PEOPLE'S DEMOCRATIC REPUBLIC OF ALGERIA

Ministry of Higher Education and Scientific Research

University of Mohamed Khider – Biskra

Faculty of Exact Sciences, Computer Science

Sciences of Natural and Life Department

Order number: ……………

THESIS
Presented in order to obtain the degree of
DOCTOR 3rd CYCLE LMD IN COMPUTER SCIENCE
Option: Artificial Intelligence

Title

An Agent-Based Approach for the

Internet of Things

Defended by
Maroua AHMID

Before the jury composed of:

Chairman Pr. Mohamed Tayeb LASKRI University of Annaba
Supervisor Pr. Okba KAZAR University of Biskra
Co- Supervisor Pr. Laid KAHLOUL University of Biskra
Examiner Pr. Faiza KHELLAF University of USTHB
Examiner Pr. Khaled REZEG University of Biskra
Invited Pr. Malika IOUALALEN University of USTHB

2021/2022
 REPUBLIQUE ALGERIENNE DEMOCRATIQUE ET POPULAIRE
Ministère de l’Enseignement Supérieur et de la Recherche Scientifique
Université Mohamed Khider – BISKRA

Faculté des Sciences Exactes,

Département
des Sciences de la Nature et de
d’informatique
la Vie

N° d’ordre: ……………

THÈSE
Présentée en vue de l’obtention du diplôme de
DOCTORAT 3ème CYCLE LMD EN INFORMATIQUE
Option: Intelligence Artificielle

Titre

Une Approche Basée Agent pour

l'Internet des Objets

Présentée par
Maroua AHMID

Devant le jury composé de:

Président Pr. Mohamed Tayeb LASKRI Université d’Annaba
Rapporteur Pr. Okba KAZAR Université de Biskra
Co-Rapporteur Pr. Laid KAHLOUL Université de Biskra
Examinateur Pr. Faiza KHELLAF Université de USTHB
Examinateur Pr. Khaled REZEG Université de Biskra
Invitée Pr. Malika IOUALALEN Université de USTHB

2021/2022
“The way of success is the way of
continuous pursuit of knowledge.”
Napoleon Hill
Acknowledgements

First and foremost, I would like to thank Almighty ALLAH, who gave me the
strength, patience, and courage to continue this work.

My sincere gratitude goes to my supervisor, Pr. Okba KAZAR for multiple

reasons that elevated the quality of this work: his patience, encouragement, advice, and
especially the confidence he had in me.

I express all my gratitude to co-supervisor Pr. Laid KAHLOUL for his advice and
support.

Also, I would like to warmly thank the jury. Starting with the president, Prof.
Mohamed Tayeb LASKRI as well as the examiners: Prof. Faiza KHELLAF, Prof. Khaled
REZEG, and Prof. Malika IOUALALEN for having done me the honor of accepting to
judge this work, also generously offering their time, guidance, and goodwill throughout
the review of this document.

Last and not least, I would like to thank all my teachers, friends, and colleagues in
the computer science department.

Maroua Ahmid
Dedication

In my life, I love,

My sisters, brothers, and friends,

But, please excuse me,

There are two people that I want to mention

My mother, the source of happiness and motivation,

My father, the source of advice and patience,

To all of you, I dedicate this dissertation.

Maroua
 ‫الملخص‬

‫ثريا‪ .‬وكقد جذبت اكثري من الهتمام بفضل‬

‫ف اكسنوات الرية‪ ،‬أصبحت ننتنت الشياء )‪ (IoT‬معروفتا ي‬
‫تببيقاتا اكقوية ف متلف ااالت‪ .‬حاكييا‪ ،‬مليارات من أجهزة ننتنت الشياء متصلة للنتنت‪ .‬يؤدي الرتفاع الائل ف‬

‫عدد الجهزة التصلة نل ننتاج الزيد من اكبيانت اكت تتبلب تقنيات تزين ومعالة عاكية‪ .‬نل جانب ذكك كدي معدات‬

‫ننتنت الشياء قدرة معالة و اكتخزين مدودة ل تسمح ععالة اكبيانت أو حت تزينها‪ .‬أيضا ل يثن كلشياء اتاذ‬

‫قرارات ذثية وسريعة كتحقيق أهدافها‪ ،‬وتعتمد قدرة اكشيء على اكتواصل مع أشياء أررى على تشابه الدمة وبروتوثولت‬

‫التصال؛ سيؤار ذكك على أداء تببيقات ننتنت الشياء وثفاءتا وتبويرها وقابليتها كلتوسع‪ .‬تقدم هذه الطروحة الث‬

‫مساهات رئيسية‪ ،‬ف الساهة ألول نقتح نج ‪ Cloud-IoT‬كتوفي حل فعال كتخزين وتليل اكبيانت اكناتة عن‬

‫تببيقات ننتنت الشياء وتسي قابلية توسعة ننتنت الشياء‪ .‬من أجل نابات جدوى اكنهج القتح ت تبوير نظام‬

‫‪Cloud-IoT‬لراقبة الريض عن ببعد‪.‬‬

‫أصبح أمن ننتنت الشياء مشثلة أثرر صعوبة بسبب حدود الوسبة والتصالت لجهزة ننتنت الشياء ما‬

‫قوً يعتمد على اكتشفي‬

‫يعل ننتنت الشياء أثرر عرضة لجمات المن والصوصية‪ .‬و من ث‪،‬ثمساهة ثنية‪ ،‬نقتح حلي ي‬
‫أيضا‪ ،‬استخدمنا آكية اكتحثم ف اكوصول كضمان رصوصية اكبيانت‪ .‬توضح اكتجربة أن‬
‫الفيف كضمان أمن اكبيانت‪ .‬ي‬
‫الل القتح أثرر ملءمة لجهزة ننتنت الشياء ‪.‬كلستفادة اكقصوى من تدفق اكبيانت أكضخمة يتاج ننتنت الشياء نل‬

‫ندارة وتليل اكبيانت ف اكوقت اكفعلي ‪.‬من أجل تعزيز ذثاء أجهزة ننتنت الشياء وتسي نجنا ألول قمنا بتبوير حل‬

‫فعال ومستقل يعتمد على اكوثيل من اجل ندارة اكبيانت وتليلها ف اكوقت القيقي ثمساهة ثكرة ‪.‬يثن كلنظام القتح‬

‫اتاذ قرارات سريعة و مناسبة ف حاكة اكبوارئ‪.‬‬

‫الكلمات الرئيسية‪ :‬ننتنت الشياء ‪،‬الوسبة اكسحابية ‪،‬نظام متعدد اكوثلء ‪،‬المن ‪،‬اكرعاية اكصحية واكتشفي‪.‬‬
Résumé

Ces dernières années, l’Internet des Objets (IdO) est devenu très connu. Il a attiré
beaucoup d’attention grâce à son application puissante dans les différents domaines.
Actuellement, des milliards d’appareils IdO sont connectés. L’augmentation massive du
nombre d’appareils connectés produit instantanément des données plus qui nécessitent
technologie élevée de stockage et de traitement. Aussi, matériel IdO a une capacité de
traitement et de stockage limité ne permet pas le traitement des données ou même le
stockage. Également, les objets ne peuvent pas prendre des décisions intelligentes et
rapides pour atteindre leurs objectifs, et la capacité de communication de l’objet avec
d’autres objets dépend de la similarité des services et protocoles de communication; cela
aura un impact sur les performances, l’efficacité, le développement et la scalabilité des
applications IdO. Cette thèse propose trois contributions principales, dans la première
contribution, nous proposons une approche Cloud-IdO pour fournir une solution efficace
pour le stockage et l’analyse des données générées par les applications IdO et améliorer
la scalabilité de l’IdO. Afin de démontrer la viabilité de l’approche proposée, un système
Cloud-IdO pour la surveillance à distance des patients est développé.

La sécurité de l’IdO devient un problème plus difficile en raison des limites de

calcul et de communication des appareils IdO qui rendent l’IdO plus vulnérable aux
attaques de sécurité et de confidentialité. Ainsi, comme deuxième contribution, nous
proposons une solution robuste basée sur une cryptographie légère pour assurer la
sécurité des données. De plus, nous avons utilisé un mécanisme de contrôle d’accès pour
garantir la confidentialité des données. L’expérimentation démontre que la solution
proposée est plus appropriée aux appareils IdO. Pour tirer le meilleur parti des flux de
données massifs, l’IdO a besoin d’une gestion et d’une analyse des données en temps réel.
Afin de renforcer l’intelligence du dispositif IoT et d’améliorer notre première approche,
nous avons développé une solution efficace, autonome basée sur les agents pour la
gestion et d’analyse des données en temps réel, comme troisième contribution. Le
système proposé peut prendre des décisions rapides et appropriées en cas d’urgence.

Mots clés: Internet des Objets, Cloud, Système Multi-Agents, Sécurité, Soins de Santé et
Cryptographie.
Abstract

In recent years, the Internet of Things (IoT) has become well known. It has
attracted much attention thanks to its powerful application in different fields. Currently,
billions of IoT devices are connected. The massive rise in the number of connected
devices instantly produces further data that requires high storage and processing
technology. Besides, the IoT equipment has limited processing, and storage capability did
not allow data processing or even storage. Also, things cannot make smart and quick
decisions to achieve their goals, and the thing ability to communicate with other things
depends on service similarity and communication protocols; this will impact the
performance, efficiency, development, and scalability of IoT applications. This thesis
provides three main contributions, in the first contribution, we propose a Cloud-IoT
approach to provide an efficient solution for the storage and analysis of data generated by
IoT applications and improve IoT scalability. In order to demonstrate the viability of the
proposed approach, a Cloud-IoT system for remote patient monitoring is developed.
IoT security becomes a more challenging problem due to the computing and
communication limits of IoT devices make IoT more vulnerable to security and privacy
attacks. Hence, as a second contribution, we propose a robust solution based on
lightweight cryptography to ensure data security. Also, we have used the access control
mechanism to ensure data privacy. The experimentation demonstrates that the proposed
solution more suitable for IoT devices. To tap the most out of massive data streams, IoT
needs real-time management and analysis of data. In order to enhance the intelligence of
IoT devices and to improve our first approach, we have developed an efficient,
autonomous, and real-time solution based on the agent for data management and analysis
as a third contribution. The proposed system can make fast and apt decisions in an
emergency case.

Keywords: Internet of Things, Cloud, Multi-Agent System, Security, Healthcare, and

Cryptography.
Table of Contents

Acknowledgements............................................................................................................. i

Dedication...........................................................................................................................ii

‫الملخص‬................................................................................................................................. iii

Résumé...............................................................................................................................iv

Abstract...............................................................................................................................v

Table of Contents.............................................................................................................. vi

List of Tables...................................................................................................................ivii

List of Figures................................................................................................................. viii

List of Algorithms............................................................................................................. ix

List of Abbreviations......................................................................................................... x

General Introduction......................................................................................................... 1

Chapter 1: Internet of Things Overview......................................................................... 5

1.1 Introduction....................................................................................................................5
1.2 IoT Definitions...............................................................................................................6
1.2.1 Internet of Things....................................................................................................6
1.2.1 Thing.......................................................................................................................7
1.3 IoT Characteristics......................................................................................................... 8
1.3.1 Connectivity............................................................................................................8
1.3.2 Heterogeneity..........................................................................................................8
1.3.3 Dynamic Nature......................................................................................................8
1.3.4 Enormous Scale...................................................................................................... 9
1.4 IoT Architecture.............................................................................................................9
1.4.1 Perception Layer................................................................................................... 10
1.4.2 Network Layer...................................................................................................... 10
1.4.3 Middleware Layer.................................................................................................10
1.4.4 Application Layer................................................................................................. 10
1.5 IoT Applications.......................................................................................................... 11
1.5.1 Healthcare............................................................................................................. 11
1.5.2 Industrial............................................................................................................... 12
1.5.3 Transportation.......................................................................................................13
1.5.4 Cities..................................................................................................................... 14
1.6 Challenges and Issues.................................................................................................. 15
 1.6.1 Green IoT..............................................................................................................15
1.6.2 Security................................................................................................................. 17
1.6.3 Data Storage and Processing.................................................................................19
1.6.4 Intelligence............................................................................................................20
1.7 Conclusion................................................................................................................... 21

Chapter 2: A Cloud-IoT System for Remote Patient Monitoring...............................23

2.1 Introduction..................................................................................................................23
2.2 Related Works..............................................................................................................24
2.3 System Architecture.....................................................................................................28
2.3.1 Devices..................................................................................................................29
2.3.2 Gateway................................................................................................................ 29
2.3.3 Cloud.....................................................................................................................29
2.3.4 Application............................................................................................................29
2.4 Implementation and Results.........................................................................................30
2.4.1 Development Tools...............................................................................................30
2.4.2 Case Study............................................................................................................ 32
2.4.3 System Interfaces..................................................................................................34
2.4.4 Results and Discussion......................................................................................... 37
2.5 Conclusion................................................................................................................... 38

Chapter 3: An Integrated Privacy Protection System for IoT.................................... 41

3.1 Introduction..................................................................................................................41
3.2 Related Works..............................................................................................................42
3.3 Security System........................................................................................................... 46
3.3.1 Data Confidentiality..............................................................................................46
3.3.2 Access Control......................................................................................................49
3.4 Experimental Evaluations............................................................................................ 51
3.4.1 Experimental Setup...............................................................................................51
3.4.2 Results and Discussion......................................................................................... 51
3.5 Conclusion................................................................................................................... 54

Chapter 4: A Cloud-IoT System Based on Smart Agent..............................................56

4.1 Introduction..................................................................................................................56
4.2 Background.................................................................................................................. 57
4.2.1 Definition of Agent...............................................................................................57
4.2.2 Characteristics of Agent........................................................................................58
4.2.3 Classification of Agent......................................................................................... 59
4.2.4 Definition of Multi Agent System........................................................................ 59
4.2.5 Characteristics of Multi Agent System.................................................................60
4.3 Related Works..............................................................................................................61
4.4 Proposed Approach......................................................................................................64
4.4.1 System Architecture..............................................................................................64
4.4.2 Agents Interactions............................................................................................... 68
4.5 Implementation and Results.........................................................................................70
4.5.1 Development Tools...............................................................................................70
4.5.2 Case Study............................................................................................................ 71
4.6 Results and Discussion................................................................................................ 73
4.7 Conclusion................................................................................................................... 74

General Conclusion..........................................................................................................75

List of Publications.......................................................................................................... 77

REFERENCES.................................................................................................................79
List of Tables

Table 2.1. Comparison of IoT-based systems......................................................... 27

Table 3.1. Comparison of security solutions in IoT systems….…………...…..… 45
Table 4.1. Comparison of agent-based solutions in IoT systems……………..….. 63
Table 4.2. Resting heart rate range……………………………...…………….….. 71
List of Figures

Figure 1.1. Internet of things dimensions…….……………………………..….… 6

Figure 1.2. Communication models in IoT..……….……………………………... 7

Figure 1.3. Internet of things architecture...……….…………………………....... 9

Figure 1.4. Top 10 IoT projects in 2020.…………….………………..……….... 11

Figure 1.5. The COVID-19 symptoms sensor……..…….…………………….... 12

Figure 1.6. The YuMi dual-arm robot……………………..………………….…. 13

Figure 1.7. BMW i8 sports car.…………………………..…………………...…. 14

Figure 1.8. Smart city applications.……………………………..……………..… 15

Figure 1.9. The green Internet of things…………...……….…..……………..…. 17

Figure 1.10. Security requirements of each IoT layer…………..………….…...… 18

Figure 1.11. The requires technologies for Intelligent Internet of Things……...… 21

