Decision Making and Security Risk Management for IoT Environments (Advances in Information Security, 106) 3031475895, 9783031475894

Citation preview

Advances in Information Security 106

Wadii Boulila · Jawad Ahmad · Anis Koubaa · Maha Driss · Imed Riadh Farah   Editors

Decision Making and Security Risk Management for IoT Environments

Advances in Information Security Volume 106

Series Editors Sushil Jajodia, George Mason University, Fairfax, VA, USA Pierangela Samarati, Milano, Italy Javier Lopez, Malaga, Spain Jaideep Vaidya, East Brunswick, NJ, USA

The purpose of the Advances in Information Security book series is to establish the state of the art and set the course for future research in information security. The scope of this series includes not only all aspects of computer, network security, and cryptography, but related areas, such as fault tolerance and software assurance. The series serves as a central source of reference for information security research and developments. The series aims to publish thorough and cohesive overviews on specific topics in Information Security, as well as works that are larger in scope than survey articles and that will contain more detailed background information. The series also provides a single point of coverage of advanced and timely topics and a forum for topics that may not have reached a level of maturity to warrant a comprehensive textbook.

Wadii Boulila • Jawad Ahmad • Anis Koubaa • Maha Driss • Imed Riadh Farah Editors

Decision Making and Security Risk Management for IoT Environments

Editors Wadii Boulila Prince Sultan University Riyadh, Saudi Arabia

Jawad Ahmad Edinburgh Napier University Edinburgh, UK

Anis Koubaa Prince Sultan University Riyadh, Saudi Arabia

Maha Driss Prince Sultan University Riyadh, Saudi Arabia

Imed Riadh Farah University of Manouba Manouba, Tunisia

ISSN 1568-2633 ISSN 2512-2193 (electronic) Advances in Information Security ISBN 978-3-031-47589-4 ISBN 978-3-031-47590-0 (eBook) https://doi.org/10.1007/978-3-031-47590-0 © The Editor(s) (if applicable) and The Author(s), under exclusive license to Springer Nature Switzerland AG 2024 This work is subject to copyright. All rights are solely and exclusively licensed by the Publisher, whether the whole or part of the material is concerned, specifically the rights of translation, reprinting, reuse of illustrations, recitation, broadcasting, reproduction on microfilms or in any other physical way, and transmission or information storage and retrieval, electronic adaptation, computer software, or by similar or dissimilar methodology now known or hereafter developed. The use of general descriptive names, registered names, trademarks, service marks, etc. in this publication does not imply, even in the absence of a specific statement, that such names are exempt from the relevant protective laws and regulations and therefore free for general use. The publisher, the authors, and the editors are safe to assume that the advice and information in this book are believed to be true and accurate at the date of publication. Neither the publisher nor the authors or the editors give a warranty, expressed or implied, with respect to the material contained herein or for any errors or omissions that may have been made. The publisher remains neutral with regard to jurisdictional claims in published maps and institutional affiliations. This Springer imprint is published by the registered company Springer Nature Switzerland AG The registered company address is: Gewerbestrasse 11, 6330 Cham, Switzerland Paper in this product is recyclable.

About the Book

The Internet of Things (IoT) refers to a network of tiny devices linked to the Internet or other communication networks. IoT is gaining popularity since it opens up new possibilities for developing many modern applications such as smart cities, smart agriculture, innovative healthcare services, etc. The worldwide IoT market surpassed $100 billion in sales for the first time in 2017, and forecasts show that this number might reach $1.6 trillion by 2025. However, as IoT devices grow, more widespread, threats, privacy, and security concerns are growing. The massive volume of data exchanged highlights significant challenges to preserving individual privacy and securing shared data. Therefore, securing the IoT environment becomes difficult for research and industry stakeholders. This book contains contemporary research that outlines and addresses security, privacy challenges, and decision-making in IoT environments. In this book, we cover a variety of subjects related to the following keywords: IoT, security, AI, deep learning, federated learning, intrusion detection systems, and distributed computing paradigms. This book provides a collection of the most up-to-date research, providing a complete overview of security and privacy-preserving in IoT environments. It introduces new approaches based on machine learning to handle security challenges providing the field with a collection of recent research not already covered in the primary literature. This book will be a valuable companion for users and developers interested in decision-making and security risk management in IoT environments.

v

Contents

Internet of Things Overview: Architecture, Technologies, Application, and Challenges . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Maroua Ahmid, Okba Kazar, and Ezedin Barka

1

IoMT Applications Perspectives: From Opportunities and Security Challenges to Cyber-Risk Management. . . . . . . . . . . . . . . . . . . . . . . . . . . . . Sondes Ksibi, Faouzi Jaidi, and Adel Bouhoula

21

Cybersecurity Challenges and Implications for the Adoption of Cloud Computing and IoT: DDoS Attacks as an Example . . . . . . . . . . . . . . . Bassam Naji Al-Tamimi, Hani Almoamari, Antonio Nehme, and Shadi Basurra Implementation of the C4.5 Algorithm in the Internet of Things Applications . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Apriandy Angdresey, Lanny Sitanayah, Tjia Valentyno Nathaniel Kairupan, and Timothy Matthew Immanuel Sumajow Intrusion Detection Systems Using Machine Learning . . . . . . . . . . . . . . . . . . . . . . William Taylor, Amir Hussain, Mandar Gogate, Kia Dashtipour, and Jawad Ahmad Multivariate Procedure for Modeling and Prediction of Temperature in Punjab, Pakistan . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Bushra Kanwal, Zeeshan Ashraf, Tahir Mehmood, Summrina Kanwal, Kia Dashtipour, and Mandar Gogate

39

53

75

99

A New Proposed Model for the Influence of Climate Change on the Tension Anticipation in Hospital Emergencies . . . . . . . . . . . . . . . . . . . . . . . 125 Nouha Mhimdi and Wahiba Ben Abdessalem Statistical Downscaling Modeling for Temperature Prediction . . . . . . . . . . . . . 147 Zeeshan Ashraf, Bushra Kanwal, Ijaz Hussain, Kia Dashtipour, Mandar Gogate, and Summrina Kanwal vii

viii

Contents

UAV-Based IoT Applications for Action Recognition . . . . . . . . . . . . . . . . . . . . . . . . 171 Selmi Mouna and Imed Riadh Farah Federated Learning for Market Surveillance . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 199 Philip Song, Summrina Kanwal, Kia Dashtipour, and Mandar Gogate Fake News in Social Media: Fake News Themes and Intentional Deception in the News and on Social Media . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 219 Hasan Idrees, Kia Dashtipour, Tassadaq Hussain, and Mandar Gogate

Internet of Things Overview: Architecture, Technologies, Application, and Challenges Maroua Ahmid, Okba Kazar, and Ezedin Barka

1 Introduction The Internet of Things is an environment that connects a large number of heterogeneous objects. Where technology is integrated into everyday objects and connects these objects to other devices, people, and services [1] using a variety of connectivity technologies such as ZigBee, WIFI, near-field communication (NFC), etc. The Internet of Things opened the way for many applications, areas, and scenarios based on the interconnectedness of the physical and virtual worlds: smart home, healthcare, smart farms, smart factories, etc. In the future, IoT applications and services will invade and impact all fields [2]. However, the Internet of Things technology is still being discussed and created, like other promising concepts; IoT

M. Ahmid () Department of Computer Science, Smart Computer Science Laboratory, Mohamed Khider University, Biskra, Algeria National School for the Application of Land Transport Technologies, Batna, Algeria e-mail: [email protected] O. Kazar College of Computing and Informatics, Department of Computer Science, University of Sharjah, Sharjah, United Arab Emirates College of Arts, Sciences & Information Technology, University of Kalba, Sharjah, United Arab Emirates e-mail: [email protected] E. Barka Department of Information Systems and Security, College of Information Technology, United Arab Emirate University, Abu Dhabi, United Arab Emirates e-mail: [email protected] © The Author(s), under exclusive license to Springer Nature Switzerland AG 2024 W. Boulila et al. (eds.), Decision Making and Security Risk Management for IoT Environments, Advances in Information Security 106, https://doi.org/10.1007/978-3-031-47590-0_1

1

2

M. Ahmid et al.

faces several technical and nontechnical obstacles and challenges [3]. According to Statista’s research department, 75 billion devices will be linked to the Internet by 2025. With the tremendous increase in the number of IoT devices, these difficulties are becoming increasingly difficult to overcome in order to enable the Internet of Things to reach its full potential; these challenges must be studied and resolved. This chapter delivers an overview of the Internet of Things and has six sections. Section 1 presented the history of the IoT. Section 2 outlines the basic concepts, definitions, and terms to describe the IoT architecture. Section 3 lists the most often used enabling technologies. Section 4 contains a collection of the most advanced IoT applications. In addition, in Sect. 5, we explore the most pressing concerns and challenges confronting the Internet of Things, followed by the conclusion in Sect. 6. The following are some of the major research contributions: • The study brings forward the concept of IoT from its origin to the present day, with periodic updates. • The detailed investigation of IoT Functional blocks at each layer, as well as the related issues. • Various communication techniques used in IoT technology have been thoroughly examined, as well as the future of IoT applications directions and market growth. • IoT challenges have been discussed towards enhancing IoT application markets.

