Adversarial Machine Learning in IoT Security: A Comprehensive Survey

Internet of Things (IoT) ecosystems are becoming increasingly ubiquitous and heterogeneous, adding extra layers of complexity to secure communication and resource allocation. IoT computing resources are often located at the network edge and distributed across many heterogeneous sensors, actuators, and controller devices. This makes it challenging to provide the proper security mechanisms to IoT ecosystems in terms of manageability and maintainability. In an IoT ecosystem, computational resources are naturally distributed and shareable among their constituency, which creates an opportunity to distribute heavy tasks to them. However, resource allocation in IoT requires secure and complex communication and coordination mechanisms, which existing ones do not adequately support. In this paper, we present Secure Actor-based Model for IoT Communication (SecIoTComm), a model for representing secure IoT communication. SecIoTComm aims to represent secure IoT communication properties and design and implement novel mechanisms to improve their programmability and performance. SecIoTComm separates the communication and computation concerns, achieving design modularity in building IoT ecosystems. First, this paper presents the syntax and operational semantics of SecIoTComm. Then, we present an IoT framework implementing the key concepts of the model. Finally, we evaluate the developed framework using various performance and scalability metrics.

Analogous to the way humans use the Internet, devices will be the main users in the Internet of Things (IoT) ecosystem. Therefore, device-to-device (D2D) communication is expected to be an intrinsic part of the IoT. Devices will communicate with each other autonomously without any centralized control and collaborate to gather, share, and forward information in a multihop manner. The ability to gather relevant information in real time is key to leveraging the value of the IoT as such information will be transformed into intelligence, which will facilitate the creation of an intelligent environment. Ultimately, the quality of the information gathered depends on how smart the devices are. In addition, these communicating devices will operate with different networking standards, may experience intermittent connectivity with each other, and many of them will be resource constrained. These characteristics open up several networking challenges that traditional routing protocols cannot solve. Consequently, devices will require intelligent routing protocols in order to achieve intelligent D2D communication. We present an overview of how intelligent D2D communication can be achieved in the IoT ecosystem. In particular, we focus on how state-of-the-art routing algorithms can achieve intelligent D2D communication in the IoT.

IoT ecosystems generates vast amount of data from interconnected devices, sensors, and networks. This vast amount of data poses significant challenges for traditional way of analysis. Deep learning comes into picture, with its ability to model complex pattern and unstructured data, addressing these challenges and elevating the IoT systems to new levels of intelligence. The integration of deep learning with Internet of Things (IoT) ecosystems provides an advancement in creating the intelligent network of connected devices. Key deep learning architectures like convolutional neural network (CNN) and Recurrent Neural Network (RNN) excel at processing sensor data for visual inspection and environmental monitoring. Looking forward, advancement in improvised learning, explainable AI, and next generation networks like 5G/6G will further enhance the capabilities of intelligent IoT ecosystems. This chapter explains how deep learning architecture facilitates the applications of IoT.

Undeniably, the Internet of Things (IoT) ecosystem continues to evolve at a breakneck pace, exceeding all growth expectations and ubiquity barriers. From sensor to cloud, this giant network keeps breaking technological bounds in several domains, and wireless sensor nodes (motes) are expected to be predominant as the number of IoT devices grows towards the trillions. However, their future in the IoT ecosystem still seems foggy, where several challenges, such as (i) device’s connectivity, (ii) intelligence at the edge, (iii) security and privacy concerns, and (iv) growing energy needs, keep pulling in opposite directions. This prospective paper offers a succinct and forward-looking review of recent trends, challenges, and state-of-the-art solutions of low-end IoT motes, where reconfigurable computing technology plays a key role in tomorrow’s IoT devices.

