B. dos Santos

Table 1 Learning and teaching styles dimensions and categories Preferred learning style (student)

Corresponding teaching style (teacher)

Categories

Dimensions

Categories

Dimensions

Sensory Intuitive

→ Perception

Concrete Abstract

→ Content

Visual Verbal

→ Input

Visual Verbal

→ Presentation

Active Reflexive

→ Processing

Active Passive

→ Student participation

Sequential Global

→ Understanding

Sequential Global

→ Perspective

Table 2 Questionnaire characterization and number of items Students Section A—Biographical and professional data (control variables: age, sex, effective service time, branch and course attended)

Total—5 items

Section B—Learning styles index: identify students’ learning preferences

11 items per dimension Total—44 items

Section C—Barriers to technology acceptance: identify students’ propensity to use technology in learning

OE—8 items PF—7 items TTF—7 items SF—4 items Total—26 items

Section D1—Identify the current teaching/learning methodologies most appropriate to the curricular context of the SOC, JSC and FOC, for each of its curricular areas

SOC—9 items JSC—16 items FOC—14 items

Section D2—Identify the new teaching/learning methodologies best suited to the curricular context of the SOC, JSC and FOC, for each of its curricular areas Teachers Section A—Biographical and professional data (control variables: age, sex, effective service time, branch, time in teaching functions, course teaching and specific teaching training)

Total—7 items

Section B—Teaching styles index: collect teachers ‘perceptions about students’ learning preferences

11 items per dimension Total—44 items

Section C—Barriers to technology acceptance: identify teachers’ propensity to use technology in learning

OE—8 items PF—7 items TTF—7 items SF—4 items Total—26 items

Section D1—Identify the current teaching/learning methodologies most appropriate to the curricular context of the SOC, JSC and FOC, for each of its curricular areas

SOC—9 items JSC—16 items FOC—14 items

Section D2—Identify the new teaching/learning methodologies best suited to the curricular context of the SOC, JSC and FOC, for each of its curricular areas

Use of Information and Communication Technologies …

463

Table 3 Characterization of the pilot test sample Type

Age

Gender

Branch

Effective service time

Years of teaching

Teaching training

Course

Number of valid samples

Teacher

36 < age < 40 = 20%

Female = 4%

Navy = 16%

16 < YoS < 20 = 16%

< 1 year = 24%

Yes = 32%

CPOS = 100%

24

41 < age < 45 = 60%

Male = 96%

Army = 52%

21 < YoS < 25 = 44%

1 < years 3 years = 24%

age > 51 = 20%

CPOG = 48%

YoS > 51 = 20% Mean = 44.28

Mean = 25.68

Min = 38

Min = 19

Max = 55 Student

Max = 37

age < 36 = 31.5%

Female = 17.7%

Navy = 38.7%

YoS < 16 = 26.6%

Yes = 32%

CPOS = 81.5%

36 < age < 40 = 35.5%

Male = 82.3%

Army = 30.6%

16 < YoS < 20 = 43.5%

No = 68%

CEMC = 9.7%

Air Force = 30.6%

21 < YoS < 25 = 6.5%

41 < age < 45 = 10.5% 46 < age < 50 = 8.9%

26 < YoS < 30 = 11.3%

age > 51 = 13.7%

YoS > 51 = 12.1%

124

CPOG = 8.9%

Mean = 19.81 Mean = 39.94

Min = 9

Min = 33

Max = 35

Max = 54

Table 4 Learning style profiles reported Sample

Perception: sensory (%)

Input: visual (%)

Processing: active (%)

Understanding: sequential (%)

N

Students SOC

91

Students JSC

92

88

70

77

101

100

75

67

Students FOC

55

82

12

64

27

11

Sample

Content: concrete (%)

Student participation: active (%)

Perspective: sequential (%)

Teachers SOC

62

60

61

60

24

Teachers JSC

50

Teachers FOC

67

58

58

59

17

64

60

78

11

Presentation: visual (%)

N

33

12

18

Teachers JSC

Teachers FOC

Mod/Str Vis (%)

Mod/Str Con (%)

Teachers SOC

Presentation: visual—verbal

82

82

54

Mild (%)

0

6

13

Mod/Str Abs (%)

45

24

50

73

45

41

17

Mild (%)

9

9

35

33

Mod/Str Ver (%)

18

0

Content: concrete—abstract

27

17

Sample

55

83

5

18

8

20

Students FOC

17

75

75

4

Students JSC

31

Mod/Str Ver (%)

65

Mild (%)

Mod/Str Vis (%)

Mod/Str Int (%)

Mod/Str Sen (%)

Mild (%)

Input: visual—verbal

Perception: sensorial—intuitive

Students SOC

Sample

Table 5 Learning style profiles reported

82

67

53

Mild (%)

18

12

50

Mod/Str Ati (%)

82

82

17

Mild (%)

Student participation: active—passive

9

33

34

Mod/Str Ati (%)

0

6

33

Mod/Str Pas (%)

9

0

13

Mod/Str Ref (%)

Processing: active—reflexive

55

75

50

Mild (%)

36

8

8

Mod/Str Glo (%)

9

12

13

Mod/Str Seq (%)

91

76

83

Mild (%)

0

12

4

Mod/Str Glo (%)

Perspective: sequential—global

9

17

42

Mod/Str Seq (%)

Understanding: sequential—global

464 N. A. R. S. Loureiro and L. A. B. dos Santos

Use of Information and Communication Technologies …

465

Table 6 ILS results conversion scale

Table 7 Index of learning/teaching styles—global Dimensions Perception Categories Sensory Intuitive Students moderate Dimensions Content Categories Concrete Abstract Teachers mild

Input Visual Verbal moderate Presentation Visual Verbal mild

Processing Active Reflexive mild Student Participation Active Passive mild

Understanding Sequential Global mild Perspective Sequential Global mild

Table 8 Index of learning/teaching styles—by gender index Dimensions Categories Students

male female

Dimensions Categories Teachers

male female

Perception Input Sensory Intuitive Visual Verbal moderate strong moderate moderate Content Presentation Concrete Abstract Visual Verbal mild mild moderate moderate

Processing Active Reflexive mild mild Student Participation Active Passive mild mild

Understanding Sequential Global mild mild Perspective Sequential Global mild moderate

Table 9 Index of learning/teaching styles—by branch of the Armed Forces Dimensions Categories

Perception Input Sensory Intuitive Visual Verbal Navy moderate moderate Army moderate moderate Students Air moderate strong Force Dimensions Content Presentation Categories Concrete Abstract Visual Verbal Navy mild moderate Army mild moderate Teachers Air mild mild Force

Processing Active Reflexive mild mild mild Student Participation Active Passive mild mild mild

Understanding Sequential Global mild mild mild Perspective Sequential Global mild mild mild

466

N. A. R. S. Loureiro and L. A. B. dos Santos

Table 10 Index of learning/teaching styles—by age group Dimensions Categories 51-55 46-50 Students 41-45 36-40 31-35 Dimensions Categories 51-55 46-50 Teachers 41-45 36-40 31-35

Perception Sensory Intuitive moderate moderate moderate moderate moderate Content Concrete Abstract mild mild mild mild mild

Input Visual Verbal moderate moderate moderate moderate moderate Presentation Visual Verbal mild mild mild mild mild

Processing Active Reflexive mild mild mild mild mild Student Participation Active Passive mild mild mild mild mild

Understanding Sequential Global mild mild mild mild mild Perspective Sequential Global mild mild mild mild mild mild

Processing Active Reflexive mild mild mild Student Participation Active Passive mild mild mild

Understanding Sequential Global mild mild mild Perspective Sequential Global mild mild mild

Table 11 Learning/teaching styles index—by course Dimensions Categories SOC Students FOC JSC Dimensions Categories SOC Teachers FOC JSC

Perception Input Sensory Intuitive Visual Verbal moderate moderate mild moderate moderate strong Content Presentation Concrete Abstract Visual Verbal mild mild mild mild mild mild

Table 12 Learning/teaching styles index—by effective time of service Dimensions Categories 31-35 26-30 Students 21-25 16-20 11-15 Dimensions Categories 31-35 26-30 Teachers 21-25 16-20 11-15

Perception Sensory Intuitive moderate moderate moderate moderate moderate Content Concrete Abstract mild mild mild mild mild

Input Visual Verbal moderate moderate moderate moderate moderate Presentation Visual Verbal mild mild mild mild mild

Processing Active Reflexive mild mild mild mild mild Student Participation Active Passive mild mild mild mild mild

Understanding Sequential Global mild mild mild mild mild Perspective Sequential Global mild mild mild mild mild

– Exercises of Computer Assisted Command Posts (CPX/CAX)

1

3

1

1

1

1

1

2

1

– Resource Management

– Communication and Leadership

– Logistics and Operational Research

– Military Organization

– Military Strategy

– Public International Law

– Military History of Portugal

– Military Units Framework

– Military Operations Planning

2

2

3

3

1

2

2

1

3

2

2

2

2

3

3

3

3

3

1

1

3

1

1

2

1

ICT-based – Individual Application Work/Group Application Work

– Digital Book

– Directed Discussion/ Learning Discussion/ Topic Discussion

– Lesson/ LessonDebate

– Lecture/ Conference/ Lecture Series

Traditional

Teaching Methodologies

Table 13 SOC results on teaching methodologies

1

2

2

2

2

2

2

– Continuing Education Online

3

3

3

3

1

3

3

3

3

– Case Study

2

2

1

– Team Based Learning (TBL)

1

– Gamification

Use of Information and Communication Technologies … 467

2

1

2

1

1

2

3

3

– Organizations Management

– Resource Management

– Command and Leadership

– Communication and Public Relations

– Public International Law

– Geopolitics

– International Context

– War and Peace

1

2

3

3

3

3

3

3

2

1

1

2

2

1

2

1

1

1

2

1

3

3

1

1

– Digital Book

– Exercises of Computer Assisted Command Posts (CPX/CAX)

ICT-based – Directed Discussion/ Learning Discussion/ Topic Discussion

– Lesson/ LessonDebate

– Lecture/ Conference/ Lecture Series

Traditional

Teaching methodologies

Table 14 JSC results on teaching methodologies

2

2

1

2

1

2

2

2

– Continuing Education Online

3

3

3

3

2

1

3

3

– Case Study

(continued)

– Team Based Learning (TBL)

468 N. A. R. S. Loureiro and L. A. B. dos Santos

1

1

2

2

1

2

1

– Strategic Planning

– Strategic Studies

– Doctrinal Operations Framework

– National Operations Planning

– Operational Management

– Exercise Planning and Conduct

3

3

3

3

2

2

1

2

1

1

2

3

3

3

2

2

2

1

2

1

– Digital Book

– Exercises of Computer Assisted Command Posts (CPX/CAX)

ICT-based – Directed Discussion/ Learning Discussion/ Topic Discussion

– Lesson/ LessonDebate

– Lecture/ Conference/ Lecture Series

Traditional

Teaching methodologies

– Military History

Table 14 (continued)

1

1

3

1

2

1

2

– Continuing Education Online

2

3

1

3

3

3

3

– Case Study

1

– Team Based Learning (TBL)

Use of Information and Communication Technologies … 469

3

3

3

3

2

– Means and Strategic Instruments

– Strands of the National Strategy

– Organizational Management and Resource Management

– Superior Organization of the Armed Forces

– Administrative Law

3

2

2

2

2

2

2

3

2

– Strategic Objectives and Interests

3

– Threats and Conflicts

– The Strategic Programming Model

– Academic Seminar/ Panel/ Symposium

1

1

1

1

1

1

1

1 3

2

1

ICT-based – Exercises of Computer Assisted Command Posts (CPX/CAX)

– Digital Book

– Directed Discussion/ Learning Discussion/ Topic Discussion

– Lesson/ LessonDebate

– Lecture/ Conference/ Lecture Series

Traditional

Teaching Methodologies

Table 15 FOC results on teaching methodologies

2

2

3

2

3

2

3

2

– Continuing Education Online

3

3

2

3

2

3

1

3

– Case Study

(continued)

– Team Based Learning (TBL)

470 N. A. R. S. Loureiro and L. A. B. dos Santos

– Academic Seminar/ Panel/ Symposium

1

3

2

2

2

2

– Command and Leadership

– Economy and Finance

– Joint and Combined Operations

– Operational Planning Exercise

– Operational Activity of the Armed Forces

3

3

1

2

3

1

1

1

3

1

2

3 3

1

3

ICT-based – Exercises of Computer Assisted Command Posts (CPX/CAX)

– Digital Book

– Directed Discussion/ Learning Discussion/ Topic Discussion

– Lesson/ LessonDebate

– Lecture/ Conference/ Lecture Series

Traditional

Teaching Methodologies

– Communication and Media Relations

Table 15 (continued)

1

3

2

2

1

2

– Continuing Education Online

2

1

3

3

2

3

– Case Study

2

1

– Team Based Learning (TBL)