2 Internet of Things History Four Carnegie Mellon University students created the ARPANET-connected coke machine in 1982. This project prompted many researchers across the world to build their network appliances. Tim Berners-Lee established the World Wide Web in 1989 [4], the universal Internet that we know today. In October 1989, John Romkey created the first IoT device, a toaster that can be controlled via Internet [5]. After 10 years, Kevin Ashton used the term “Internet of Things” for the first time at the Massachusetts Institute of Technology (MIT) in 1999 [6]; to describe objects equipped with microchip identification by radio frequency identification (RFID chip). As shown in Fig. 1, LG announced its smart refrigerator plans in 2000, which would assess whether or not the food products kept in it needed to be refilled. In 2005, the International Telecommunications Union (ITU) issued its first report on IoT. After 3 years, in 2008, the number of IoT devices linked to the Internet exceeded the world’s population for the first time, which caused a problem with the exhaustion of IP addresses, which was resolved by the introduction of the Internet Protocol version 6 (IPv6). According to Cisco’s Business Solutions Group, the IoT was born in 2008– 2009, and in 2009, the Commission of the European Communities published an article about an action plan for the IoT in Europe, thus confirming that the IoT had

Internet of Things Overview: Architecture, Technologies, Application, and Challenges

3

Fig. 1 IoT timeline history

arrived at an elevated level of attention among European researchers, politicians, and commercial and industry partners [7]. After 6 years, in 2015, the Alliance for the Internet of Things (AIOTI) was created by the European Commission, to foster interaction and collaboration between IoT stakeholders [8]. The number of devices linked to the Internet surpassed 30 billion in 2020, and this number is expected to quadruple by 2025. According to studies, there will be more than 75 billion linked devices to the Internet by 2025 [9]. This statistic excludes PCs, smartphones, and tablets by seeing the amazing growth of the connected device number to the Internet over the past decade; these figures seem more realistic rather than an overstatement. .

3 Internet of Things Architecture The IoT layers are differentiated by their functions, techniques, and devices, which led to disagreement between experts and researchers about whether to divide this structure into three or four layers [10]. Many researchers divide three layers: the perception layer, the network layer, and the application layer. The fourth layer, which some researchers add, is referred to as the support layer. Scalability, interoperability, and decentralization of heterogeneous devices must be considered in IoT architecture design since IoT devices may need to move or interact with their surroundings in real time. Thus, dividing IoT architecture into four levels is a more effective maens of ensuring interoperability among IoT devices in diverse contexts

4

M. Ahmid et al.

Fig. 2 IoT detailed architecture

[11, 12]. Figure 2 shows the detailed architecture of IoT. Each layer’s functioning is outlined below.

3.1 Perception Layer The perception layer, or “device layer,” is the first layer in IoT architecture and consists of diverse material objects, actuators, and sensors. It contains two parts [13].

3.1.1

Perception Node

The motor, actuator, and sensors represent the perception node, which detects, gathers, and controls data. Depending on the sensor type and use of the Internet of Things, the data can be blood pressure, sugar level, oxygen level in the blood, or other variables. The sensors sense environment status and get information from the environment. The actuator changes the environment status.

3.1.2

Perception Network

The perception network connects the Internet of Things devices to the network layer. It is also in charge of controlling, connecting, and sharing data with the network layer as well as between devices. One of the most important protocols in the perception network is the data sensor acquisition protocols which enable the interaction between IoT devices in an organized and significant way, such as MQ Telemetry Transport (MQTT).

Internet of Things Overview: Architecture, Technologies, Application, and Challenges

5

3.2 Network Layer The network layer, or “transmission layer,” is responsible for transferring data from the perception layer to the next layer, connecting all IoT devices, and allowing them to share data. Further, it sends data to different IoT gateways and switches, which act as a bridge between numerous IoT devices to facilitate data aggregation and transfers to/from other IoT devices [7]. Through a wired or wireless transmission medium, 4G, 5G, Wi-Fi, ZigBee, etc. are the most commonly used technologies [14]. Network layer networks must be built to support the communication requirements for security, latency, and bandwidth; it can be public, private, or hybrid models.

3.3 Middleware Layer The middleware layer, or “service layer,” offers integration services and application operations into IoT [12]. Usually, it refers to the cloud, which offers a low-cost platform for IoT data storage and analytics [15]. The cloud offers efficient platforms, protocols, and APIs to support IoT applications. The primary functions of the Middleware layer are communications and service administration, in addition to data exchange and storage [12]. The Middleware layer consists mostly of four elements:

3.3.1

Service Discovery

Service discovery uses relationships between different items to locate the ideal service and discover objects that can appropriately provide the desired information and services [12]. There are two types: client-side and server-side. In client-side service discovery, client applications search for services by exploring or querying the service registry, which includes service instances, whereas server-side service discovery allows the clients applications to explore services via a load balancer or router.

3.3.2

Service Composition

Service composition is the ability to dynamically identify and combine component services or collections of services to construct new services to schedule the more convenient services and to obtain the most reliable services to meet demand [12]. Service composition improves the reusability and provides an automatic way of reusing existing services.

6

3.3.3

M. Ahmid et al.

Trustworthiness Management

In order to develop a trustworthy framework, the trustworthiness management phase carefully identifies the confidence mechanisms to analyze and use the data from various service [12]. The trust values may be calculated based on three factors: the previous successful transactions between entities, complaints an entity has accumulated from previous transactions, or the passage of time.

3.3.4

Service APIs

Application Programming Interface (API) services allow an IoT application to communicate with a server-side system in order retrieve and/or change data within it. It also supports interactions between services, resulting in a flexibility and adaptability IoT system.

3.4 Application Layer The application layer, or business layer, is the highest in IoT architecture, and it is visible to the end user. It is primarily responsible for managing applications based on data maintained and addressed by the Middleware layer. Until now, there has been no universal standard for developing and building IoT apps for this layer [16]. Depending on the service provided, the application layer can be structured in several ways. Smart buildings, healthcare monitoring systems, highway monitoring systems, and many other IoT applications have been developed [17]. The various protocols used in the application layer are distributed through various end systems, where the application uses a protocol in one end system for exchanging data packets with another application in another end system [18]. The application layer protocols are several, like HTTPS, DDS, MQTT, Web Sockets, CoAP, XMPP, AMQP, etc. [19].

4 The Required Technologies The Internet of things can be realized in the real world with the integration of several technologies and protocols as shown in Fig. 3. In this section, we present some the relevant enabling technologies.

Internet of Things Overview: Architecture, Technologies, Application, and Challenges

7

Fig. 3 IoT required technologies

4.1 RFID Radiofrequency identification (RFID) is a technology based on the exchange of information using electromagnetic signals. It is a noncontact communication technology, generally used to follow and identify IoT devices without any contact. It supports data exchange over a short distance via radio signals. It uses frequency ranges of 125 kHz for low frequency (LF), 13.56 MHz for high frequency (HF), [433, 860–960] MHz for ultrahigh frequency (UHF), and [2.45, 5.8] GHz for microwave. The RFID system consists of an RFID tag, an RFID reader, and an antenna [20]. An RFID tag is a small chip that has a unique identification number; it is connected to an antenna; each RFID tag is attached to an IoT device. The RFID reader identifies the device and gets the information by querying the RFID tag via the appropriate signals. The RFID antenna is adjusted to cover only a small range of carrier frequencies centered on the RFID system.