Figure 2.1. The general architecture of the proposed system……….….….….… 28

Figure 2.2. The wiring diagram of heart rate measurement device…….……...… 31

Figure 2.3. A Cloud-IoT based system for patient heart rate monitoring………. .32

Figure 2.4. The heart rate measurement device of the proposed system………… 33

Figure 2.5. Cloud data structure of the proposed system………………….…..… 34

Figure 2.6. The login interface of the mobile application……………………..… 34

Figure 2.7. The signup interface of the mobile application…………………....… 35

Figure 2.8. The patient profile interface of the mobile application……………… 36

Figure 2.9. The patient list interface of the mobile application…………………. 36

Figure 2.10. Cloud data of the proposed system…………………………….….… 37

Figure 2.11. The heart rate readings of devices………………………….……..… 37

Figure 2.12. Comparison of heart rate readings between devices…………….….. 38

Figure 3.1. Interaction diagram of the ECC ElGamal cryptosystem…………..… 48

Figure 3.2. Comparative analysis of algorithms' key size………………..……… 52

Figure 3.3. Comparative analysis of algorithms’ encryption time…….………… 53

Figure 3.4. Comparative analysis of algorithms’ decryption time……….……… 53

Figure 4.1. System global A-IoT architecture…………………………………… 64

Figure 4.2. The architecture of device agent………………..…………………… 65

Figure 4.3. The architecture of gateway agent…………………………...……… 66

Figure 4.4. The architecture of the manager agent……………………….……… 66

Figure 4.5. The architecture of the analyzer agent………………….…………… 67

Figure 4.6. The architecture of the user agent…………………………………… 67

Figure 4.7. General diagram of interactions between agents……………….…… 69

Figure 4.8. A Cloud-IoT health monitoring system based on smart…………..… 71

Figure 4.9. Smart device agent's algorithm……………………………………… 72

Figure 4.10. Alert message………………………………………………...……… 73

Figure 4.11. Accurate emergency case detection……………….………………… 74

List of Algorithms

Algorithm 1. Initialization……………………………………………….……….. 47

Algorithm 2. Key Generation of device………………………………………...… 47

Algorithm 3. Encryption………………………………………………………..… 48

Algorithm 4. Decryption……………………………………………………….…. 49

Algorithm 5. Access Control………………………………………………….….. 50

List of Abbreviations

IoT Internet of Things

MIT Massachusetts Institute of Technology

MEMS Micro-Electro-Mechanical Systems

CERP-IoT Cluster European research projects on the Internet of things

NFC Near Field Communication

GSM Global System for Mobile Communications

LTE Long Term Evolution

LoRa Long Range

IIoT Industrial Internet of Things

CO2 Carbon dioxide

G-IoT Green Internet of Things

WSN Wireless Sensor Network

DC Data Center

QoS Quality of Service

M2M Machine-to-Machine

AI Artificial Intelligence

ML Machine Learning

NLP Natural Language Processing

DL Deep Learning

AIoT Artificial Intelligence of Things

BMI Body Mass Index

GPRS General Packet Radio Service

CHF Chronic Heart Failure

ECG Electrocardiography

GSM Global System for Mobile Communications

WBSN Wireless Body Sensor Networks

IDE Integrated Development Environment

SDK Software Development Kit

WHO World Health Organisation

BPM Beats per Minute

MAC Message Authentication Codes

DES Data Encryption Standard

OCB mode Offset codebook mode

AES Advanced Encryption Standard

PKI Public-Key Encryption

CP-ABE Ciphertext-Policy Attribute-based Encryption

RSA Rivest-Shamir-Adleman

MEMK Memory Efficient Multi Key

CRT Chinese Remainder Theorem

ECC ElGamal Curves Cryptosystem ElGamal algorithm

UN Username

Password PW

NIST National Institute of Standards and Technology

MAS Multi-Agent Systems

DA Device Agent

GA Gateway Agent

WAN Wide Area Network

PAN Personal Area Network

LAN Local Area Network

MA Manager Agent

AA Analyzer Agent

UA User Agent

JADE Java Agent DEvelopment Framework

GUI Graphical User Interface

General Introduction

In the introduction of this thesis, we will first present the context and problematic
of the research work, followed by an overview of the thesis's contributions. Finally, we
will describe the thesis organization.

Context and Problematic

The Internet of Things is a new Internet revolution. It contributes connects the

digital and the physical world by using information and communication technologies. It
also provided powerful applications that permeating into practically all areas that play a
leading role and directly affect the lives of humans, such as healthcare, industry,
transportation, and logistics. IoT facilitating people’s daily lives accelerated economic
growth. However, many issues and challenges still represent significant concerns for IoT
growth and development and widespread adoption. The IoT challenges are related to
many factors, such as a heterogeneous environment, increased demands for data storage
and processing, security and privacy threats, etc.

The huge increase of connected devices number generates more data immediately
that requests for more storage and processing technologies. Total data volume for IoT
devices worldwide is expected to exceed tens of zettabytes in the next few years, this will
exacerbate the challenges of storing, processing, and managing large volumes of data. In
addition, data security and privacy issues must be considered seriously in IoT
applications due to the volume and the sensitivity of data created and distributed through
the network, accessible not only by users but also by devices and services. Furthermore,
the IoT device's limitations in terms of processing and storage make IoT more vulnerable
to protection and privacy attacks. While the loT system shows capable of making
decisions and interacting intelligently with its devices, the real situation is that these
devices are still not smart enough to make quick and smart decisions in dangerous
situations. It works so based on specific business rules and does not take into account
unexpected environmental changes. Furthermore, the ability of devices to communicate
with other devices is dependent on service similarity and contact protocols. Addressing

1
these problems would be the key to achieving the long-term efficiency and stability of
IoT.

Contributions

In view of all the above problems, this thesis has three contributions; each of them
addresses a particular issue:

 As the first contribution, a Cloud-IoT approach for real-time remote monitoring is

proposed. We concentrate on the concept of integrating IoT and cloud computing
technologies to enhance the IoT ability of processing and storage by the cloud's
limitless power and capability, as well as to improve IoT scalability. In order to prove
the feasibility of our approach, a Cloud-IoT system for remote heart rate monitoring is
developed.

 As the second contribution, an integrated privacy protection system for IoT is

proposed. The proposed system ensures data confidentiality in the perception layer
and controls access to the data in the cloud.

 As the third contribution, a Cloud-IoT system based on smart agents is proposed,

represent an integrated system that combines IoT and intelligent agent technology to
enhance IoT intelligence. In order to prove the feasibility of our approach, a Cloud-
IoT system based on smart agents for the health monitoring of cardiovascular patients
is proposed. This approach conforms to the approaches proposed in the previous
contribution.

Thesis Organization

This thesis is contained four chapters, organized in the "articles" format. So, some
chapters are transcriptions of papers that have been published in or submitted to
international journals and conferences.

 Chapter 1 gives an overview of the Internet of Things. Here, we provide definitions,

characteristics, architecture, and applications of IoT. In addition, we highlight the most
critical issues and challenges of IoT.

2
 In chapter 2, we propose a Cloud-IoT approach to improve IoT scalability. In order to
prove the feasibility of the proposed system, a Cloud-IoT system for remote patient
monitoring is developed. Also, we carry out the experimental analysis of the proposed
approach.

 In chapter 3, we propose an integrated privacy protection system for IoT. In order to

verify the efficiency of the proposed system, a study of the performances of the
proposed approach is elaborated by experimentation.

 In chapter 4, we propose a Cloud-IoT system based on smart agents has been

proposed. It represents an integrated system that combines IoT and intelligent agent
technology to enhance IoT intelligence. We evaluate our proposal by experimentation.

 Finally, the thesis conclusion synthesizes the contributions and highlights the future
perspectives of this research.

3
“The Internet of Things has the potential
to change the world, just as the Internet
did. Maybe even more so.”
Kevin Ashton

4
Chapter 1

Internet of Things Overview

1.1 Introduction

Over the past several years, many research have contributed to the emergence of
the Internet of Things (IoT). In October 1989, John Romy developed a toaster that could
be switched on/off from Internet, which was considered the first IoT device [1]. After one
decade, the term Internet of Things was first used in 1999 by Kevin Ashton at the
Massachusetts Institute of Technology (MIT) [2]. Over several years, the Internet of
Things evolved due to the integration of different technologies, ranging from cellular
networking to the Internet and from embedded devices to micro-electromechanical
systems (MEMS) [3], which leads to increase the number of devices, for the first time,
the number of devices connected to the Internet outnumbered the number of people in the
world in 2008 [4]. In recent years, IoT research has become very popular and has
attracted much attention from research teams. It contributed to interconnecting the
physical world and the virtual world by using multiple information and communication
technologies. It also provided powerful applications and systems in different fields,
including healthcare, industry, transportation, and logistics. Consequently, it facilitated
people’s daily lives and accelerated economic growth. However, IoT technology is still
facing several issues that need to be addressed.

For more understanding IoT, the present chapter provides an overview of the
Internet of Things. It introduces the different definitions and characteristics of IoT,
describes the architecture of IoT and explains its layers, and most useful IoT applications
fields, also highlight the most critical issues and challenges.

5
1.2 IoT Definitions

The Internet of Things is a new Internet revolution; it connects the digital and the
physical world. The IoT allows things to connect everything and anyone, at any time,
anywhere, using any network. As shown in Figure. 1.1. It has introduced a new
dimension to information and communication technologies, in addition to the time, space,
and network dimensions that allow anybody to connect at any time and from anywhere,
through any network. Now we have a new thing dimension that allows us to connect to
anything can be tagged, monitored, and connected.

Figure 1.1. Internet of things dimensions.

1.2.1 Internet of Things

Since coined IoT term for the first time, attempts have been made to define the
Internet of Things by many different groups, including researchers, academicians,
developers, and corporate people. However, until now, there is no universal definition of
the Internet of Things. In the following, we mention the most known definitions of the
Internet of things.

Haller, Karnouskos and Schroth define the Internet of Things as "A world where
physical objects are seamlessly integrated into the information network, and where the
physical objects can become active participants in business processes. Services are

6
available to interact with these smart objects over the Internet, query their state and any
information associated with them, taking into account security and privacy issues." [5]

The Cluster European research projects on the Internet of things (CERP-IoT)

defines the Internet of things as: "a dynamic global network infrastructure with self-
configuring capabilities based on standard and interoperable communication protocols
where physical and virtual things have identities, physical attributes, and virtual
personalities and use intelligent interfaces, and are seamlessly integrated into the
information network." [6]

1.2.1 Thing

A thing in the context of IoT is a physical device with a unique identifier

associated with at least one address that can be machine-readable and used to
communicate with it. It has an embedded system, has at least one communication model
shown in Figure 1.2, some essential computing skills; this can range from the ability to
sensing to the ability to perform complex calculations. It role is to detect physical
phenomena (temperature, light, level of electromagnetic radiation, etc.) from the physical
environment around them by using embedded sensors, collect data, then sent it to the
desired destination through the Internet. Also, it may have the ability to act or react in the
face of situations and changes in the environment.

Figure 1.2. Communication models in IoT.

7
1.3 IoT Characteristics

The IoT has several characteristics; those characteristics differ depending on the
domain. In the following, some of the common fundamental characteristics of the IoT.

1.3.1 Connectivity

In the IoT system, devices, services, and even people connected to the Internet, to
collect, store and exchange the data that will be analyzed later. The connection of IoT
devices is pivotal as these simple interactions contribute to the collective intelligence of
the IoT system. Each IoT system has specific needs that determine the most suitable
connectivity technology. The choice of connectivity technology depends on different
elements, such as the application domain, network type, distance, and the environment,
etc [7].

1.3.2 Heterogeneity

Heterogeneous IoT refers to allow communicating with a wide variety of devices

using multiple protocols. Today, there is a large collection of different IoT devices. and
applications that interact with other devices, services, or platforms through various kinds
of networks [8]. Heterogeneity is an important feature of the IoT system as it enhances
IoT scalability, but at the same time, managing a huge number of heterogeneous devices
is a big challenge.

1.3.3 Dynamic Nature

IoT dynamic nature means any changes in devices state such as

connected/disconnected, location, speed, etc. Also, any change around the device [9],
such as a change in temperature and humidity degree, or even a change in the number of
devices by connecting new things and applications, disconnecting others, add new
services, segregate services, discontinue services, etc. IoT technology is constantly
evolving and ​ progressing at a fast pace.

8
1.3.4 Enormous Scale

Internet-connected devices are increasing day by day. For the first time, the
number of Internet-connected devices outnumbered the world's population in 2008 [4],
with their number reached 7 billion devices in 2018. By 2022, Cisco visual networking
index expects that 12.3 billion devices will be connected to the Internet, exceeding the
world’s projected population at that time (8 billion) by 4.3 billion devices [10]. Thus,
management, analysis, and interpretation of the generated data will become more
challenging.

1.4 IoT Architecture

The IoT has special characteristics that must take into consideration when
designing the IoT architecture; for this reason, many researchers have proposed various
architectures for IoT with three layers, four layers, and five layers [11]. The most popular
and widely applied model is four-layer architecture [12] as shown in Figure 1.3.

Figure 1.3. Internet of things architecture.

9
1.4.1 Perception Layer

The perception layer or "device layer" represents the bottom layer of the Internet
of things architecture, consisting of different sensors, actuators, and real-world objects
[13]. The perception layer's primary responsibilities are to measure, detect, collect, and
extract data related to devices. This data may be related to the object's location,
movement, and direction or related to environmental factors such as humidity,
temperature, and wind speed, etc.

1.4.2 Network Layer

The network layer or "transmission layer" plays an essential role in transferring

the information collected from sensors to the middleware layer's processing unit, in
addition to addressing, and routing functions [13]. Its principal goal is to connect all
things together and allow things to share the data with other connected things via the
Internet [14].

1.4.3 Middleware Layer

The primary responsibility of the middleware layer or "service layer" is the

service configuration and management. Also, determine the most appropriate service or
device to satisfy the user request, because IoT devices can implement specific services
and can only connect and communicate with those that implement the same type of
service [15]. Moreover, the middleware layer aggregate, filter, save, process, and analyze
the data coming from the network layer then make decisions [16].

1.4.4 Application Layer

The application layer is the highest layer in IoT architecture, and it is visible to
the end-user. Mainly it managing the applications based on the data managed and treated
by the middleware layer, also, supports the service requests by users [17]. It includes all
applications using IoT technology; it can be healthcare, manufacturing, and transportation,
etc.

10
1.5 IoT Applications

The goal of IoT is to increase the efficiency of the environment in order to make
them more profitable and comfortable for people; IoT solutions can provide competitive
advantages over current solutions. Currently, there are numerous and diverse IoT
applications, permeating into practically all areas that play a leading role and directly
affect the lives of humans. According to the IoT analytics research tracking IoT platforms,
the most IoT projects in 2020 were developed in manufacturing, industrial, transportation,
energy, retail, cities and healthcare as shown in Figure 1.4, where the analysis was based
in 1414 actual IoT projects.

Figure 1.4. Top 10 IoT projects in 2020 [60].

1.5.1 Healthcare

In the healthcare field, IoT has gained more attention because of its vital role in
this field. It moves the medical process away from hospitals and provides patients the
ability to access care remotely and receive assistance in emergency cases. It also allows
healthcare providers to monitor patient health status in real-time, at home, work, or
anywhere [18]. The real-time monitoring system helps healthcare professionals to
understand the variations that can influence patients' health status and help them detect
illness early. Also, it may reduce complications and help the healthcare providers propose
the appropriate treatment before things go wrong [19]. Thus, save patients' lives.

11
Moreover, IoT improves the quality and effectiveness of service at a lower cost and
improves patients' lifestyle and supports independent or semi-autonomous living,
especially in the case of the elderly, disabled, or people who have a chronic illness
requiring constant supervision.

The advance in sensor technology, including wearable, implantable, and ambient

sensors, allows healthcare providers to monitor patient health status and vital information
such as heart rate, temperature, oxygen saturation, etc. Healthcare application growth is
growing in tandem with the advancement in sensor technologies [20]. There are currently
various healthcare applications such as mood monitoring, oxygen saturation monitoring,
human activity recognition, etc. One of the most promising applications is a small sensor
shown in Figure 1.5, that sticks to a person's throat and detects heart rate, body
temperature, measuring the vibrations of coughs and chest movements for detecting early
signs of COVID-19 symptoms.

Figure 1.5. The COVID-19 symptoms sensor [21].

1.5.2 Industrial

The Industrial Internet of Things (IIoT), known as the fourth industrial revolution
(Industry 4.0) [22], shapes the future of innovative and modern manufacturing. It aims to
automate the cycle of manufacturing from material warehousing to marketing. IIoT
facilitates the production flow in a manufacturing plant, helps stakeholders monitor the
supply chain, manages warehouses. IIoT gives industrial managers a more accurate view
of how industrial operations are moving along and find dark places in the manufacturing

12
process such as defects in machinery, or a machine goes down [29]. Also, help them
monitor the factors that might lead to operating conditions that are less than optimal and
make informed business decisions to improve productivity, thus providing faster time to
market and increase profits [23].

Today, with the advancement of IIoT, several IIoT applications already on the
market. One of them is YuMi robot, it is a dual-arm robot with work side-by-side with
people on the same tasks, and it is tireless the precise and repetitive tasks such as small
parts assembly, as shown in Figure 1.6.

Figure 1.6. The YuMi dual-arm robot [24].

1.5.3 Transportation

Transportation is the backbone of our lives, and the need for it is increasing day
by day. Thus, transportation must be improved not only for passenger safety and security
but also for the goods transported and vehicles. Currently, IoT makes a revolution in the
transportation industry and makes transportation safer by integrating sensors in roads [25],
railways, and vehicles will lessen the chances of sudden breakdowns and helps to predict
the possible failures of the machine parts, also help the maintenance department for the
preventive measures. As a result, lesser maintenance breakdown delays. Also, IoT allows
better monitoring and analysis of traffic flow through devices at all traffic observation
points; it helps drivers with suggestions of alternative roads in case congestion, helps
them get parking spots, and informs them of the fees required [26]. They are thus leading

13
to minimize traffic congestion and improve road safety. Now, there are many IoT
applications in transport, such as real-time location monitoring of the fleet, tracking of
cargo, fuel and mileage statistics, smart cars, etc.

BMW i8 sports car shown in Figure 1.7 is equipped with several sensors. These
sensors collect a set of data such as location, mileage, destination routing, battery voltage,
fuel, and fluid levels, in addition to the health and status of the car components, etc. Soon,
it will also pick up more weather data through sensors on the vehicle. This data is
available to drivers through the BMW Connect app, which provides time-to-leave
notifications, car remote control, monitoring, etc.

Figure 1.7. BMW i8 sports car.