Use of Information and Communication Technologies … 471

472

N. A. R. S. Loureiro and L. A. B. dos Santos

References 1. Garrido, A.: Errar é humano! A vivência de erros e seus efeitos na produção oral sob a perspectiva do aluno de Inglês como Língua Estrangeira.pdf (2006) 2. Lewin, C., McNicol, S.: Criar a Sala de Aula do Futuro: conclusões do projeto iTEC (2014) 3. Moskos, C.C., Williams, J.A., Segal, D.R.: The Postmodern Military: Armed Forces After the Cold War. Oxford University Press on Demand (2000) 4. Exército Português: Manual Didático (MD) 240-01 Qualidade da Formação (2014) 5. Downes, C.: Rapidly evolving, digitally-enabled learning environments: implications for institutional leaders, educators and students. In: Doughty, R., Wells, L., and Hailes, T. (eds.) Innovative Learning: A Key to National Security. pp. 101–132. Army Press (2015) 6. Denning, P.J., Higgins, S.L.: Being in uncertainty: cultivating a new sensibility in military education. In: Doughty, R., Wells, L.I., Hailes, T. (eds.) Innovative Learning: A Key to National Security. pp. 133–154 (2015) 7. Neal, D.J.: Technology and its impact on defense/security thinking and learning intervention issues. In: Doughty, R., Wells, L.I., Hailes, T. (eds.) Innovative Learning: A Key to National Security. pp. 155–170 (2015) 8. Schatz, S., Fautua, D.T., Stodd, J., Reitz, E.: The Changing Face of Military Learning. The Army Press 9. Foon Hew, K., Sum Cheung, W.: Use of Web 2.0 Technologies in K-12 and Higher Education: The Search for Evidence-Based Practice (2013) 10. Bower, M., Hedberg, J., Kuswara, A.: A framework for Web 2.0 learning design. Educ. Media Int. 47, 177–198 (2010) 11. Felder, R., Silverman, L.: Learning and teaching styles in engineering education. Eng. Educ. 78, 674–681 (1988) 12. Pinel, S.: Bringing It All Together: Literacy, ICT and the 21st Century Skills 13. Low, M.L.: Technology That Will Shape Education In 2017 14. Felder, R., Spurlin, J.: Application, reliability and validity of the index of learning styles. Int. J. Eng. Educ. 21, 103–112 (2005) 15. Soloman, B.A., Felder, R.: Index of Learning Styles Questionnaire, pp. 1–5. NC State University. North Carolina State University (1996) 16. Soloman, B.A., Felder, R.: Index of Learning Styles Questionnaire (1999) 17. Felder, R.M., Soloman, B.A.: Learning Styles and Strategies (2012) 18. Felder, R., Silverman, L.: Learning and Teaching Styles In Engineering Education (2002) 19. Davis, F.D.: A Technology Acceptance Model for Empirically Testing New End-User Information Systems: Theory and Results (1985) 20. Venkatesh, V., Smith, R.H., Morris, M.G., Davis, G.B., Davis, F.D., Walton, S.M.: User Acceptance of Information Technology: Toward a Unified View (2003) 21. Gu, X., Zhu, Y., Guo, X.: Meeting the “digital natives”: understanding the acceptance of technology in classrooms. Educ. Technol. Soc. 16, 392–402 (2013) 22. Kim, C., Jahng, J., Lee, J.: An empirical investigation into the utilization-based information technology success model: Integrating task-performance and social influence perspective. J. Inf. Technol. 22, 152–160 (2007) 23. Lee, M.C.: Explaining and predicting users’ continuance intention toward e-learning: an extension of the expectation-confirmation model. Comput. Educ. 54, 506–516 (2010) 24. Goodhue, D.L.;, Thompson, R.L.: Task-technology fit and individual performance (1995) 25. Ajzen, I.: The theory of planned behavior. Organ. Behav. Hum. Decis. Process. 50, 179–211 (1991) 26. Ajzen, I.: The theory of planned behavior. Orgn. Behav. Hum. Decis. Process. 50, 179–211 (1991) 27. Lewis, W., Agarwal, R., Sambamurthy, V.: Sources of influence on beliefs about information technology use: an empirical study of knowledge workers. MIS Q. 657–678 (2003)

Use of Information and Communication Technologies …

473

28. Thompson, R., Compeau, D., Higgins, C., Lupton, N.: Intentions to use information technologies: an integrative model. In: End User Computing Challenges and Technologies: Emerging Tools and Applications: Emerging Tools and Applications vol. 18, pp. 79–101 (2007) 29. Creswell, J.: Research Design: Qualitative, Quantitative, and Mixed Methods Approaches (2013) 30. Quivy, R., Campenhoudt, L.: Manual de Investigação em Ciências Sociais. Gradiva, Lisboa (2008) 31. Creswell, J.: Educational research: Planning, Conducting, and Evaluating Quantitative and Qualitative Research. University of Nebraska–Lincoln (2012) 32. Bryman, A.: Quantity and Quality in Social Research. Taylor & Francis e-Library, 2004, New York (1988)

Using Augmented Reality for Learning Naval Operations Anacleto Correia , António Gonçalves

and Marielba Zacarias

Abstract Augmented reality (AR) is a technology that extends the perception of human reality by allowing, for example, inserting virtual objects in real environments. With augmented reality, learning can be more effective with graphics and interactivity in the manipulation of content. With this work, we intend to apply the AR technology in a process of developing a solution in learning context. For this purpose, different types of devices were analyzed in order to select the type of equipment with better characteristics for use in teaching/learning solutions. An agile development process was tailored to be used in the development of AR applications. The process was then instantiated in the development of an AR solution for learning purposes of ships’ maneuvers in formations in the context of naval operations. The effectiveness of the AR solution was then validated through statistical tests of data collected from a survey to a sample of users. Keywords Augmented reality · Mobile applications · Learning application · Handheld display · Naval operations

1 Introduction The augmented reality (AR) aims for a more interactive communication, closer to a real environment, such as the simulation of the actual presence of people and objects, or the presentation and manipulation of virtual objects close to the interlocutor with whom the communication is done. The characteristics that define AR are the combination of real and virtual, real-time interactivity, and the 3D experience [1]. With AR, it is possible to improve the human senses by enriching the real world with virtual information (dynamic images, spatial sounds, and haptic sensations), generA. Correia (B) · A. Gonçalves CINAV—Alfeite, 2810-001 Almada, Portugal e-mail: [email protected] M. Zacarias Universidade do Algarve, FCT/DEEI· Gambelas, 8000-139 Faro, Portugal © Springer Nature Singapore Pte Ltd. 2020 Á. Rocha and R. P. Pereira (eds.), Developments and Advances in Defense and Security, Smart Innovation, Systems and Technologies 152, https://doi.org/10.1007/978-981-13-9155-2_37

475

476

A. Correia et al.

ated by computer in real time and positioned in space through special technological devices [1]. So, AR aims at greater integration between the user and the display device. This symbiosis intends more easiness in information sharing of presented in the devices, namely with greater authenticity in the projection of game scenarios and teaching/training contents [2]. The benefits obtained with AR are translated on an improved performance of everyday tasks when people wearing AR device access, for example, for an on-screen projection of contextualized information that help them to perform complex tasks. In real or simulated war scenarios, the potentialities of AR are evident, namely in the situational awareness of a conflict scenario and in support of the decision-making process of the forces in the operations theater. In the navy, the garrison of a ship needs to train in different scenarios, as closer as possible to real situations in order to obtain a high degree of readiness. The training on board of a ship enables the consolidation of knowledge, previously acquired, through practice. One of the important skills to exercise is the maneuvers of the ship in the context of a naval force. Training is thus a basic requirement for the readiness of the ship’s crew. When the training is performed in real environment, it is usually carried out at the stations on board, whether the ship is part of a fleet or is standalone. Being this the ideal training situation, however, there are obstacles to its full realization, namely the cost that is involved navigation of several ships in open sea. An alternative to overcome this drawback is the training in simulators or in simulated environments. As examples of the creation of total or partially virtualized scenarios, we can have a ship (isolated or as part of a fleet of virtual or real ships), training procedures of the ship or fleet regarding, among others, damage limitation, battle stations for one or multiple threats (aerial, submarine, and surface), maneuvers and evolutions, tactical operations, refueling operations, or navigation in restricted waters. This paper intends to report a work carried out aimed at the construction of an AR application, to support the teaching and learning of vessel’s maneuvers in naval formations. The article was developed in six sections. In this section, we presented the motivation for the theme. In the second section, the technical characteristics of AR are presented, as well as some of the existing applications in the context of the armed forces. The methodology used for the development of the AR application is also presented. The third section describes the process of collecting functionalities from the target users and details the steps developed in the construction of the proposed AR solution. The fourth section presents the process of validation of the built solution, based on a survey designed to obtain feedback from users. Finally, in the last section, we present the conclusions and the future work.

Using Augmented Reality for Learning Naval Operations

477

2 Context of Augmented Reality 2.1 AR Systems and Devices Nowadays, it is possible to visualize virtual information over real objects or scenarios, in a context of AR, through two main techniques: optical combination or video mixing. Through optical combination, the user has a direct view of the virtual object; that is, the object is generated on the environment that the user observes. In video mixing, as the name implies, there are two moments in the process of composing the final image: (1) video recording, which captures the real environment, and (2) the subsequent addition of objects on the video that, in the end of the mixing process, will be presented to the user, thus merging the real environment and the virtual objects [3]. AR display devices are classified into three categories: handheld displays, spatial displays, and head-attached displays [3]. This devices’ classification is based on how the user interface is performed, or, in other words, how the user uses the devices. Although the devices have the same objectives, they use different technologies and methods of visualization—varying their optical, mechanical and electronic components—with each of the devices being more appropriate for specific situations. In addition, the distance of the viewer to the real object is variable on the different devices. In the context of the present work, we chose to use handheld displays as preferred devices, since they are highly versatile equipment for learning, capable in most of them to support AR technology with acceptable quality, thus conferring a good experience of use. In addition, the fact that most students have a mobile phone makes it easier to spread out the application. Although the head-attached displays are able to ensure a more immersive environment, they still have the drawback of their higher cost.

2.2 AR Marine Applications The AR technology offers applications for purposes both recreational and professional, related, the latter, with military areas. AR began by demonstrating its potential, especially in the area of aviation, both military and civil. The U.S. Air Force took the first steps in using this technology with the creation of teams to work on the creation of systems to assist the pilot during the flight and in the definition of firing angles. With the heads-up display (HUD) system, it was possible for the pilots to make a more precise firing line on the targets, performing the necessary calculations for the effect [4]. This system displays the information in a transparent fixed area, with the observer having only one viewing angle for information reading. In military aviation, head-mounted displays (HMDs) are replacing HUDs with other advantages and features, such as an allowed higher viewing angle [4].

478

A. Correia et al.

In the case of the navy, there are numerous developments in this field. A helmet for divers allows the clarification of the information in an underwater environment, as well as the help and explanation in the assembly of equipment [5]. Another project, GunnAR [6], allows a remotely transmitted firing order to appear on the visor of the helmet intended for machine gun operators aboard ships. Also, for land forces there are helmet projects for aiding and simplifying tactical scenarios, making training more realistic [7]. In the specific area of navigation, AR applications were developed for navigation aids (see Fig. 1). With the common objective of making the navigation more secure and avoiding accidents at sea, these programs mainly aim to identify dangers and ships in transit, allowing a better clarification of the situation. This type of aids facilitates the work of the ship, mainly the larger ones, especially in situations of greater traffic such as the entry and exit of ports [8]. Another project is the one of building a ship bridge equipped with AR technology, capable of notifying warnings to navigation, dangers in the area and meteorological warnings, in order to guarantee the safety and facilitate the crossing of the ships, especially in situations of increased danger as in arctic navigation [9]. In areas important for navigation and subsistence of a ship at sea, assistance can be provided with the support of AR goggles allowing technicians an efficient installation or assembly of intended equipment (e.g., antennas) [11].

Fig. 1 AR application for sailing (Source [10])

Using Augmented Reality for Learning Naval Operations

479

2.3 Development Methodology In an AR project, it is necessary to adopt a methodology that allows the definition of the users’ requirements and the construction of the solution. In multimedia software development projects, the use of agile methodologies by application development teams is often due to the positive results that are generally associated with it [12]. Because the development of AR solutions presents very similar contours to the creation of multimedia software, the Scrum (see Fig. 2) was used in the realization of the AR solution of the present work. Thus, the construction project of the AR solution has started with a list of needs, that is, the requirements that had been considered in the development of the solution, being this list the execution’s guide of the project. The evaluation and estimation of the effort and time spent in each task, presented in the list, were also carried out, in order to verify whether the activities were feasible in the forecasted execution time. When the list was completed, the cycle of the sprints began, ending when the proof of concept was completed. Each sprint began with its planning, in which the tasks to be performed by each individual during the sprint were outlined. The preparation of a Scrum board was the solution for accounting of time. In this board, three columns were represented: (1) the work developed; (2) in development; and (3) what was lacking to develop. The goal at the end of each sprint was that all tasks to be located in the third column.

Fig. 2 Scrum methodology (Source [13])

480

A. Correia et al.

Frequent meetings were held with the project coordinator to monitor the work. At the end of each sprint, a meeting was held with all the elements that contributed to the sprint, team and users, where the result obtained in the sprint was presented. Finally, the completed solution was been demonstrated to users for experimentation and feedback. This feedback will contribute to the subsequent development of the work in order to improve, in the following sprints and in a new cycle, aspects less achieved in the solution to date.

3 Solution Development The AR application being build was intended to be a proof of concept for learning ship maneuvers in orderly formations, according to ATP I Vol. II [14]. The aim of the application was also, besides teaching the trainees about maneuvers of ships, a demonstration of the potential of AR on naval training. In order to ensure that the requirements of the solution were aligned with the needs of the learning process, a requirements’ collecting was made among seasoned students of a navy academy. The information was collected through brainstorming sessions with answers obtained, validated later through a questionnaire. The collected requirements, for the application of AR, to be developed were as follows: (a) the minimum number of vessels, in order to make the movement of vessels and the final formation disposition noticeable, should be 3 vessels; (b) the perspective (position and viewing angle) to observe the evolution of the formation should be flexible, allowing different positions and angles (bow, stern, and cross); (c) the virtual image should emerge from a sheet, allowing to be observable from all angles; (d) the user should be able to interact with the virtual objects; (e) it should be possible to select different weather conditions without affecting the visibility of the objects in the scenario or the response time of the application; (f) the representation of the guide vessel should be different from the other ships of the formation, since the latter takes different behaviors from the other vessels in the different formations; (g) before the ships move, they should take a position at an initial in-line formation and the virtual ships maneuver, adopting trajectories to occupy the selected formation, with each vessel acting independently; (h) the type of mark used to trigger the display of the virtual formation should be a mark inserted imperceptibly in the scenario; (i) the characteristics of the devices that users prefer to use are head-mounted displays and handheld displays (smart phones), because they are portable. Based on the user requirements raised, sketches, on the form of use cases, were drawn up regarding the maneuvers of the virtual vessels in formations. The movement of the ships was to be triggered after the reading, by the smart phone, of artificial mark specific of each use case. An example of a use case, the outline of which is shown in Fig. 3 is the FORM E formation, where the vessels maneuver in order to arrive at the depicted final formation. Starting from a straight formation, ships will slightly open the bow and hold it, passing to a final position on the front ship’s wing. The first ship of the

Using Augmented Reality for Learning Naval Operations

481

formation, which will be the guide ship, keeps its movement and bow. Paired vessels will be placed on the port side of the guide, with a 4° opening in the wing. The odd ships will form with a 2° aperture in the starboard wing of the guide ship. In the sketch presented in Fig. 3, the ships appear viewed from above; however, in the developed application, they can be seen from all positions, including the vertical. Based on the Scrum methodology, the solution was built through several sprints, in which it was possible to improve the functionality and graphic design of the application, with the contribution and feedback received from the users. The requirements collected were implemented in an AR application developed with the support of the Unity platform with Vuforia [15]. This integrated development environment (IDE) supports multimedia development and possesses a wide range of functionalities that facilitate the creation of AR/VR products. The implementation of the solution was based on the following technical options: (a) creation of the ship formation by mixing the real and virtual scenarios (see Fig. 4); (b) recognition of the marks of the formations residing in a database (see Fig. 5); (c) movement of vessels in the formations; (d) precipitation and rough sea in predefined scenarios (see Fig. 6); (e) observation of the formation within the mark detection distance (see Fig. 4).