4.2 WSN A wireless sensor network (WNS) is an infrastructure composed of computing, sensing, and communication devices that allow an administrator to instrument, observe, and react to events and phenomena in specified environments. WSN can play an important role in the IoT since it can support a significant number of sensor nodes while retaining adequate battery life. RFID and WSN can be used for data

8

M. Ahmid et al.

acquisition in the IoT, but mainly RFID is used for device identification. In contrast, WSN is used for the parameter perception of the environment around them [19].

4.3 RFID Sensor Network RFID sensor network (RSN) is a wireless sensor network and RFID system integration. Integration of these two technologies can extend the area of applications and also offers added value to existing applications. Different implementations of RSN were successfully achieved [21]. For example, the integration of RFID and WSN provides continuous data monitoring throughout the food supply chain, ensuring that retailers meet requirements during product delivery and storage, such as maintaining the temperature and humidity requirements [22].

4.4 NFC Near-field communication (NFC) is based on the technology used for RFID. It provides short-range communication between devices by utilizing the highfrequency 13.56 MHz RFID band. NFC allows communication between two devices without contact at a maximum distance of approximately 20 cm or less [23]. NFC technology enables simple and safe two-way interactions between electronic devices. The incorporation of NFC into consumer electronics products has created prospects for IoT applications such as contact exchange, electronic ticketing, electronic payment, etc.

4.5 Arduino The Arduino Integrated Development Environment (Arduino IDE) is a collection of open-source software and development boards. Arduino development boards may act as a minicomputer by receiving input and modifying the outputs of various electrical devices. Also, it can be easily programmed using the C language, erased, and reprogrammed at any time in the Arduino IDE [24].

4.6 Raspberry Raspberry Pi is similar to Arduino board open hardware, with excepting the essential chip on the Raspberry Pi. It is designed for the Linux operating system, but currently,

Internet of Things Overview: Architecture, Technologies, Application, and Challenges

9

Table 1 Comparison of IoT wireless technologies Network standard Range Frequency band/wavelength

Data protection

RFID IEEE 802.15

NFC ISO/IEC 13157

ZigBee IEEE 802.15.4

6LoWPAN IEEE 802.15.4

Up to 3 m 125 kHz, 13.56 MHz, [433, 860–960] MHz, [2.45, 5.8] GHz Symmetric encryption

Up to 10 cm 13.56 MHz

500 m 868/915 MHz

200 m 868/915 MHz

AES and RSA

AES-128

16-bit CRC and FEC

it has Linux optimized versions, the most popular of which is Raspbian. The Raspberry Pi board has RAM, a processor, a graphics chip, different interfaces and connectors for external devices, as well as support for numerous input and output peripherals [25].

4.7 ZigBee ZigBee is a wireless networking technology that is intended for short-distance communication with low power consumption. It is a high-level protocol for the communication of personal or household devices. It is based on the IEEE 802.15.4 personal wireless network standard (WPAN) and uses low-power digital radio signals. ZigBee transmission distances are between 10 and 100 meters, as shown in Table 1, line of sight, and a specified rate of 250 kbit/s, suitable for intermittent data transfer from input device [26]. Its benefits include energy efficiency, low complexity, low cost, a low data rate, and security. It can use Advanced Encryption Standard (AES)-128 encryption while talking with its peers [27].

4.8 6LowPAN The IPv6 over Low-Power Wireless Personal Area Network (6LoWPAN) system is a wireless mesh network with low power consumption where each node has its own Internet Protocol version 6 (IPv6) address, which allows it to connect directly to the Internet. Based on the IPv6 routing protocol for low-power and lossy networks (RPL), each node chooses its parent. There are three sorts of nodes [28]: a sink node, an intermediate node, and a leaf node.

10

M. Ahmid et al.

5 Applications IoT offers the potential to develop many applications in different fields, of which just a small portion is now available to our society. Soon, the IoT field will have enormous and rewarding growth in applications in the domestic and commercial fields [29, 30]. According to the Precedence Research website, the industrial IoT market will continue to grow through 2030, with the industrial IoT market valued at USD 392.85 billion in 2022 and expected to be worth roughly USD 1742.8 billion by 2030, as shown in Fig. 4.

5.1 Transportation and Logistics Many cities have implemented smart transportation to improve public transportation, reduce infrastructure costs, exchange data on the current traffic situation, and alleviate traffic accidents and congestion. Also, airlines, railroad companies, and public transportation agencies can collect massive amounts of data and process them in the fog/edge computing infrastructure. This enables them to utilize them to enhance their services and give consumers the most efficient and safe flights possible, as well as to minimize their expenditures and boost their profitability. Online reservation sites, in-vehicle Google mapping, and smart cars are some of IoT transportation applications [31]. The Internet of Things played a crucial role in the logistics and supply chain industry as it provided many advantages, including monitoring storage conditions

Market size in USD Billion 1742.8 1446.67 1200.86 996.82

326.1

2021

392.85

2022

473.27

2023

570.14

2024

686.85

2025

Fig. 4 Industrial IoT market size projection [31]

827.44

2026

2027

2028

2029

2030

Internet of Things Overview: Architecture, Technologies, Application, and Challenges

11

such as humidity and temperature, discovering the freshness of the product without looking at specific indicators, discovering storage incompatibility, making arrangements during transport of products to maintain product quality, etc. [18].

5.2 Healthcare A lot of advancements and significant changes are occurring in the field of IoT healthcare; like connected things anywhere, anytime, with anyone correctly using any network and any service. The network of embedded or worn sensors on the body collects vital information on patient health [32], which may be used to create beneficial changes in the healthcare area. The goal of the IoT application in healthcare is to facilitate the examination and monitoring of patients and exit the traditional area of hospital visits and long waits. By applying the IoT in healthcare, the medical staff can sense and treat the behavioral, biological, and social characteristics exclusive to the patient [33] so that they can work on them. Also, it reduce medical visits and healthcare costs by monitoring the health status of patients remotely and from anywhere and at any time; this leads to excellent results while making healthcare cost-effective. Many applications and devices are designed for healthcare purposes, such as blood glucose monitoring systems for elderly people and smart baby sleep temperature monitors. However, the IoT impact on healthcare is still in its initial development phases [34].

5.3 Smart Environments The realization of smart environments (SEs) has been an ongoing hot topic since its inception. The smart environment is an environment rich in sensing, actuation, operating, communication, and computation capabilities that aim to gain knowledge about the environment and exploit it to adapt to the preferences and needs of its residents. For these reasons, SEs collect information, process it, and act on it; different types of smart environments do so at different degrees. Moreover, the various areas come with various requirements and then technical choices, which affect the tactics of how and where to process data and how to act based on information in a specific context [35]. As shown in Fig. 5, there are various IoTbased smart environments. We can categorize and classify based on the application domain. The following are the primary categories:

5.3.1

Smart Grid

The smart grid is an electricity grid that intelligently integrates the procedures of all connected users, including generators, machines, and consumers, in order to

12

M. Ahmid et al.

Fig. 5 Smart environments applications

provide a sustainable, safe, and economical electrical supply. While the traditional grid can only transfer or distribute electrical power. The smart grid uses many types of advanced technologies, digital computing services, and communications in the energy system infrastructure [35].

5.3.2

Smart Cities

The smart city is a technology-intensive city that can provide the collection, analysis, and distribution of information to improve services provided to citizens, meet their requirements, increase operational efficiency, and help make better decisions at the municipal level. For this reason, it is equipped with various electronic components and applications for monitoring systems, transmission systems, street lighting systems, etc. [36].

5.3.3

Smart Home

Smart homes are a network of electronic devices, sensors, software, installed detection and control devices, and network communication within the home. In smart homes, all devices communicate with each other smoothly [37] in order to control air conditioning, ventilation, heating, lighting, hardware, monitoring, and safety systems. Smart homes have changed human lives by providing comfort, security, and recreation while lowering energy use [38].

