Survey for Anomaly Detection of IoT Botnets Using Machine Learning Auto-Encoders

摘要

This study focus on Machine Learning techniques for Internet of Things security threats detection. It seeks to investigate the feasible of using auto-encoders to detect IoT botnets. Botnets can develop DDoS attacks and present a major security concern in IoT networks, as there is no single method has demonstrated the potential to address this security threat. These methods often fail to meet IoT environments requirements, such as processing power and energy consumption. Auto-encoders offers one of the solutions to botnet detection. Future research needs to explore the opportunities that auto-encoders present in the detection of IoT botnets.