Stability Study of Internet of Things Against Botnets Attacks

The Internet of Things (IoT) is expanding day by day and has become an integral part of the modern world. The network system is complex and versatile, capable of connecting different types of devices. Basically, the concept of IoT is to connect anything with anyone, anywhere, and any time. The elements, i.e., objects or things of IoT, are sensors, actuators, tags, RFID, and readers that facilitate the connection between the physical and virtual domains. As IoT coverage is expanding, the various types of operational challenges are surfacing, one of which is malware attacks, where maleware-infected IoT devices collectively form a botnet. And the botnet poses a significant threat to IoT operation. The IoT is facing various types of security threats due to inadequate regulatory standards, intrinsic security flaws, a lack of security awareness among the consumers, and other contributing factors. The botmaster exploits the deficiencies of IoT, targets a node/device of the network,infect it and forms a botnet. Then the infected node starts to spread bots in the entire network. Botnet malware spreads from host to host, often targeting IoT devices, similar to how biological viruses spread in susceptible populations. The bot's spreading capability is also affected by its memory requirements. Therefore, to secure the IoT against botnet attacks, it is important to understand and analyze the propagation features of botnets in networks. This paper proposes a mathematical model based on an epidemiological principles. The proposed model is an enhancement over an IoT-SIA (IoT-Susceptible-Infectious-Abandon) model. An IoT-Susceptible-Exposed-Infectious-Abandoned (IoT-SEIA) model is proposed to address the limitations of the prior model. The proposed model includes an idea of exposed state, which helps in the determination of bot presence in IoT at an early stage. The basic reproduction number, a critical threshold parameter governing malware propagation in IoT networks, is derived, as well as the malware-free equilibrium points of the system, and it is discussed. Further, comparative analysis shows that the proposed model is more effective in combating botnet attacks in comparison to the IoT -SIA model. The analytical and numerical findings demonstrate that the proposed IoT-SEIA model achieves better mitigation of botnet malware propagation than the previous model.