Intrusion Detection Systems (IDS) are vital for protecting connected and autonomous vehicles (CAVs) from cyber threats by monitoring network traffic for suspicious activities. The increasing connectivity of these vehicles amplifies the potential attack surface, making robust security measures essential. Federated Learning (FL) enhances IDS by allowing multiple vehicles to collaboratively train detection models while keeping their data localized, thus preserving privacy. This decentralized approach enables real-time adaptation to emerging threats without sharing sensitive information. By leveraging insights from a diverse range of vehicles, FL-based IDS can improve detection accuracy and response times. Ultimately, integrating federated learning with intrusion detection strengthens the overall security framework for CAVs, ensuring safer and more resilient automotive systems. This talk presents an overview of Federated Learning-based Intrusion Detection Systems and related challenges in vehicular use cases, as well as the development of a test platform to evaluate such use case, modeled by a network of Raspberry-Pi’s.
Robert Akinie is currently pursuing a PhD student in Electrical and Computer Engineering at North Carolina A&T State University. He received a BSE in Electrical and Computer Engineering from Calvin University, Grand Rapids, Michigan, in 2021. His current research interests focus on federated learning, autonomous vehicle architecture, embedded devices and intrusion detection systems.