Fig. 3 Formation of ships in FORM E

482

A. Correia et al.

Fig. 4 AR solution: FORM E visible in the application running on Android phone

Fig. 5 AR solution: vessels aligned at the formation corresponding to their mark

4 Results’ Validation In the end of the development process, it was intended to verify if the built solution satisfies the requirements defined by the users. For this purpose, data obtained from a survey of 40 respondents were analyzed, most of them, seasoned students from the naval academy. Most of the respondents had already accompanied naval operations on the bridge of a ship, where they can have a better visual perception of the maneuvers of the ships. In an initial phase of the survey, the knowledge of the respondents was scored regarding how well they knew the maneuvers and evolutions, without consulting any supporting material. Through a set of questions, where one could visualize images of

Using Augmented Reality for Learning Naval Operations

483

Fig. 6 AR solution: scenario with rain and rough sea

different naval formations, it was intended that the respondents were able to match the images and the name of the formation. Most of the answers were incorrect, revealing lack of consolidated knowledge, of the sample of respondents, regarding the depicted maneuvers and evolutions of ships. The following questions of the survey were answered after a previous visualization of the AR solution and observation of the evolution of the formations evolving. It was intended to verify whether the respondents understood the operation with the AR solution, had seized the presented information, and improved their knowledge on the subject. After viewing the solution, the correct response rate increased considerably. The percentage of correct answers was above 80%. The results seemed to indicate effectiveness in the transmission of the concepts through the AR solution and greater knowledge capture by the respondents. In order to statistically validate the respondents’ answers, we proceeded to the development of hypothesis tests. It was intended to test, if it is right to generalize to the overall population, the average opinion of the answers obtained in the sample. As random variable of the statistical distribution, it was considered the answer to questions posed questions, with values ranging from 1 (weak) to 6 (very good). For the questions posed, it was intended to validate statistically if the mean of responses was above the value 3 (satisfies). In the presence of a sample and ignoring the mean and the variance of the population, the t-test was used in hypotheses tests. The tests were unidirectional, with an alpha significance level of 0.05. With these assumptions, we tested the hypothesis regarding the question: Is the AR built solution effective for learning naval operations? In order to evaluate the statistical significance of the results, hypothesis tests were performed (see Table 1), to the answers obtained to the following questions in the

484 Table 1 Statistics of hypothesis tests, for AR solution validation

A. Correia et al.

T -test 1P

2P

3P

Average

4.85

5.15

4.65

Variance

1.207692

0.64359

1.053846

Responses

40

40

40

Hypothesis of average difference

3

3

3

gl

39

39

39

Stat t

10.64691

16.94977

10.16542

P(T 1.685, which happens for all the questions, so H 0 was rejected. Likewise, P one-tailed must be less than alpha (0.05) to reject H 0 , which is also true for the three questions. It can therefore be concluded that for the collected sample, H 0 should be rejected for the three questions, and therefore, H 0 should also be rejected for Q1. The results allowed to conclude that one cannot reject the hypothesis that the constructed AR solution is effective for the teaching of naval operations.

5 Conclusion This paper describes the work carried out in order to build an AR application to support the teaching and learning of vessel maneuvers in naval formations. In the context of the present work, a handheld display (smart phone) was used as an AR visualization device, due to the versatility of this type of equipment for its purpose, namely to be able to withstand AR technology with acceptable quality and good user experience. In order to collect user requirements and develop the AR solution, the agile methodology Scrum was used, given its versatility, namely for the construction of multimedia applications. As a platform for the development of the AR solution, was used Unity with Vuforia, due to the functionalities that it makes available for the construction of games and multimedia applications. The built solution was validated through a survey of a sample of 40 users. The answers obtained were the object of

Using Augmented Reality for Learning Naval Operations

485

statistical analysis to evaluate the susceptibility of generalization of the results, which proved to be promising. Future work intends to iterate the development of the proof of concept with new sprints and extend the AR development to new applications in the naval context. Acknowledgements The work was funded by the Portuguese Ministry of Defense and by the Portuguese Navy/CINAV.

References 1. Azuma, R.T.: A survey of augmented reality. Presence: Teleoper. Virtual Environ. 6(4), 355–385. ACM Digital Library (1997) 2. Peddie, J.: Augmented Reality, Where We all Live. Springer International Publishing, Tiburon (2017) 3. Bimber, O., Raskar, R.: Spatial Augmented Reality: Merging Real and Virtual Worlds. AK Peters/CRC Press (2005) 4. Aukstakalnis, S.: Pratical Augmented Reality. Addison-Wesley, Crawsfordville (2016) 5. Wheeler, D.: U.S. Navy developing new augmented-reality diver’s helmet. https://bit.ly/ 2SG0L4J. Accessed Apr 2018 (2016) 6. Owens, K.: The Navy tests new augmented reality helmet, automation. https://bit.ly/2GA4oCQ. Accessed Apr 2018 (2017) 7. Seck, H.: Augmented reality training systems destined for marine schoolhouses. https://bit.ly/ 2E6EncD. Accessed Apr 2018 (2016) 8. Baldauf, M., Procee S.: Augmented reality in ships bridge operation. In: ISIS 2014—International Symposium Information on Ships, Hamburg (2014) 9. Wingrove, M.: Augmented reality is coming to ship bridges. https://bit.ly/2DsXD2E. Accessed Apr 2018 (2018) 10. Azadux: The UX development story. http://azadux.com/sailing/. Accessed Apr 2018 (2018) 11. Russel, K.: Cobham explores augmented reality for maritime satcom. https://bit.ly/2I4ZFLC. Accessed Apr 2018 (2017) 12. Schwaber, K., Sutherland, J.: The Scrum Guide. Scrum Alliance (2011) 13. Medlock, J.: A short introduction to the scrum methodology. https://bit.ly/2THpnqV. Accessed Apr 2018 (2018) 14. NATO, ATP I Vol. II, Allied Maritime Tactical Signal and Maneuvering Book (2016) 15. Unity. https://unity3d.com. Accessed Apr 2018 (2018) 16. Elvas, F.: Realidade Aumentada Aplicada a Panoramas Táticos. M.Sc. Thesis, Portuguese Naval Academy. https://bit.ly/2Ducn10. Accessed Dec 2018 (2018)

Communication of the Quality of Higher Education Institutions of Ecuador Guillermo Santa María, Carina Rey, Lilian Molina and Arturo Clery

Abstract The objective of this work was to determine the fulfillment of the attributions and functions of the Ecuadorian agency, which regulates the management of quality assurance of higher education, declared in its Communication Plan, through an exploratory investigation of the web pages of all regulated universities. For this purpose, standards were designed from the Plan, which collected the pertinent information through a scale. The results are classified based on six elements that the Plan itself highlighted, namely communication management, its dissemination and positioning, the image of the authorities, links with other universities, management with international institutions and the empowerment of the mission and institutional vision. Each of these elements presents its own characteristics and is analyzed in the present work. Keywords Communication · Quality · University

1 Introduction Quality, as a concept, has been developed in response to the demands of contexts. There is agreement, among specialists, that it is a complex concept, relative and consequently impossible to conceptualize univocally [13]. G. S. María Universidad de Guayaquil, Guayaquil, Ecuador e-mail: [email protected] C. Rey Universidad de Barcelona, Melcior de Palau, Barcelona, Spain e-mail: [email protected] L. Molina (B) · A. Clery Universidad Estatal Península de Santa Elena, Santa Elena, Ecuador e-mail: [email protected] A. Clery e-mail: [email protected] © Springer Nature Singapore Pte Ltd. 2020 Á. Rocha and R. P. Pereira (eds.), Developments and Advances in Defense and Security, Smart Innovation, Systems and Technologies 152, https://doi.org/10.1007/978-981-13-9155-2_38

487

488

G. S. María et al.

The application of the theories of quality to higher education in Latin America, after initial resistance, has been assumed as necessary and essential. From a vision of public policies, the existence of a new pact between the State and the universities is now admitted: the accountability of the universities through the fulfillment of certain quality standards [11]. During the second half of the last century and the first two decades of the present, HEIs in Latin America, adjusted their reality to the evolution experienced by this concept. In the 1990s, the issue of quality as an ingredient of reforms to higher education appears on the agenda of Latin-American countries [1]. In Ecuador, the Constitution that came into effect in 1998, determined, in order to ensure the quality objectives, that Higher Education Institutions—IES, would be obliged to socially account, and for this purpose established an autonomous system of evaluation and accreditation. As a result, the Higher Education Law issued in April 2000, in accordance with the constitutional mandate, contemplates a National System of Evaluation and Accreditation, which would be directed by a National Council for Evaluation and Accreditation, CONEA. In order to achieve quality in higher education, the CONEA considered a dynamic network of interactions: the quality of the input (input), the quality of the process (output), and the quality of the result (outcome), arguing that in that network it was necessary to take into account some elements, such as the quality of the educational impact in the context, the quality of the offer, the curriculum, the demand, and the quality of the design and the implementation. In 2008, with the coming into force of the twentieth Constitution of the Republic, it was determined that the higher education system would be governed by the principles of responsible autonomy, co-government, equality of opportunities, quality, relevance, integrality, and self-determination for the production of thought and knowledge, within the framework of the dialogue of knowledge, universal thought and global scientific and technological production. With this Constitution, the governance of the system is in charge of a public entity of planning, regulation and internal coordination of the system, and the relationship between its various actors with the Executive Power, this is called Higher Education Council—CES; and, a second public technical entity for accreditation and quality assurance of institutions, careers and programs, which is called the Council for Evaluation, Accreditation and Quality Assurance of Higher Education—CEAACES, which cannot be made up of representatives of the institutions that are the objects of regulation. In 2010, the new and current Organic Law on Higher Education—LOES is approved, and the purpose of the regulation are to define its principles and guarantee the right to quality higher education that strives for excellence [8]. It is not the right to any higher education; the text of restrictive form places the word “quality”. Admission to public institutions of higher education began to be regulated through a system of leveling and admission, to which all applicants submit themselves; is in charge of the Secretariat of Higher Education, Science, Technology and Innovation—SENESCYT, in coordination with the Ministry of Education, regarding its

Communication of the Quality of Higher Education Institutions …

489

articulation between the bachelor’s degree or its equivalent and higher education, seeking only the entrance of students that meet minimum standards. With the validity of the LOES, the State regains control of the university system, while initiating efforts for the public policies to constitute instruments of communication between the State and universities [3], and a way through which rulers seek to change the direction, in a positive way, of the conditions of poverty and marginality of the population [6]. The LOES enforceable, from 2010 to August 2018, defines the principle of quality in its Article 93, as “the constant and systematic search for excellence, relevance, optimal production, transmission of knowledge and development of thought through self-criticism, external criticism and permanent improvement.” The CEAACES until 2018, in addition to organizing the processes of improvement, had in its task a clear orientation toward the purification. The purge ended with the approval by the National Assembly of the Organic Law of Extinction of Universities and Polytechnic Schools, suspended by CEAACES. The CES, motivated by the resolution that approved the regulation for the designation and powers of the board members of the trust called “Mandate 14—more quality,” defined that period as the final closure of a stage that represented the existence of institutions that did not meet the quality parameters [9], was also considered a way to guarantee the right to quality higher education and establish mechanisms to ensure social accountability, distribution and efficient use of public resources in favor of the institutions of the higher education system. In a second reform to the LOES (the first one was carried out in December 2016), promoted by a new Government in August 2018, quality is redefined as “the continuous, self-reflective search for improvement, assurance and collective construction of the culture of higher educational quality with the participation of all levels of higher education institutions and the higher education system, based on the balance of teaching, research and innovation and the link with society, guided by relevance, inclusion, the democratization of access and equity, diversity, responsible autonomy, integrality, democracy, the production of knowledge, the dialogue of knowledge, and citizen values.” This new definition establishes that quality should be guided by the principles, which are already determined by the norm as natural to the system: relevance, responsible autonomy, integrality and dialogue of knowledge, and also incorporates new guiding principles of quality: inclusion, equity, diversity, democracy, and citizen values. This definition is close to that of the United Nations Educational, Scientific and Cultural Organization—UNESCO, which at its Regional Conference, held in the city of Havana in 1996, agreed to define quality as “the adequacy of the being and the to do of higher education, to its duty to be.” Where the being must be understood as contained in the vision, mission, principles, and values of the university and what to do corresponds to the operative and procedural aspects. These last approach universities to the proposed ideals, through systems of learning, research, management, selfevaluation, and external evaluation, linked to the context [5].

490

G. S. María et al.

Extrapolating the concept of UNESCO to the Ecuadorian University System of 2018, it can be said that the being is contained in the principles that appear in Article 351 of the current Constitution and those declared in Article 93 of the LOES of 2018, the to do is subject to the self-evaluations in HEIs and the advance or retreat toward these ends, should be monitored by CEAACES.

2 The Evaluation System of Higher Education in Ecuador An autonomous system of evaluation and accreditation was established in Ecuador with the constitutional norm of 1998, of coordination and cooperation with the National Council of Higher Education—CONESUP. It is the first occasion in which a system is declared to address, officially, in its integrality, the higher education [10]. Up until that moment, the Ecuadorian State had maintained a traditional relationship with its universities. University autonomy was the main concern and, in some cases, it had become an alibi to avoid social accountability. It was inevitable that the State initiate efforts to articulate a response to the expansion of enrollment, the proliferation of new centers of higher education and the political, economic, and social changes that the very dynamics of Ecuadorian society promoted. The Higher Education Law of the year 2000 indicated that this system would be directed by the National Council for Evaluation and Accreditation—CONEA and that it would be governed by its regulations. In 2002, the General Regulations were issued and the actions of the system were made viable. Four years had elapsed since the mandate of the constitutional norm. The National System of Evaluation and Accreditation of Higher Education, was in charge of institutional self-evaluation, external evaluation, and accreditation. This process began with the institutional request to CONEA by the university or polytechnic school that decided to start the process. If solitude was not found, the process did not start. By 2008, the year in which Mandate 14 of the Constituent Assembly mandates CONEA to prepare the report on the level of institutional performance of Ecuadorian HEIs, from a total of 62 universities: 5 were accredited, 13 were in the process of evaluation external, 18 had carried out a self-assessment exercise and delivered a report, 11 were in the design stage of the self-evaluation project, 9 had not started the self-assessment process, and 7 did not comply with the legal time [14]. The slowness with which the system had operated, the absence of relevant changes, more allegations of corruption in the delivery of academic titles, were elements of concern and would give rise to the decision of the National Assembly expressed in Constituent Mandate No. 14. The CONEA was composed of the following: 2 academics appointed by the President of the Republic, from three candidates drawn up by the University Assembly and qualified by the CONESUP, 2 academics appointed by the National Congress, from lists prepared by the University Assembly and qualified by the CONESUP; 1 member appointed by the Minister of Education and Culture; 1 member appointed by

Communication of the Quality of Higher Education Institutions …

491

the National Federations of the Professional Associations; 1 member designated by the state science and technology agency; and 1 member appointed by the Federation of the Chambers of Production. The integration reflects a predominance of academics from the organizations subject to evaluation and accreditation, as well as a marked interference by the regulatory body, CONESUP. After 10 years, from 1998 to 2008, the new constitutional norm puts the accent on the organisms that will govern the higher education system. One of them will be a technical body for accreditation and quality assurance of institutions, careers and programs, and cannot be made up of representatives of the institutions subject to regulation. The denomination of the constitutional text system, on this occasion, is used to substantiate higher education; we must remember that in the Constitution of 1998 the denomination of system was used to substantiate the evaluation and accreditation processes. This new Constitution of 2008, imperatively states that, within five years, all HEIs, their careers and programs and postgraduate programs are evaluated and accredited, and warns that if they do not pass the evaluation and accreditation, they will be left out of the system of higher education. In replacement of the CONEA, the LOES issued in 2010, will give rise to the Council for the Evaluation, Accreditation and Quality Assurance of Higher Education—CEAACES. For the first time, the assurance of the quality in the denomination of the organism appears. The qualified evaluation and accreditation system is attached into the faculties of the new Council. CEAACES implements the Twentieth Disposition of the Constitution and carries out the evaluations and accreditations established for the universities, Polytechnic Schools and their extensions. It also complies with the CONEA report and evaluates, within 18 months after the enactment of the Law, the universities and Polytechnic Schools that were placed in category E. The constitutional norms establish that the Law defines the coordination mechanisms for the higher education with the Executive Power. The LOES to comply with this provision creates the National Secretariat for Higher Education, Science and Technology and Innovation, SENESCYT, to coordinate actions between the Executive Function and the institutions of the Higher Education System. This Secretariat is headed by an official appointed by the Executive. In this way, the presence of the State, through the officials it designates for the regulation, planning, coordination, evaluation, accreditation, and assurance of the quality of the higher education system, dominates almost the entire spectrum of the system. Before 2008, the State had avoided having a predominant presence, with the enforcement of the new Constitution and the new LOES, the State’s role became central in the integration of the entities that govern the so-called higher education system The reform of the LOES of August 2018 incorporates SENESCYT as part of the higher education system. And the task to ensure the effective compliance of the principle of quality is delivered to the Inter-institutional Quality Assurance System—SIAC.