1.5.4 Cities

A smart city is a city connected, equipped with a significant number of sensors

and actuators, this enables it to track, monitor, manage, and control the city. Also, it
accurately identifies the points that need improvement and implements most of the
changes remotely, to anticipate any anomalies and improve the quality of life for citizens
[26]. Smart cities applications can make cities more effective and efficient. It provides
efficient management of waste, water supply, and energy to reduce economic and
environmental costs [27], fast assistance in emergency cases due to surveillance systems
anywhere, etc. As shown in Figure 1.8, there are many IoT applications in smart cities
such as the smart grid, smart governance, smart education, and smart buildings, etc [28].
In the last few years, various smart city projects have been launched, such as Oslo, San

14
Francisco, London, Barcelona, and Singapore. In 2018 Algerian government started an
ambitious project to transform Algiers city into a smart city by 2035 gradually.

Figure 1.8. Smart city applications.

1.6 Challenges and Issues

IoT affects various aspects of human lives, as we mentioned in the previous

section. However, many issues and challenges still represent significant concerns for IoT
growth and development and widespread adoption. Therefore, problems and concerns
related to IoT need to be addressed from different aspects, such as services, social and
environmental impacts, etc [30]. The IoT challenges are related to many factors, such as a
heterogeneous environment, increased demands for data storage and processing, security
and privacy threats, etc., overcoming them will be the key to creating lasting productivity
and prosperity.

1.6.1 Green IoT

The rapid IoT growth accompanies further deterioration of the environment

resulted in high energy consumption, the accelerated exhaustion of traditional energy
resources, waste, the greenhouse effects, carbon dioxide (CO2) emissions, etc. Therefore,

15
there is a need to minimize the negative influence of IoT technology on the environment
and reduce the implementation resources of applications and to make sustainable IoT, a
green Internet of Things (G-IoT) is proposed [31]. The responsibility lies in the hands of
the system engineer and designer to design, program, and configure the IoT devices,
services, and applications in the most sustainable way [32]. In the following, we will
explain the critical points of green IoT shown in Figure 1.9.

 A wireless sensor network (WSN) consists of many sensor nodes with limited power,
powered by the battery. In some cases, a low power level may not be determined until
a very late stage. At this point, the node or the entire WSN crashes due to failure to
meet power requirements [33], because of high power consumption and the daily need
to replace the battery has made WSN challenging and costly to maintain. There is a
need to save energy consumption by using renewable energy mechanisms for charging
and utilization purposes and generating energy from environmental elements [34] such
as vibrations, kinetic, sun, etc.

 With growing application utilization of cloud, more data center (DC) needs to be
deployed, thus more power consumption, energy dissipation in the environment by
these DC, and carbon footprints [35]. Moreover, continuous use of processor chips
produces more heat, more heat requires cooling, and the cooling process produces heat
again. There is an important need to develop software solutions consuming less energy
with minimum resource utilization with high quality of service (QoS) [36]. Also,
design and manufacture devices that consume less energy with the same computing
speed.

 The machine-to-machine (M2M) technology involves a large number of machines

communicating and sharing information with us; the M2M communication paradigm
supports a variety of IoT applications, especially real-time monitoring applications.
However, the vast number of massive machines involved in M2M communications
will consume much energy [37].

 Today, the economy, energy, and environment gain much attention, with a particular
focus on keywords like sustainability and environment friendly for creating a greener
start for the future [38]. Renewable alternatives such as hydro, solar, and wind energy

16
 will gradually be the major resource of energy in the future [39], to meet growing
energy demands and mitigating climate change.

Figure 1.9. The green Internet of things.

1.6.2 Security

Security and privacy become more critical and challenging; they must be
considered seriously in IoT due to the quantity and the sensitivity of data created and
distributed through the network, accessibility to it not only by users but also by devices
and services [40]. Also, the computing and communication limits of IoT devices make
IoT more vulnerable to security and privacy attacks; this problem is further compounded
by the exponential growth of the number of connected devices. According to the HP
company report, nearly 70% of IoT devices are exposed to attacks and violations [41].
Numerous security threats exist on each layer in IoT architecture that must be addressed
appropriately to avoid causing damage. As shown in Figure 1.10, each IoT layer has its
security requirements that must be ensured at each level of IoT architecture to implement
a completely secure IoT system [42]. In the following, we will briefly explain the key
security threats of each IoT layer:

 The most devices in the perception layer are usually worked in outside environments,
which facilitates the tamping with hardware components, change the measurements,
shapes, or mechanical properties of these parts. The wireless network is the most used
in this layer, which makes it easily penetrated through disturbing waves [43]; thus, the

17
 attackers can control the devices and catch all the data, send malicious data, or even
compromising the confidentiality of information [42].

 The attacker can penetrate privacy and confidentiality in the network layer by
eavesdropping, surveillance, and traffic information analysis. Also, he can get
unapproved access to IoT devices, and modifying critical data values utilize it for
criminal goals [44]. The attacks may be external attacks by people from outside of the
network, they do not have access permission to IoT network or even internal attacks
by people who have permission and privileges to enter the network and they use it
illegally [42].

 The most security problems facing the middleware layer are in cloud computing due to
the data centralization, like loss of subtle data, data escape, cloning, intercept data-in-
motion, and moves data ownership [45]. Therefore, individuals and service providers
lose control over sensitive data. Also, breaching the security of any component in the
virtual infrastructure affects the security of other components and, thus, affects the
overall system security [46], [47], thus resulting in personal, and financial harm.

 Like the other layers, there are many security challenges related to the application
layer, especially access control and privacy challenges, where every application has
many users with varying degrees of access privileges. Also, various applications with
different authentication mechanisms, making it extremely difficult to merge them all
[48]. The privacy challenges are related to the data volume that will be uncovered,
who, and when to use it [49].

Figure 1.10. Security requirements of each IoT layer.

18
1.6.3 Data Storage and Processing

According to Statista research department, the total data volume of IoT devices
worldwide is projected to exceed 79.4 zettabytes by 2022 [61]. The huge increase of
connected devices number generates more data immediately that requests for more
storage and processing technologies, this makes the combination of cloud and IoT has
become inevitable to enhance the IoT ability of storage and processing with unlimited
capabilities and resources of the cloud [50]. IoT and cloud integration offers new
possibilities for the development of IoT applications. However, various challenges arise
during the storage, processing, and management of massive amounts of data in cloud
environments.

In IoT data storage, the key problems are the transferring and isolating of data
within the cloud platform. In most cloud platforms, different tenants share the same
computing resources for improving the utilization of cloud resources; this may cause a
difference in data authority and require data isolation in the cloud platform [51]. Another
issue, transferring massive data from edge devices to the cloud infrastructure causes
certain network performance problems such as delays, congestion, availability, etc. Also,
the costs of transferring data over the Internet and storing data on cloud servers [52].

The heterogeneity is one of the IoT characteristics. At the same time, it is a

challenge to integrate massive heterogeneous data from multiple devices with different
types of structured, semi-structured, and unstructured data that should be processed and
integrated to create a comprehensive and meaningful view for further utilities. However,
the oversimplification of data could lead to unfounded conclusions [53]. Traditional
sequential pattern algorithms can pose the challenge of scalability when working with big
data. On the other hand, the current parallel systems in the cloud are incapable of
effectively supporting complex parallel processing. Furthermore, high costs, and lack of
interaction ability in real-time processing [54].

19
1.6.4 Intelligence

While the loT system shows the ability to make decisions and interact smartly
between the devices, it does so based on specific business rules without considering
unintended environmental changes [55]. Also, things cannot make smart and quick
decisions to achieve their goals and the thing ability to communicate with other things
depends on service similarity and communication protocols [56]. As we mentioned
before, IoT devices produce massive data streams with varied data sources and types,
changes continuously. To tap the most out of this overwhelming these data, IoT needs
real-time management and analysis of data and create accurate insights from these data;
this will challenge the data management and analysis capability of IoT.

The use of autonomous technologies such as artificial intelligence (AI), makes

IoT more intelligent. From a technological point of view, AI is composed of a set of
technologies such as and deep learning (DL), machine learning (ML), natural language
processing (NLP). While IoT allows devices to interact via the Internet, AI lets them
learn from their data and experience. AI allows companies to analyze and wring insights
from data quickly, better interpret data, respond to events more rapidly in the case of a
risk, mishap or failure, and make decisions similar to how a human brain does the same
[57]. The use of AI they can avoid these failures in the first instance by using IoT-based
predictive analytics to predict which the conditions result in failure of the devices in an
automated manner without human intervention [58]. As a result, preventive action can be
taken in time, thus avoiding costly repairs and avoid setbacks in their business. More to
the point, increasing operational efficiency, save on unnecessary expenses, and making
huge profits.

The combination of IoT and AI technologies together build a new concept to IoT,
it is the artificial intelligence of things (AIoT) where IoT devices are the digital nervous
system and AI is the brain of the system. As shown in Figure 1.11, AI with the cloud and
big data technologies can make IoT applications realize their full potential when devices
can analyze data, make decisions, and act on that data without humans' interaction [59].
Soon, it will be rare to find an IoT application that does not use AI in any way.

20
 Figure 1.11. The requires technologies for Intelligent Internet of Things.

1.7 Conclusion

The Internet of things has changed the world; where it has introduced new
dimensions to information and communication technologies that enable the connection to
anything and interconnect the physical world and the virtual world. With the Internet of
Things expanding every day, soon everything will be linked to the Internet. Moreover, it
provided powerful applications and systems in different fields; that facilitated people's
daily lives and accelerated economic growth, including healthcare, industry, and
transportation, etc.

In order to understand the context of work, the presented chapter provides an

overview of the Internet of Things. An introduction presents the evolution of embedded
systems, information, and communication technology towards the new paradigm IoT
over several years. The different definitions and characteristics of IoT are covered. Then,
a detailed explanation of IoT architecture and its layers. Currently, there are numerous
and diverse IoT applications, permeating into practically all areas, we have selected the
most useful fields, that profoundly affect human lives. Since their unique characteristics
pose new problems, we have highlighted the most critical issues and challenges.

21
“I believe, if you zoom out into the future,
and you look back, and you ask the
question, ‘What was Apple’s greatest
contribution to mankind?’ it will be about
health.”
Tim Cook

22
Chapter 2

A Cloud-IoT System for Remote Patient Monitoring

2.1 Introduction

According to the Statista Research Department, by 2025, 75.44 billion IoT

devices will be connected [62]. The massive rise in the number of connected devices
instantly produces further data that requires more storage and processing technology.
Currently, IoT is one of the main big data sources. However, IoT devices have limited
storage and processing capacities that do not enable on-site data storage and processing;
this can impact IoT application development and scalability. Where the collection,
storage, and processing of data will generate new challenges regarding performance and
efficiency [6, 9]. For these reasons, special attention must be given to the storage,
analysis, and access of the vast amount of generated data. Especially in the healthcare
field, that requiring real-time analysis of vital patient data as a critical necessity to
identify abnormal conditions, avoid possible illnesses, and act as fast as possible to save
lives [63]. In terms of processing and storage capacity, the cloud has unlimited capacities.
Also, cloud computing provides ubiquitous access to storage and computing power [64].
The combination of IoT and cloud will provide an efficient solution for the storage and
analyses of data generated by IoT applications [65]. Considering all the above issues, this
chapter aims to propose a Cloud-IoT approach to improve IoT scalability. In order to
prove the feasibility of the proposed approach, a Cloud-IoT system for remote patient
monitoring is developed.
The remainder of the chapter is organized as follows: Section 2.2 presents and
discusses the related works. Section 2.3 presents the details of the proposed approach.
Section 2.4 presents the experimental results. Section 2.5 concludes this chapter.

23
2.2 Related Works

The rapid growth of IoT applications in different fields leads to create a large
amount of information. However, the IoT equipment has limited processing, and storage
capability didn’t permit field data processing or even storage. Thus, the collection,
acquisition, processing, sharing, searching operation for large amounts of data represents
a big challenge. The combination of IoT and cloud computing is desperately required to
take advantage of almost infinite cloud capacity and resources to overcome technological
constraints and to provide new services in a wide variety of real-life scenarios. However,
previous research focuses mostly on the individual development of a single technique.
Recently, several authors [66–78] have proposed health monitoring systems. In
[66], the authors developed an android application and IoT device for monitoring the
pulse rate and Body Mass Index (BMI). The authors of [67] developed a wearable device
for monitoring the blood oxygen saturation and heart rate of athletes. This device can
send the data to the coaches and doctors via General Packet Radio Service (GPRS) and
WiFi networks to provide them with the athlete’s condition. In another work [73], the
authors proposed a smartphone-based system for monitoring physical activity of patients
with chronic heart failure (CHF), then send the collected data to a central server. In [68]
project, the developed device can monitor a patient's pulse and body temperature, then
transfer the data to a smartphone application. The proposed system in [69] can monitor
electrocardiography (ECG), heartbeat, and temperature parameters. Then it sends results
to the server via the Global System for Mobile Communications (GSM), then the doctors
can observe the results on the website. Similar work to this work was proposed in [70]. In
[71], the authors developed a heart monitoring system that provides an interface between
the doctor and the patients for two-way communication. In [77], the authors proposed a
smartphone-integrated ECG monitoring system that monitors the ECG of users during
smartphone utilization. In [74], the authors developed a device that can measure the
physiological parameters of the human body. The results are displayed directly on the
Smartphone and used Bluetooth as wireless serial communication. In [78], the authors
developed a wearable device for monitoring the heart rate and temperature of the human
body. Then stores the results in a computer.

24
The proposed monitoring system in [72] uses wireless body sensor networks (WBSN) to
sense the blood pressure and heart rate. It consists of multiple wireless relay nodes that
transfer the data sent by the coordinator to the base station. In [75], the authors developed
a personal health monitor for psychophysiological evaluation. The proposed system
measures the heart rate variability through WBSN to quantify stress levels before and
during intense training sessions of military members. In [76], the authors developed a
health monitoring system based on a wearable device for monitoring the
electrocardiogram, respiration, and activity, then send all signals to the remote monitoring
center.

In the last few years, much research has been proposed for monitoring farms. In
[80], the authors proposed an agriculture monitoring system based on WSN for
monitoring the environmental factors and soil properties continuously such as moisture,
temperature, and leaf wetness. In [81], the authors developed an automatic agriculture
system based on a remote-controlled robot for doing agriculture tasks such as weeding,
moisture sensing, irrigation, spraying, bird and animal scarring.. In [84], the authors
proposed a farming system that measures the physical parameters such as soil moisture
content, pH of the soil and nutrient content, for irrigator control, the quantity of green
manure, and compost. The proposed system farmer to control the farming through the
mobile phone. In [85], the authors developed an automatic irrigation system for
controlling irrigation via the smartphone. The proposed system contains a portable device
that measures soil moisture, air humidity, and air temperature, embedded into the
irrigation system. For smart homes, several systems have been proposed for monitoring
and control homes. In [79] project, the authors developed a smart home monitoring
system for the elderly. They named (SEHMS) to help the elderly and individuals with
disabilities live independently in their homes. In [82], the authors proposed a home
lighting control system based on a smartphone. In [83], the authors designed a home
automation system that allows old aged people to monitor and control the home via the
android application.

Although several studies have developed monitoring systems in vital fields, little
attention has been given to storage and analysis data issues that require the utilization of
hug storage and processing resources. Knowing that IoT devices have limited storage and
25
processing capacity, that prevent on-site data processing and storage. Consequential
issues of performance, security, and privacy will therefore affect the future development
of applications. Also, scalability is a big challenge without sufficient infrastructure. Some
previous studies support remote monitoring by using GSM and GPRS technologies.
However, the problem with those approaches is that they are not providing real-time
monitoring, GSM can only use SMS to transmit data. GPRS can send and receive short
message service texts over the network, and it is data traffic speed is far greater than the
GSM mode. However, it is not in real-time, and it is slower compared to wifi.
Additionally, connecting to a mobile network consumes more battery life than connecting
the mobile devices to a wireless network. Those approaches may not be practical in all
situations, especially in critical cases that need real-time monitoring.
For these purposes, the present chapter aims to propose a Cloud-IoT approach for
real-time remote monitoring. In order to verify the validity of the proposed approach, we
developed a Cloud-IoT system for remote patient monitoring. It can be an effective way
to benefit from limitless cloud capacities in terms of storage and processing power and
improve IoT scalability and performance. Also, deliver real-life IoT-based services in a
distributed, dynamic, and responsive manner. Also, handle a wide range of connections
from millions of devices for enabling pervasive and ubiquitous computing scenarios.

Table 2.1 illustrates the differences between our work and the previously listed
works from the aspect of the application area, communication technology, mobile
application, support remote monitoring, and cloud. Also, it helps to identify the different
technologies used to address the same issues.