The Internet of Things (IoT) has expanded rapidly, allowing for a network of sensors and gadgets to collect and share information to make people's lives easier and more convenient. As the Internet of Things (IoT) grows, however, energy efficiency becomes a major issue, especially for portable and wireless gadgets. Low-power, long-range communication capabilities are needed, and Long-Range Wide Area Network (LoRaWAN) has emerged as a viable solution to meet this need. This study provides an in-depth analysis of the LoRaWAN-based, low-power Internet of Things. The suggested network architecture is optimized for low power consumption and high connectivity for numerous Internet of Things (IoT) use cases. This low-power Internet of Things network relies on LoRaWAN gateways, end devices, and a server to function. LoRaWAN is a technology that enables the low-power, long-range transmission of data packets. The results show that the optimized case and non-optimized case have a delivery ratio of 0.85 to 0.73 from node 100 to 500. LoRaWAN significantly reduces energy usage compared to conventional IoT connectivity alternatives, making it a fantastic option for battery-powered devices in far-flung or limited-resource locations. Finally, the adoption of LoRaWAN provides a viable solution to address the energy efficiency concerns in IoT networks, hence allowing for sustainable, long-lasting IoT installations and enabling a wide variety of new applications within the IoT ecosystem. Furthermore, addresses the potential applications of this technology in the future, including upgrades and integration with other technologies like edge computing and 5G networks.

The proliferating popularity of Internet of Things (IoT) devices has led to wide‐scale networked system implementations across multiple disciplines, including transportation, medicine, smart homes, and many others. This unprecedented level of interconnectivity has introduced new security vulnerabilities and threats. Ensuring security in these IoT settings is crucial for protecting against malicious activities and safeguarding data. Real‐time identification and response to potential intrusions and attacks are essential, and intrusion detection systems (IDS) are pivotal in this process. However, the dynamic and diverse nature of the IoT environment presents significant challenges to existing IDS solutions, which are often based on rule‐based or statistical approaches. Deep learning, a subset of artificial intelligence, has shown great potential to enhance IDS in IoT. Deep learning models can identify complex patterns and characteristics by utilizing artificial neural networks, automatically building hierarchical representations from data. This capability results in more precise and efficient intrusion detection in IoT‐based systems. The primary aim of this survey is to present an extensive overview of the current research on deep learning and IDS in the IoT domain. By examining existing literature, discussing mainstream datasets, and highlighting current challenges and potential prospects, this survey provides valuable insights into the prevailing scenario and future directions for using deep learning in IDS for IoT. The findings from this research aim to enhance intrusion detection techniques in IoT environments and promote the development of more effective antimalware solutions against cyber threats targeting IoT device systems.

The rapid growth of Internet of Things (IoT) technologies has enabled transformative applications in smart cities, environmental monitoring, and sustainable infrastructure. However, the pervasive deployment of IoT devices exposes these ecosystems to significant security and privacy risks, thereby necessitating robust intrusion detection systems (IDS). Conventional IDS approaches often impose high computational and energy demands, which limit their applicability in resource-constrained and environmentally sensitive settings. To address these challenges, this study proposes a privacy-preserving and energy-efficient intrusion detection framework designed specifically for IoT-enabled environments. The framework integrates lightweight cryptographic mechanisms with optimized feature selection and machine learning-based detection to safeguard user data while minimizing resource utilization. Experimental evaluation demonstrates that the proposed model achieves superior detection accuracy, reduced false alarm rates, and up to 25% lower energy consumption compared to conventional IDS approaches. These results highlight the potential of the framework to enhance security resilience in IoT ecosystems while contributing to environmentally sustainable computing practices.

Recently, extensive research into healthcare services and their technological advancements is being carried out globally. The Internet of Things (IoT) has proven to be a predominantly potential candidate in these advancements. IoT supports infrastructure development in connecting a wide range of medical devices, sensors, and health practitioners in order to ensure the best medical treatment for patients in both cities and remote places. This has enabled the IoT to provide major support in medical application, giving rise to assisted healthcare facilities. The IoT ecosystem has a complicated architecture, with several components interacting and an evolving variety of applications for users. This is a comprehensive and interdependent system that involves real-time data analysis, interconnections of device, data transfers, and analysis. The IoT creates a connected ecosystem developed from several intelligent systems that integrates respondents with machines and allows for statistics decision making. A network of interconnected devices that perceive vital data related to medical conditions of patients in real time is referred to as the Internet of Medical Things (IoMT), a healthcare application of IoT technology. Sensors used for sensing real-time data related to monitoring temperature, blood pressure levels, heartbeat levels, specialized signal converters, and communication interfaces can be added to existing medical facilities to turn them into IoMT devices. IoMT can be used to create wearable devices, home-care and personal-care medical devices, and mobile healthcare apps, along with creating communication with medical specialists in a given area. The IoMT system’s challenges and ecosystem interface requirements are also a matter of concern considering the complexity of the systems. This study offers a comprehensive evaluation of the various domains of use of IoMT, as well as its future scope and challenges. Recent advances in sensor technology, networks, cloud storage and processing, mobility, and big data analytics have made it possible to create affordable smart medical devices and a connected healthcare environment.