492

G. S. María et al.

This new system is integrated by the CES, the CACES, and the IES. The selfevaluation is given a permanent role and turned into the cornerstone, which must be carried out in reference to the purposes of the institutions. Thus, it must be understood that the missions and visions of the universities would become the referents of the aforementioned self-process. The process will be accredited by CACES, who will plan and coordinate the operations. In this way, the State remains at the forefront of the system even when the self-evaluation processes, which are initiatives of the university institutions, acquire centrality. Harmonizing the dynamics of state agencies with universities will be a process to be built. Internal processes, such as self-evaluations, and external ones, such as accreditations, must interact, mesh, and function effectively. This interaction is on the horizon of the Ecuadorian system.

3 Quality Assurance The CONUEP in 1989 carried out the first diagnostic study of the Ecuadorian university, and one of the basic problems detected was the lack of a higher education system; instead there existed a set of independent institutions, with weak cooperation mechanisms. The island scenario of university institutions to which the previous study alludes had in the defense of university autonomy a unifying link: Defending autonomy was indispensable and there ones that when, invoking it, attributed to it thaumaturgical properties and associated it with quality were few. Nine years passed so that, in the Constitution issued in 1998, the constitutional norm warns that to ensure quality objectives, universities should perform an accountability; for its fulfillment, an evaluation and accreditation system is formed. This is the first normative evidence of concern for quality and its assurance. Two years later, in 2000, the Higher Education Law ratifies that the objective of the evaluation and accreditation system is to ensure quality, encourage ongoing processes to improve academic quality and management, and to do so, appeals to the processes of institutional self-evaluation, external evaluation, and accreditation [12]. The following objective, established in the Law, was for Ecuadorian society to be informed of the institutional performance of higher education establishments in order to guarantee quality, promoting their depuration and improvement. The wording of the norm’s text suffers from lack of clarity; however, its concern for the depuration and improvement is noticed. The 14th Mandate, in 2008, includes this intention and expresses it in its resolutions. After the 21 universities, which in 1989 had been subject to evaluation by the CONESUP, the creation of private and public universities had been approved in a number close to 70. An increase of more than 200%.

Communication of the Quality of Higher Education Institutions …

493

In June 2006, only the Universidad del Azuay had fulfilled all the requirements and was accredited. A context of application of neoliberal policies in the university system, of commercialization of higher education and indifference of the State in relation to definitions and regulations, had introduced chaos and discredit [7]. Reconciling with the concepts of quality was urgent as well as the need for depuration and improvement. Reality, the political and economic context imposed their logic, while the norm remained a dead letter and organisms limited themselves to contemplate the decadence. The 14th Mandate reflects the disappointment of the process entrusted to CONESUP and CONEA and obliges CONESUP to prepare a report on the academic and legal situation [2], and CONEA is required to submit a technical report to comply with the depuration and the improvement as established in Article 91 of the Law of Higher Education of that time. In November 2009, CONEA delivers its report, which consists of an initial classification of Ecuadorian universities and a radical recommendation: Purge the Ecuadorian university system of the group of universities that did not meet the basic conditions of a university institution [4]. These universities had been created during the 1990s and the early years of the twenty-first century. The LOES of 2010 establishes for these universities identified in the 14th Mandate, type E, the prohibition of offering new academic programs and/or postgraduate courses; at the same time, they have to be evaluated within a period of 18 months. This new evaluation was in charge of the new body that the LOES created in substitution of the CONEA: the Council of Evaluation, Accreditation and Quality Assurance of Higher Education—CEAACES. The assurance of quality defines it as actions that guarantee efficient and effective management. And this definition extends to the Councils and accrediting bodies.

4 Final Considerations Quality assurance appears as a component of the name of the new body; it is assumed as one more process of the system. The concern in the previous stage had been: evaluation and accreditation. Now, the assurance is constituted in a process to be implemented, it stops being an objective without concretion. It is essential to shield what has been achieved in order to avoid backing down, and in parallel, it is forced to reach new levels of improvements. In the period from 2010 to 2018, CEAACES carries out evaluations of careers, exams for qualification of professional practice, depurates extensions, participates in interventions in universities, re-categorizes, and accompanies processes of improvement plans, institutional strengthening, and quality. And finally, in 2018, we find ourselves with a quality assurance which results show an asymmetric reality hidden in the accreditation of 55 universities out of a total of 59, extensions 39, and 56 evaluated careers. The qualification exams for

494

G. S. María et al.

professional practice have been applied in careers that could compromise the public interest, putting at risk essentially the life, health, and safety of the citizens; between 2015 and 2018, the qualification exams for professional practice—EHEP have been applied on nine occasions for students of medical careers, seven times for the career of dentistry, and twice for students of the nursing career. The reform introduced, in August 2018 to the LOES, has the HEIs whose quality has been accredited from the asymmetry, as a base line. Thus, on the official website of CEAACES, there are 18 universities in category C, 28 in category B, and 8 in category A. The universities categorized C, which correspond to 30% of the HEIs, should accelerate their improvement processes in order to narrow the existing gap; in a similar situation are categorized B of the A. On the other hand, the norm did not contemplate or contemplates a reassurance process for the system. Neither CONEA nor CEAACES from 2010 to 2018 have been evaluated technically by independent bodies to ensure that their work effectively meets acceptable standards. Acknowledgements Thanks are due to the Universidad de Guayaquil and the Universidad Estatal Península de Santa Elena for their administrative and economic contributions to the development of this research, through its Public and Social Policy Observatories.

References 1. Cielo, C., Ospina, P., Vega, C.: Reforma y renacimiento: conversacones docentes sobre la reforma universitaria en Ecuador. Universidad Andina Simón Bolívar, Quito (2016) 2. Ecuador: Legislación comparada de nueve países de América Latina y españa. Consejo Nacional de Educación Superior, Quito (2009) 3. Ecuador: Ley Orgánica de Educación Superior (2010) 4. Ecuador: La evaluación de la calidad de la Universidad ecuatoriana: la experiencia del Mandato 14. Consejo de Evaluación, Acreditación y Aseguramiento de la Calidad de la Educación Superior—CEAACES, Quito (2014) 5. Gutiérrez-García, R., Sainz-Venegas, P., Contreras-Cornejo, A., Lugo-Carvajal, M., PreciadoAlatorre, T., Ornelas-Jiménez, J.: Universidad Sustentable. Instituto Tecnológico y de Estudios Superiores de Occidente, Jalisco (2017) 6. López, M., Cifuentes, L., Palma, A.: La gestión de las políticas públicas en las universidades: una aproximación a su caracterización en el Ecuador. Dominio de las Ciencias, 447–464 (2018) 7. Lucas, K.: De los bancos de papel a las universidades de garaje: impactos de las políticas neoliberales en la educación superior en Ecuador. Consejo de Educación Superior CES, Quito (2015) 8. Martínez, A., Vásquez, P.: La importancia de la evaluación en las instituciones educativas conforme a la nueva Ley Orgánica de Educación Superior en el Ecuador. Revista Iberoamericana de Evaluación Educativa, 174–180 (2012) 9. Molina, L.: Evaluación de las biblotecas universitarias ecuatorianas: análisis del entorno y propuesta de un catálogo de indicadores. Universidad Estatal Península de Santa Elena, La Libertad (2015) 10. Rama, C.: La educación a distancia y virtual en Ecuador. Universidad Técnica Particular de Loja UTPL-Virtual Educa, Loja (2015)

Communication of the Quality of Higher Education Institutions …

495

11. Toscanini, M., Aguilar, A., García, R.: Diagnóstico de las políticas públicas de la educación superior en el Ecuador. Revista Cubana de Educacaión Superior, 161–178 (2016) 12. Valdés, M., Villegas, T.: Responsabilidad social universitaria: fundamento para la gestión sociocultural. Revista Cubana Educación Superior, 55–62 (2017) 13. Villarroel, C.: Gerencia, planificación y evaluación universitarias. La Revista Venezolana de Educación Educere, 513–522 (2005) 14. Villavicencio, A.: De la universidad funcional a la universidad de la razón. Universidad Andina Simón Bolívar, Quito (2013)

The Four Dimensions of the GDPR Framework: An Institutional Theory Perspective Isabel Maria Lopes, Teresa Guarda and Pedro Oliveira

Abstract The EU general data protection regulation (GDPR) is the most important change in data privacy regulation in 20 years. The regulation will fundamentally reshape the way in which data are handled across every sector. The organizations had two years to implement it. Despite this, it has been observed that, in several sectors of activity, the number of organizations having adopted that control is low. This study aimed to identify the factors which condition the adoption of the GDPR by organizations. Methodologically, the study involved interviewing the officials in charge of information systems in 18 health clinics in Portugal. The factors facilitating and inhibiting the implementation of GDPR are presented and discussed. Based on these factors, a set of recommendations are made to enhance the adoption of the measures proposed by the regulation. The study used Institutional Theory as a theoretical framework. The results are discussed in light of the data collected in the survey, and possible future works are identified. Keywords Regulation (EU) 2016/679 · General data protection regulation · Institutional Theory · Health clinics

I. M. Lopes (B) · T. Guarda Centro ALGORITMI, Universidade do Minho, Braga, Portugal e-mail: [email protected] T. Guarda e-mail: [email protected] I. M. Lopes UNIAG (Applied Management Research Unit), Polytechnic Institute of Bragança, Bragança, Portugal I. M. Lopes · P. Oliveira School of Technology and Management, Polytechnic Institute of Bragança, Bragança, Portugal e-mail: [email protected] T. Guarda Universidad Estatal Peninsula de Santa Elena—UPSE, La Libertad, Ecuador © Springer Nature Singapore Pte Ltd. 2020 Á. Rocha and R. P. Pereira (eds.), Developments and Advances in Defense and Security, Smart Innovation, Systems and Technologies 152, https://doi.org/10.1007/978-981-13-9155-2_39

497

498

I. M. Lopes et al.

1 Introduction In recent years, data protection has become a forefront issue in cybersecurity. The issues introduced by recurring organizational data breaches, social media and the Internet of Things (IoT) have raised the stakes even further [1, 2]. The EU GDPR, enforced from May 25, 2018, is an attempt to address such data protection. The GDPR makes for stronger, unified data protection throughout the EU. Not all data is protected by the GDPR, it is only applicable to personal data. This is defined in Article 4 as follows [3]: “personal data” means any information relating to an identified or identifiable natural person (‘data subject’); an identifiable natural person is one who can be identified, directly or indirectly, in particular by reference to an identifier such as a name, an identification number, location data, an online identifier or to one or more factors specific to the physical, physiological, genetic, mental, economic, cultural or social identity of that natural person.

In a study [4] conducted in 57 health clinics (the survey was sent to 190 clinics, but only 57 gave an effective answer) about whether they had started or completed the process of implementation of the measures laid down in the GDPR, 39 (69%) gave a negative answer and 18 (31%) said to have started or completed the adoption of such measures. Among the 18 clinics which are in the implementation stage, 7 (39%) consider that the process is completed, while the remaining 11 (61%) still have some measures to implement. These figures seem insufficient when bearing in mind that the GDPR had already come into force six months before the conduction of this study. Thus, the following two research questions were formulated in order to guide the research work: 1. Which dimensions condition the adoption of the GDPR? 2. Which recommendations might be put forward so as to enhance the adoption of the GDPR by health clinics? The answer to the first question aims to know the positive and negative conditioning dimensions influencing the adoption of the GDPR by health clinics. In possession of these elements, it will be relevant to produce a set of recommendations which enable the adoption of the GDPR measures by clinics. The structure of the present work consists of an introduction, followed by a desk review on the general data protection regulation and the desk review of Institutional Theory. Section 4 focuses on the research methodology, identifying the target population and the structure of the survey. The results of the study are presented in Sect. 5, and Sect. 6 consists of the conclusions drawn from the study. Finally, the limitations of this research work are identified and possible future studies are proposed.

The Four Dimensions of the GDPR Framework: An Institutional …

499

2 General Data Protection Regulation Personal data are owned by a natural person and are as such inalienable. However, with the development of technology and the Internet, personal data are becoming increasingly exposed and accessible to those who could use them for their own, unlawful purposes [5]. The GDPR is a cornerstone regarding personal data treatment regulation, which aims to respond to the new challenges in the field of data protection generated by the evolution of technology and the globalization of markets. It originated from the European Union package on the reform of data protection and came into force directly and mandatorily on May 28, 2018, impacting significantly the life of organizations. Both the European Parliament and the Council of the European Union saw the need to implement a more coherent and solid data protection framework, anchored in a rigorous application of the rules, since it is crucial to generate the trust needed for the development of digital economy in the whole internal market. The obligatory adaptation of the organizations to the GDPR will imply a set of legal, technological, and functional changes with a direct impact on the daily life of the organizations, as a result of their increased responsibility with the data protection subjects, which has been reinforced by the new legislation. On the other hand, the transfer of responsibilities from national authorities to the organizations obliges them to prove, at all times, full compliance with the legislation. Organizations are subject to high fines when non-compliance is detected [6]. The GDPR aims to protect the personal data of EU residents and not just secure it. There is no single solution for the GDPR due to its complex remit and enterprise-wide impact. The GDPR compliance requires a 360° perspective [7]: 1. There is no silver bullet to the GDPR: • The GDPR is the most complex piece of regulation in recent times, and it impacts everybody. • The depth and breadth of activity required to demonstrate and maintain compliance are significant. • There is no single solution to the GDPR. 2. The GDPR impacts the enterprises: • The GDPR is an enterprise-wide business problem, and it requires enterprises solutions. • Approached in the right way, the GDPR will allow organizations to build trust with customers/stakeholders and positively differentiate within the digital age. 3. Fragmented Approach: • The GDPR requires a 360° perspective, and organizations that approach this regulation from a single viewpoint are liable to an increased risk exposure. • There are over 20 defined areas to the GDPR, all of which require a distinct solution.