26
 Table 2.1. Comparison of IoT-based systems.
Support
Application Communicatio Mobile
Refernece Date remote Cloud
area n technology application
monitoring
[66] 2018 Healthcare Wi-Fi Yes No No
GPRS/ WiFi/
[67] 2015 Healthcare No Yes No
Zigbee
[68] 2017 Healthcare Wi-Fi Yes No No

[69] 2016 Healthcare GSM No Yes No

[70] 2016 Healthcare GSM No Yes No

[71] 2015 Healthcare WiFi Yes Yes No

[72] 2015 Healthcare WiFi No Yes No

[73] 2015 Healthcare / No Yes No

[74] 2019 Healthcare Bluetooth Yes No No

[75] 2003 Healthcare Bluetooth No No No

[76] 2005 Healthcare GPRS No Yes No

[77] 2016 Healthcare Bluetooth Yes No No

[78] 2012 Healthcare / No No No

[79] 2013 Smart home Wi-Fi No Yes No

Smart
[80] 2017 Wi-Fi Yes Yes No
agriculture
Smart
[81] 2016 Wi-Fi No Yes No
agriculture
[82] 2013 Smart home Bluetooth Yes No No

[83] 2013 Smart home Bluetooth Yes No No

Smart
[84] 2015 GSM No Yes No
agriculture
Smart
[85] 2014 Wi-Fi Yes Yes No
agriculture

27
2.3 System Architecture

Currently, millions of devices are connected now and are forecast to increase
dramatically in the coming years, this number of devices needs a massive storage
capacity. Also, the huge increase of connected devices number generates more data
immediately, which requests for more storage and processing technologies that are not
provided in IoT devices, this will generate many problems regarding performance and
efficiency. Thus, it will impact IoT application development and scalability. In contrast,
the cloud has unlimited resources and capabilities of processing and storage. Also, it
provides and manages services at any time and everywhere with minimal management
effort and expense. Moreover, it has many advantages, such as flexibility, high reliability,
virtualization, high automation, high efficiency, scalability, low cost, and fast service
providing.
For this purpose, we focus on the idea of integration between IoT and cloud
computing technology to enhance the IoT ability of processing and storage with
unlimited resources and capabilities of the cloud also improve IoT scalability, we propose
a new hybrid solution to address the increasing demands for data storage and processing.
As shown in Figure 2.1, the proposed system architecture is mainly composed of four
components.

Figure 2.1. The general architecture of the proposed system.

28
2.3.1 Devices

Sensors and actuators can be purely integrated with devices or even attached to
devices, machines, or items that have support for internet connectivity, in order to
observe the physical world, measure the different surroundings factors, such as
temperature, humidity, speed, etc., then transmit data over the internet or other networks.
IoT devices can also be managed and controlled remotely when required. There are two
possibilities, the data gathered from IoT devices will be sent first to the gateway then to
the cloud, or sent it directly to the cloud, this depending on IoT communication capability.

2.3.2 Gateway

The gateway works as a communication point between various IoT devices and
the cloud; it can be a smartphone, a router, a computer, etc. All data that moves to the
cloud or vice versa will pass through the gateway if the device cannot connect directly to
the Internet. Its key responsibilities are data collection and pre-processing, and uploading
only the necessary data to the cloud in order to reduce the network traffic.

2.3.3 Cloud

The cloud server receives data from the IoT gateway (or IoT devices) and stores
the data into the cloud database. The cloud providers are responsible for providing and
managing computational and storage resources. They are also responsible for the
accessibility to data from anywhere at any time and served data to several customers at a
time. Thanks to the cloud, the users can monitor and/or control IoT devices from
anywhere at any time via an application.

2.3.4 Application

The application can be a mobile application and/or web application. For ease of
remote management of monitoring systems and remotely control the IoT devices, we
have developed a mobile application updated in real-time and provides a visualization of

29
the sensor data to allow users to stay up-to-date 24/7 with the environment and device
status.

2.4 Implementation and Results

In order to demonstrate the efficiency of the proposed system, the implementation

of the proposed solution follows a series of steps to achieve a system capable of solving
the problems posed in this chapter. The following shows the development tools, the
configuration used to develop this system, the case study, system interfaces, and the
experimental results.

2.4.1 Development Tools

We have used a set of hardware and software development tools to build our IoT
healthcare solution.
 Android studio: is Google proprietary Integrated Development Environment (IDE)
for Android application development, launched in 2013. It is based on IntelliJ
powerful code editor and developer tools, free, available for Windows, Mac OSX,
and Linux. Android Studio offers powerful features such as a raft of tools and
Software Development Kit (SDK) for testing, debugging, and integrating with other
SDKs, Built-in support for Google Cloud Platfor, and flexible Gradle-based build
system [88].
 Firebase: is a set of hosting services for different types of applications. Launched in
2011 under the name Envolve, and was acquired by Google in October 2014.
Firebase offers useful features for developing high-quality applications and extending
the user base, such as real-time databases where firebase programs remain responsive
even when offline; when connectivity is restored, the client computer receives any
changes it has missed and automatically syncs with the new server status. Also, the
application can be run by millions of users at the same time without encountering any
bugs or losing responsiveness. Besides, it allows the developer the flexibility to
upload and download files directly from clients via the Firebase SDKs [89].

30
 Raspberry Pi: The Raspberry Pi 3 B + is a credit-card-sized computer that can
connect to a monitor, a keyboard, and a mouse assembly [90]. Ultra-compact
motherboard with 1.4 GHz ARM Cortex-A53 Quad-Core processor, 1 GB RAM,
HDMI, 4x USB 2.0, Gigabit Ethernet, Extended 40-pin GPIO header, 5V/2.5A DC
power input, and Micro SD port for storing data and loading the operating system, we
have installed Raspbian Stretch as an operating system. The strong point of this new
version of Raspberry Pi 3 is at the level of network connectivity. The Cypress
network chip of the B + model combines Bluetooth 4.2 and Wi-Fi Dual-Band b / g / n
/ ac for more speed and less latency.
 ADS1115: is a 4-channel analog-to-digital converter, low-power, 16-bit. The
Raspberry Pi is a digital-only computer; in other words, it cannot read analog inputs
and does not include a hardware analog-to-digital converter. For this purpose, we
have connected the ADS1115 between the Raspberry 3 Pi B+ and sensor to convert
the out signal of the sensor (analog signal) to a digital signal with high accuracy, as
shown in Figure 2.2.
 Pulse Sensor is a heart rate sensor integrated with an amplifier and noise cancellation
circuit. It is based on the principle of photoplethysmography to detect volumetric
variations of blood in peripheral circulation at the surface of the skin; by measuring
small differences in the intensity of light transmitted through or reflected from the
tissue. These variations in intensity are associated with changes in blood flow through
the tissue [91]. The sensor can clip onto a fingertip or earlobe for measuring heartbeat

Figure 2.2. The wiring diagram of heart rate measurement device.

31
2.4.2 Case Study

Nowadays, around the world, there are many people whose health could be
impaired by a lack of adequate health care monitoring, such as elderly patients with heart
disease, where cardiovascular diseases are the world's leading cause of death, 17.9
million deaths per year have been recorded [86]. Besides, patients with coronavirus
disease, that needs daily examination in quarantine. More to the point, the shortage of
healthcare workers compared to the global population, according to figures from the
World Health Organisation (WHO), more than 40 % of WHO Member States report
having less than 10 physicians per 10 000 population and more than 26 % report having
fewer than 3 [87]. The increase in the elderly population, the prevalence of chronic
diseases, and the spread of coronavirus disease have further exacerbated the situation.

Urgently, healthcare needs to be transformed from a hospital-centered structure to

a person-centered environment with a focus on patient wellbeing. Recent advances in IoT
technologies can improve healthcare systems. Also, the possibility of a remote
monitoring device will help the caregivers to have pre-diagnosis of the patient's condition
and early detection before the diseases develop into a crisis stage. For these purposes, we
developed a Cloud-IoT based system for patient heart rate monitoring. Healthy people
can also benefit from the proposed system to check their health and wellness. The
proposed system consists of four units, as shown in Figure 2.3.

Figure 2.3. A Cloud-IoT based system for patient heart rate monitoring.

32
 Patient unit: consists of a wearable device prototype showing in Figure 2.4 and a
mobile application. The device is responsible for sensing the patient heart rate and
automatically upload the sensing data to the cloud. The patient can easily measure
and monitor his heart rate anytime, anywhere. He needs to open the mobile
application, log in to his account, put his finger on the heart rate sensor, and stay still.
The device takes less than five milliseconds to get an accurate heart rate measurement.

Figure 2.4. The heart rate measurement device of the proposed system.
 Family unit: consists of a mobile application that supports multiple user profiles,
allowing each of the patient family members to have profiles to have their own
individual measurement, and easily monitor the heart rate of family members remotely
anytime, anywhere with a single app.

 Health providers unit: consists of a mobile application that allows the medical staff to
easily monitor patient heart rate remotely anytime, anywhere with a single app and in
real-time, see the patient medical record. Thus they can know the factors leading to the
patient's deterioration, get accurate disease diagnosis, take preventive actions
accordingly, identify the emergency condition for risky patients, and prepare for an
immediate medical response (suggest medicines, sent the ambulance, etc. )

 Cloud unit: Cloud stores the data inside the collection (myhealthapp), each patients'
medical record has its own document (user ID number). As shown in Figure 2.5, the
document contains key-value (birthyear, cityCountry, code, etc.). The cloud provides
patients to upload, store, and retrieve their medical data. Further, it allows health

33
 providers and patients' family members to have ubiquitous access to patients' medical
records away from hospitals and health clinics through their smartphones.

Figure 2.5. Cloud data structure of the proposed system.

2.4.3 System Interfaces

To benefit from the proposed system, we have developed a mobile phone

application that provides users with various features. The mobile application interfaces
are presented below. As shown in Figure 2.6, the first interface is the login interface,
which enables the user to log in by using his username and password.

Figure 2.6. The login interface of the mobile application.

34
 For security and control of the user information, only the registered users are
allowed to use our system; thus, login requires signup. In order to create a new account,
the user must insert his personal information, password, and code (the measurement
device identifier) into the second interface, as shown in Figure 2.7.

Figure 2.7. The signup interface of the mobile application.

As shown in Figure 2.8, the patient interface shows a selected portion of the
patient's personal information, current location, and real-time heart rate for allowing
patients, their family members, and health providers to keep track of important health
information.

35
 Figure 2.8. The patient profile interface of the mobile application.
The patient list interface supplies healthcare providers with easy access to patient
health information, remote and real-time heart rate tracking, as shown in Figure 2.9. Also,
it helps healthcare providers to understand the patient's health situations and needs. The
healthcare providers can click on the patient item in the list, then get more information in
the patient profile interface.

Figure 2.9. The patient list interface of the mobile application.

36
2.4.4 Results and Discussion

The effectiveness of the proposed system was validated by experiments in remote

heart rate monitoring. For a group of volunteers contain 20 people between the age of 5
and 75 years. As shown in Figure 2.10, all user data was stored in the cloud, where each
user's data was stored in a separate document.

Figure 2.10. Cloud data of the proposed system.

In order to ensure the accuracy of the readings of our device, we have compared
the heart rate reading of our device and the smartwatch Huawei gt 2, as shown in Figure
2.11.

Figure 2.11. The heart rate readings of devices.

37
 The results show that there is a good match between the two devices' readings. As
shown in Figure 2.12, the mean difference between them was (± 2) beats per minute
(bpm). This difference is due to the different pressure of the index finger on the sensor.

Figure 2.12. Comparison of heart rate readings between devices.

As the proposed system has the ability to monitor the patient’s heart rate in the
environment where the patient was located and allows healthcare providers to stay
informed 24/7 through the mobile application. It would be very important for healthcare
providers to understand and consider the differences that will impact the health status of
these patients. The measured parameter has been shown in digital form in the smartphone
application, making it easy for all users to understand in emergency and normal situations.
Since it can move the treatment process away from healthcare buildings and provide
patients with remote access to care services and support in emergency situations. It will
reduce the regulatory burden on healthcare buildings and solve the lack of beds problem.

2.5 Conclusion

In this chapter, a Cloud-IoT approach to improve IoT scalability is proposed. In

order to prove the feasibility of our approach, a Cloud-IoT system for remote heart rate
monitoring is developed. The effectiveness of the proposed system was validated by
experiments in remote heart rate monitoring with a voluntary group of 20 participants
between the ages of 5 and 75. The experimentation showed a good match between the
heart rate reading of our device and the smartwatch Huawei gt 2. The method is an
effective way to improve healthcare services since it can move the treatment process

38
away from healthcare buildings, thus will reduce the regulatory burden on healthcare
buildings and solve the lack of beds problem. Also, it can help healthcare providers to
understand and consider the differences that impact the health status of patients in
everyday life. On the basis of the promising findings presented in this chapter, work on
the remaining issues is continuing and will be presented in the next chapters.

39
“Despite continued security problems, the
IoT will spread and people will become
increasingly dependent on it. The cost of
breaches will be viewed like the toll taken
by car crashes, which have not persuaded
very many people not to drive.”
Richard Adler

40
Chapter 3

An Integrated Privacy Protection System for IoT

3.1 Introduction

IoT involves connecting a big number of smart devices and sensors embedded in
everyday objects which represent things. These things are often used to sense, collect and
transmit data through insecure public networks, making data more vulnerable to security
and privacy concerns. Also, IoT device's limits in terms of computing, memory, and
energy capacities make the adoption of the traditional methods for data protection
extremely difficult and can affect the system performance parameters such as speed, size,
and power consumption [92], especially in systems with strict time constraints such as
real-time monitoring system, where a few seconds of delay can lead to dangerous
consequences. Also, the exponential growth of connected devices and data volume
aggravates security problems further [93].

IoT-based healthcare systems deal with human beings' data collected from
wearable sensors, the wireless communications nature of these devices makes data more
vulnerable to security and privacy concerns [94]. Medical data stored may also be
accessed or stolen illegally. Which has a direct impact on the patient's privacy. Identity
management, access control, and data confidentiality are three key challenges in IoT
security [95]. In view of all the above problems, the purpose of this chapter is to propose
an integrated privacy protection system for IoT. The proposed system consists primarily
of two phases. The first phase ensures data confidentiality in the perception layer by
using ECC ElGamal to secure the IoT devices' communication. The second phase
controls the access to the data in the cloud through three steps: identification,
authentication, and authorization.

41
 The remainder of the chapter is organized as follows: Section 3.2 presents and
discusses the related works. Section 3.3 presents the details of the proposed security
system. Section 3.4 presents the experimental results. Section 3.5 concludes this chapter.

3.2 Related Works

Currently, IoT applications permeate all fields, directly affect our lives, and
collect a huge amount of sensitive data that is distributed across unsecured networks,
which increases the importance of data confidentiality, authentication, and access control
within the IoT system. However, IoT device limits in terms of storage, computation,
communication, and power make it more vulnerable to attacks. Moreover, it is hard to
implement traditional security methods in IoT systems, especially in the real-time
monitoring system, where a few seconds of delay can lead to dangerous consequences.

Over the years, researchers have proposed several security methods based on
encryption for embedded devices to preserve information security and protect it against
unauthorized access. Several authors [96-100] have proposed security models based on
symmetric encryption, where the same key is used for both encryption and decryption of
messages. In [96], the authors have proposed an improved preserving communication
protocol for the smart home systems. The proposed protocol is based on a symmetric
encryption scheme and Message Authentication Codes (MAC) to guarantee data security,
integrity, and authenticity. The authors of [97] have developed a security model to secure
the healthcare systems. The proposed system is based on symmetric cryptography using a
hash function with a key management system. In another work [98], the authors have
developed an encryption algorithm improved from Data Encryption Standard (DES) for
the healthcare system's privacy protection. In [99], the authors have proposed a secure
IoT-based healthcare system using BSN, which they named BSN-Care. The proposed
system is based on Offset codebook mode (OCB mode) for data security. The authors of
[100] have used the Advanced Encryption Standard (AES) to secure IoT camera's
captured images before sending them to the cloud.
Very few publications used homomorphic encryption in the internet of things,
which allows doing operations on encrypted data without decrypting it first and get the

42
result on the encrypted form. One of them is presented in [101]; the authors have
developed an IoT data security system which they named Talos. The proposed system is
based on partially homomorphic encryption. In [102], the authors have proposed a data-
sharing framework for the smart grid. The proposed system is based on homomorphic
encryption to protect consumer privacy in the analysis of electricity consumption reports.
In contrast, several research works [103-107] have been based on asymmetric
encryption in the internet of things. Asymmetric encryption uses pairs of keys, the public
key for information encryption and private key for information decryption. In [103], the
authors proposed a monitoring system that collects health data from sensors, then
transmitted it from gateways to back-end infrastructure. The proposed framework used
public-key encryption (PKI) to secure data transmission. The authors of [104] have
proposed a novel delegation-based mechanism, namely OEABE for Outsourcing
mechanism for the Encryption of Ciphertext-Policy Attribute-based Encryption (CP-
ABE), by delegating the most costly computations to a semi-trusted party in the CP-ABE
encryption process. In [105], the authors have developed a new version of Rivest-Shamir-
Adleman (RSA) to secure the information exchange between IoT devices, named the
Memory Efficient Multi Key (MEMK) generation scheme. In another work [106], the
authors have implemented the RSA algorithm to ensure network layer security in the
smart city. In a recent paper [107], the authors have proposed a secure outsourcing
scheme for RSA decryption in IoT based on the Chinese remainder theorem (CRT) to
concurrently conceal the private keys and the plaintext.
The two key limitations of [96-100] research are that security keys would be lost
if the key sharing is insecure. Thus, both sides of the communications get compromised,
and all the encrypted data can be easily decrypted. It requires a protected channel for
secret key exchange before starting data transportation through an unsecured channel.
This approach may not be practical in IoT because if exists a secure channel to share the
key, so no need to be using encryption in the first place. Additionally, it necessitates a
large number of keys; each pair of participants wishing to share encrypted messages
needs a new key, which makes the security and management of the key very challenging.
The problem with the [101, 102] approaches is that it is the ciphertexts are much
bigger than the plaintexts. So the processing of these big ciphertexts is usually slower and

43
computationally expensive than the computation of the plaintext itself, thus making it
impractical for most IoT applications. The major drawbacks of the [103-107] approaches
are relatively complicated and computationally intensive, which takes more time and
requires more computation resources. Asymmetric cryptography may not be practical in
IoT devices with limited processing and storage capabilities.
Although all studies have studied data confidentiality, little attention has been
given to real-time encryption. The speed of the cryptography algorithm is very important
for monitoring systems. Since it directly affects the performance of the system.
Especially in systems with strict time constraints, such as real-time monitoring systems,
where a few seconds of delay can lead to dangerous consequences. Also, all the previous
studies do not take into account users' authentication and authorization. Access control in
data security is crucial to reduce the chance of data theft and ensure data privacy. Access
controls managing user access privileges, authenticate and authorize users to access the
data they are allowed to see and use based on their position in the system, and keeps
unauthorized users from accessing sensitive information.
For these reasons, the present chapter proposes an integrated privacy protection
system for IoT to provide a security solution at the same security level as traditional
cryptography. The present work based on Lightweight Cryptography that can work over
IoT devices constrained devices due to the limitation of their resources. For more data
security, we have implemented an access control system to ensure data privacy. Table 3.1
summarizes the differences between our work and the previous security solutions
implemented in the Internet of things, from the aspect of the encryption algorithms name
and type, the key security features, and system performance.

