The term internet of things (IoT) denotes the advanced phase of the internet, wherein a worldwide communication infrastructure is established between individuals and machines. The IoT is now being developed as a worldwide infrastructure that has the potential to significantly transform various aspects of human life. Nevertheless, the interconnected nature of IoT systems and the involvement of other disciplines in their implementation have presented novel security obstacles. However, the utilization of deep learning techniques holds significant importance in enhancing the security of IoT systems. This transformation goes beyond the facilitation of secure communication between devices, as it enables the development of intelligent systems that prioritize security. In this chapter, the basic architecture of IoT has been discussed with its applications and security challenges. Different deep-learning techniques for securing IoT devices with their strengths and weaknesses have also been explored.
TopIot Overview
The Internet of Things (IoT) is a concept that involves interconnected devices that can communicate, collect data from their surroundings, analyze that data, and share it to accomplish specified objectives. Currently, the IoT has played a pivotal role in facilitating automation across all facets of our daily existence. This encompasses the incorporation of intelligent home technologies such as air conditioning, security systems, lighting, and several other services and gadgets (Aldahmani et al., 2023). Furthermore, it has significantly influenced multiple facets of human existence, resulting in the development of intelligent education systems, advanced healthcare solutions, innovative agricultural practices, and sophisticated industrial processes, among others. The IoT sector is anticipated to experience significant expansion, with an estimated 22 billion intelligent gadgets projected by the end of 2025 (Yunana et al., 2021). Figure 1 illustrates the concept of IoT, where people and things can be connected through various paths/networks and services, regardless of time or location.
The concept of the IoT encompasses the interrelationships of many objects, such as industrial systems, intelligent sensors, autonomous vehicles, mechanisms, terminals, and mechanical systems (Sharma et al., 2020; Zhang et al., 2023). IoT devices possess the capability to store a wide range of crucial information, including but not limited to sound data, light intensity, temperature readings, electricity consumption, mechanical motions, chemical reactions, impact measurements, biological changes, and geo-location data (Gaurav et al., 2023). IoT devices are utilized to facilitate machine-to-machine communication, machine-to-human engagement, and human-to-human activity.
IoT systems are intricate and consist of integrated configurations. Hence, it is resilient to uphold the security need in the extensive attack surface of the IoT system. To meet the security requirement, solutions must incorporate comprehensive considerations (Sarker et al., 2023). Nevertheless, IoT devices primarily operate in an unsupervised setting. As a result, an unauthorized individual may gain physical access to these devices. IoT devices typically connect over wireless networks, which might be vulnerable to eavesdropping, allowing unauthorized individuals to intercept private information transmitted across the communication channel (Salman & Arslan, 2023). The dissemination of information will occur within the public domains, specifically the network layer and application layer (Yazdinejad et al., 2023). If there is a lack of an efficient mechanism for safeguarding information, it may be susceptible to theft, hence posing a risk to privacy. Consequently, ensuring security and privacy are important considerations for IoT-enabled devices.