Real-Time Adaptive Intrusion Detection System [RTPIDS] for Internet of Things Using Federated Learning and Blockchain
Sunil Raj Y, E. Helen Parimala, V. S Jayakumar Paul Bosco, Aarav Kannan J, Prafulla Kumar, Vinothkumar Kolluru
Abstract
In the growing field of the Internet of Things (IoT), ensuring security and privacy has become a critical concern, particularly in sensitive domains such as healthcare. This paper presents a Real-Time Adaptive Intrusion Detection System (IDS) specifically designed for Internet of Medical Things (IoMT) networks. The proposed system leverages Federated Learning (FL) and Blockchain technologies to address key challenges, including handling non-independent and identically distributed (non-IID) data, ensuring secure model updates, and providing real-time detection capabilities. By employing FL, the system enables IoMT devices to collaboratively train models without sharing sensitive data, thereby preserving privacy and improving scalability. Blockchain integration further enhances the system by ensuring the integrity and security of model updates, mitigating risks of tampering. The system demonstrates strong performance in detecting simpler intrusions in real-time, though additional enhancements are required to manage more complex attacks and large-scale IoT deployments. This work contributes a robust and scalable IDS framework, particularly valuable for protecting critical IoT environments such as healthcare networks.