500

I. M. Lopes et al.

Fig. 1 Four dimensions of the GDPR framework [7]

• Cognizant has a comprehensive framework that covers all the required GDPR areas and which can be traced back to each individual component of the legislation. The GDPR implementation program covers four distinct dimensions which are critical to the GDPR compliance: technology, data, process, and people (see Fig. 1). Organizations must work on these four dimensions, each of which having its own specificities, starting the process by “preparing,” followed by “protecting,” and finally by “maintaining,” so that the GDPR adoption can be a reality and organizations can be in compliance with the regulation.

3 Institutional Theory Changes in legislation, technology, and the economy generate modifications in the organizational environment. In the face of this, the search for innovation represents one way for the survival of organizations. The success of the organization is then measured by the capacity to survive, change, and anticipate the market needs [8]. Therefore, organizations gradually institutionalize organizational practices in order to face new realities, which cannot be faced using the previously existing organizational practices. The Institutional Theory considers the processes through which structures (e.g., frameworks, rules, norms, and routines) are established as trustworthy guidelines for social behavior. Also, it accounts for the way these elements are created, spread,

The Four Dimensions of the GDPR Framework: An Institutional …

501

adopted, and adapted throughout time and space, as well as the way they fall into decline and disuse [9]. Institutions may be conceived as high resilient social structures that enable and constrain the behavior of social actors and that provide stability and meaning to social life [10–12]. The authors [13] outlined the processes inherent to institutionalization as consisting of four stages, namely innovation, habitualization, objectification, and sedimentation. The institutionalization process starts at “Innovation,” which occurs due to external forces such as technological change, legislation, or market forces. In this sense, the word innovation means structural rearrangements or new organizational practices aimed at solving organizations’ problems. Following this comes a sequential process of three stages which enables the evaluation of the institutionalization degree of a certain social reality. In an organizational context, the process of “Habitualization” involves the creation of new structural arrangements in response to specific organizational problems or sets of problems, shaped through policies and procedures of a specific organization or set of organizations with similar problems. Hence, this is the pre-institutionalization stage. After the solution for the problem has been generated, it is possible to move on to the “Objectification” process, which accompanies the spreading of the new structure, expanding its use. Objectification implies the development of a certain degree of social consensus regarding the structure and its growing adoption, based on that consensus, by the organization. This process configures the semi-institutional stage. The stage in which institutionalization is complete is called “Sedimentation,” and it is characterized by the adoption of the structure or organizational practice by the whole organization for a long period of time. Figure 2 presents a diagram with a summarized overview of the causal critical forces in the institutionalization process. To better understand this diagram, we applied another concept also proposed by [14] to the stages of institutionalization to observe the institutionalization process across different stages or levels. For those authors, the institutionalization process

Fig. 2 Component process of institutionalization [14]

502

I. M. Lopes et al.

occurs in three distinct stages (or phases): the pre-institutional, semi-institutional, and total institutionalization stages. As shown in Fig. 2, in the pre-institutional stage, causal forces external to the organization arise from changes in technology, legislation, and market forces, which may lead to the introduction of innovations in organizations. These forces may also lead to the creation of new habits. The author [12] discusses the distinction between studies focusing on the creation of institutions and studies focusing on the change of institutions. The first ones focus on the process and the conditions which give place to new rules, understandings, and practices. The second ones examine the way a set of beliefs, norms, and practices is attacked, becomes “non-legitimate,” or falls into disuse, being then replaced by new rules, ways, and scripts. Deep down, these two processes are related, as the institutional creation implies the change of the existing institutions and the institutional change implies the creation of new institutions.

4 Research Methodology The survey (first study) was sent to 190 clinics, but only 57 gave an effective reply, which corresponds to a response rate of 30%. The sample subjects were selected randomly based on the kind of clinic and their location distributed throughout the 18 inland Portuguese districts as well as Madeira and the Azores [4]. Considering the research questions previously formulated, and in articulation with the selected research method (Institutional Theory), a primary data collection tool emerged: the interview. For a better understanding of the results, we can group the clinics into three clusters (Table 1): Table 1 Clusters of GDPR implementation Cluster

Stage

Number of clinics n (%)

Number of clinics interviewed

1

In compliance with the regulation

7 (12.2)

6

2

Which are implementing the measures imposed by the regulation

11 (19.3)

6

3

Which are not in compliance with the regulation

39 (68.5)

6

The Four Dimensions of the GDPR Framework: An Institutional … Table 2 Clinics surveyed and interviewed

503

Type of clinic

Clinics surveyed

Clinics interviewed

Nursing

12

4

Dental

16

6

Ophthalmology

4

1

Medical and diagnostics

15

4

Orthopedics

3

1

Physiotherapy

7

2

• Cluster 1—Clinics in compliance with the regulation; • Cluster 2—Clinics which are implementing the measures imposed by the regulation; • Cluster 3—Clinics which are not in compliance with the regulation. Altogether, 18 health clinics were interviewed, distributed equitably among the three clusters (each cluster contributed to six interviews). Table 2 shows the characterization of the clinics involved in the study according to their type. The structure of the survey resulted from a desk review on personal data protection and the study of the legal framework Regulation (EU) 2016/679 of the European Parliament and of the Council of April 27—General Regulation on Data Protection [3].

5 Results Compliance with the GDPR must be a priority and must be seen by organizations as an opportunity. Since we are talking about health clinics, therefore, dealing with “sensitive data,” such priority becomes even more pressing. From the work conducted within the 18 clinics under study, the following measures can be recommended in order to improve the adoption of the propositions laid down in the regulation, as far as the components of the institutionalization process are concerned (Fig. 3). The major challenge is faced in the habitualization stage, since it is when a bigger quantity of factors considered to be critical occurs within the institutionalization process as far as effort, cost, and complexity are concerned. With regard to the four dimensions presented in Fig. 1, their taking into account increases the patients’ trust toward the health clinic. The dimension—technology, responsible for data architecture, is crucial in order to monitor activities in an automatized way, managing and protecting contents and data. Thus, compliance with the GDPR implies the implementation of data protection technology.

504

I. M. Lopes et al.

Fig. 3 Component process of institutionalization/adoption of GDPR

In the dimension—data, where management and security are conducted, and bearing in mind that most data treatment activities are based on the use of information systems and automatized means, it is necessary to assess the systems used, the levels of access and the security measures implemented. In the dimension—process, responsible for the consent of rights, the processes are honed so as to respond to the data holders’ rights and manage data processing. In the dimension—people, responsible for governance and oversight, it is mandatory to involve all the stakeholders a cultural change toward data protection by all—with well-defined rules and responsibilities. The conjunction of these actions to enhance the implementation of the GDPR in the health clinics can be summarized in six essential points: 1. Review periodically—data protection is not a destination, it’s a journey! So, review on a regular basic and correct your course accordingly. 2. Raise awareness—conduct staff training and awareness sessions. Most breaches occur due to staff ignorance so make them aware and mitigate the risk.

The Four Dimensions of the GDPR Framework: An Institutional …

505

3. Review policies and related documents—review existing policies and create new ones where needed. These policies allow you to formulate processes and procedures for staff to follow. 4. Make a plan—based on the audit and risk assessment, determine a roadmap to achieve compliance. While it is important to address high-risk areas, do not ignore the low hanging fruit. Small, easy wins can get the project off to a positive start! 5. Identify and assess privacy-related risks—assess the risks associated with how data are being processed. Compile a risk register. 6. Conduct an audit—know your data—how you got it; who can access it; where it is stored; how long you keep it, etc. Create a list of recommendations to address areas of concern.

6 Conclusions This paper brought forward guidelines which are believed to enhance the institutionalization of measures imposed by the GDPR in organizations. We also argue that the use of Institutional Theory as a support to the interpretation of the adoption stage of implementation by organizations and as a support to the projection of guidelines which enhance the institutionalization of these GDPR measures in organizations represents a promising means for research. In Portugal, there is still some vagueness concerning some of the issues regarding data protection, due to the fact that the Portuguese legislation has not yet been approved (note that although the GDPR is a EU regulation and therefore is directly enforced in Portugal and in all member states, the truth is that certain issues demand a legal definition by the Portuguese legislator). A considerable effort is being made by companies to be in compliance with the GDPR. However, it is a long path and the instructions regarding the four dimensions (technology, data, process, and people) must be properly implemented and, above all, maintained bearing in mind that periodic improvements are necessary. This process is never completely closed, and reviews must be conducted whenever necessary. As a possible future work, we suggest assessing how far the certification of the information security management system by ISO 27001 grants companies’ compliance with the GDPR, since the implementation of an information security management system by a company must ensure that all the relevant controls of risk containment associated with confidentiality, integrity, and availability are implemented and kept functional. Acknowledgements UNIAG, R&D unit funded by the FCT—Portuguese Foundation for the Development of Science and Technology, Ministry of Science, Technology and Higher Education. Project n.º UID/GES/4752/2019. This work has been supported by FCT—Fundação para a Ciência e Tecnologia within the Project Scope: UID/CEC/00319/2019.

506

I. M. Lopes et al.

References 1. Mäkinen, J.: Data quality, sensitive data and joint controller ship as examples of grey areas in the existing data protection framework for the Internet of Things. Inf. Commun. Technol. Law 24(3), 262–277 (2015) 2. Nurse, J.R.C., Creese, S., De Roure, D.: Security risk assessment in Internet of Things systems. IEEE IT Prof. 19(5), 20–26 (2017) 3. European Parliament and Council, Regulation (EU) 2016/679 of the European Parliament and of the Council of 27 April 2016, Official Journal of the European Union (2016) 4. XXXX 5. Skendzic, A., Kovacic, B., Tijan, E.: General data protection regulation—protection of personal data in an organization. In: 41st International Convention on Information and Communication Technology, Electronics and Microelectronics, pp 1370–1375 (2018) 6. Da Conceição Freitas, M., Mira da Silva, M.: GDPR in SMEs, vol. 2018, pp. 1–6. In: 13th Iberian Conference on Information Systems and Technologies (2018) 7. West, I.: The big scan thing!—How the EU General Data Protection Regulation (GDPR) will affect your business! https://www.slideshare.net/CraigShipley1/digital-enterprise-festivalbirmingham-130417-ian-west-cognizant-vp-data-management-the-implications-of-the-euglobal-data-protection-regulation-on-every-business-and-their-digital-service-providers. Last accessed 1 Dec 2018 8. Brown, S.L., Eisenhardt, K.M.: Competing on the edge: strategy as structured chaos. Harvard Business School Press, Boston (1998) 9. Scott, W.: Institutional Theory. Encyclopedia of Social Theory, pp. 408–414. Thousand Oaks, Sage (2004) 10. DiMaggio, P. Powell, W.: Introduction. In: Powell, W.W., DiMaggio, P.J. (eds.) The New Institutionalism in Organizational Analysis, pp. 1–38. University of Chicago Press, Chicago (1991) 11. North, D.: Institutions, Institutional Change and Performance. Cambridge University Press, Cambridge (1990) 12. Scott, W.R.: Institutions and Organizations: Ideas and Interests, 3rd edn. Sage, Thousand Oaks (2008) 13. Tolbert, P.S., Zucker, L.G.: The institutionalization of institutional theory. In: Handbook of Organization Studies. Sage, London (1996) 14. Tolbert, P.S., Zucker, L.G.: A institucionalização da teoria institucional. In: Clegg, S., Hardy, C., Nordy, W (eds.) Handbook de estudos organizacionais (pp. 196–219). Tradução de Humberto F. Martins e Regina Luna S. Cardoso, v.1. Atlas, São Paulo (1999)

Metrics and Indicators of Information Security Incident Management: A Systematic Mapping Study Alyssa Cadena, Franklin Gualoto, Walter Fuertes, Luis Tello-Oquendo, Roberto Andrade, Freddy Tapia and Jenny Torres

Abstract The number of threats and vulnerabilities has increased rapidly in recent years. For this reason, organizations are in need of providing improvements in their computer security incident management (CSIM), in order to safeguard their intellectual capital. Therefore, the identification and use of both metrics and indicators are a crucial factor to manage security incidents. In this context, organizations try to improve their level of CSIM based on standards or only according to their criteria based on their experience. This article aims at carrying out a systematic mapping study of academic articles conducted in this research area, in order to present a document that describes metrics and indicators of security incidents in organizations. The results of this work show and describe several key indicators and metrics related to the cost, quality, and service (time) involved in dealing with such incidents. Also, it is expected that this study serves as a strategic reference for organizations.

A. Cadena · F. Gualoto · W. Fuertes · F. Tapia Universidad de las Fuerzas Armadas ESPE, Sangolquí, Ecuador e-mail: [email protected] F. Gualoto e-mail: [email protected] W. Fuertes e-mail: [email protected] F. Tapia e-mail: [email protected] L. Tello-Oquendo (B) Universidad Nacional de Chimborazo, Riobamba, Ecuador e-mail: [email protected] R. Andrade · J. Torres Escuela Politécnica Nacional, Quito, Ecuador e-mail: [email protected] J. Torres e-mail: [email protected] © Springer Nature Singapore Pte Ltd. 2020 Á. Rocha and R. P. Pereira (eds.), Developments and Advances in Defense and Security, Smart Innovation, Systems and Technologies 152, https://doi.org/10.1007/978-981-13-9155-2_40

507

508

A. Cadena et al.

Keywords Security incidents · Metrics · Indicators · Security incident management · Key performance indicators · Empirical study

1 Introduction According to Miloslavskaya, “Information security (IS) incidents have become not only more numerous and diverse but also more damaging and disruptive. Preventive controls based on the IS risk assessment results decrease the majority but not all the IS incidents” [1]. One way to control incidents is to manage them through proper identification and analysis of key metrics and indicators. Therefore, it is necessary to emphasize all those metrics and indicators of computer security incidents that can help organizations in the quest to safeguard their intellectual capital from threats. In the literature, several related works can be found on this matter. For instance, in [2] the authors present a method that evaluates the ability in which the organization handles security incidents; in addition, they perform the analysis of various security metrics based on detection, response, and cost of incidents and focus on early warning indicators based on resilience. It is important to visualize from the technical aspect, where the experiences obtained from the field of action are reflected when dealing already in organizations with the management of computer security incidents; a clear example of this is [3] in which the authors propose a methodology based on “goal, question, metrics (GQM)” that establishes steps for the detection and resolution of security violations; they also recognize metrics already proposed by ISO 27004:2009 and NIST 800-55. “What is not defined cannot be measured. What is not measured cannot be improved. What cannot be improved, is always degraded” [4]. Hence, the importance of discovering key metrics and indicators for CSIM to help measure the effectiveness and efficiency of such management. Under this perspective, the purpose of this work is to conduct a systematic mapping study (SMS), which allows organizations to guide in the definition of metrics and indicators of CSIM. To achieve this, an empirical study was carried out in three databases: IEEE, ACM, and SpringerLink to later analyze the articles found and translate, according to the criteria of the authors of this study, metrics, and indicators of computer security incidents. The results obtained emphasize those focused on cost, quality, and service. The remainder of this paper is structured as follows. Section 2 documents the SMS process, while Sect. 3 presents the results of the systematic mapping. On the other hand, Sect. 4 introduces a proposal for CSIM, specifically in higher education institutions. Finally, Sect. 5 draws the conclusions and future work.