44
 Table 3.1. Comparison of security solutions in IoT systems.

Encryption
Ref Date Type Confidentiality Access control Real-time
algorithms

Symmetric
[96] 2017 AES Yes No /
cryptography

Symmetric
[97] 2014 / Yes No No
cryptography

Improved Symmetric
[98] 2015 Yes No Yes
DES cryptography

Symmetric
[99] 2015 OCB Yes No /
cryptography

Symmetric
[100] 2016 AES Yes No /
cryptography

Homomorphic
[101] 2015 / Yes No No
encryption

Homomorphic
[102] 2016 / Yes No No
encryption

Asymmetric
[103] 2012 / Yes No No
cryptography

Asymmetric
[104] 2018 CP-ABE Yes No /
encryption

Asymmetric
[105] 2017 MEMK Yes No /
encryption

Asymmetric
[106] 2019 RSA Yes No No
encryption

Asymmetric
[107] 2020 RSA Yes No No
encryption

45
3.3 Security System

Currently, IoT security is a big point of discussion; one of the key issues in IoT
security is data confidentiality. However, IoT devices are unable to accommodate
conventional cryptography algorithms due to IoT devices' limits in terms of storage,
computation, communication, and power. In addition to that, the majority of IoT devices
use wireless technologies in communication, which makes them more vulnerable to
attacks such as disturbing waves and ultimately leads to steal sensitive data. Another
issue facing IoT is information access control. With the increasing number of Internet of
Things users and the amount of confidential data, this issue gets more complicated. The
users' identity must be verified, and the access level granted to each user must be
determined. The data access control refers to the authentication and authorization to
prevent unauthorized users from accessing sensitive information. For these reasons, we
proposed an integrated privacy protection system for IoT. The proposed system consists
primarily of two phases:

3.3.1 Data Confidentiality

Data confidentiality is critical in IoT systems, where it has become a major

concern for IoT users. The best way to ensure data confidentiality is cryptographic
algorithms; they are commonly used to encrypt data in order to allow devices to
communicate safely even over an unsafe medium. Even if the transmitting medium has
been hacked, the encrypted information is virtually worthless to unauthorized individuals
without the correct decryption keys. Thus, the hacker would not be able to reach the
contents. However, it is not easy to enforce adequate cryptographic functions on IoT
devices due to their limited capabilities. In contrast to conventional cryptographic
algorithms, the Curves Cryptosystem ElGamal algorithm (ECC ElGamal) has very small
keys at the same security level. These advantages make ECC ElGamal more suitable to
be used in IoT devices' communication security. For those reasons, we have adopted ECC
ElGamal to secure the IoT devices' communication.

For system initialization, let an elliptic curve group Ep (a,b), using the elliptic
curve over a finite field (p) where P is a large prime, α is a primitive root of P as

46
described in Algorithm 1, such that satisfy the equation (3.1) for 0 < x < P, a and b
determine the curve shape and satisfy the equation (3.2) P and α are public numbers.

y2 = x3 + a x + b (3.1)

4 a3 + 27 b2 ≠ 0 (3.2)

Algorithm 1 Initialization
1: P = large prime ();
2: α = Base (); // Base point
3: ECC_2_Equation (a, b); // (3.2)
End
In order to establish secure communications between device and gateway (D, G),
device as a sender and gateway as a reserve, D creates a key pair {d, βd} as (3.3) where d
is the private key and βd is the public key of D device as described in Algorithm 2. The
same thing for G (3.4).

βd = d α (3.3)

βg = g α (3.4)

Algorithm 2 Key Generation of device

1: d = random () ; // private key
2: βd = d * α ; // public key
3: k = random () ; // secret number
End
When the device wants to send the information i to the gateway, the device
converts the information i to a point I on the elliptic curve E by using a mapping function
map(). Also, choose a random integer k, and keep it secret, calculate the pair point {Y1,
Y2}. Then, mapping the pair point {Y1, Y2} into the pair point {Y1m, Y2m} by using the
mapping function map() and sends them to the gateway as shown in Figure 3.1.

Y1 = k α (3.5)

Y2 = I + k βg (3.6)

47
 D G
{d, βd} {g, βg}
Y1 = k α
I←i
Y2 = I + k βg
Y1m ← Y1
Y2m ← Y2
{Y1m, Y2m}

i = Y2m - g Y1m
Figure 3.1. Interaction diagram of the ECC ElGamal cryptosystem.

ECC ElGamal algorithm can only encrypt and decrypt a point on the curve, not
the information. Based on that we have encrypted the information characters one by one
by mapping the character to a point on the curve where each character of the information
represents a point in ECC ElGamal. Then encrypt it and map it again to another point as
illustrated in Algorithm 3.

Algorithm 3 Encryption
Input: bytes[] and point βg
Output: Pair of point Y1, Y2
1: var s : integer; s = size (bytes []);
2: var result : array [s * 2] of byte;
3: var p : array [s] of point;
4: var j, i : integer; j = 0;
5: For i = 0 to s do
6: p [i] = map (bytes [i]);
7: Y1 = k * α; Y2 = p [i] + (k * βg); // encryption of point p [i]
8: result [j] = map (Y1);
9: j = j + 1;
10: result [j] = map (Y2);
11: j = j + 1;
12: End

48
 13: Return result;
14:End
When the gateway receives the ciphertext, it uses his private key g to decrypt the
information I point by point, then recovers the readable information as shown in
algorithm 4.

i = Y2 - g Y1 (3.7)
Algorithm 4 Decryption
Input: bytes[] and g;
Output: result array [] of byte;
1: var s : integer; s = size(bytes []);
2: var result : array [s / 2] of byte;
3: var j, i : integer; j = 0;
4: For i = 0 to size (result []) do
5: j = j + 2;
6: // Decryption by point
7: result [i] = bytes [j + 1] - (bytes [j] * g )
8: End
9: Return result;
10:End

3.3.2 Access Control

Access control on IoT cloud systems is crucial to ensure the privacy of health
information and personally identifiable information of users, also limiting information
access to only authorized people. The access control systems can avoid unauthorized
access in the cloud data, reduce chances of the privacy breach and data leaks occurring.
To provide a high-security level, we have developed an access control system that can
authenticate and authorize users to access the health data they are allowed to see and use
according to their roles. As shown in algorithm 4, the proposed system has three steps:
identification, authentication, and authorization.

49
 First, we verify the users' identities that request access to a system. The user must
have an identity in the system so that the authentication method can interrogate the
account. The authentication process has two steps and if one of these steps fails, the
access will be denied. First, the mobile application will prompt the user to enter a
username and password that will be submitted to the cloud server to check whether the
username is already in the users' list. Once the provided password matches the user's
password stored in the cloud, the authorization process begins.

In the authorization process, the system will verify whether or not the user is
authorized to access data through verifying the access permissions for this user that
determined by the metadata concerning the user’s account. Once a user is given the
authorization to access a specific part of data on the cloud, he can perform a set of actions
according to his role. The proposed system has different levels of access control,
healthcare providers have access to all patient information, the patient can monitor his
heart rate, and it can also monitor their family members' heart rate if they invited him.
Algorithm 5 Access Control
Input: identification, password, and query;
Output: query result;
1: The user provides a username (UN) and password (PW) to the mobile applicatio;
2: The mobile application will pass the UN and PW to cloud servers;
3: Cloud servers research the username;
4: If exists and PW matches the user's password stored do
5: If the user is a healthcare provider do
6: Show all health information of patients;
7: Else
8: Show only the health information of the user
Show list of family members who invited him
9: Else
10: Denied the access and shows an error message
11: Return query result;
12:End

50
3.4 Experimental Evaluations

In this section, we demonstrate the efficiency of our system by comparing it to

other systems in terms of computational and storage costs. The experimental results ​ are
presented in charts.

3.4.1 Experimental Setup

For experimental evaluation, we have implemented the proposed system in

raspberry pi 3 B +, with 1 GB RAM, 1.4 GHz ARM Cortex-A53 Quad-Core processor,
micro SD 32 GB, and Raspbian Stretch as an operating system. In order to test the
performance of our ECC ElGamal implementation, we also implemented public key
algorithms ElGamal and RSA, with keys' long based on the recommendation of the
National Institute of Standards and Technology (NIST) [108], and we measured the
encryption and decryption time of the same message with 160 bits long on three different
security levels.

3.4.2 Results and Discussion

The bar chart compares the key size between the three algorithms RSA, ElGamal,
and our implementation of ECC ElGamal in three security levels. Overall, both
asymmetric algorithms RSA and ElGamal have long key size, while ECC ElGamal key
size is shorter than other algorithms in all security levels as shown in Figure 3.2. RSA
and ElGamal key size was 1024 bits long in the first security level, being longer than
ECC ElGamal's key size by approximately 864 bits. Then, it rose to 2048 bits in the
second security level and continued to increase steadily to 3072 bits in the last security
level. However, ECC ElGamal's key size was 160 bits in the first security level. The bar
chart showed a gradual increase to 224 bits in the second security level and continued to
increase steadily to 256 bits, being shorter than other algorithms' key size by
approximately 2816 bits in the last security level.

51
 Figure 3.2. Comparative analysis of algorithms' key size.
As shown in Figure 3.3, the bar chart shows the encryption time taken to encrypt
the same message by the three algorithms in three different security levels. Overall, both
asymmetric algorithms RSA and ElGamal take longer, while ECC ElGamal takes much
less time than other algorithms in all security levels. In the first security level, the
encryption time taken by the RSA and ElGamal was 0,231 seconds and 0,145 seconds
respectively. Then, RSA and ElGamal's encryption time continued to increase sharply
throughout the security levels. Where RSA rose significantly to 4,78 seconds, being
longer than ElGamal's encryption time by approximately 1,833 seconds in the last
security level. However, ECC ElGamal's encryption time was 0,011 seconds in the first
security level. Then, it gradually increased to 0,017 seconds in the last security level,
being shorter than RSA and ElGamal's encryption time by approximately 4,77 seconds
and 2,93 seconds respectively.
The bar chart illustrates the decryption time taken to decrypt the same message by
RSA, ElGamal, and our implementation of ECC ElGamal algorithms in three different
security levels, as shown in Figure 3.4. Units are measured in seconds. Overall, the
decryption time of RSA and ElGamal algorithms was higher than that of ECC ElGamal.
In the first security level, the decryption time taken by the RSA and ElGamal algorithms
was 0,483 seconds and 0,203 seconds respectively. Then, RSA and ElGamal's decryption

52
time continued to increase sharply throughout the security levels. Where RSA rose
significantly to 7,626 seconds, being longer than ElGamal's decryption time by
approximately 2,467 seconds, in the last security level. However, ECC ElGamal's
decryption time was 0,003 seconds in the first security level. Then, it gradually increased
to 0,006 seconds in the last security level, being shorter than RSA and ElGamal's
decryption time by approximately 7,620 seconds and 7,153 seconds respectively.

Figure 3.3. Comparative analysis of algorithms’ encryption time.

Figure 3.4. Comparative analysis of algorithms’ decryption time.

53
 In contrast to RSA and ElGamal algorithms, ECC ElGamal offers high security
using less computing power, and memory, resulting in much faster response times and
better performance. This benefit would be very important for IoT devices with limited
processing power, memory, and energy capabilities. ECC ElGamal can be the next-
generation implementation of public key cryptography, and the best alternative for IoT
devices.

3.5 Conclusion

In this chapter, an integrated privacy protection system for IoT is proposed. In

order to ensure the user's data privacy, we have encrypted the data by ECC ElGamal to
secure the IoT devices' communication in the perception layer. We have also developed
an access control system for the data in the cloud. In order to verify the efficiency of the
proposed system, we have compared our implementation of ECC ElGamal with other
systems in terms of computational and storage costs in three different security levels. The
experimentation shows that ECC ElGamal was more suitable to be used in IoT devices'
communication security, where ECC ElGamal takes much less time than other algorithms
in all security levels.

54
“Without intelligence there is no value.”
Kiva Allgood

55
Chapter 4

A Cloud-IoT System Based on Smart Agent

4.1 Introduction

Currently, one of the most key problems that facing IoT is heterogeneous device
management. While the loT system shows capable of making decisions and interacting
intelligently with its devices, the real situation is that these devices are still not smart
enough to make quick and smart decisions in dangerous situations [109], [110]. It works
so based on specific business rules and does not take into account unexpected
environmental changes. Furthermore, the ability of devices to communicate with other
devices is dependent on service similarity and contact protocols. When IoT devices can
process data, make decisions, and act on that data without requiring human intervention,
IoT apps can reach their full potential [111], [112]. In view of all the above problems, an
integrated system that combines IoT and intelligent agent technology to enhance IoT
intelligence has been proposed. Specially, we have proposed a Cloud-IoT system based
on smart agents. The proposed system consists of a group of agents distributed in the four
IoT architecture layers; a group of agents collaborates to collect, process, and store data
in the cloud. The remaining agents are deployed on IoT devices.
The remainder of the chapter is organized as follows: Section 4.2 introduces the
different definitions, classifications, and characteristics of agents, as well as the multi-
agent system. Section 4.3 presents and discusses the related works. Section 4.4 presents
the details of the proposed approach. Section 4.5 presents the experimental results.
Section 4.6 concludes this chapter.

56
4.2 Background

In this section, we present a brief background of the agent and the multi-agent
system technologies.

4.2.1 Definition of Agent

Although the term "agent" is commonly used, it has numerous contradictory

definitions. The more detailed definition of the agent is given by Jacques Ferber where it
combines the agent's functioning and its capacities [113]:

"An agent can be a physical or virtual entity that has the capacity to:

 Act in an environment,
 Communicate directly with other agents,
 Be equipped with a set of trends (in the form of individual objectives or satisfaction or
even survival function, which it seeks to optimize),
 Has its own resources,
 Perceive (but in a limited way) his environment,
 Only have a partial representation of this environment (and possibly none),
 Have skills and offer services,
 To reproduce, possibly,
 Behave which tends to meet its objectives, taking into account the resources and skills
at its disposal, and according to its perception, its representations, and the
communications it receives."

According to Jennings, Sycara, and Wooldridge [114]: "An agent is a computer

system, situated in some environment, that is capable of flexible autonomous action in
order to meet its design objectives."

According to these definitions, an agent is a physical entity acts in an environment,

endowed with sensors and shareholders; it decides for itself what it must do to achieve its
objectives (determined at its design), considering the capabilities at its disposal.

57
4.2.2 Characteristics of Agent

The agent community defines a set of agent characteristics. In fact, they are
frequently true, but not always the case [115].

 Autonomy: is the most interesting characteristic of the agent in which he can

independently take the appropriate decision without the intervention of other agents or
even human intervention. Also, execute the necessary action, and control its' own
internal state [117].
 Flexibility: A flexible agent must be [116]:
o Reactive: In this case, the agent is able to perceive the environment in which he
has situated, which may be an agent container, an application, a physical device,
and Internet, then react to changes in time.
o Proactive: The agent's behavior is determined by his goals, and therefore he can
produce goal-directed actions, as well as take the necessary initiative, not just
responses to his environment.
o Social: An agent must have the ability to interact with external sources (other
agents, system or even humans) when he needs to ask for help, share/obtain
knowledge, and negotiate in order to meet his goals.
 Situation: The agent is located in an environment; if he can perceive his environment
thanks to the sensory inputs he receives from this same environment and modify his
behavior accordingly at all levels (objective, plan, action, etc.) [116].
 Reasoning: An agent can deduce and create knowledge based on the information he
has collected. Then, he can decide which goal to pursue, [118] which event to react to,
how to act to accomplish a goal, or even suspend or abandon others [116].
 Adaptive: an agent has this property if he can use his knowledge to modify his
behavior and automatically adapts to changes. Also, he can control his skills according
to the agent or the environment with which he interacts [115].