5.3.4

Smart Industry

The term “smart industry” or “Industry 4.0” refers to an industry that integrates ubiquitous sensing capabilities with industrial infrastructure for automating different industrial operations, like production monitoring, material detection, tracking,

Internet of Things Overview: Architecture, Technologies, Application, and Challenges

13

and many more. In the future, as urban populations rise and natural resources become scarce, the smart industry is expected to contribute significantly to maintaining the quality of life [39].

5.3.5

Smart Building

In order to adapt to the daily lives of users or residents, the development of energy management systems for network systems, monitoring and protection systems, etc. makes smart buildings integrate smart information infrastructure into the building’s architecture. It is outfitted with a large number of sensors, controls, operators, and applications to do this [39]. Its applications include smart parking, conference room scheduling, logistical tracking, energy, water efficiency system, motion detectors installed at the lift lobby, etc.

6 The Challenges and Problems Like any new technology, some issues are revealed. This section outlines and discusses the main challenges and issues related to the Internet of Things that must be addressed to accommodate the trillions of IoT devices.

6.1 Security IoT solutions must be secure to protect sensitive data and vital physical infrastructure. However, the broad deployment of IoT devices and the special nature of the data collected and transmitted by these devices have produced significant security issues. In order to understand these challenges, we briefly discuss the most important IoT security challenges.

6.1.1

Authentication

Authentication techniques are essential for the full and the broad deployment of IoT, and they are also a critical factor for encouraging people to use new technologies and giving them safe access to different IoT resources. To ensure authentication, all IoT devices must be able to authenticate other objects easily and use effective authentication techniques to establish a reliable and trusted connection. These actions can be challenging due to the IoT structure and entities (devices, clients, services, etc.) [40].

14

6.1.2

M. Ahmid et al.

Authorization

Authorization is the specification and limits access rights to different resources, only for authorized resources by using access control mechanisms. Each IoT node may only support some access verification procedures. That may differ from node to other on the same network. Therefore, deploying and managing a variety of authorization and access control mechanisms designed explicitly for limited node capabilities, presents a huge challenge. Another issue is that IoT users should know about the data management procedures that will be used and guarantee that data is secure throughout the process [41].

6.1.3

Privacy

Services provided by IoT applications offer excellent benefits for human life. However, they can come at a high cost in terms of a person’s privacy, which remains a major concern for the IoT because IoT devices not only gather personal information but also monitor users’ actions and habits without their knowledge. Moreover, questions arise regarding user privacy regarding the external companies that manage, create, track, or organize these devices and their data [42].

6.1.4

Confidentiality

Secrecy or confidentiality of data is an access limitation to information except for authorized parties and prevents disclosure of information to unauthorized parties by using various mechanisms. For example, the Internet of Things network should not disclose sensor readings to its neighbors unless it is configured and designed to do so. If personal data become public, confidentiality and hence privacy are lost. Data confidentiality must be guaranteed in the full data life cycle [43].

6.2 Cloud The Internet of Things applications are helping to speed up the growth of data. According to Statista’s research department, 175 zettabytes of data will be generated by 2025, as shown in Fig. 6. It necessitates huge storage and processing capacity, which is not available on IoT devices, driving the usage of the cloud not just by individuals but also by companies. Because the cloud has limitless resources; provides on-demand, easy, and scalable network access; and may be extremely useful to IoT, integration of both technologies is required, and this integration is named the “Cloud of Things” (CoT) [44]. However, the CoT has many issues that must be addressed. A high data transfer rate is required to handle the vast amount of data created from IoT devices and send

Internet of Things Overview: Architecture, Technologies, Application, and Challenges

15

Data volume in zettabytes 175 129.5 101 50.5

2020

64.5

2021

79.5

2022

2023

2024

2025

Fig. 6 Annual size of global datasphere

it to the cloud. As a result, obtaining appropriate network performance to transfer data to cloud environments becomes a critical problem; also, broadband growth is not consistent with the evolution of storage and computing. In addition to that, the full performance of the application largely depends on the characteristics of this data management service, and finding an ideal data management solution that allows CoT to manage vast amounts of data is still a big problem [45]. In addition to that, real-time applications are susceptible to the efficiency of performance because the timing may be affected by unforeseen matters, so services and data provided must be achieved with high efficiency. The network at CoT must be flexible enough to accommodate different sorts of data and services based on their needs. Also, timesensitive data should be stored in the nearest physical location of the user, so that the delay is reduced, as the storage location is also essential for sensitive services.

6.3 Communication In complex and heterogeneous systems such as the Internet of Things, communication links must fulfill demanding throughput, range, and latency requirements while maintaining high levels of security and sticking to a limited energy budget [46]. The number of continuous contracts increases day by day and is expected to reach billions during the next few years. Regulating such a large number of devices is a significant problem, particularly in terms of efficiently regulating access to devices on the channel with little cost. Energy autonomy is a critical difficulty in wireless IoT systems. This requirement grows as the number of devices rises, as support for regular battery swaps becomes impossible.

16

M. Ahmid et al.

6.4 Intelligence Currently, billions of IoT devices are connected to the Internet, which allows them to communicate and exchange information, thus creating new services and applications. While the entire linked loT system demonstrates the ability to communicate intelligently and make decisions, it does so according to certain predefined business principles and without regard for unexpected environmental changes. We may say the entire system is intelligent, but not the IoT devices, because most of them, if they are disconnected from the system or the Internet, cannot make the right decisions in critical cases. They are unable to reason in their settings or make wise and timely judgments and decisions to achieve their objectives. The connected devices must become smart and be able to make decisions in an organized way to make the network self-configurable, self-organized, and self-adaptive [4]. Artificial intelligence (AI) technology has the potential to address difficult issues that demand rapid and intelligent response. Integrating the Internet of Things with artificial intelligence will result in a strong technology, with the extensive analytical capabilities of artificial intelligence, efficiently analyzed the IoT data to obtain valuable information. Also, artificial intelligence may assist IoT devices in intelligently interacting with humans and other objects. A smart system can collect data, represent, reason about, and interpret it. It can also learn and extract patterns and meanings, derive new information, learn from experience, and develop intelligent strategies and behaviors. Also, some smart processes include behavioral detection, data mining, a rule-based analytical process, a recognition system, recommendation, and optimization solutions.

7 Conclusion IoT has radically changed the way we live. It has added a new communications dimension by allowing communications with or between things, thereby starting the “any things” vision. In this chapter, we have covered the main concepts and key points of IoT technology. First, we have presented the origin of this technology, followed by the Internet of Things architecture. Also, we have presented the required technologies and some IoT application areas. Finally, we have studied the critical challenges of the IoT by emphasizing what the facts are and what are the issues that require a more thorough search. Although there are some challenges, the Industrial Internet of Things market continues to grow especially in manufacturing/industrial applications, followed by transportation/mobility and energy IoT applications. Declarations Conflict of Interest: None Funding: None Availability of data and materials: None Code availability: None

Internet of Things Overview: Architecture, Technologies, Application, and Challenges

17

Authors’ contributions: This chapter delivers an overview of the Internet of Things, tracing the concept of IoT from its origins to the present day with periodic updates. The detailed investigation of IoT functional blocks at each layer and the related issues and various communication techniques used in IoT technology have been thoroughly examined, as well as the future of IoT applications’ directions and market growth. IoT challenges have been discussed in the context of enhancing IoT application markets so that the reader may get a general understanding of IoT. All authors read and approved the final manuscript.