Metrics and Indicators of Information Security Incident …

509

2 Systematic Mapping Study Process (Research Methodology) Petersen defines a SMS as a secondary study aimed at constructing a classification scheme and structuring a software engineering field of interest [5]. To carry out the study, he suggests a five-step process as follows: (1) Define the research questions; (2) search the relevant documents; (3) select the primary studies; (4) analyze the abstracts and extract keywords and data; and (5) map the selected primary studies. Each of the steps in the process has an outcome, with the systematic mapping being the final outcome of the overall processes. The steps mentioned above were followed, and a protocol is constructed for the purpose of avoiding bias [6].

2.1 Research Questions In order to frame the research on the problematic raised, research questions (RQs) were established, which were defined following the objectives of this study: to analyze and define metrics and indicators of information security incidents employing this SMS. In this way, we follow the respective guidelines leading to the RQs presented below: RQ1: What are the metrics related to information security incidents in organizations? RQ2: Are there indicators related to information security incidents? RQ3: Are there Key Performance Indicators (KPI’s) related to cost, quality, and service? RQ4: In most metrics, what standards apply to information security incident management? RQ5: Are there studies on SMS aimed at Information Security Incidents? This information will lead to obtaining the metrics and key indicators of information security incidents as well as a formal study that validates all the data shown.

2.2 Inclusion and Exclusion Criteria For the selection of the studies, the corresponding titles and abstracts were taken into account in which the following inclusion criteria will be denoted: – – – –

Studies in the field of computer security; Studies in the field of information security; Articles published since 2010; Scientific articles and conference papers. Within the exclusion criteria, the following parameters were considered:

510

– – – –

A. Cadena et al.

Studies that were not included in the selected databases; Duplicate studies; Course articles, books, or early access articles; Studies that did not present contents in English or Spanish.

2.3 Digital Libraries and Search String The following digital libraries were considered for the literature review: IEEE Xplore, ACM Digital Library, and Springer Link. The reason for selecting them is because they cover a wide range of publications in the field of engineering science maintaining a complete and consistent database. For conducting the initial search, we used keywords and search terms closely related to: – “Security information management”; – “Security incident metrics and indicators”; – “Key performance indicators or KPIs.” Then, to determine the optimal search string for research, several options have been considered taking into account that they conform with the terms as mentioned above. The proposed inclusion and exclusion criteria were also applied generating diverse possibilities. The first search approach can be seen in Table 1. Finally, once the search string was optimized, we selected the one that contains the most significant amount of relevant information and that returned a manageable number of documents for the research. The optimized search string is shown in Table 2.

Table 1 Search strings or chains, databases, and results Suggested search string

Academic database result

((((Information Security Incident) OR Security Information Management) AND Metrics) AND Indicators OR Key Performance Indicators)

IEEE Xplore

(((Information Security Incident) OR Security Information Management) AND Indicators)

2077

ACM

893

Springer Link

24,126

IEEE Xplore

391

ACM

13,277

Springer Link

7102

((Security Information Management) AND (Informatic Security OR Security Incidents) AND (Metrics OR Indicators) AND (Key performance Indicators OR KPIs))

IEEE Xplore

3

((((Security Information Management) AND Informatic Security OR Security Incidents) AND Metrics OR Indicators) AND Key performance Indicators OR KPIs)

ACM

2509

Springer Link

9

IEEE Xplore

2180

ACM

27

Springer Link

23,941

Metrics and Indicators of Information Security Incident …

511

Table 2 Selected search string ((Information Security Incidents) AND (Indicators OR Metrics OR Key Performance Indicators))

3 Results of the Systematic Mapping Study Aiming at refining the information obtained from the search string, an information crossing was performed in two ways: manually and using the academic databases. By doing so, the amount of literature obtained for reviewing was reduced to 10 relevant articles as shown in Table 3. This result can also be visualized in Fig. 1 which illustrates the number of articles selected from each academic database for conducting this study. In addition, Fig. 2 depicts the mechanism for determining the relevance of each study. This was done by means of validation of the researchers, by simple selection and equitable weights selection. Note that we also analyzed the years in which the articles were indexed. This information can be seen in Fig. 3. It reflects that, in the year 2015, a greater emphasis was made in the generation of scientific articles focused on the analysis of metrics and indicators of CSIM in the organizations. Then, it decreases until the year 2018 in which the tendency to study this subject increases again.

Table 3 Metrics and indicators of information security incidents: systematic mapping study Item study title S1

[7]

“A practical experience on evaluating intrusion prevention system event data as indicators of security issues”

S2

[2]

“Forewarned is forearmed: indicators for evaluating information security incident management”

S3

[8]

“A comprehensive survey on machine learning for networking: evolution, applications and research opportunities”

S4

[9]

“Adoption of security as a service”

S5

[1]

“Security operations centers for information security incident management”

S6

[10]

“Information security considerations for protecting NASA mission operations centers (MOCs)”

S7

[11]

“Establishing national cyber situational awareness through incident information clustering”

S8

[12]

“The β-time-to-compromise metric for practical cybersecurity risk estimation”

S9

[13]

“Integration of IT frameworks for the management of information security within industrial control systems providing metrics and indicators”

S10

[14]

“Application of security metrics to instrument systems that use distributed processing”

512

A. Cadena et al.

2

SpringerLink

19 0

ACM

3 8

IEEE Xplore

44 0

10

20

30

40

50

All Articles

Selected Articles

Fig. 1 Results of the research string application

S1

Fig. 2 Relevance of the literature considered for this study

S10

100% 80%

S2

60% 40%

S9

S3

20% 0%

S8

S4

S7

S5 S6

Number of publications

3.5 3

3

2.5 2

2

2

1.5 1

1

1

1

2016

2017

0.5 0 2010

2011

2012

2013

2014

2015

Publication year

Fig. 3 Distribution of the publications by year

2018

2019

Metrics and Indicators of Information Security Incident …

513

3.1 Answers to Research Questions The selected documents, result of the SMS, were used to answer the research questions posed in Sect. 2.1, which are the focus of this research. Regarding the research questions RQ1 and RQ2, several metrics and indicators related to information security incidents were analyzed. However, a significant number of articles focused solely on information security were found through the search; therefore, an exhaustive analysis was carried out, and the results are presented in Table 4. It is worth noting that the metrics and indicators considered by the studies analyzed are based on standards such as [14]. On the other hand, [13] is based on the metrics proposed by other authors as [15–18]; in the same way, [12] is based on [19], and [9] is based on indicators proposed by various authors. Conversely, articles were found that support their indicators or metrics throughout their research process as is the case of [1, 7, 8, 10]. However, there are several studies that use indicators based on the REWI method [20] (e.g., [2]) and that present metrics in the form of facts (e.g., [11]). Regarding the research question RQ3, cost, quality, and service are three entities of great importance in organizations, which is why the work of [1, 2, 13] focuses on these attributes. With regard to the research question RQ4, both the metrics and indicators of information security incidents analyzed in the standards were proposed by their authors based on ISO/IEC 27035 [21] or NIST SP 800:61 [22], among others. However, the indicators presented in [14] are based on the ANSI/ISA-99.01.0 1-2007 [23] and ISA-99.02.01-2009 [24] standards. Finally, regarding the research question RQ5, there are studies on SMS, as is the case of [2], where the authors present a method to evaluate an organization’s capacity to manage security incidents. They mention that this method is based on the form of response to change (resilience) and describe how to identify, select, and implement early warning indicators for the management of information security incidents.

4 Proposal Several institutions worldwide have presented standards for CSIM, highlighting the standards ISO/IEC 27004:2009 (International Organization for Standardization/International Electro-technical Commission) [25] and NIST SP 800-55 Revision 1 (National Institute of Standards and Technology) [26], as guidelines that provide guidance and advice on the development and use of measures to evaluate the effectiveness of the information security incident management. However, there is the problem that as long as organizations continue to focus on isolated technology solutions without the respective balance between processes and management, uncertainty will continue to be generated as to why the increase in incidents despite the investments made in information security. Organizations dedicated to IS management, as is the case of higher education institutions (HEIs), propose to carry out an

– – – – – –

– – – – – –

No metrics

[7]

[2]

[1]

Presence and fulfillment of plans Incident statistics Detection and response statistics Consequences Incident management cost and performance Culture and learning aspects

Number of alerts per week Number of (distinct) attackers per week Number of (distinct) objectives per week Number of (distinct) signatures per week Number of alerts per attacker per week Number of attackers per target per week

Metrics

Study

Table 4 Metrics and indicators of information security incidents

– Unauthorized user on the network or shared credentials – Unauthorized access to sensitive data, personally identifiable information (PII), and financial data – Unauthorized connection of the internal host (client or server) to the Internet – Excessive access from a single or multiple internal host to the external malicious Web site (from known blacklists) – Detection of malware and user activity outside of the hour (at night or on weekends) – Multiple logins with the same ID from different locations in a short time – Multiple alarms from a single host or duplicate events on multiple computers on the same subnet over a 24-h period – Repeated attack from a single source or on a single host – Access to the Internet service account or to an unauthorized device, among others (continued)

(Based on resilience) – Risk awareness – Responsiveness – Support

No indicators

Indicators

514 A. Cadena et al.

Performance metrics for precision validation [8] – True positive rate or recall – False positive rate – True negative rate

No metrics

– Number of new {incidents | vulnerabilities (discovered) | plan of actions and milestones (opened)} in the past 30 days – Number of open {incidents (i.e., under investigation) | vulnerabilities (not yet mitigated) | plan of actions and milestones (not yet closed)} in the past 30 days – Number of closed {incident investigation(s) | vulnerabilities (mitigated) | plan of actions and milestones} in the past 30 days

Action Damage euro Damage score Incident date Incident ID System Time until discovery Time until report Vulnerability date Business area

Time-to-compromise

[8]

[9]

[10]

[11]

[12]

Table 4 (continued)

No indicators

No indicators

No indicators

Indicators focused on vulnerability and threat

No indicators

(continued)

Metrics and Indicators of Information Security Incident … 515

Project metrics and indicators specified by PMI-PMBOK – Program of progress of all projects in the portfolio – The active risks and the amount of money used compared to that budgeted for that date Management of ITIL services related to processes, services and technology: – The number of closed and pending tickets for each of the processes of the service operation stage – The number of closed and pending tickets for the incident process – The number of tickets closed and the total pending – The response times of each of the Ethernet ICS nodes per month – The percentage of uptime of each of the Ethernet ICS nodes per month – The number of closed tickets per day for each area administered by ICS – The ticket seniority times for each service associated with the established SLA – The percentage of compliance of each ICS engineer’s priorities for the week associated with the established SLA – The average time taken to complete a ticket – The average cost of a ticket

– Number of times [in a specified time interval] an error condition occurs – Total number and number of unsuccessful connection attempts to authenticate devices communicating via untrusted networks per [unit of time determined by the asset owner] – Current number of untrusted network connections – Number of access retries associated with a source address recorded by each access point

[13]

[14]

Table 4 (continued)

No indicators

516 A. Cadena et al.

Metrics and Indicators of Information Security Incident …

517

objective evaluation of the incident records in order to find useful information that feeds their knowledge. Based on the above, according to a study presented in 2017 by Consortium for Advanced Internet Development (CEDIA) in its article “State of Information and Communication Technologies in Ecuadorian Universities,” in the area of ICT security, 70% of HEIs are supported by CEDIA’s CSIRT, 8% have their Committee, and 22% none. Similarly, 54% of the HEIs surveyed rely on CEDIA’s vulnerability analysis tools, 22% on their own, and 24% none [27]. This shows evident shortcomings in HEIs concerning CSIM. Besides, there is a lack of studies on information security incidents in HEIs and how these could benefit management decision-making and the continuous improvement of the IS. For this reason, it is proposed to perform a study on the incidents aiming at achieving cognitive learning that allows analyzing the problem from another perspective. On the one hand, the generation of dashboards of the preeminent indicators related to security incidents is highly advised. Besides, the discovery of behavioral patterns of incidents through data mining techniques will provide relevant information. This information will be transformed into knowledge that will drive better business strategies that will translate into a competitive advantage for any organization. It is for this reason that some research is planned where metrics and key indicators will be established in the management of security incidents in HEIs in order to improve their management, using benchmark surveys (KPI benchmark surveys). Additionally, a business intelligence model required to evaluate these metrics and indicators will be implemented through the “cross-industry standard process for data mining” (CRISPDM) or through Ralph Kimball’s methodology and where the effectiveness of the model will finally be evaluated, leveraged with the action research methodology, to determine the confidence level of the results.

5 Conclusions The objective of this study was to conduct a systematic literature review of academic articles focused on security incident management, in order to identify and describe metrics and indicators of security incident management in organizations. To achieve this, an empirical study was carried out on three indexed databases: IEEE, ACM, and Springer. The articles found were then analyzed along with their metrics and indicators. The methodological procedure consisted of a systematic mapping study with their respective research questions, inclusion and exclusion criteria, selection of digital libraries, and analysis of the respective search strings. Among the main findings were metrics and key indicators focused on cost, quality, service, products, incident management, resilience, and early warning indicators. It should be noted that some of them were paired with the following international standards: ISO/IEC 27035, NIST SP 800:61, ANSI/ISA-99.01.0 1-2007, and ISA-99.02.01-2009. This information could serve for the analysis of their behavior as input for decision-making in the context of information security incident management in organizations. All this will

518

A. Cadena et al.

result in reducing threats, vulnerabilities, and computer risk, which will influence the increase in the level of information security in institutions. As a future work, we have planned to create a knowledge database to design and develop a decision support system of the management of security incidents. Acknowledgements The authors would like to thank the financial support of the Ecuadorian Corporation for the Development of Research and the Academy (RED CEDIA) in the development of this work, under Research Team GT-II-Cybersecurity.