58
4.2.3 Classification of Agent

According to Ferber Jacques, there are two main types of agents [119]. This
classification is based essentially on the agent's decision-making process and the
representation of the environment at his disposal.

 Cognitive agent: endowed with reasoning and planning skills and communication
skills, can solve complicated problems in a relatively individual way [119], thanks to
his knowledge base, which comprises all the information and knowledge necessary for
carrying out his task and managing interactions with other agents and with his
environment [120].
 Reactive agent: can perceive the environment and only reacts to changes in the
environment [119]. In other words, such an agent does not do any deliberation,
knowledge, or planning. Reactive agent decision-making based on Stimulus/Response;
he simply acquires perceptions and reacts to them. Since there is no reasoning, these
types of agents can act and react very quickly [121].

The combination of the two previous models of agents results in a new hybrid agent
that combines reactive and cognitive abilities, allowing them to adapt their behavior in
real-time to changing environments. In this model, the agent has a multi-layered
architecture, each layer being either a cognitive component or a reactive component [122].

4.2.4 Definition of Multi Agent System

There are several attempts to define multi-agent systems (MAS). One of the most
common definitions is proposed by Jacques Ferber [113]:

"We call a multi-agent system (or SMA), a system composed of the following elements:

 An environment E, i.e., space generally having a metric.

 A set of objects O. These objects are situated, i.e., for any object, it is possible, at a
given moment, to associate a position in E. These objects are passive, i.e., they can be
perceived, created, destroyed, and modified by agents.
 A set of agents A, which are particular objects (A includes O), which represent the
active entities of the system.
59
 A set of relations R which unite objects (and therefore agents) between them.
 A set of operations Op allowing agents of A to perceive, produce, consume, transform
and manipulate objects of O.
 Operators responsible for representing the application of these operations and the
world's reaction to this attempted modification, which will be called the laws of the
universe. "

4.2.5 Characteristics of Multi Agent System

The MASs are used efficiently in several fields, thanks to these characteristics. In
the following, we listed the most important characteristics of MAS.

 Distribution: The system is decomposable, with the agent as the primary component.
The system's agents may be located in different locations [123].
 Leadership: there are two types [124]:
o Leader-following or centralized MAS: has an agent as a leader; he establishes
goals and tasks for the other agents based on a single global goal.
o Leaderless or decentralized MAS: the agents are independent, and there are no
centralized decisions valid for the whole system.
 Heterogeneity: MAS can be classified into two categories based on agent
heterogeneity [124]:
o Homogeneous MAS refer to agents with similar characteristics and functionalities,
o Heterogeneous MAS refer to agents with a variety of characteristics and
functionalities.
 Topology: based on the agent's location and relations can be classified into two
categories [124]:
o Static topology: an agent's location and relationships remain the same throughout
his lifetime.
o Dynamic topology: in this topology, the agent's location and relationships change;
an agent can move, exit, or join the MAS; or even forms new communications
with other agents.

60
4.3 Related Works

Currently, IoT represents a revolutionary concept, rich in potential as well as in

development problems such as decentralization, intelligence, and scalability. Very few
publications have been based on the decentralized MAS. Agents in the decentralized
MAS can work independently and collaborate with each other to accomplish the global
goal [133]. In [125], the authors have proposed an Agent-based architecture for smart
objects-oriented Cloud-IoT systems, which they named CA-IoT. The proposed
framework consists of a group of agents embedded in a cyber-physical world, combined
with a cloud, and interfaced with external IT systems. The authors of [132] have
proposed a framework for monitoring the civil infrastructure based on agents for remote
access and autonomously process collected information. In [135], the authors proposed
an agent-based traffic administration system for traffic monitoring and managing.
In contrast, several research works [103-107] have been based on the centralized
MAS as shown in Table 4.1, where a single agent acts as a central agent in centralized
MAS, determining the system's state depending on the information available from all
other agents. The central agent makes an optimal decision based on the information
obtained [133]. The authors of [126] have developed an agent-based health care
monitoring system. The proposed system can sense vital signs of the patient, such as
body temperature, blood pressure, heart rate, and respiratory rate by using WSNs. In a
recent paper [127], the authors have developed a multi-agent-based mobile healthcare
system for enhancing interaction among patients, caretakers, doctors, and other
healthcare providers. The authors of [128] have designed a distributed information
platform based on mobile multi-agent for support the intensive and distributed nature of
wide-area monitoring environment, named MADIP. MADIP is made up of two groups of
agents: static agents and mobile agents.
In [129], the authors have proposed an activity monitoring system for promoting
the elders' healthy lifestyles and diagnose psychomotor abnormalities early. In [130], the
authors proposed a mobile gateway solution based on agents for mobile health scenarios.
This gateway autonomously collects patient's information. Then it forwards the collected
data to the healthcare providers. The authors of [131] have proposed an agent-based

61
architecture for mobile health monitoring, which includes a set of intelligent agents who
gather patient data, reason together, and suggest actions to patients and medical staff. In
[134], the authors develop an automatic diagnosis engine based on mobile agents to
support intensive and distributed ubiquitous fetal surveillance without interruption with
the normal routines of the patient and without limiting her mobility.
The major drawback of the [126-134] approaches is it based on centralized MAS.
In centralized MAS, a single agent performs all the major processing, has complete
control over all aspects of the system, and makes decisions based on analyses of
information gathered by other agents. This approach may not be practical in all situations,
it is computationally intensive, less flexible in terms of analysis and decision making.
Also, it is more vulnerable to failure in case of the central agent goes down/dead for some
reason and the other agents are unable to handle requests; it would most likely bring the
whole IoT system down with it.
In contrast, the [125], [132], and [135] approaches are based on decentralized
MAS, where each agent within the system acts as an autonomous entity with decision-
making authority. Another significant benefit of decentralized MAS is efficiency; there is
no single point of failure since each agent acts as an individual system. If one agent fails,
the remaining agents will continue to function.
However, most of the previous studies do not take into account parallel
computation. The system must be able to keep up with multiple customer demands in
order to have an effective IoT system. The agent's parallel behavior allows the execution
of simultaneously multiple actions, unlike serial behavior. It can reduce runtime, speeds
up a system's operation, and increase operational efficiency.
For these reasons, the present chapter proposes a Cloud-IoT system based on
smart agents for the health monitoring of cardiovascular patients. This approach
conforms to the approaches proposed in the previous chapters. The proposed system can
collect, retrieve, store, and analyze the heart rate of the patient in a secure way. It can also
detect the critical case immediately and take a quick and smart decision. It can be used
not only by the hospital but also by any health care provider.

62
 Table 4.1 summarizes the differences between our work and the previous agent-
based solutions implemented in the Internet of things, from the aspect of the agent
behavior and MAS leadership characteristic.
Table 4.1. Comparison of agent-based solutions in IoT systems.

Agent Behavior
Ref Date Leadership
Parallel Autonomy

[125] 2014 No Yes Decentralized

[132] 2012 No Yes Decentralized

[126] 2013 No No Centralized

[127] 2019 No No Centralized

[128] 2011 No No Centralized

[129] 2012 No No Centralized

[130] 2016 No / Centralized

[131] 2008 No No Centralized

[134] 2014 No / Centralized

[135] 2015 No Yes Decentralized

63
4.4 Proposed Approach

For dealing with a dynamic system like IoT, a multi agent system is good
solutions, to enhance the IoT system intelligence, encapsulate the complex functionalities,
underlying heterogeneous hardware, and implementation details, maximizing
interoperability among heterogeneous resources, increasing scalability. For this purpose,
a Cloud-IoT system based on smart agents is proposed, and we named it A-IoT. In the
following, we will present the details of the proposed system.

4.4.1 System Architecture

The main idea of this approach is to enhance the IoT system's intelligence by
adding a vertical layer of agent to the IoT architecture. The agent layer shown in Figure
4.1., is a set of agents distributed in the four horizontal IoT architecture layers; a group of
agents collaborates to collect, process, and store data in the cloud. The remaining agents
are deployed on IoT devices. The agents can directly control the hosting device and its
physical elements. It can also communicate and collaborate with other agents to complete
their acts and objectives. It also reacts intelligently with complex scenarios without
requiring active human interaction. In the following, we will explain in detail the
different agents' roles.

User Agent
Manager Agent
Analyzer Agent
Gateway Agent
Device Agent

Figure 4.1. System global A-IoT architecture [136].

64
 Device Agent (DA): an embedded agent in each IoT device, as shown in Figure 4.2,
this agent is composed of three modules: communication module with other agents, a
rule base that includes all rules of this agent, and the reasoning module. Device agent
responsibilities are as follows:
o Environment perception, save the perception data, which will be encrypted before
being sent to the gateway agent.
o Control of IoT device and adjust the device state based on the user and
environment needs.
o Communicate with other agents.

Figure 4.2. The architecture of device agent.

 Gateway Agent (GA): an embedded agent in the IoT gateway, as shown in Figure 4.3,
this agent is composed of the following modules: two communication modules, the
Wide Area Network (WAN) communication module to communicate with the
manager agent in the cloud, and Personal Area Network (PAN) and local area network
(LAN) communication module for the communication with the device agents. Security
module handles data flowing in two directions from/to devices agent, in a secure way.
Treatment module for the treatment of sensing data before sending it to the cloud for
reducing the amount of data that must be forward to the cloud to speed up the reply
times. This agent responsibilities are as follows:

65
 o Aggregate the sensing data from device agents, decryption/encryption of the
sensing data, and pre-analysis them.
o Save data for a short time in case of network failure.
o Sent the sensing data to the manager agent in the cloud.
o Communicate with other agents.

Manager Agent

WAN communication module

Security Treatment
module module

LAN/PAN communication module

Device Agents

Figure 4.3. The architecture of gateway agent.

 Manager Agent (MA): an included agent in the cloud, as shown in Figure 4.4, this
agent is composed of the following modules: communication module to communicate
with the other agent, treatment module for the data treatment before saving it in the
cloud database, management module for data management. This agent responsibilities
are as follows:
o Data collection, storing, recovery, and processing.
o Communication with other agents.

Management
module
Other Communication
Agents module Data base
Security
module

Figure 4.4. The architecture of the manager agent.

66
 Analyzer agent (AA): an included agent in the cloud, as shown in Figure 4.5, this
agent is composed of the following modules: communication module to communicate
with the other agent, analyzes module for analyzes of user's data, and security module
for verification of user's privilege. This agent responsibilities are as follows:
o User’s authentication and authorization verification.
o Analyzes the user request, then sends it to the manager agent for request execution,
and returns the result to the user agent.
o Communication with other agents.

Analyze
module
Other Communication
Agent module Data base
s
Security
module

Figure 4.5. The architecture of the analyzer agent.

 User agent (UA): an embedded agent on the user device, as shown in Figure 4.6, this
agent is composed of the following modules: communication module to communicate
with the analyzer agent, interface module to provide a graphical interface to interact
with our system and treatment module for user requests treatment. This agent
responsibilities are as follows:
o Acquire user requests.
o Send those requests to the analyzer agent.
o Present the results to the users.
o Communicate with analyzer agents.

Figure 4.6. The architecture of the user agent.

67
4.4.2 Agents Interactions

A shown in Figure 4.7., the diagram contains three alternative that shows the
different agents and their interactions in the proposed scheme, as well as the chronology
of these interactions; we will explain it in depth below:

The user sends the login request to a user agent, who then sends the user login
information (user Id and password) to the analyzer agent. After that, the analyzer agent
returns the response to the user agent. As shown in the authentication alternative, the
request will be passed only if the access information is correct, at which point the user
agent will load the monitoring interface of the application; otherwise, the request will be
abandoned.

After successfully signing in, if the user wants to know critical information or do
an action that needs authorization, the user agent then sends the authorization request to
the analyzer agent, which contains the user Id and query type. Then, the analyzer agent
checks if the user is authorized or not, as shown in the authorization alternative; if the
user is authorized, the analyzer agent sends the request to the manager agent; otherwise,
the request will be ignored.

As shown in the request alternative, if the request type is get state, the manager
agent sends a request to the gateway agent containing the device Id in order to obtain the
device state. Otherwise, the manager agent sends a request to the gateway agent
containing the system Id and the new state to change the device state. In all instances, the
gateway agent encrypts the request before sending it to the device agent, and the device
agent receives and decrypts the encrypted request. The device agent will then execute the
request and transmit the device state to the gateway agent. The gateway agent receives
the decrypted device state and sends it to the manager agent. The manager agent then
collects, stores, processes, and passing the device state over to the analyzer agent. The
analyzer agent collects the device state and transfers it to the user agent to display the
result in the user application.

68
Figure 4.7. General diagram of interactions between agents [137].

69
4.5 Implementation and Results

In this section, we present the implementation and experimental process used to

demonstrate the efficacy of the proposed system, including development tools, the case
study, and the experimental results.

4.5.1 Development Tools

In order to ensure rapid and efficient development, proper functioning of our

system based on agents, we have used two agent frameworks, osBrain for the device and
Java Agent DEvelopment Framework (JADE) for the cloud and the mobile application.

 JADE: is free software distributed by Telecom Italia that was fully implemented with
the JAVA programming language by the TILAB laboratory, intended for developers
who want to build multi-agent systems [138]. A JADE-based architecture can be
distributed across devices that do not even need to use the same operating system. The
configuration can be controlled remotely through a graphical user interface (GUI), and
when the agents are running, it can be changed by moving them from one device to
another if required. In addition to the agent abstraction, JADE provides a simple but
effective task execution and composition model. Also, JADE provides peer-to-peer
agent communication based on the asynchronous message passing paradigm [139].
We have used JADE version 4.5.0. In addition to JADE, we used JADE-Android to
develop agents in mobile applications, which is formed by the combination of the
standard version of JADE and the android plug-in, intended for developers who are
interested in creating JADE-based applications running on the Android Operating
System [140].
 OsBrain: is free software Framework creation and execution agent that handles MAS
in distributed environments such as IoT systems, fully implemented with the Python
programming language by OpenSistemas. OsBrain agents operate autonomously and
asynchronously communicate with each other through message passing. Remote
agents may be treated and reconfigured even when they are running, not only variables
but also new methods can be developed in remote agents [141]. We have used osBrain

70
 platform version 0.6.5, which is based on the Python 3 programming language to build
and run the agent that was embedded in the user device.

4.5.2 Case Study

In order to enhance the intelligence of our Cloud-IoT system for remote patient
monitoring, described in the second chapter, we have integrated the agents in each layer,
as shown in Figure 4.8. We have configured our agents for data analysis, and automatic
real-time reasoning using parallelBehavior, which allows executing several behaviors in
parallel to run code effectively and save time.

Family module

Health care providers module

Patient module

User Agent Device Agent Gateway Agent

Analyzer Agent Manager Agent
Figure 4.8. A Cloud-IoT health monitoring system based on smart.
Since the heart rate varies from person to person depending on factors like age,
sex, and exercise, our algorithms are based on resting heart rate data from the Center for
Disease Control and Prevention in the United States, shown in Table 4.2.
Table 4.2. Resting heart rate range [142].

Age group Male Female

4-5 87-104 84-100

6-8 79-94 76-92
9-11 76-91 70-86
12-15 70-87 70-87
16-19 69-85 66-83

71
 Age group Male Female

20-39 66-82 61-78

40-59 64-79 61-77
60-79 64-78 60-75
 Device Agent: embedded agent in a wearable device prototype. The device agent can
sense the heart rate, immediately send sensing data to the cloud, and securely
communicate with the gateway agent in parallel.
 Gateway agent: embedded in the patient mobile application. It is responsible for data
pre-treatment based on Table 4.2, detects the abnormal case, and sends a warning
SMS to the healthcare providers and patient family if the patient's pulse is outside the
normal range, and securely communicates with the other agents as shown in Figure 4.9.

Figure 4.9. Smart device agent's algorithm.

72
 User agent: embedded in the medical staff mobile application. It does partial
manipulation of the data based on Table 4.2, so if the patient's heartbeat is outside the
normal level, the user agent changes the heart rate color to red color in the graphical
interface. In addition to acquiring user requests and send/ receive those requests to the
analyzer agent.
 Analyzer agent: embedded in the cloud, it is responsible for data analyzes
(authorization and authorization) and request execution of the user. In addition to
sends/receive requests to the user agent.
 Manager agent: embedded in the cloud, it is responsible for data collection, storing,
processing, and data error detection.

4.6 Results and Discussion

Experiment results reveal that the proposed device can make fast and intelligent
decisions in emergencies, launch the alarm to get help from the closest persons on the
place and send an alarm message to the patient's family and the health care providers. As
shown in Figure 4.10, the alarm message contains the patient's details, heart rate, and
precise GPS coordinates. Thus, the health care providers take fast and appropriate
decisions and apply the required actions to protect the patient life quickly.

