Artificial intelligence-enabled anomaly IDS for IoT network
Farhood Nishat · 2024
Internet of things can be applied to many different fields, including health care, education, smart cities, smart grid, future transportation, and others. Therefore, due to vulnerabilities in the topological structure of IoT networks, a flood of fake data packets can be sent easily. Traditional techniques for an intrusion detection system (IDS) fail to identify cyberattacks. Accordingly, this chapter provides detailed information on artificial intelligence (AI) enabled techniques for an IDS to detect threats like DoS/DDoS. AI-/ML-based techniques provide better solutions for IoT networks to identify cyber threats. Moreover, in the literature, traditional techniques are discussed, but they have many problems in threat detection. An IDS is of three main types: signature, anomaly, and hybrid. This chapter is mainly based on AI-based anomaly IDS. Also, AI-based anomaly IDS’s real-time applications, future trends, and solutions are properly incorporated. In addition, technical issues regarding the IDS is well explained, which will help AI/ML engineers, researchers, practitioners in the near future.