References 1. Miloslavskaya, N.: Security operations centers for information security incident management. In: 2016 IEEE 4th International Conference on Future Internet of Things and Cloud (FiCloud), pp. 131–136 (2016) 2. Bernsmed, K., Tondel, I.A.: Forewarned is forearmed: indicators for evaluating information security incident management. In: 2013 IEEE Seventh International Conference on IT Security Incident Management and IT Forensics, pp. 3–14 (2013) 3. Hajdarevic, K., Allen, P.: A new method for the identification of proactive information security management system metrics. In: 2013 IEEE 36th International Convention on Information & Communication Technology Electronics & Microelectronics (MIPRO), pp. 1121–1126. (2013) 4. Thomson, W., Kelvin, L.: Baltimore Lectures. CJC a. Sons, Ed., London (1904) 5. Petersen, K., Vakkalanka, S., Kuzniarz, L.: Guidelines for conducting systematic mapping studies in software engineering: an update. Inf. Softw. Technol. 64, 1–18 (2015) 6. Elberzhager, F., Münch, J., Nha, V.T.N.: A systematic mapping study on the combination of static and dynamic quality assurance techniques. Inf. Softw. Technol. 54(1), 1–15 (2012) 7. Miani, R.S., Zarpelao, B.B., Sobesto, B., Cukier, M.: A practical experience on evaluating intrusion prevention system event data as indicators of security issues. In: 2015 IEEE 34th Symposium on Reliable Distributed Systems (SRDS), pp. 296–305 (2015) 8. Boutaba, R., Salahuddin, M.A., Limam, N., Ayoubi, S., Shahriar, N., Estrada-Solano, F., Caicedo, O.M.: A comprehensive survey on machine learning for networking: evolution, applications and research opportunities. J. Internet Serv. Appl. 9(1), 16 (2018) 9. Senk, C.: Adoption of security as a service. J. Internet Serv. Appl. 4(1), 11 (2013) 10. Takamura, E., Mangum, K., Wasiak, F., Gomez-Rosa, C.: Information security considerations for protecting NASA mission operations centers (mocs). In: 2015 IEEE Aerospace Conference, pp. 1–14 (2015) 11. Skopik, F., Wurzenberger, M., Settanni, G., Fiedler, R.: Establishing national cyber situational awareness through incident information clustering. In: 2015 IEEE International Conference on Cyber Situational Awareness, Data Analytics and Assessment (CyberSA), pp. 1–8 (2015) 12. Zieger, A., Freiling, F., Kossakowski, K.P.: The β-time-to-compromise metric for practical cyber security risk estimation. In: 2018 IEEE 11th International Conference on IT Security Incident Management & IT Forensics (IMF). pp. 115–133 (2018) 13. Bustamante, F., Fuertes, W., Díaz, P., Toulkeridis, T.: Integration of IT frameworks for the management of information security within industrial control systems providing metrics and indicators. In: 2017 IEEE XXIV International Conference on Electronics, Electrical Engineering and Computing (INTERCON), pp. 1–4 (2017) 14. Munro, J.K.: Application of security metrics to instrument systems that use distributed processing. In: Future of Instrumentation International Workshop (FIIW), 2011, pp. 5–8 (2011) 15. Stouffer, K., Falco, J., Scarfone, K.: Guide to industrial control systems (ICS) security. NIST Spec. Publ. 800(82), 16 (2011)

Metrics and Indicators of Information Security Incident …

519

16. Rose, K.H.: A guide to the Project Management Body of Knowledge (PMBOK® Guide)—Fifth Edition. Proj. Manag. J. 44(3), e1–e1 (2013) 17. Lloyd, V.: ITIL Continual Service Improvement (Best Management Practices). The Stationery Office (2011) 18. ISACA: COBIT 5: A business framework for the governance and management of enterprise IT. ISACA (2012) 19. McQueen, M.A., Boyer, W.F., Flynn, M.A., Beitel, G.A.: Time-to-compromise model for cyber risk reduction estimation. In: Quality of Protection, pp. 49–64. Springer (2006) 20. Øien, K., Massaiu, S., Tinmannsvik, R.K.: Guideline for implementing the REWI method; Resilience based Early Warning Indicators. SINTEF report A 22026 (2012) 21. Information Technology—Security Techniques—Information Security Incident Management. Standard, International Organization for Standardization, Geneva, CH (2011) 22. Cichonski, P., Millar, T., Grance, T., Scarfone, K.: Computer security incident handling guide. NIST Spec. Publ. 800(61), 1–147 (2012) 23. ANSI/ISA: Security for Industrial Automation and Control Systems Part 1: Terminology, Concepts & Models. Tech. rep., American National Standards Institute/International Society of Automation (ANSI/ISA) (2007) 24. ANSI/ISA: Security for Industrial Automation and Control Systems: Establishing an Industrial Automation and Control Systems Security Program. Tech. rep., American National Standards Institute/International Society of Automation (ANSI/ISA) (2009) 25. ISO/IEC: Information Technology—Security Techniques—Information Security Management—Measurement (ISO/IEC 27004: 2009). ISO/IEC (2009) 26. Chew, E., Swanson, M., Stine, K., Bartol, N., Brown, A., Robinson, W.: NIST Special Publication 800–55 Revision 1. Performance Measurement Guide for Information Security, National Institute of Standards and Technology, US Department of Commerce. Computer Division, Gaithersburg, MD 20899, 8930 (2008) 27. Verdugo, R.P.: Estado de las tecnologías de la información y la comunicación en las universidades ecuatorianas. CEDIA (2017)

Virtual Environment for Remote Control of UGVs Using a Haptic Device F. Roberto Segura, Pilar Urrutia-Urrutia, Z. Andrea Sánchez, C. Tomás Núñez, T. Santiago Alvarez, L. Franklin Salazar , Santiago Altamirano and Jorge Buele

Abstract This paper presents a virtual reality environment designed for military training personnel, focused on the remote control of unmanned land vehicles. The environment design has been made in the V-REP software, where a prototype of an explorer robot based on the kinematic model of a unicycle is presented. This vehicle is attached with proximity sensors to detect obstacles and thus be able to avoid them. Instead of operating with a conventional joystick that only allows the use of push buttons, a haptic device with force feedback is used with which the user experiences a more realistic immersive situation. In this context, the person can manipulate the unmanned vehicle direction and perceive when there is a collision with a nearby object as if it were on the site. To link the input device (Novint Falcon) with the virtual interface, the device mathematical modelling is carried out, and through MATLAB, the respective processing and the implementation of the P. Urrutia-Urrutia · Z. Andrea Sánchez · L. Franklin Salazar · S. Altamirano · J. Buele (B) Universidad Técnica de Ambato, Ambato 180103, Ecuador e-mail: [email protected] P. Urrutia-Urrutia e-mail: [email protected] Z. Andrea Sánchez e-mail: [email protected] L. Franklin Salazar e-mail: [email protected] S. Altamirano e-mail: [email protected] F. Roberto Segura · T. Santiago Alvarez Instituto Tecnológico Superior Guayaquil—Ambato, Ambato 180205, Ecuador e-mail: [email protected] T. Santiago Alvarez e-mail: [email protected] C. Tomás Núñez CELEC EP, Baños 180250, Ecuador e-mail: [email protected] © Springer Nature Singapore Pte Ltd. 2020 Á. Rocha and R. P. Pereira (eds.), Developments and Advances in Defense and Security, Smart Innovation, Systems and Technologies 152, https://doi.org/10.1007/978-981-13-9155-2_41

521

522

F. Roberto Segura et al.

proportional–integral–derivative (PID) control algorithm for the displacement are made. The after-scenario questionnaire (ASQ) test is used, and a general average of 1.78/7 is obtained. Being a value close to 1, it shows the acceptance that the system has for the users. Keywords Haptic interface · Force feedback · Training · Teleoperation · Unmanned ground vehicle · Virtual reality

1 Introduction In recent years, the advance of technology has brought with it several tools that facilitate the daily tasks of the human being [1]. He wants to acquire more knowledge, explore, and innovate, and thanks to this, science expands covering more fields of application [2]. This is how robotics was born, a topic of interest that has allowed us to carry out several multidisciplinary researches [3]. The main interest of researchers is to design robots with autonomy that can be used in health care, space exploration, military, and rescue applications, among others [4]. Within mobile robotics, the unmanned ground vehicles (UGV) stand out, and they can be teleoperated by specialists and thus develop dangerous activities, where it would be risky to enter a person [5]. Time and effectiveness, in addition to safety, of these robotic systems is highly compared to actions developed by a human being where the aim is not intended to supplant it, but complement it with an important instrument [6]. Additionally, by attaching sensors and video devices to a robot, reconnaissance and exploration tasks can be performed in known and unknown territories. The knowledge of space can then be defined in two parts: structured spaces and unstructured spaces. On the one hand, the path to be followed and the objects that must be avoided are already defined; on the other hand, several characteristics that can change must be contemplated and therefore must be studied to obtain satisfactory executions [7]. Whether directing a real or a simulated robot, the operator’s learning is very important for the integration of new knowledge and skills. Traditionally, these vehicles have been operated by control levers such as joysticks which only have push buttons, but do not give the feeling of contact with the environment in question [8]. This is the reason why the use of haptic devices with force feedback allows the user to feel the effect of being inside the environment where actions are executed through a slave robot. Another important technological tool that is very popular is virtual reality (VR). It consists in a platform that facilitates the implementation of interactive environments, allowing the user to relate to objects and situations that simulate the real world. Several areas of application of the VR are come to combine including entertainment, education, research, rehabilitation, military applications, and others. For example, in [9] a study is carried out in which the relationship between presence and performance is evaluated, while a psychomotor task is executed in a virtual environment. Although virtual simulation has been validated empirically to be effective in the transfer of skills to the live environment, what is still unknown is the effect of

Virtual Environment for Remote Control of UGVs …

523

the user sense of presence in the simulation performance. In this way, it can be known in a better way the behavior of the user before an immersive experience. For their part in [10] using a VR application, war veterans with post-traumatic stress disorder are trained so that they can obtain a job after serving in their military work. Preliminary results showed that participants who attended 95% of the sessions showed improvement in their abilities when facing a job interview and in their self-confidence. In parallel, by merging graphic computing with other areas of knowledge, staff can be trained to perform tasks that require hand–eye coordination, such as surgery, maneuvers in dangerous spaces, or teleoperation. In medical area, there are some researches like [11], where the application of virtual reality is described to optimize the acquisition and retention of military medical skills. This system not only develops the surgical skills of the person, but also incorporates an intelligent mechanism that recognizes the deficiencies of the skills and generates an optimal program of adaptive training. In the proposal developed by Ibari et al. [12], a VR system is presented for the remote control of a unicycle robot through an intuitive multimode man–machine interface. This system manages the tracking of a path by the real robot through the Internet and a wireless network. The result analysis shows the system performance in terms of precision, stability and convergence, and the ability to run from any machine connected to the Internet. In this context, the prior research carried out by Tikanmäki et al. [13] presents a teleoperation system, where the user can control semiautonomous mobile robots simulated by a joystick, replacing the conventional video screens with an immersive virtual environment. The operator can see and inspect the situation by collecting data from an appropriate perspective without suffering motion sickness due to latency in traditional teleoperations. The user experiences are positive when highlighting the possibility of having multiple virtual screens to display images of the camera without experiencing dizziness and without requiring prior knowledge. In the work of Lwowski et al. [14], two experiments were carried out with unmanned ground vehicles (UGV). The first one is manually driven in square form with a Vive Tracker where an estimated position error in average odometry is visualized. Second experiment has a PID controller used to drive the UGV in two different test patterns: a square and an eight figure. The usage of virtual reality as a simulation environment shows as result the usefulness of the tool for systems researchers. Related works have shown the use of a joystick to drive simulated and real vehicles but without obtaining a complete immersive experience. In this context, this paper seeks to develop the skills and abilities of military personnel to maneuver exploration UGVs in a semi-structured environment. To achieve this goal, a virtual reality environment is presented, where objects that the vehicle must evade are found in order to carry out the desired trajectory and task. In addition to having a visual perception, it is required that the user determines the exact force that must be applied to command a robotic mechanism based on the kinematic model of a unicycle, and therefore, the Novint Falcon haptic device is used. For the connection between the physical and virtual media, the MATLAB software is used, where through the kinematic study of the Falcon Master Robot its behavior can be simulated. Through a PID control algorithm, the desired trajectory is tracked. The experimental tests and

524

F. Roberto Segura et al.

results demonstrate the validity of the proposal, since the control actions are executed efficiently, thus representing an important teaching tool. This project is divided into six sections, including the introduction in Sects. 1 and 2 presenting the description of system. Section 3 shows the elements of system. Section 4 talks about the implementation; Sect. 5 shows the results; the conclusions are presented in Sect. 6.

2 Methodology The methodology used to solve the problems encountered follows a set of hierarchical steps shown in Fig. 1. In this figure, all the sections that will contain the methodology to solve the problem are detailed. First, a mathematical analysis of the equations representing the Robotic Master System is considered, analyzing the kinematics of the device. All the necessary libraries and drivers are also considered to connect the haptic device to the computer, so it must have to be installed the corresponding packages for the MATLAB and V-REP software. Mathematical modeling and the implementation of a control method allow to simulate the behavior of the device in MATLAB, prior to the execution of real tests. Also, the control method allows the accomplishment of tasks by the unicycle-type mobile robot. This scenario contains primary elements and/or equipment that represents a training scenario for the skill of the user, where the acquired information is converted to forces. The conversion to forces allows to fulfill the purpose of this work, which focuses on presenting a clearer perspective of the interaction between the slave robot and the environment with which it interacts.

3 Modeling 3.1 Geometric Analysis of the Novint Falcon Kinematic Model Geometrically, the mechanism of the haptic device can be analyzed as a parallel robot with a triangular shape at the base and at the end, the joints. Three reference systems are required for analysis, where A is the reference system of the base of the three arms, C is the reference system of the workspace, and B1−2−3 are the reference systems relative to each arm of the haptic device.

Virtual Environment for Remote Control of UGVs …

525

Fig. 1 Methodology to solve the problem

3.2 Analysis of the Proposed Controller In order to achieve the goal of teleoperation, a control law is proposed to notify, by force feedback, the type of excitations that users must do with the haptic device. In order to feedback an adequate force, the kinetic model of Novint Falcon is used as shown in Eq. (1) and using inverse kinematics the Eq. (2) is obtained. P˙C = R −1 J¯q(t) ˙

(1)

q˙ = R J¯−1 P˙C

(2)

In this context, the control law is shown in Eq. (3).    q˙ = R J¯−1 P˙Cd + PID P˜C

(3)

where the expression PID represents the control algorithm to compensate for the error that occurs in carrying out the displacement task, and it is defined in (4). 



