Machine Learning techniques optimized by Practical Swarm optimization for Intrusions Detection in IoT

Particle swarm optimization (PSO) is a robust and straightforward to implement optimization technique. It has been used to improve and optimize the performance of machine learning techniques. In this paper, we compare different machine learning techniques for intrusion detection in IoT using PSO. The application of machine learning for cybersecurity in IoT requires substantial data on IoT attacks, but the lack of data on IoT attacks is a significant problem. Our study used the Cooja IoT simulator to generate high-fidelity attack data in IoT 6LoWPAN networks. We provide a performance analysis of different machine learning models optimized by the PSO algorithm for detecting IoT routing attacks with high accuracy and precision. The efficient network architecture for all machine models is chosen based on various network topologies and network scenarios. The experimental results give a brief idea about the cost of each Machine learning model for intrusion detection, especially in terms of energy consumption overhead and memory occupation.