Designing A Self-Learning And Secured Intrusion Detection System (Ids) For A Distributed Iot- Architecture (Similar To Smart Homes) To Propulsion Systems Using Federated Learning And Sensor Data Analytics.

Abstract

The rapid usage  of Internet of Things (IoT) devices equipped with sensors involves larger amounts of data that supports for intelligent data analytical applications. However, centralizing this data raises critical privacy and security concerns. Federated Learning (FL) offers a promising decentralized approach, enabling edge devices to collaboratively train machine learning models without sharing raw data. This work proposes a federated learning framework tailored for IoT sensor networks, integrating lightweight security mechanisms to ensure data privacy and secure communication under resource constraints typical of IoT environments.