PID( P˜C ) = kp P˜C + ki

t  0

 ∂˜  PC P˜C dt + kd ∂t

(4)

526

F. Roberto Segura et al.

Fig. 2 Experimental execution of the desired trajectory

3.3 Device Simulation The simulated object representing the Novint Falcon allows to validate both the modeling and the proposed PID controller. Next, the operation of the controller is demonstrated by tracking the trajectory described by: xd = r ∗ cos( p), yd = (2 ∗ π ) ∗ 4, z d = π ∗ sin( p), components that form a circumference. To perform the validation of the experimental control with the haptic device is considered a control gain of k p = 0.1, and this gain is applied for the control of the errors in x, y, z. Figure 2 shows the simulation of the stroboscopic movement of the robotic system to fulfill the desired trajectory, where the path to be followed and the execution that the operative end executes are shown.

4 Develop of the Virtual Environment 4.1 Virtual Environment The robot interaction with an environment limited by walls can allow the increase in the dexterity of an operator located at a remote site. In this way, the use of V-REP properties is proposed to create a room, in which multiple objects are located in different positions. These are chairs, tables, columns, and different furniture that can interfere with the path of the robot and should be avoided by the user.

Virtual Environment for Remote Control of UGVs …

527

Fig. 3 UGV simulated in the V-REP software

Created the simulation environment, a differential robot is incorporated that moves horizontally on the plane X-Y. Its design is based on the kinematic model of a unicycle. In Fig. 3, the configuration of this device located on the work space is indicated, denoting the speeds that produce a variation in its displacement. These speeds are two: linear μ (forward displacement) and angular ω (counterclockwise rotation), when modified, allow intuitive control of the robot’s direction. To incorporate the design into the simulation environment, the V-REP libraries of a base prototype (Pioneer 3-DX) are used. This one has two motors (each motor is associated with a wheel that allows it to move according to its dynamics), and ultrasonic sensors are also incorporated to determine the presence of nearby objects.

4.2 MATLAB and V-REP Interaction In the application development, the main step is to adapt the configuration of the simulation software and the mathematical software. Through this configuration, both packages are intercommunicated to test the operation of the haptic device in the control process. To maneuver the direction of the vehicle using the Falcon Novint, the sending and receiving of feedback data is constant and its configuration is detailed in Fig. 4.

4.3 Control of the Mobile Robot In order to avoid collisions between the robot and the elements of the environment, forces are fed into the operating end of the master robot when the proximity to an obstacle exceeds a threshold. In order to carry out the proposed task, the position differences between the robot and the neighboring elements are calculated through the incorporation of a virtual laser. This information allows taking actions for evasion. The implementation of a PID control algorithm allows to determine the feedback force to the haptic device so that the operator can avoid collisions. This prevents damage to the robotic mechanism through the injection of proportional speeds to

528

F. Roberto Segura et al.

Fig. 4 Flowchart of the interaction between MATLAB, Falcon Novint, and V-REP simulator

each wheel (depending on the location of the obstacle in the plane). This process is supervised by visual feedback, since a camera is placed on the front of the robot to visualize the space where it moves, allowing to validate the feedback obtained by the haptic device.

5 Experimental Tests The aim to conduct a UGV within a semi-structured environment is the main requirement in order to carry out the teleoperation experimentation through the use of the Falcon. Considering that the robot is in an initial position, it is proposed to perform an exploration task and take it to a desired point. To do this, the operator must apply linear and angular velocities through the haptic device, while experiencing force feedback. If an object is close and a collision will be made, a force is generated, informing the operator in a direct way, so that it makes the required corrections and keeps getting closer to the desired point. The used hardware to perform these tests has an Intel Core i7-7500U at a 2.7 GHz speed, with 16 GB RAM and an Ethernet network device: NIC Fast Ethernet family RTL8101E, wireless network device: Realtek RTL8187B supporting the Wireless 802.11g 54 Mbps. For the optical device, it has been used the Novint Falcon with 3D Touch Workspace of 4 × 4 × 4, with 2 lbs. of force capabilities and 400 dpi of position resolution. Figure 5a shows the system in execution, at the starting point, and in Fig. 5b, the user is shown performing teleoperation tasks. For the validation of this proposal, 10 tests were carried out with three different users. The experimental data taken from the teleoperation task of the mobile vehicle

Virtual Environment for Remote Control of UGVs …

529

Fig. 5 a Virtual environment presented to the user, b Experimental tests performed by a user

are within a considerable margin of error; in each test, it is seen that the average errors for x, y, z oscillate depending on the movement and position of the mobile. Depending on the distance between the robot and nearby obstacles, the user modifies the return forces to avoid collision. In Fig. 6, the correction force injected into the haptic device is presented as a response to the interaction between the robot and the environment on the X, Y, Z axes. Based on these variations, various errors in each of the axes are calculated, as shown in Fig. 7. After users have completed the task, they are given the ASQ test developed by Lewis [15] which consists of three questions, with a rating of 1–7 each. The lower the selected score, the higher the satisfaction of the usability of the user with his/her system. This is presented in Table 1. Results are determined by calculating the average (arithmetic mean) of the three questions. In this way, the test of this work obtains 1.33, 2, and 2, respectively, for three people, so the general average was 1.78/7.

Fig. 6 Feedback forces on X, Y, Z axes as a response between the robot-scenario interaction

530

F. Roberto Segura et al.

Fig. 7 Control errors obtained when executing the application Table 1 ASQ test Question

Score

1. Overall, I am satisfied with the ease of completing the tasks in this scenario

1–7

2. Overall, I am satisfied with the amount of time it took to complete the tasks in this scenario

1–7

3. Overall, I am satisfied with the support information (online help, messages, documentation) when completing the tasks

1–7

6 Conclusions and Future Works Through the use of a powerful mathematical software and a 3D application development engine, it has been possible to present an intuitive and interactive technological tool for the development of skills of military personnel. In this prototype is integrated mathematical modeling, autonomous control, the use of a haptic device, and a simulated environment. The simulation of robots provides an optimization of economic resources and the opportunity to change its characteristics according to the needs of the final application. The V-REP software allows, based on a virtual environment, to teleoperate the displacement of a UGV while the haptic device performs the feedback forces that provide a complete immersive experience. For this, it is recommended to work on 32-bit-based operating systems to improve compatibility with the libraries of the device, or to develop libraries that allow to generate *.dll files that are easily executable by 64-bit computer packages. When applying the ASQ test, the average values obtained were part of the best range of the evaluation; it shows the virtual environment acceptance for the user, time, complexity, and supplementary material of the application. After the use of this system, weaknesses were detected in the learning process, so that let the authors to propose teaching strategies, improving the skill and reflexes of the student. The users were often practicing and having fun, which contributes to the

Virtual Environment for Remote Control of UGVs …

531

stress reduction to which they are constantly subjected. As the tests are developed, the dexterity and self-motivation increase, which is evidenced in lower error values when maneuvering the vehicle. Based on this, the authors are already working on future research, where the first instance is to create new environments, which include several types of obstacles, to prevent monotony. In addition, it is priority to use the same principle to develop more military and industrial applications and thus cover more areas. Acknowledgements To the authorities of Universidad Técnica de Ambato (UTA), Dirección de Investigación y Desarrollo (DIDE), Instituto Tecnológico Superior Guayaquil—Ambato and CELEC EP., for supporting this work and future research.

References 1. Mosey, S., Guerrero, M., Greenman, A.J.: Technology entrepreneurship research opportunities: insights from across Europe. J. Technol. Transfer 42(1), 1–9 (2017) 2. McCarthy, J., Wright, P.: The Enchantments of Technology. Human–Computer Interaction Series, pp. 359–373. Springer (2018) 3. Zieli´nski, C., Winiarski, T., Kornuta, T.: Agent-based structures of robot systems. In: The 19th Polish Control Conference, KKA 2017, pp. 493–502. Springer (2017) 4. Kunze, L., Hawes, N., Duckett, T., Hanheide, M., Krajník, T.: Artificial intelligence for longterm robot autonomy: a survey. IEEE Rob. Autom. Lett. 3(4), 4023–4030 (2018) 5. Choi, S.Y.: Agent-based human-robot interaction simulation model for the analysis of operator performance in the supervisory control of UGVs. Int. J. Precis. Eng. Manuf. 19(5), 685–693 (2018) 6. Aguilar, W.G., Rodríguez, G.A., Álvarez, L., Sandoval, S., Quisaguano, F., Limaico, A.: Onboard visual SLAM on a UGV using a RGB-D camera. In: 10th International Conference on Intelligent Robotics and Applications, ICIRA 2017, pp. 298–308. Springer (2017) 7. Arbanas, B., et al.: Decentralized planning and control for UAV–UGV cooperative teams. Auton. Rob. 42(8), 1601–1618 (2018) 8. Wang, H.: Adaptive control of robot manipulators with uncertain kinematics and dynamics. IEEE Trans. Autom. Control 62(2), 948–954 (2017) 9. Stevens, J.A., Kincaid, J.P.: The relationship between presence and performance in virtual simulation training. Open J. Model. Simul. 3(2), 41–48 (2015) 10. Smith, M.J., et al.: Virtual reality job interview training for veterans with posttraumatic stress disorder. J. Vocat. Rehabil. 42(3), 271–279 (2015) 11. Siu, K.C., Best, B.J., Kim, J.W., Oleynikov, D., Ritter, F.E.: Adaptive virtual reality training to optimize military medical skills acquisition and retention. Mil. Med. 181(5), 214–220 (2016) 12. Ibari, B., Ahmed-Foitih, Z., Reda, H.E.A.: Remote control of mobile robot using the virtual reality. Int. J. Electr. Comput. Eng. 5(5) (2015) 13. Tikanmäki, A., Bedrník, T., Raveendran, R., Röning, J.: The remote operation and environment reconstruction of outdoor mobile robots using virtual reality. In: 2017 IEEE International Conference on Mechatronics and Automation (ICMA), pp. 1526–1531 (2017) 14. Lwowski, J., Joordens, M., Majumdar, A., Benavidez, P., Prevost, J.J., Jamshidi, M.: The utilization of virtual reality as a system of systems research tool. In: 2018 IEEE 13th Annual Conference on System of Systems Engineering (SoSE), pp. 535–540 (2018) 15. Lewis, J.R.: IBM computer usability satisfaction questionnaires: psychometric evaluation and instructions for use. Int. J. Hum. Comput. Interact. 7(1), 57–78 (1995)

Author Index

A Abreu, António, 39 Aguilar, Wilbert, 205 Altamirano, Santiago, 231, 521 Amorim, Pedro H. O., 425 Andrade, Roberto, 507 Andrango, Raúl, 439 Andrea Sánchez, Z., 231, 521 Aroca, Katherine, 371, 383 Augusto, Maria Fernanda, 87

Degani, M. H., 363 de O. Gonçalves, Mariana C., 425 Domínguez, Miguel A., 397 Dornellas, Rafael M., 425 dos Santos, Lúcio Agostinho Barreiros, 451 Duque, Juan F., 397 Dutra, Max Suell, 307

B Barbosa, Jorge, 29, 65 Barbosa, Joseane R., 425 Barbosa Junior, José Ailton L., 149 Benavides, Eduardo, 51 Brandão, Ana Paula, 257 Brito-Filho, Francisco A., 149 Buele, Jorge, 231, 439, 521

F Fernández, Rafael Caballero, 161 Fernández-Villacañas Marín, Manuel A., 293 Fonseca, Miguel, 321 Forero, Sergio Iván Rueda, 345 Franklin Salazar, L., 231, 439, 521 Fuertes, Walter, 51, 507

C Cadena, Alyssa, 507 Canchignia, Christian, 205 Carrera, Enrique V., 411 Carrillo, Juan M., 397 Carrillo, Sandra, 439 Carvalho, João Vidal, 39 Clery, Arturo, 487 Coelho, Maria Manuela Martins Saraiva Sarmento, 185 Correia, Anacleto, 321, 475 D Daniel Tenezaca, B., 205 de Barros Paes, Carlos Eduardo, 333

E Edisson, Jordan-H., 439

G Garcés, Andrés Gordón, 135 Gonçalves, António, 475 Gonçalves, Rafael, 281 Gualoto, Franklin, 507 Guarda, Teresa, 87, 497 Guerra, Lesslie, 363 H Hallo, Vicente, 371, 383 Huertas, Harley Lovato, 135 J Jácome, Richard Navas, 135 Jakomin, R., 363 Jiménez, Arturo, 371, 383

© Springer Nature Singapore Pte Ltd. 2020 Á. Rocha and R. P. Pereira (eds.), Developments and Advances in Defense and Security, Smart Innovation, Systems and Technologies 152, https://doi.org/10.1007/978-981-13-9155-2

533

534 K Kaczynska, Manuela, 281 L Lopes, Isabel, 87 Lopes, Isabel Maria, 497 López, Victoria M., 439 Loureiro, Nuno Alberto Rodrigues Santos, 451 M Maialle, M. Z., 363 María, Guillermo Santa, 487 Marco Antonio, Zurita C., 123 Mendoza, Dario, 205, 371, 383 Menezes, Sofia, 245, 281 Molina, Lilian, 87, 487 Mourão, R. T., 363 Moura, Ricardo, 321 N Narváez, Celio Humberto Puga, 161 Neto, Valdemar Vicente Graciano, 333 O Oliveira, Pedro, 497 P Palomeque, David, 397 Paredes, Manolo, 411 Penello, G. M., 363 Pereira, Pedro, 363 Pereira, Robson P., 425 Pérez-Aloe, Raquel, 397 Pérez, Bruno, 397 Pinto, L. D., 363 Pires, M. P., 363 Procel, Patricia Constante, 135 R Ramamoorthi, Lokesh, 15 Ramírez-Cabrales, Fabián, 345 Raptopoulos, Luciano Santos Constantin, 307

Author Index Reinoso, Cristina, 439 Reis, João, 245, 281 Rey, Carina, 487 Ribeiro, Constantino G., 307 Roberto Segura, F., 231, 521 Rocha, Álvaro, 39, 87 S Sanchez, Manuel, 51 Sanchez, Sandra, 51 Santiago Alvarez, T., 231, 521 Santos, Lúcio Agostinho Barreiros dos, 185 Sarkar, Dilip, 15 Semaan, Felipe S., 425 Silveira, João Almeida, 269 Soria, Angel, 439 Souza, P. L., 363 T Tapia, Freddy, 507 Téllez, Osvaldo Fosado, 161 Tello-Oquendo, Luis, 507 Tomás Núñez, C., 231, 521 Torres, Jenny, 507 U Urrutia-Urrutia, Pilar, 231, 439, 521 V Valverde, José Mª, 397 Velasco, Nancy, 371, 383 Vélez, Alex Fernando Jimenez, 161 Viana, Nayana L. M., 149 Victor, Avelino, 39 Victor, José Avelino, 87 Victor Xavier, Enríquez C., 123 W Wilbert Geovany, Aguilar C., 123 Z Zacarias, Marielba, 475