Internet of Things (IoT) devices face increasing security threats while operating under severe energy constraints. This research addresses the critical challenge of implementing effective cybersecurity measures within sustainable energy frameworks for resource-constrained IoT environments. The research presents Sustainable Micro-Neural Energy-Efficient Security Intelligence (SMEESI), a novel TinyML-enabled intrusion detection system (IDS) that significantly reduces energy consumption while maintaining robust security capabilities. The research innovation combines lightweight neural network architectures with energy-aware anomaly detection algorithms specifically optimized for microcontroller deployment. The SMEESI framework includes an adaptive power management module that dynamically adjusts computational intensity based on threat levels, achieving energy efficiency without compromising security posture. Performance evaluation demonstrates a 78% reduction in power consumption compared to traditional IDS implementations while maintaining 94.3% detection accuracy across multiple attack vectors. Memory footprint requirements decreased by 65%, enabling deployment on severely resource limited IoT sensors and actuators. The system has been successfully tested in smart buildings, industrial monitoring, and healthcare IoT applications, proving its versatility across critical infrastructure domains. This research contributes to green cybersecurity by enabling sustainable security monitoring in IoT ecosystems, extending device battery life, reducing electronic waste through prolonged hardware lifecycles, and minimizing the carbon footprint of security operations while maintaining essential protection against evolving cyber threats in the digital age.

The introduction of Internet of Things (IoT) ecosystems into personal homes and businesses prompts the idea that such ecosystems contain residual data, which can be used as digital evidence in court proceedings. However, the forensic examination of IoT ecosystems introduces a number of investigative problems for the digital forensics community. One of these problems is the limited availability of practical processes and techniques to guide the preservation and analysis of residual data from these ecosystems. Focusing on a detailed case study of the iHealth Smart Scale ecosystem, we present an empirical demonstration of practical techniques to recover residual data from different evidence sources within a smart scale ecosystem. We also document the artifacts that can be recovered from a smart scale ecosystem, which could inform a digital (forensic) investigation. The findings in this research provides a foundation for future studies regarding the development of processes and techniques suitable for extracting and examining residual data from IoT ecosystems.

In today’s competitive landscape, achieving customer-centricity is paramount for the sustainable growth and success of organisations. This research is dedicated to understanding customer preferences in the context of the Internet of things (IoT) and employs a two-part modeling approach tailored to this digital era. In the first phase, we leverage the power of the self-organizing map (SOM) algorithm to segment IoT customers based on their connected device usage patterns. This segmentation approach reveals three distinct customer clusters, with the second cluster demonstrating the highest propensity for IoT device adoption and usage. In the second phase, we introduce a robust decision tree methodology designed to prioritize various factors influencing customer satisfaction in the IoT ecosystem. We employ the classification and regression tree (CART) technique to analyze 17 key questions that assess the significance of factors impacting IoT device purchase decisions. By aligning these factors with the identified IoT customer clusters, we gain profound insights into customer behaviour and preferences in the rapidly evolving world of connected devices. This comprehensive analysis delves into the factors contributing to customer retention in the IoT space, with a strong emphasis on crafting logical marketing strategies, enhancing customer satisfaction, and fostering customer loyalty in the digital realm. Our research methodology involves surveys and questionnaires distributed to 207 IoT users, categorizing them into three distinct IoT customer groups. Leveraging analytical statistical methods, regression analysis, and IoT-specific tools and software, this study rigorously evaluates the factors influencing IoT device purchases. Importantly, this approach not only effectively clusters the IoT customer relationship management (IoT-CRM) dataset but also provides valuable visualisations that are essential for understanding the complex dynamics of the IoT customer landscape. Our findings underscore the critical role of logical marketing strategies, customer satisfaction, and customer loyalty in enhancing customer retention in the IoT era. This research offers a significant contribution to businesses seeking to optimize their IoT-CRM strategies and capitalize on the opportunities presented by the IoT ecosystem.