References 1. M. Ahmid, O. Kazar, S. Hamida, L. Kahloul, P. Ghoudous, Internet of things new challenges in distributed artificial intelligence. Int. J. Comput. Sci. Electron. Eng. 3(5), 375–377 (2015) 2. G. Gardaševi´c, M. Veleti´c, N. Maleti´c, D. Vasiljevi´c, I. Radusinovi´c, S. Tomovi´c, M. Radonji´c, The IoT architectural framework, design issues and application domains. Wirel. Pers. Commun. 92(1), 127–148 (2017) 3. B.L.R. Stojkoska, K.V. Trivodaliev, A review of Internet of Things for smart home: Challenges and solutions. J. Clean. Prod. 140, 1454–1464 (2017) 4. M.S. Gal, D.L. Rubinfeld, Data standardization. NYUL Rev. 94, 737 (2019). https://doi.org/ 10.2139/ssrn.3326377 5. J. Romkey, Toast of the IoT: The 1990 interop internet toaster. IEEE Consum. Electron. Mag. 6(1), 116–119 (2016) 6. M. Kayyali, Internet of Things (IoT): Emphasizing its applications and emergence in environmental management—The profound cases, in Environmental Informatics, (Springer, Singapore, 2022), pp. 201–212 7. Q.F. Hassan, S.A. Madani, Internet of Things: Challenges, Advances, and Applications (Chapman and Hall/CRC, 2017), Boca Raton, Fla. United States 8. C. Brewster, I. Roussaki, N. Kalatzis, K. Doolin, K. Ellis, IoT in agriculture: Designing a Europe-wide large-scale pilot. IEEE Commun. Mag. 55(9), 26–33 (2017) 9. A. Ghimire, Brief survey and testbed development for blockchain based Internet of Things, Doctoral dissertation, The University of Mississippi, 2020 10. T. Yousuf, R. Mahmoud, F. Aloul, I. Zualkernan, Internet of Things (IoT) Security: Current status, challenges and countermeasures. Int. J. Inform. Secur. Res. 5(4), 608–616 (2015) 11. L. Atzori, A. Iera, G. Morabito, Understanding the Internet of Things: Definition, potentials, and societal role of a fast evolving paradigm. Ad Hoc Netw. 56, 122–140 (2017) 12. M. Ahmid, O. Kazar, A comprehensive review of the internet of things security. J. Appl. Secur. Res. 18(3), 289–305 (2021) 13. A.A. Laghari, K. Wu, R.A. Laghari, M. Ali, A.A. Khan, A review and state of art of Internet of Things (IoT). Arch. Comput. Methods Eng., 1–19 (2021) 14. S.L. Peng, S. Pal, L. Huang, Principles of Internet of Things (IoT) ecosystem: Insight paradigm (Springer, 2020). Midtown Manhattan, New York City, United States 15. G. Ambika, D.P. Srivaramangai, A study on security in the Internet of Things. Int. J. Sci. Res. Comput. Sci. Eng. Inform. Technol. 5(2), 12–21 (2017) 16. O. Yousuf, R.N. Mir, A survey on the Internet of Things security. Inform. Comput. Secur. 27(2), 292–323 (2019) 17. F.J. Ferrández-Pastor, J.M. García-Chamizo, M. Nieto-Hidalgo, J. Mora-Martínez, Precision agriculture design method using a distributed computing architecture on internet of things context. Sensors 18(6), 1731 (2018) 18. J. Dizdarevi´c, F. Carpio, A. Jukan, X. Masip-Bruin, A survey of communication protocols for internet of things and related challenges of fog and cloud computing integration. ACM Comput. Surv. (CSUR) 51(6), 1–29 (2019)

18

M. Ahmid et al.

19. J.A. Scheibmeir, Y.K. Malaiya, Social media analytics of the Internet of Things. Discover Internet Things 1(1), 1–15 (2021) 20. D.B. Baxter, J.D. Scalzo, J. Ram, A. Sitar, R. Das, U.S. Patent No. 10,087,750. Washington, DC: U.S. Patent and Trademark Office, 2018 21. A. Lozano-Nieto, RFID Design Fundamentals and Applications (CRC Press, 2017), Boca Raton, Florida, United States 22. K.M. Modieginyane, B.B. Letswamotse, R. Malekian, A.M. Abu-Mahfouz, Software defined wireless sensor networks application opportunities for efficient network management: A survey. Comput. Electr. Eng. 66, 274–287 (2018) 23. S. Olenik, H.S. Lee, F. Güder, The future of near-field communication-based wireless sensing. Nat. Rev. Mater. 6(4), 286–288 (2021) 24. J. Boxall, Arduino Workshop: A Hands-on Introduction with 65 Projects (No Starch Press, 2021) San Francisco, California, United States 25. J.W. Jolles, Broad-scale applications of the Raspberry Pi: A review and guide for biologists. Methods Ecol. Evol. 12(9), 1562–1579 (2021) 26. S. Madakam, R. Ramaswamy, S. Tripathi, Internet of Things (IoT): A literature review. J. Comput. Commun. 3(05), 164 (2015) 27. P.I.R. Grammatikis, P.G. Sarigiannidis, I.D. Moscholios, Securing the Internet of Things: Challenges, threats and solutions. Internet of Things 5, 41–70 (2019) 28. H.A. Al-Kashoash, M. Hafeez, A.H. Kemp, Congestion control for 6LoWPAN networks: A game theoretic framework. IEEE Internet Things J. 4(3), 760–771 (2017) 29. G.C. Hillar, MQTT Essentials-A Lightweight IoT Protocol (Packt Publishing Ltd, 2017), Birmingham, England 30. M. Ahmid, O. Kazar, S. Hamida, L. Kahloul, P. Ghodous, Internet of Things new challenges in distributed artificial intelligence, in 2015 International Conference on Future Computational Technologies (ICFCT’2015), (2015, December), pp. 158–162 31. Precedence Research, Packaging automation market size to surpass USD 128.2 Bn by 2030 (2022, October 11). https://www.globenewswire.com/en/news-release/2022/11/17/2557889/0/ en/Packaging-Automation-Market-Size-to-Surpass-USD-128-2-Bn-by-2030.html 32. M. Ahmid, O. Kazar, S. Benharzallah, L. Kahloul, A. Merizig, An intelligent and secure health monitoring system based on agent, in 2020 International Conference on Informatics, IoT, and Enabling Technologies (ICIoT’ 2020), (IEEE, 2020), pp. 291–296 33. M. Ahmid, O. Kazar, A cloud-IoT health monitoring system based on smart agent for cardiovascular patients, in 2021 International Conference on Information Technology (ICIT), (IEEE, 2021), pp. 1–6 34. M. Ahmid, O. Kazar, L. Kahloul, A secure and intelligent real-time health monitoring system for remote cardiac patients. Int. J. Med. Eng. Inform. 14(2), 134–150 (2022) 35. C. Gomez, S. Chessa, A. Fleury, G. Roussos, D. Preuveneers, Internet of things for enabling smart environments: A technology-centric perspective. J. Ambient Intell. Smart Environ. 11(1), 23–43 (2019) 36. A.H. Sodhro, S. Pirbhulal, Z. Luo, V.H.C. de Albuquerque, Towards an optimal resource management for IoT based Green and sustainable smart cities. J. Clean. Prod. 220, 1167–1179 (2019) 37. N. Javaid, A. Sher, H. Nasir, N. Guizani, Intelligence in IoT-based 5G networks: Opportunities and challenges. IEEE Commun. Mag. 56(10), 94–100 (2018) 38. T. Hargreaves, C. Wilson, R. Hauxwell-Baldwin, Learning to live in a smart home. Build. Res. Inf. 46(1), 127–139 (2018) 39. M.S. Sruthi, IOT based real time people counting system for smart buildings. Int. J. Emerg. Technol. Innov. Eng. 5(2) (2019) 40. M. El-hajj, A. Fadlallah, M. Chamoun, A. Serhrouchni, A survey of internet of things (IoT) Authentication schemes. Sensors 19(5), 1141 (2019) 41. F.A. Alaba, M. Othman, I.A.T. Hashem, F. Alotaibi, Internet of Things security: A survey. J. Netw. Comput. Appl. 88, 10–28 (2017)

Internet of Things Overview: Architecture, Technologies, Application, and Challenges

19

42. M. Yamin, Y. Alsaawy, A.B. Alkhodre, A. Sen, A. Ahmed, An innovative method for preserving privacy in Internet of Things. Sensors 19(15), 3355 (2019) 43. M.A. Razzaq, S.H. Gill, M.A. Qureshi, S. Ullah, Security issues in the internet of things (IoT): A comprehensive study. Int. J. Adv. Comput. Sci. Appl. 8(6), 383–388 (2017) 44. H. Atlam, R. Walters, G. Wills, Internet of things: State-of-the-art, challenges, applications, and open issues. Int. J. Intell. Comput. Res. 9(3), 928–938 (2018) 45. H.F. Atlam, R.J. Walters, G.B. Wills, Fog computing and the internet of things: A review. Big Data Cogn. Comput. 2(2), 10 (2018) 46. A. Burg, A. Chattopadhyay, K.Y. Lam, Wireless communication and security issues for cyber– physical systems and the internet-of-things. Proc. IEEE 106(1), 38–60 (2017)