Figure 4.10. Alert message.

73
 As shown in Figure 4.11, the real-time analysis of heart rate values helps the
medical staff to know the patient's situation, prepare the appropriate manner for treating
the patient, and take the appropriate decisions in emergency cases. The widespread
adoption of this system can reduce the annual number of human deaths from
cardiovascular disease.

Figure 4.11. Accurate emergency case detection.

4.7 Conclusion

In this chapter, an integrated system that combines IoT and intelligent agent
technology to enhance IoT intelligence has been proposed. The proposed system consists
of a group of agents distributed in the four IoT architecture layers, as a vertical layer of
agents; a group of agents collaborates to collect, process, and store data in the cloud. The
remaining agents are deployed on IoT devices. We have configured our agents for data
analysis and automatic real-time reasoning using parallel behavior, which allows
executing several behaviors in parallel to run code effectively and save time. Experiment
results reveal that the proposed system can make fast and intelligent decisions in
emergencies without human intervention.

74
General Conclusion

Currently, IoT has gained wide popularity and attention thanks to its powerful
applications in different fields. However, IoT technology is still facing several challenges
that become more complicated with the huge increase in the number of connected devices
to the Internet. In this regard, this thesis provided three main contributions. In the first
contribution, we focused on the idea of integration between IoT and cloud computing
technology to enhance the IoT ability of processing and storage with unlimited resources
and capabilities of the cloud and improve IoT scalability. We have proposed a Cloud-IoT
approach for real-time remote monitoring. In order to verify the validity of the proposed
approach, we developed a Cloud-IoT system for remote patient monitoring. Healthy
people can also benefit from the proposed system to check their health and wellness. Also,
it helps the caregivers to easily measure and monitor patient heart rate anytime, anywhere
for the pre-diagnosis of the patient's condition and early detection before the diseases
develop into a crisis stage.

As a second contribution, we propose an integrated privacy protection system for

IoT. The proposed system provided a high-performance and security level. It consists
primarily of two phases: the first phase ensures data confidentiality in the perception
layer based on lightweight cryptography. In the second phase, we have used the access
control mechanism to ensure data privacy in the cloud through three steps: identification,
authentication, and authorization. The experimentation demonstrates that the proposed
solution more suitable for IoT devices that have computing and communication limits. In
order to tap the most out of massive data streams, to enhance the intelligence of IoT
devices, and to improve our first approach, we have developed an efficient, autonomous,
and real-time solution based on the smart agent for data management and analysis in a
secure way as a third contribution. The proposed system can collect, retrieve, store, and
analyze the patient's heart rate securely. It can also detect the critical case immediately
and take a quick and smart decision without requiring active human interaction.

75
 Although the proposed framework is very promising and capable of providing
very satisfactory results, there are other significant details that we want to carry out
further in our future applications and research.

 Supporting patients' mobility by attaching a battery to the device and reduce the
device's size.
 Developing the proposed system to meet other medical standards like ECG, body
temperature, blood pressure, etc.
 Improve the agents' algorithm to predict and diagnose disease before it occurs, based
on patients' physical information stored in a cloud.

76
List of Publications

International Journals

Ahmid, M., Kazar, O., Hamida, S., Kahloul, L., & Ghodous, P. (2015). Internet of Things
New Challenges in Distributed Artificial Intelligence. International Journal of Computer
Science and Electronics Engineering (IJCSEE), 3(5), 375-377.

Ahmid, M., Kazar, O., & Kahloul, L. (in press). A Secure and Intelligent Real-Time Health
Monitoring System for Remote Cardiac Patients. International Journal of Medical
Engineering and Informatics.

Ahmid, M., & Kazar, O. (in press). A Comprehensive Review of the Internet of Things
Security. Journal of Applied Security Research (JASR).

International Conferences

Ahmid, M., Kazar, O., Hamida, S., Kahloul, L., & Ghodous, P. (2015, Novembre) An
agent based approach for the Internet of Things. In 2nd International Workshop on
Artificial Intelligence and Information & Communication Technologies (IWIICT 2015).
Biskra, Algeria.

Ahmid, M., Kazar, O., Hamida, S., Kahloul, L., & Ghodous, P. (2015, December). Internet
of Things New Challenges in Distributed Artificial Intelligence. In 2015 International
Conference on Future Computational Technologies (ICFCT’2015) (pp. 158-162). Dubai,
UAE.

Ahmid, M., Kazar, O., Benharzallah, S., Kahloul, L., & Merizig, A. (2020, February). An
Intelligent and Secure Health Monitoring System Based on Agent. In 2020 International
Conference on Informatics, IoT, and Enabling Technologies (ICIoT' 2020) (pp. 291-296).
Doha, Qatar. IEEE.
77
Ahmid, M., & Kazar, O. (2021, July). A Cloud-IoT Health Monitoring System Based on
Smart Agent for Cardiovascular Patients. In 2021 International Conference on Information
Technology (ICIT 2021) (pp. 1-6). Amman, Jordan. IEEE.

National Conferences

Ahmid, M., Kazar, O., & Kahloul, L. (2017, Novembre). IoT and Security challenge.
Doctoral Day LINFI (JDL’2017). Biskra, Algérie.

Ahmid, M., Kazar, O., & Kahloul, L. (2018, January). Une approche basée agent pour
l’internet des objets. Journées d’Etudes Informatique Théorique et Appliquée
(JEITA’2018). Biskra, Algeria.

78
REFERENCES

[1] Radenkovic, B., & Kocovic, P. (2020). From ubiquitous computing to the Internet
of things. In Securing the Internet of Things: Concepts, Methodologies, Tools, and
Applications (pp. 1523-1556). IGI Global.

[2] Zhou, M., Fortino, G., Shen, W., Mitsugi, J., Jobin, J., & Bhattacharyya, R. (2016).
Guest editorial special section on advances and applications of Internet of Things
for smart automated systems. IEEE Transactions on Automation Science and
Engineering, 13(3), 1225-1229.

[3] Whitmore, A., Agarwal, A., & Da Xu, L. (2015). The Internet of Things—A
survey of topics and trends. Information systems frontiers, 17(2), 261-274.

[4] Evans, D. (2011). The internet of things: How the next evolution of the internet is
changing everything. CISCO white paper, 1(2011), 1-11.

[5] Haller, S., Karnouskos, S., & Schroth, C. (2008, September). The internet of
things in an enterprise context. In Future Internet Symposium (pp. 14-28).
Springer.

[6] Sundmaeker, H., Guillemin, P., Friess, P., & Woelfflé, S. (2010). Vision and
challenges for realising the Internet of Things. Cluster of European Research
Projects on the Internet of Things, European Commision, 3(3), 34-36.

[7] Vannieuwenborg, F., Verbrugge, S., & Colle, D. (2018). Choosing IoT
connectivity? A guiding methodology based on functional characteristics and
economic considerations. Transactions on Emerging Telecommunications
Technologies, 29(5), e3308.

[8] Xiao, G., Guo, J., Da Xu, L., & Gong, Z. (2014). User interoperability with
heterogeneous IoT devices through transformation. IEEE Transactions on
Industrial Informatics, 10(2), 1486-1496.

[9] Patel, K. K., & Patel, S. M. (2016). Internet of things-IOT: definition,

characteristics, architecture, enabling technologies, application & future
challenges. International journal of engineering science and computing, 6(5),
6122-6131.
79
[10] Forecast, G. M. D. T. (2019, February). Cisco visual networking index: global
mobile data traffic forecast update, 2017–2022. CISCO white paper. (pp. 1-33).

[11] Li, S., Da Xu, L., & Zhao, S. (2015). The internet of things: a survey. Information
Systems Frontiers, 17(2), 243-259.

[12] Sha, K., Wei, W., Yang, T. A., Wang, Z., & Shi, W. (2018). On security challenges
and open issues in Internet of Things. Future Generation Computer Systems, 83,
326-337.

[13] Yaqoob, I., Ahmed, E., Hashem, I. A. T., Ahmed, A. I. A., Gani, A., Imran, M., &
Guizani, M. (2017). Internet of things architecture: Recent advances, taxonomy,
requirements, and open challenges. IEEE wireless communications, 24(3), 10-16.

[14] Sri, T. S., Prasad, J. R., & Vijayalakshmi, Y. (2016). A review on the state of art of
Internet of Things. International Journal of Advanced Research in Computer and
Communication Engineering, 5(7), 189-193.

[15] Razzaque, M. A., Milojevic-Jevric, M., Palade, A., & Clarke, S. (2015).
Middleware for internet of things: a survey. IEEE Internet of things journal, 3(1),
70-95.

[16] Chelloug, S. A., & El-Zawawy, M. A. (2017). Middleware for internet of things:
survey and challenges. Intelligent Automation & Soft Computing, 1-9.

[17] Asim, M. (2017). A survey on application layer protocols for Internet of Things
(IoT). International Journal of Advanced Research in Computer Science, 8(3),
966-1000.

[18] Ahmid, M., Kazar, O., Hamida, S., Kahloul, L., & Ghodous, P. (2015). Internet of
Things New Challenges in Distributed Artificial Intelligence. International Journal
of Computer Science and Electronics Engineering (IJCSEE), 3(5), 375-377.

[19] Ahmid, M., Kazar, O., Benharzallah, S., Kahloul, L., & Merizig, A. (2020,
February). An Intelligent and Secure Health Monitoring System Based on Agent.
In 2020 International Conference on Informatics, IoT, and Enabling Technologies
(ICIoT' 2020) (pp. 291-296). IEEE.

80
[20] Dey, N., Ashour, A. S., & Bhatt, C. (2017). Internet of things driven connected
healthcare. In Internet of things and big data technologies for next generation
healthcare (pp. 3-12). Springer, Cham.

[21] Jeong, H., Rogers, J. A., & Xu, S. (2020). Continuous on-body sensing for the
COVID-19 pandemic: Gaps and opportunities. Science Advances, 6(36),
eabd4794.

[22] Boyes, H., Hallaq, B., Cunningham, J., & Watson, T. (2018). The industrial
internet of things (IIoT): An analysis framework. Computers in industry, 101, 1-12.

[23] Mumtaz, S., Alsohaily, A., Pang, Z., Rayes, A., Tsang, K. F., & Rodriguez, J.
(2017). Massive Internet of Things for industrial applications: Addressing wireless
IIoT connectivity challenges and ecosystem fragmentation. IEEE Industrial
Electronics Magazine, 11(1), 28-33.

[24] Robotics, A. B. B. (2019). YuMi—IRB 14000.

[25] Herrera-Quintero, L. F., Vega-Alfonso, J. C., Banse, K. B. A., & Zambrano, E. C.

(2018). Smart its sensor for the transportation planning based on iot approaches
using serverless and microservices architecture. IEEE Intelligent Transportation
Systems Magazine, 10(2), 17-27.

[26] Ahmid, M., Kazar, O., Hamida, S., Kahloul, L., & Ghodous, P. (2015, December).
Internet of Things New Challenges in Distributed Artificial Intelligence. In 2015
International Conference on Future Computational Technologies (ICFCT’2015)
(pp. 158-162).

[27] Kim, T. H., Ramos, C., & Mohammed, S. (2017). Smart city and IoT. Future
Generation Computer Systems, 76, 159-162

[28] Hammi, B., Khatoun, R., Zeadally, S., Fayad, A., & Khoukhi, L. (2017). IoT
technologies for smart cities. IET Networks, 7(1), 1-13.

[29] Schneider, S. (2017). The industrial internet of things (iiot) applications and
taxonomy. Internet of Things and Data Analytics Handbook, 41-81.

81
[30] Sisinni, E., Saifullah, A., Han, S., Jennehag, U., & Gidlund, M. (2018). Industrial
internet of things: Challenges, opportunities, and directions. IEEE Transactions on
Industrial Informatics, 14(11), 4724-4734.

[31] Solanki, A., & Nayyar, A. (2019). Green internet of things (G-IoT): ICT
technologies, principles, applications, projects, and challenges. In Handbook of
Research on Big Data and the IoT (pp. 379-405). IGI Global.

[32] Georgiou, K., Xavier-de-Souza, S., & Eder, K. (2017). The IoT energy challenge:
A software perspective. IEEE Embedded Systems Letters, 10(3), 53-56.

[33] Lu, W., Gong, Y., Liu, X., Wu, J., & Peng, H. (2017). Collaborative energy and
information transfer in green wireless sensor networks for smart cities. IEEE
Transactions on Industrial Informatics, 14(4), 1585-1593.

[34] Mahapatra, C., Sheng, Z., Kamalinejad, P., Leung, V. C., & Mirabbasi, S. (2016).
Optimal power control in green wireless sensor networks with wireless energy
harvesting, wake-up radio and transmission control. IEEE Access, 5, 501-518.

[35] Radu, L. D. (2017). Green cloud computing: A literature survey. Symmetry, 9(12),
295.

[36] Mishra, S. K., Puthal, D., Sahoo, B., Jena, S. K., & Obaidat, M. S. (2018). An
adaptive task allocation technique for green cloud computing. The Journal of
Supercomputing, 74(1), 370-385.

[37] Pereira, C., & Aguiar, A. (2014). Towards efficient mobile M2M communications:
Survey and open challenges. Sensors, 14(10), 19582-19608.

[38] Oncel, S. S. (2017). Green energy engineering: Opening a green way for the
future. Journal of cleaner production, 142, 3095-3100.

[39] Gibson, L., Wilman, E. N., & Laurance, W. F. (2017). How green is
‘green’energy?. Trends in ecology & evolution, 32(12), 922-935.

[40] Mendez Mena, D., Papapanagiotou, I., & Yang, B. (2018). Internet of things:
Survey on security. Information Security Journal: A Global Perspective, 27(3),
162-182.

82
[41] Enterprise, H. P. (2015). Internet of things research study: 2015 report. Online:
[Link] hp. com/h20195, 2.

[42] Ahmid, M., & Kazar, O. (in press). A Comprehensive Review of the Internet of
Things Security. Journal of Applied Security Research (JASR).

[43] Khattak, H. A., Shah, M. A., Khan, S., Ali, I., & Imran, M. (2019). Perception
layer security in Internet of Things. Future Generation Computer Systems, 100,
144-164.

[44] Ali, F., Khan, M. S., & Akhtar, H. (2019). Security Review in Internet of Things.
Internet of Things and Cloud Computing, 7(3), 80-87.

[45] Al-Issa, Y., Ottom, M. A., & Tamrawi, A. (2019). eHealth cloud security
challenges: A survey. Journal of Healthcare Engineering, 2019, 1-16.

[46] Subramanian, N., & Jeyaraj, A. (2018). Recent security challenges in cloud
computing. Computers & Electrical Engineering, 71, 28-42.

[47] Bhushan, K., & Gupta, B. B. (2017). Security challenges in cloud computing:
state-of-art. International Journal of Big Data Intelligence, 4(2), 81-107.

[48] Lu, Y., & Da Xu, L. (2018). Internet of things (iot) cybersecurity research: A
review of current research topics. IEEE Internet of Things Journal, 6(2), 2103-
2115.

[49] Tewari, A., & Gupta, B. B. (2020). Security, privacy and trust of different layers
in Internet-of-Things (IoTs) framework. Future generation computer systems, 108,
909-920.

[50] Aazam, M., & Huh, E. N. (2016). Fog computing: The cloud-iot\/ioe middleware
paradigm. IEEE Potentials, 35(3), 40-44.

[51] Kobusińska, A., Leung, C., Hsu, C. H., Raghavendra, S., & Chang, V. (2018).
Emerging trends, issues and challenges in Internet of Things, Big Data and cloud
computing. Future Generation Computer Systems, 87, 416-419.

83
[52] Cai, H., Xu, B., Jiang, L., & Vasilakos, A. V. (2016). IoT-based big data storage
systems in cloud computing: perspectives and challenges. IEEE Internet of Things
Journal, 4(1), 75-87.

[53] Jirkovský, V., Obitko, M., & Mařík, V. (2016). Understanding data heterogeneity
in the context of cyber-physical systems integration. IEEE Transactions on
Industrial Informatics, 13(2), 660-667.

[54] Rashid, Z. N., Zebari, S. R., Sharif, K. H., & Jacksi, K. (2018, October).
Distributed Cloud Computing and Distributed Parallel Computing: A Review.
In 2018 International Conference on Advanced Science and Engineering
(ICOASE) (pp. 167-172). IEEE.

[55] Javaid, N., Sher, A., Nasir, H., & Guizani, N. (2018). Intelligence in IoT-based 5G
networks: Opportunities and challenges. IEEE Communications Magazine, 56(10),
94-100.

[56] Zhang, Y., Ma, X., Zhang, J., Hossain, M. S., Muhammad, G., & Amin, S. U.
(2019). Edge intelligence in the cognitive Internet of Things: Improving
sensitivity and interactivity. IEEE Network, 33(3), 58-64.

[57] Mohamed, E. (2020). The Relation of Artificial Intelligence with Internet Of

Things: A survey. Journal of Cybersecurity and Information Management, 1(1),
30-24.