In current years, Internet of Things (IoT) has come a long way and is well emerged within many organizations and fields covering the sector of healthcare. Continuously, encompassing of the Internet of Medical Things (IoMT) in the field of the medical sector helps in the rise of medical data and analysis of that. Smart healthcare plays important role in human life and various research projects are running on it. It consists of diverse ingredients like wireless communication technology, cloud computing, edge computing, medical sensors, artificial intelligence (AI), IoMT, and IoT. The chapter introduces IoMT with respect to the multiple challenges like privacy, security and low-power operation faced while incorporating IoT with healthcare sector. For emergency treatment, a cloud-based healthcare framework provides a platform to support patients over internet communication by medical experts. Privacy and security to the medical records are on higher priority due to the very sensitive nature of the content. Moreover, a comprehensive survey on IoMT-based cloud and edge-intelligent healthcare system is also described. The current advancement in the field of IoT, IoMT, cloud computing, and AI related to privacy and security, some pathways for forthcoming work in collaboration of healthcare sector and IoT based on a collection of challenges and open issues are also addressed.

The rapid proliferation of Internet of Things (IoT) devices has ushered in a new era of interconnected technologies, promising unprecedented convenience and efficiency in our daily lives. However, the exponential growth of IoT also presents an alarming challenge in terms of cyber security. This review paper delves into the intricate landscape of cyber security in the IoT era, providing a comprehensive analysis of existing research, key challenges, and innovative solutions. The paper explores the vulnerabilities inherent in IoT ecosystems, the evolving threat landscape, and the implications for privacy, data integrity, and critical infrastructure. By examining the state-of-the-art security mechanisms, including encryption, authentication, and intrusion detection, this review sheds light on the ongoing efforts to safeguard IoT systems. Furthermore, it delves into the policy and regulatory frameworks that govern IoT security and the ethical considerations surrounding data privacy. Through an in-depth evaluation of research articles, case studies, and industry best practices, this review offers valuable insights for researchers, policymakers, and practitioners working to fortify the security of IoT ecosystems and ensure a safer and more resilient future for the Internet of Things. So, this paper talks about how to make sure these gadgets are safe and that the bad people can't do anything bad with them. It's like putting locks on your doors to keep your toys safe. They talk about how to make strong locks and rules for these gadgets so that they stay safe and you can keep using them without any worries. It's all about keeping the Internet of Things safe!

The Internet of Things (IoT) is a new technology that improves the quality of people's lives by dominating a wide range of applications and service sectors. Fifth-generation (5G) wireless networks will soon be able to link this IoT ecosystem. It has been thoughtfully created to support the IoT industry's explosive development. One of the essential technologies in the 5G architecture, network slicing allows for dividing the physical network into several logical networks (i.e., slices) with various network properties. As a result, network slicing is a crucial driver of IoT in 5G. Through dedicated slices, network slicing can meet the different networking needs of heterogeneous IoT applications. We thoroughly review network slicing's use in IoT realization in this study. We cover network slicing in various IoT application scenarios and the technical problems that can be resolved using network slicing. This paper also discusses open research issues and integration challenges linked to network slicing in the IoT ecosystem.

The integration of blockchain technology and Internet of Things (IoT) computing has emerged as a promising approach for secure and decentralized data management. Here we explore integrating the blockchain with IoT computing to enhance data security, integrity, and privacy. By leveraging the immutability, transparency, and distributed consensus of blockchain, the study aims to investigate how this integration can overcome the limitations of traditional centralized data management approaches in the IoT ecosystem. The chapter conducts a comprehensive literature review to identify the key advantages and challenges associated with integrating blockchain and IoT computing. It examines real-world use cases, implementation strategies, and the security and privacy considerations involved in the integration. Through this exploration, we uncover these two techniques to foster a new era of data management that prioritizes security, transparency, and autonomy.