Dr. Maroua Ahmid received a PhD degree in artificial intelligence from the University of Mohamed Khider Biskra (UMKB) in 2021, where he was a member of the Smart Computer Sciences Laboratory. Dr. Ahmid has published over ten journals and conference papers. Dr. Ahmid acts as a reviewer in several journals and conferences. Her areas of expertise include cybersecurity, privacy, Internet of Things, Internet of Health Things, Artificial Intelligence, mobile application development, Intelligent Systems, and Cloud Computing. Prof. Okba Kazar Obtained his engineer diploma in 1987 at Constantine University (Algeria) and Magister degree in 1997 followed by PhD degree from the same university in 2005. He is member of editorial board of some international journals and author of more than 370 publication in international journals and conferences. He participates as a member of program committee and co-chair for international conferences. He is interested and working in artificial intelligence field and multi-agents systems with their applications and also advanced information systems, web services, semantic web, Big data, IoT, robotics, cloud computing, and information security. He is a Full Professor since 2011 at Computer Science Department of Biskra University. He was a visiting professor at the United Arab Emirate University for two years and half, now he is at the University of Kalba at Sharjah. Dr. Ezedin Barka is currently an Associate Professor at the United Arab Emirate University. He received his PhD in Information Technology from George Mason University, Fairfax, VA in 2002, where he was a member of the Laboratory for Information Security Technology (LIST). His current research interests include Access Control, where he published a number of papers addressing delegation of rights using RBAC. Other research areas include digital rights management (DRM), large-scale security architectures and models, trust management, security in UAVs, and network “wired and wireless” and distributed systems security. Dr. Barka has published over 50 Journals and conference papers. Dr. Barka is an IEEE member, a member of the IEEE Communications Society, and a member of the IEEE Communications and Information Security Technical Committee (CISTC). He serves on the technical program committees of many international IEEE conferences such as ACSAC, GLOBECOM, ICC, WIMOB, and WCNC. In addition, he has been a reviewer for several international journals and conferences.

IoMT Applications Perspectives: From Opportunities and Security Challenges to Cyber-Risk Management Sondes Ksibi, Faouzi Jaidi, and Adel Bouhoula

1 Introduction Countries face a significant increase in chronic diseases due to an aging population and changing living habits. Therefore, healthcare expenditures are rising and becoming unbearable. The emergence of IoT brings a paramount solution to a large scale medical care in a cost effective way. Hence, a growing demand on IoTbased home healthcare has been reported [1]. Innovative companies aligned their product portfolios to this demand by proposing smart devices such as smart watches, electrocardiograms, smart thermometers, etc. As a consequence, the number of IoT connections in healthcare increased significantly. As per statistics, in the European Union, it was at 2.79 million in 2019 and expected to reach 10.34 million connections by 2025 [2]. The rapid expansion of connected health is also driven by a great market opportunity, in fact according to All The Research [3], the global IoMT market was valued in 2018 at $44.5 billion and is expected to raise to $254.2

S. Ksibi () University of Carthage, Higher School of Communications of Tunis (Sup’Com), Digital Security Research Lab, Tunis, Tunisia e-mail: [email protected] F. Jaidi University of Carthage, Higher School of Communications of Tunis (Sup’Com), Digital Security Research Lab, Tunis, Tunisia University of Carthage, National School of Engineers of Carthage, Tunis, Tunisia e-mail: [email protected] A. Bouhoula Department of Next-Generation Computing, College of Graduate Studies, Arabian Gulf University, Manama, Kingdom of Bahrain e-mail: [email protected] © The Author(s), under exclusive license to Springer Nature Switzerland AG 2024 W. Boulila et al. (eds.), Decision Making and Security Risk Management for IoT Environments, Advances in Information Security 106, https://doi.org/10.1007/978-3-031-47590-0_2

21

22

S. Ksibi et al.

billion in 2026. This report shows that the smart wearable device segment of IoMT (including smart watches and sensor-laden smart shirts) made up the largest share of the global market in 2018 at roughly 27%. Moreover, the IoMT area is poised for even further growth as Artificial Intelligence (AI) is integrated into smart devices with improvement of real-time capabilities, remote measurement, and data analysis. IoMT interconnects medical devices, care services, and software applications. It is a highly distributed intelligent system for medical information exchange. These connected devices offer time saving, mobility, and flexibility to patients by permanently monitoring their vital signs and well-being. From fitness tracking to complex surgeries, IoMT is providing more accurate and timely results [4]. However, IoMT is vulnerable to many cyber-attacks because medical devices lack security or are poorly secured against various attacks. Cyber-attacks can have severe consequences on patients’ safety and privacy and, therefore, decelerate the large scale deployment of IoMT. Privacy preserving is a critical issue because of the sensitivity of health related data. The challenge consists of providing a high level of security in resource-limited devices. Many industrials and researchers worked on securing connected medical devices. The aim is to improve trustworthiness and promote the adoption of the smart health paradigm. Risk assessment and mitigation were proposed as efficient solutions to evaluate the impact of cyber-security attacks on IoMT nodes and users. In this context, the aims of the current research paper are as follows. Throughout the paper, we present and discuss basic IoMT applications opportunities and challenges. We deeply study and synthesize security issues within IoMT applications. To handle the aforementioned issues, we discuss the effectiveness of risk management solutions and highlight factors of risk assessment and analysis for IoMT systems. Based on the conducted study, a framework to reinforce trustworthiness between IoMT communicating nodes is then proposed. Our proposal aims to help with decision making based on a distributed risk management approach. The reminder of this paper is organized as follows. In Sect. 2, we present a synthesis of e-health security challenges. In Sect. 3, we discuss related works. In Sect. 4, we outline the idea of our proposal for risk management in distributed ehealth systems. Finally, in Sect. 5, we conclude the paper and highlight future works.

2 IoMT Cyber-Security Challenges 2.1 IoMT Security Requirements Data handled in an IoMT environment is precious for attackers that is why the risk landscape is getting wider with the widespread development of the technology. Patients’ data is at high risk and their lives also in certain cases. As reported in [5] in 2017, the healthcare industry was targeted by “of almost 32,000 intrusion attacks per day.” Therefore, securing IoMT communication has become an issue of paramount

IoMT Applications Perspectives: From Opportunities and Security Challenges. . .

23

concern to the healthcare industry. Medical information requires a high level of protection because it is directly related to patients’ health and safety. Exchanged data in medical systems is fragmented and shared between health service providers over a very heterogeneous infrastructure. Hence the security of data in such systems is quite complex. Moreover, in emergency events, data arrangement needs to be done quickly and in a trustworthy manner. On the other side, healthcare professionals are disinclined to use ICT (Information and Communication Technologies), especially open networks and systems, due to the related security risks. Indeed, security mechanisms should encompass the layered architecture of an IoMT network [6]. Security of IoMT, similarly to other computing systems, targets the CIA triad to protect the system hardware, software, communications, and data. The CIA triad refers to the (i) Confidentiality: secrecy and access restrictions to data, (ii) Integrity: maintaining data trustworthiness and coherence, and (iii) Availability: ensuring data and resources accessibility when needed.

2.2 Technical Issues Security and privacy are very challenging for an e-healthcare system since it has to deal with sensitive and private data. A lot of works have shown that security shortcomings in connected medical devices systemically affect patient’s health and safety [7–9]. In literature, security issues in healthcare have been addressed from diverse perspectives: devices embedded security functions, networks issues, system architecture and protocol stack security, malware detection and prevention, etc. Moreover, in [10], the authors highlighted that there is a lack of governance mechanisms and standards, regulations and laws as well as industry best practices. This has led to great difficulties in the implementation of basic security requirements. So, the need for security risk management is prevalent. The security risk in Information System (IS) corresponds to the risk that occurs due to loss of data or systems confidentiality, integrity, or availability, defined as main security requirements. This risk considers potential adverse impacts to the organization (including assets, mission, functions, image, or reputation), users, other organizations, and the country [11]. IoMT environments are populated by heterogeneous devices that can be manipulated by different stakeholders. These environments might be exposed to cyber risks. Many vulnerabilities are viable with the connected devices (i.e., sensors, wearables, and actuators) used in the IoMT systems since: (a) traditional security mechanisms are not suitable because of the resource constrained nature of the devices notably wearables and implantable ones [12]; (b) security is not supported by design within on-site medical equipment [13]; (c) the diversity of stakeholders with different security objectives adds a new security challenge to IoMT; (d) a boundless diversity in use cases makes the security issue crucial; (e) a lack of standards leading to a providers proprietary security measures [13]; as well as (e) the security issues linked to firmware of the medical devices. Hence the attack surface with networked medical devices is being widened.