[58] González García, C., Núñez Valdéz, E. R., García Díaz, V., Pelayo García-Bustelo,
B. C., & Cueva Lovelle, J. M. (2019). A review of artificial intelligence in the
Internet of Things. International Journal of Interactive Multimedia and Artificial
Intelligence, 5, 9-20.

[59] Ghosh, A., Chakraborty, D., & Law, A. (2018). Artificial intelligence in Internet of
things. CAAI Transactions on Intelligence Technology, 3(4), 208-218.

[60] Scully, P. (2020). Top 10 IoT applications in 2020. Retrieved from [Link]
[Link]/top-10-iot-applications-in-2020/

[61] O'Dea, S. (2020). Data volume of IoT connected devices worldwide 2019 and
2025. Retrieved from [Link]
connected-devices-data-size/

84
[62] O'Dea, S. (2020). Internet of Things (IoT) connected devices installed base
worldwide from 2015 to 2025. Retrieved from
[Link]
worldwide/

[63] Forkan, A., Khalil, I., & Tari, Z. (2014). CoCaMAAL: A cloud-oriented context-
aware middleware in ambient assisted living. Future Generation Computer
Systems, 35, 114-127.

[64] Armbrust, M., Fox, A., Griffith, R., Joseph, A. D., Katz, R., Konwinski, A., &
Zaharia, M. (2010). A view of cloud computing. Communications of the ACM,
53(4), 50-58.

[65] Psannis, K. E., Xinogalos, S., & Sifaleras, A. (2014). Convergence of Internet of
things and mobile cloud computing. Systems Science & Control Engineering: An
Open Access Journal, 2(1), 476-483.

[66] Jadhav, S. R., Shinde, R. P., Patil, P. S., Thakar, G. B. (2018). Health Detector
Android Application using IoT for Rural Area. International Journal on Future
Revolution in Computer Science & Communication Engineering, 4(5), 131-136.

[67] Fu, Y., & Liu, J. (2015). System design for wearable blood oxygen saturation and
pulse measurement device. Procedia manufacturing, 3, 1187-1194.

[68] Vuha, C. A. H. K., Rajani, M., & Vineeth, J. M. (2017). Smart Health Care
Monitoring using Internet of Things and Android. International Journal of
Advanced Research in Electronics and Communication Engineering, 6(3), 101-
104.

[69] Navdeti, P., Parte, S., Talashilkar, P., Patil, J., & Khairnar, V. (2016). Patient
parameter monitoring system using Raspberry Pi. International Journal Of
Engineering And Computer Science, 5(3).

[70] Reddy, N. A., Krishnan, G. H., & Raghuram, D. (2016). Real Time Patient Health
Monitoring Using Raspberry PI. Research Journal of Pharmaceutical Biological
And Chemical Sciences, 7(6), 570-575.

[71] Kakria, P., Tripathi, N. K., & Kitipawang, P. (2015). A real-time health monitoring
system for remote cardiac patients using smartphone and wearable sensors.
International journal of telemedicine and applications, 2015, 1-11.

85
[72] Aminian, M., & Naji, H. R. (2013). A hospital healthcare monitoring system using
wireless sensor networks. International Journal of Health & Medical Informatics,
4(02), 121.

[73] Aranki, D., Kurillo, G., Yan, P., Liebovitz, D. M., & Bajcsy, R. (2016). Real-time
tele-monitoring of patients with chronic heart-failure using a smartphone: Lessons
learned. IEEE Transactions on Affective Computing, 7(3), 206-219.

[74] Suprayitno, E. A., Marlianto, M. R., & Mauliana, M. I. (2019). Measurement

device for detecting oxygen saturation in blood, heart rate, and temperature of
human body. In Journal of Physics, 1402(3), 1-6.

[75] Jovanov, E., Lords, A. D., Raskovic, D., Cox, P. G., Adhami, R., & Andrasik, F.
(2003). Stress monitoring using a distributed wireless intelligent sensor system.
IEEE Engineering in Medicine and Biology Magazine, 22(3), 49-55.

[76] Paradiso, R., Loriga, G., & Taccini, N. (2005). A wearable health care system
based on knitted integrated sensors. IEEE transactions on Information Technology
in biomedicine, 9(3), 337-344.

[77] Kwon, S., Lee, D., Kim, J., Lee, Y., Kang, S., Seo, S., & Park, K. (2016). Sinabro:
A smartphone-integrated opportunistic electrocardiogram monitoring system.
Sensors, 16(3), 361.

[78] Malhi, K., Mukhopadhyay, S. C., Schnepper, J., Haefke, M., & Ewald, H. (2012).
A zigbee-based wearable physiological parameters monitoring system. IEEE
sensors journal, 12(3), 423-430.

[79] Lee, J. V., Chuah, Y. D., & Chieng, K. T. (2013). Smart elderly home monitoring
system with an android phone. International Journal of Smart Home, 7(3), 17-32.

[80] Suma, N., Samson, S. R., Saranya, S., Shanmugapriya, G., & Subhashri, R. (2017).
IOT based smart agriculture monitoring system. International Journal on Recent
and Innovation Trends in computing and communication, 5(2), 177-181.

[81] Gondchawar, N., & Kawitkar, R. S. (2016). IoT based smart agriculture.
International Journal of advanced research in Computer and Communication
Engineering, 5(6), 838-842.

86
[82] Yan, M., & Shi, H. (2013). Smart living using Bluetooth-based Android
smartphone. International journal of wireless & mobile networks, 5(1), 65.

[83] Javale, D., Mohsin, M., Nandanwar, S., & Shingate, M. (2013). Home automation
and security system using Android ADK. International journal of electronics
communication and computer technology (IJECCT), 3(2), 382-385.

[84] Jagannathan, S., & Priyatharshini, R. (2015, July). Smart farming system using
sensors for agricultural task automation. In 2015 IEEE Technological Innovation
in ICT for Agriculture and Rural Development (TIAR) (pp. 49-53). IEEE.

[85] Kaewmard, N., & Saiyod, S. (2014, October). Sensor data collection and
irrigation control on vegetable crop using smart phone and wireless sensor
networks for smart farm. In 2014 IEEE Conference on Wireless Sensors (ICWiSE)
(pp. 106-112). IEEE.

[86] Cardiovascular Diseases. (2017, May 17). WHO. Retrieved from

[Link]

[87] Medical doctors (per 10 000 population). (2015, October 20). WHO. Retrieved
from [Link]
details/GHO/medical-doctors-(per-10-000-population)

[88] Smyth, N. (2017). Android Studio 2.3 Development Essentials-Android 7 Edition.

PayloadMedia, Inc..

[89] Stonehem, B. (2016). Google Android Firebase: Learning the Basics (Vol. 1). First
Rank Publishing.

[90] Upton, E., & Halfacree, G. (2014). Raspberry Pi user guide. John Wiley & Sons.

[91] Lindberg, L. G., Tamura, T., & Öberg, P. Å. (1991). Photoplethysmography.

Medical and Biological Engineering and Computing, 29(1), 40-47.

[92] Conti, M., Dehghantanha, A., Franke, K., & Watson, S. (2018). Internet of Things
security and forensics: Challenges and opportunities, 78(2), 544-546.

87
[93] Alaba, F. A., Othman, M., Hashem, I. A. T., & Alotaibi, F. (2017). Internet of
Things security: A survey. Journal of Network and Computer Applications, 88, 10-
28.

[94] Kumar, J. S., & Patel, D. R. (2014). A survey on internet of things: Security and
privacy issues. International Journal of Computer Applications, 90(11), 20-26.

[95] Maras, M. H. (2015). Internet of Things: security and privacy

implications. International Data Privacy Law, 5(2), 99-104.

[96] Song, T., Li, R., Mei, B., Yu, J., Xing, X., & Cheng, X. (2017). A privacy
preserving communication protocol for IoT applications in smart homes. IEEE
Internet of Things Journal, 4(6), 1844-1852.

[97] Rghioui, A., L'aarje, A., Elouaai, F., & Bouhorma, M. (2014, October). The
internet of things for healthcare monitoring: security review and proposed
solution. In 2014 Third IEEE International Colloquium in Information Science
and Technology (CIST) (pp. 384-389). IEEE.

[98] Gong, T., Huang, H., Li, P., Zhang, K., & Jiang, H. (2015, December). A medical
healthcare system for privacy protection based on IoT. In 2015 Seventh
International Symposium on Parallel Architectures, Algorithms and Programming
(PAAP) (pp. 217-222). IEEE.

[99] Gope, P., & Hwang, T. (2015). BSN-Care: A secure IoT-based modern healthcare
system using body sensor network. IEEE sensors journal, 16(5), 1368-1376.

[100] Kulkarni, S., Durg, S., & Iyer, N. (2016, March). Internet of things (iot) security.
In 2016 3rd International Conference on Computing for Sustainable Global
Development (INDIACom) (pp. 821-824). IEEE.

[101] Shafagh, H., Hithnawi, A., Dröscher, A., Duquennoy, S., & Hu, W. (2015,
November). Talos: Encrypted query processing for the internet of things.
In Proceedings of the 13th ACM conference on embedded networked sensor
systems (pp. 197-210). ACM.

[102] Alharbi, K. N., Lin, X., & Shao, J. (2016). A privacy-preserving data-sharing
framework for smart grid. IEEE Internet of Things Journal, 4(2), 555-562.

88
[103] Doukas, C., Maglogiannis, I., Koufi, V., Malamateniou, F., & Vassilacopoulos, G.
(2012, November). Enabling data protection through PKI encryption in IoT m-
Health devices. In 2012 IEEE 12th International Conference on Bioinformatics &
Bioengineering (BIBE) (pp. 25-29). IEEE.

[104] Nguyen, K. T., Oualha, N., & Laurent, M. (2018). Securely outsourcing the
ciphertext-policy attribute-based encryption. World Wide Web, 21(1), 169-183.

[105] Thirumalai, C., & Kar, H. (2017, April). Memory Efficient Multi Key (MEMK)
generation scheme for secure transportation of sensitive data over Cloud and IoT
devices. In 2017 Innovations in Power and Advanced Computing Technologies (i-
PACT) (pp. 1-6). IEEE.

[106] Guo, A., Xu, M., Ran, F., & Wang, H. (2019). A novel medical internet of things
perception system based on visual image encryption and intrusion
detection. Cluster Computing, 22(6), 13405-13413.

[107] Zhang, H., Yu, J., Tian, C., Tong, L., Lin, J., Ge, L., & Wang, H. (2020). Efficient
and secure outsourcing scheme for RSA decryption in Internet of Things. IEEE
Internet of Things Journal, 7(8), 6868-6881.

[108] Barker, E., & Dang, Q. (2016). Nist special publication 800-57 part 1, revision
4. NIST, Tech. Rep, 16.

[109] Arsénio, A., Serra, H., Francisco, R., Nabais, F., Andrade, J., & Serrano, E.
(2014). Internet of intelligent things: Bringing artificial intelligence into things
and communication networks. In Inter-cooperative collective intelligence:
Techniques and applications (pp. 1-37). Springer, Berlin, Heidelberg.

[110] Hu, F., Xie, D., & Shen, S. (2013, August). On the application of the internet of
things in the field of medical and health care. In 2013 IEEE international
conference on green computing and communications and IEEE Internet of Things
and IEEE cyber, physical and social computing (pp. 2053-2058). IEEE.

[111] Blake, M. B. (2015). An internet of things for healthcare. IEEE Annals of the
History of Computing, 19(04), 4-6.

[112] Dimitrov, D. V. (2016). Medical internet of things and big data in healthcare.
Healthcare informatics research, 22(3), 156.

89
[113] Ferber, J. (1997). Les systèmes multi-agents: vers une intelligence collective.
InterEditions.

[114] Jennings, N. R., & Wooldridge, M. (1998). Applications of intelligent agents.

In Agent technology (pp. 3-28). Springer, Berlin, Heidelberg.

[115] Jennings, N. R. (2000). On agent-based software engineering. Artificial

intelligence, 117(2), 277-296.

[116] Russell, S., & Norvig, P. (2010). Intelligence artificielle: Avec plus de 500
exercices. Pearson Education France.

[117] Suzanne Barber, K., Goel, A., & Martin, C. E. (2000). Dynamic adaptive
autonomy in multi-agent systems. Journal of Experimental & Theoretical
Artificial Intelligence, 12(2), 129-147.

[118] d'Inverno, M., Luck, M., Georgeff, M., Kinny, D., & Wooldridge, M. (2004). The
dMARS architecture: A specification of the distributed multi-agent reasoning
system. Autonomous Agents and Multi-Agent Systems, 9(1), 5-53.

[119] Ferber, J., & Weiss, G. (1999). Multi-agent systems: an introduction to distributed
artificial intelligence. Reading: Addison-Wesley.

[120] Huhns, M. N., & Singh, M. P. (1998). Cognitive agents. IEEE Internet
computing, 2(6), 87-89.

[121] Cruces, A. L. L., & De Arriaga, F. (2000). Reactive agent design for intelligent
tutoring systems. Cybernetics & Systems, 31(1), 1-47.

[122] Nwana, H. S. (1996). Software agents: An overview. Knowledge engineering

review, 11(3), 205-244.

[123] Sycara, K. P. (1998). Multiagent systems. AI magazine, 19(2), 79-79.

[124] Dorri, A., Kanhere, S. S., & Jurdak, R. (2018). Multi-agent systems: A
survey. Ieee Access, 6, 28573-28593.

90
[125] Fortino, G., Guerrieri, A., Russo, W., & Savaglio, C. (2014, May). Integration of
agent-based and cloud computing for the smart objects-oriented IoT.
In Proceedings of the 2014 IEEE 18th international conference on computer
supported cooperative work in design (CSCWD) (pp. 493-498). IEEE.

[126] Vaidehi, V., Vardhini, M., Yogeshwaran, H., Inbasagar, G., Bhargavi, R., &
Hemalatha, C. S. (2013). Agent based health monitoring of elderly people in
indoor environments using wireless sensor networks. Procedia Computer
Science, 19, 64-71.

[127] Polu, S. K. (2019). Modeling of efficient multi-agent based mobile health care
system. Int J Innov Res Sci Technol, 5(8), 10-14.

[128] Su, C. J., & Wu, C. Y. (2011). JADE implemented mobile multi-agent based,
distributed information platform for pervasive health care monitoring. Applied
Soft Computing, 11(1), 315-325.

[129] Naranjo-Hernández, D., Roa, L. M., Reina-Tosina, J., & Estudillo-Valderrama, M.

Á. (2012). SoM: A smart sensor for human activity monitoring and assisted
healthy ageing. IEEE transactions on biomedical engineering, 59(11), 3177-3184.

[130] Santos, J., Rodrigues, J. J., Silva, B. M., Casal, J., Saleem, K., & Denisov, V.
(2016). An IoT-based mobile gateway for intelligent personal assistants on mobile
health environments. Journal of Network and Computer Applications, 71, 194-
204.

[131] Chan, V., Ray, P., & Parameswaran, N. (2008). Mobile e-Health monitoring: an
agent-based approach. IET communications, 2(2), 223-230.

[132] Smarsly, K., Law, K. H., & Hartmann, D. (2012). Multiagent-based collaborative
framework for a self-managing structural health monitoring system. Journal of
computing in civil engineering, 26(1), 76-89.

[133] Sharma, A., Srinivasan, D., & Kumar, D. S. (2016, July). A comparative analysis
of centralized and decentralized multi-agent architecture for service restoration.
In 2016 IEEE congress on evolutionary computation (CEC) (pp. 311-318). IEEE.

[134] Su, C. J., & Chu, T. W. (2014). A mobile multi-agent information system for
ubiquitous fetal monitoring. International journal of environmental research and
public health, 11(1), 600-625.

91
[135] Al-Sakran, H. O. (2015). Intelligent traffic information system based on
integration of Internet of Things and Agent technology. International Journal of
Advanced Computer Science and Applications (IJACSA), 6(2), 37-43.

[136] Ahmid, M., & Kazar, O. (2021, July). A Cloud-IoT Health Monitoring System
Based on Smart Agent for Cardiovascular Patients. In 2021 International
Conference on Information Technology (ICIT 2021) (pp. 1-6). IEEE.

[137] Ahmid, M., Kazar, O., & Kahloul, L. (in press). A Secure and Intelligent Real-
Time Health Monitoring System for Remote Cardiac Patients. International
Journal of Medical Engineering and Informatics.

[138] Chmiel, K., Gawinecki, M., Kaczmarek, P., Szymczak, M., & Paprzycki, M.
(2005). Efficiency of JADE agent platform. Scientific Programming, 13(2), 159-
172.

[139] Bellifemine, F., Poggi, A., & Rimassa, G. (2000, July). Developing multi-agent
systems with JADE. In International Workshop on Agent Theories, Architectures,
and Languages (pp. 89-103). Springer, Berlin, Heidelberg.

[140] Bergenti, F., Caire, G., & Gotta, D. (2014, September). Agents on the Move:
JADE for Android Devices. WOA, 1260, 1-5.

[141] osBrain. (2019). What is osBrain?. Available at:

[Link] (accessed 12 August 2019)

[142] Ostchega, Y. (2011). Resting pulse rate reference data for children, adolescents,
and adults: United States. US Department of Health and Human Services, Centers
for Disease Control and Prevention, National Center for Health Statistics.

92