24

S. Ksibi et al.

Fig. 1 IoMT vulnerabilities mapped to layers

Exploitable vulnerabilities of IoMT are in three main parts (layers) of the system: • Things layer or perception layer is composed typically of medical devices (things) with sensing capabilities and transmission interfaces. They gather health data and send it to higher layers. Regarding their placement medical things are classified to wearable, implantable, ambient, and stationary devices. • Data transfer layer ensures the transmission of the collected data to various destinations relying on communication protocols (BLE, Wi-Fi, 4G/5G, etc.). • Application layer processes the data and presents it in customized views or stores it for further analysis. Figure 1 shows some of the security vulnerabilities in the three layers. In several research works, the authors have demonstrated that connected medical devices lack efficient security mechanisms and they are vulnerable to various types of attacks. Table 1 presents some of demonstrated attacks.

2.3 Regulatory Issues In connected medical devices industry, manufactures claim that their products are compliant with standard protocols and rules from the design phase. Since devices are manufactured at a large scale and are of high risk to the patient, governmental regulatory authorities are competent to verify and validate their compliance. The Health Insurance Portability and Accountability Act (HIPAA), the General Data Protection Regulation (GDPR), and the Food and Drug Administration (FDA) are European and US authorities that aim to regulate privacy and security of medical devices. FDA conducts risk analysis and categorizes medical devices in three classes according to their associated risks. Like presented in Fig. 2, FDA categorizes

IoMT Applications Perspectives: From Opportunities and Security Challenges. . .

25

Table 1 Examples of risk medical devices attacks Research work [14] [15] [16]

[17]

[18]

Medical device Implantable medical devices (IMDs) Fitness tracker ´ Withings activité´ IMD blood glucose monitor, the Dexcom G4 Implantable CardioverterDefibrillator (ICD)

Vulnerability Device hardware

Demonstrated attack Battery-depletion

Impact High

Device software

Attack on firmware update User tracking and data alteration

Medium

High

Wearable smart watches

Device software

Reverse engineering the communication protocol between the device and the programmer to carry out DOS attack Traffic sniffing

Device hardware/software Device software/hardware

High

Low

Fig. 2 FDA medical devices classification

medical devices in 3 classes according to their associated risks. Class III regroups around 10% of medical devices of the highest security risks [19]. FDA works on updating its pre-market cyber-security requirements for medical devices. It supervises devices security incidents in order to keep patients safe. Guidelines are released and manufacturers are requested to follow the recommendations during the design phase to safeguard medical devices against cyber-security vulnerabilities. FDA recommends the implementation of a cyber-security risk management program for pre-market and post-market lifecycle steps. The National Institute of Standards and Technology (NIST) defines cyber-security standards for connected devices. NIST publishes guidelines regarding to security of connected devices. It provides a list of controls to deal with security vulnerabilities. The Inter-

26

S. Ksibi et al.

national Organization for Standardization (ISO) as a standardization organization publishes risk management standards applicable to medical devices. Although the evolution of regulation and risk analysis guidelines for IoMT, many applications and devices on the market that handle medical data do not fall inside the purview of the FDA. Manufacturers are facing problems to implement robust security algorithms and follow standards when designing tiny and resource-limited devices. Moreover, in large scale deployment, real-time risk assessment is a complex task.

3 Related Works IoMT is a flourishing surface for medical care applications. Nevertheless, it is also creating enormous and various security threats. The design of novel automated risk assessment mechanisms would effectively mitigate harmful impacts on patients’ safety and privacy. Risk analysis uses data and resources characteristics to measure the associated danger with medical devices and estimate the potential consequences. Trust and risk assessment was proposed in industry and academia to enhance security within IoMT. In the following, we discuss risk management standards, guidelines, and best practices and review related models and approaches defined in literature.

3.1 Standards, Guidelines, and Best Practices ISO has published standards and guidelines for medical devices risk assessment ISO 14971 [20]. It aims to support manufacturers to evaluate the medical devices associated risks, apply mitigation controls, and monitor the effectiveness of these controls. The standard uses qualitative impact metrics. The National Institute of Standards and Technology (NIST) published SP 800-30 [21] to offer guidance for risk assessment conducting and mitigation controls developing. In its special publication SP 1800-8 [22], NIST focused on security of wireless infusion pumps and their risk mitigation. The Open Web Application Security Project (OWASP) [23] gives a checklist of common vulnerabilities that should be verified for medical devices. The Factor Analysis of Information Risk Institute (FAIR) [24] aims to establish a standard reference based on a quantitative approach for the risk assessment. It is not based on consensus, but it promotes commercial software such as RiskLens [25] and CyVaR [26]. Although they are based on quantitative assessment approach, they represent black-box tools. The MITRE’s Common Vulnerability Scoring System (CVSS) [27] performs vulnerability severity assessment for medical devices based on mathematical approximations that translate expert’s opinions to numerical scores. Several risk assessment frameworks rely on CVSS to compute the impact of system vulnerabilities.

IoMT Applications Perspectives: From Opportunities and Security Challenges. . .

27

Table 2 Summary of risk assessment and management standards Organization ISO

Type Standard

NIST

Standard

OWASP FAIR MITRE

Standard Standard Guide

Scope Risk analysis and risk management Risk analysis and risk management Risk management Risk analysis Risk analysis

Approach Qualitative and quantitative Guidelines

Application Pre and post deployment

Qualitative Quantitative Qualitative

Post deployment Pre and post deployment Pre-deployment

Pre and post deployment

Like depicted in Table 2, most of the publications are guidelines and best practice guides addressing especially post-deployment risks. They mainly provide guidance about how to conduct risk assessment and use qualitative scales to access the impact of cyber threats on medical devices (i.e., high, medium, and low).

3.2 Review of Existent Solutions Numerous research works focused on trust and risk assessment within e-health, IoT, and IoMT applications. Nurse et al. in [28] discussed the applicability of traditional security risk assessment approaches and methodologies in the context of IoT applications. The authors highlighted that traditional solutions are not adequate with the context of IoT and emphasized the necessity of new approaches for IoT risk assessment. Radanliev et al. in [29] introduced a quantitative approach, which relies on the coupling of the MicroMort (MM) and Value-at-Risk (VaR) models for the assessment of the economic impact of IoT cyber-risk. MM is a technique used to assess risk of death; it was adapted to evaluate the statistical value of preventing fatality. VaR allows assessing a financial loss for a project. It was adapted to evaluate monetary losses of IoT cyber risks over a period of time. Malik et al. in [30] focused on security vulnerabilities identification and mitigation within IoT. The authors worked to list common vulnerabilities based on a smart software vendor to provide possible mitigating solutions. In [31], Akinrolabu et al. introduced a model for quantitative risk assessment of SaaS applications. The proposal identifies weak links in the provider supply chain, evaluates its security risks, and presents the risk value in monetary terms. Stine et al. in [32] proposed a risk scoring system for medical devices based on the potential harm to the patient. The proposed system considers a doctor’s worst case assessment and rely on STRIDE (a model developed by Microsoft for threats classification) to generate the risk scores. Low cost, ease of use, and uncomplicated results are the main objectives of this system. In [33], medical risk scenarios are modeled based on a static fault tree analysis. Then a Bayesian inference was used to dynamically quantify the risk of medical devices over the time. The model aims to capture the impacts of factors in case of an adversary event. A dynamic model for IoT risk assessment was presented in [34].

28

S. Ksibi et al.

The proposed risk assessment method is based on Artificial Immune System (AIS). Distributed attack detectors use the immune mechanisms to identify attacks and an agent quantifies the device risk score based on the detected attacks.

3.3 Discussions Despite the considerable research efforts, IoMT applications security assessment is still a complex and challenging task. With regard to several contextual and resources constraints, well established risk assessment approaches cannot be directly applied and simply used in the IoMT context. Therefore, new context-aware approaches and models to assess IoMT applications risks are required. Main research works for risk assessment within an IoMT context deal mainly with the assessment of the financial impact of cyber-security threats. They often focus on post-deployment risks and target device providers and manufactures. Moreover, from an access control perspective, they fail to handle both aspects: assessing risks of access requests and risks associated with policies critical defects, anomalies, and attacks. In the following, we introduce a dynamic and comprehensive approach to deal with the discussed limitations and respond to new needs.

4 The Security Risk Management Approach 4.1 Principle IoMT environments are characterized by their ubiquity, scarcity of computing resources, dynamic infrastructures, and changing threat models. This involves a variety of contexts for managing security risks and strategies for risk mitigation. IoMT communicating nodes are prone to cyber threats due to various vulnerabilities. Intrinsic defects in medical device coupled with data networking and processing components result in a wide attack surface. To deal with this problematic, we rely on a dynamic, modular, and quantified risk management approach [6]. Our proposal performs static and dynamic risk assessment. Based on a set of contextual information and devices capabilities, the approach provides a risk estimation in pre-deployment scenarios. For post-deployment risk analysis, an overall risk is accumulated from the other parts. To do so, three main areas are considered: data acquisition area, data transmission area, and finally data analysis and storage area [35]. The main objectives are: (i) Establishing a fine-grained risk management framework based on contextual and specific risk metrics, qualifiers, thresholds, factors, etc., (ii) Evaluating the cumulative risk for a global IoMT service delivery process, (iii) Automating the update procedure for risk mitigation response, and (iv) combining pre-deployment and post-deployment risk assessment.

IoMT Applications Perspectives: From Opportunities and Security Challenges. . .

29

Fig. 3 Dynamic-quantified risk management approach

The approach, like presented in Fig. 3, consists of three basic subsystems (agents) and a centralized module. The Device Risk Manager (DRM) performs risk management in the data acquisition layer; it deals with risk factors related to devices and their computing capabilities. The Network Risk Manager (NRM) is an agent for the communication protocols and connectivity. The Storage and Processing Risk Manager (SPRM) focuses on the components responsible for data storage and processing (mainly the cloud or local databases). The central unit called Core Risk Manager (CRM) performs an end-to-end risk assessment via collecting information from the other subsystems. It performs risk classification, prioritization and correlates results in order to refine the decision making process. The CRM updates the elementary databases of the other units if optimized metrics are obtained. The subsystems can either: (i) make decisions autonomously regarding specific pre-defined risk thresholds or (ii) delegate the decision making to the orchestrator which has a global vision of the whole system.

4.2 Risk Vectors Generally, smart devices, wearable devices, and sensor nodes are resource constrained because of size and mobility. IoMT rely on several nodes to accomplish the healthcare service. Many stakeholders with variable security awareness levels are involved. As a consequence, the capability of organizing different components/endpoints that should operate in a coherent way to deliver applications of

30 Table 3 Examples of potential attacks

S. Ksibi et al. Layer Physical

Network

Application

Some potential attacks Side channels Tag cloning Fault injection Tampering Sensor tracking Poisoning ... Eavesdropping Man in the middle Rogue access Replay DoS Sinkhole ... SQL injection XSS injection Account hijacking Ransomware Brute force ...

interest is of the most disruptive challenges. Hence, the overall security level of IoMT is upper-limited by the security features of the weakest component. The attack surfaces concern mainly medical devices, communication channels, and data. Let us consider a basic 3 layers architecture of IoMT which is composed of application, network, and perception layers. The IoT architecture is defined as a basic 3 layers which can be extended to 4 or 5 layers. IoMT, as a sub-variant of IoT, presents security issues in all the layers. Some of the possible attacks mapped to the medical device in each one of the IoMT layers are as listed in Table 3. Various proposals dealt with IoMT risk analysis and considered medical devices vulnerabilities as one of the most important risk factors. The common vulnerability and exposure (CVE) publishes discovered medical devices vulnerabilities. In some research works other vulnerabilities are presented. Attack difficulty and number of stakeholders are also considered as main factors to estimate the probability of an attack against IoMT. For the impact of a potential attack, scores are mainly mapped to financial losses and level of harm for the patient.

IoMT Applications Perspectives: From Opportunities and Security Challenges. . .

31

4.3 Risk Ratings To classify the evaluated risk values, each module is able to estimate and re-estimate a risk threshold or a risk rating. In this context, we define the following initial risk rating that will be automatically updated based on the evolution of the risk factors: [Minor (.≥0% and .|z|) values. For example, with the coefficient  representing the individuals’ number, the value is 0.271. On the other hand, ANT-TENS has a smallest value 0.000 which is |z|) less than 0.05 which shows that the independent variables used in the model are statistically significant and related to the response variable ET. So we can conclude at this level that the ANTTENS model is the most significant. Now, we will introduce the residual statistical parameters. The most appropriate model is the one with the smallest residual standard error values. The residue is the difference between the observed values and the estimated ones. The best value of residue is the closest to zero. Table 6 summarizes the different metrics related to the residual (minimum, Q1 , Median, Q3 , Max) [24]. ANT-TENS has a smallest median value close to zero which is equal to 0.29; it is the most optimal. This shows that the prediction is perfect. We will calculate the interquartile range which is a criterion of dispersion data series. It is the difference between the third quartile and the first quartile (Q3 − Q1 ), and its value is significant and more robust with low values. The interquartile is very low and close to zero which shows the prediction efficiency for the three models. Also, we calculated the null deviance which is a measure of how well the response variable is predicted by the model. A low null deviance implies that the

A New Proposed Model for the Influence of Climate Change on the Tension. . .

139

Table 5 The ANT-TENS characteristic model Variables Patient age

Individual number Emergency visit number Hospitalization’s need Emergency reason

Season

Max temperature Noon temperature Max humidity percent Max pressure Cloud cover AVG percent Max heat index Constantt Table 6 The residual measures metrics

Coefficient α 1 : grpage[20,40] α 2 : grpage[40,60] α 3 : grpage[60,90] : Individual number β: nbr_visit_emergency ϒ: hospitalization’s need μ1 : work accident μ2 : violent act μ3 : another disease μ4 : intoxication accidental μ5 : voluntary intoxication μ6 : chronic disease pattern λ2 : Season 2 λ3 : Season 3 λ4 : Season 4 X0 : MT X1 : NT X2 : MHP X3 : MP X4 : CAP X5 : MHI β0

Z-value −4.187 −3.333 −1.156 1.1 3.519 −8.334 0.162 −3.342 0.886 −0.693 1.822 1.341 2.830 0.917 3.714 −2.595 2.139 −2.530 3.633 2.889 2.097 −3.524

Pr(>|z|) 0.0283 0.00085 0.2476 0.2713 0.000433 0.05: variable not significant. 0.05 ≥ p > 0.01: significant variable represented by (*). 0.01 ≥ p > 0.001: highly significant variable represented by (**). p ≤ 0.001: very highly significant variable represented by (***).

In Fig. 4, the hollow points represent an insignificant value, and the horizontal lines represent the confidence intervals. The abscissa axis represents the values of the OR (odds ratio), and each point is located in a confidence interval, i.e., there is a 97.5% chance that the predicted value is true and the risk of error only 2.5%. An odds ratio of 1 corresponds to no effect. In the event of a beneficial effect, the odds ratio is less than 1, and it is greater than 1 in the event of a deleterious effect. The further the odds ratio is from 1, the greater the effect. Our model is well presented, the number of hollow points is low, and more than 95% of the variables are significant. To see the prediction results, we have established the following figure (Fig. 4) which shows the predicted probabilities values of emergency type for each variable used in the model with their confidence interval. As illustrated, two graphic types of representation are marked; for the qualitative variables, the confidence interval is in vertical form, and on the other hand, for the quantitative variables, the confidence interval is presented in the margin form value. Interpreting Fig. 4 the anticipated tension situation arises with a high probability percentage that the emergency type (medical, surgical, or pediatric) is visited.

A New Proposed Model for the Influence of Climate Change on the Tension. . .

Age

141

[0,20) [20,40) (p