T5 – Federated Learning for Autonomous and Connected Vehicles: Architectures, Challenges, and 6G-Enabled Opportunities

Co-presenter: Momina Shaheen, University of Roehampton, UK,
Co-presenter: Weiwei Jiang, Beijing University of Posts and Telecommunication, China
Co-presenter: Umar Khokhar, Georgia Gwinnett College, Georgia, USA,

Abstract: The rapid advancement of Autonomous and Connected Vehicles (ACVs) has triggered an explosion of data generated by on-board sensors, LiDAR, and V2X communications. While this data is essential for training robust deep learning models for perception and path planning, traditional centralized cloud-based training faces critical bottlenecks: high communication overhead, stringent latency requirements for safety-critical tasks, and growing data privacy regulations (e.g., GDPR). Federated Learning (FL) has emerged as a transformative decentralized machine learning paradigm. It allows vehicles and Roadside Units (RSUs) to collaboratively train a global model while keeping raw data locally on the device. However, the high mobility of vehicles, intermittent wireless connectivity, and heterogeneous hardware capabilities pose significant hurdles to standard FL protocols. This tutorial provides a comprehensive deep dive into the integration of FL within the vehicular ecosystem. We will explore tailored architectures, such as hierarchical FL (Vehicle-to-RSU-to Cloud) and peer-to-peer (P2P) FL for vehicular ad-hoc networks (VANETs). The session will further analyze the role of 6G technologies—including Terahertz (THz) communications, Reconfigurable Intelligent Surfaces (RIS), and native AI—in enabling ultra-reliable, low-latency FL. Attendees will gain a clear understanding of communication-efficient aggregation, privacy-preserving techniques (Differential Privacy, Secure Multiparty Computation), and the practical application of FL in real-world scenarios like cooperative perception and trajectory prediction.

Co-presenter’s Bios:

Momina Shaheen:

Dr. Momina Shaheen is an academic and researcher in computer science with over 8 years of experience in teaching and research. She currently serves as a Senior Lecturer in Computing and Programme Leader at the University of Roehampton London. Her work focuses on edge computing, the Internet of Things, federated learning, and cybersecurity, with applications across smart cities, healthcare, finance, and education. Shaheen holds a B.Sc. in Information Technology, an M.Eng. in Software Engineering, and defended her Ph.D. in Computer Science, focusing on improving deep learning performance in federated machine learning. Her academic contributions include over 46 peer-reviewed publications in Q1 journals, multiple book chapters, and editorial work. She is also serving as Editor of Robotics and AI Review journal. She has edited 5 books in the field of algorithms and computing, and served as a reviewer for prominent journals such as Nature (under Early Career Reviewer Program), IET Information Security, Springer Scientific Reports, MDPI Electronics, IEEE Access, PLOS ONE, and ACM Transactions, and has chaired international conferences.

Weiwei Jiang:

Dr. Weiwei Jiang received the B.Sc. and Ph.D. degrees from the Department of Electronic Engineering, Tsinghua University, Beijing, China, in 2013 and 2018, respectively. He is currently an assistant professor with the School of Information and Communication Engineering, Beijing University of Posts and Telecommunications, and Key Laboratory of Universal Wireless Communications, Ministry of Education. His current research interests include artificial intelligence, machine learning, big data, wireless communication and edge computing. He has published more than 70 academic papers in IEEE Trans and other journals, with more than 5500 citations in Google Scholar. He is one of 2023, 2024 and 2025 Stanford’s List of World’s Top 2% Scientists.

Umar Khokhar:

Dr. Umar Khokhar is an Associate Professor of Information Technology at Georgia Gwinnett College (GGC), USA. He holds a Ph.D. in Information Security with a focus on ultralightweight cryptography for IoTs, and he was awarded a Presidential Gold Medal during his MS studies for academic excellence. Dr. Khokhar has over 15 years of experience in academia and industry, previously serving as an Assistant Director of IT and holding various faculty positions where he taught courses in Information Security, Wireless Communication, and Ethical Hacking. A prolific researcher, he has authored or co-authored over 50 peer-reviewed journal and conference publications and has been a lead investigator on multiple grants funded by the National Science Foundation (NSF) and the University System of Georgia. His technical expertise spans cryptographic algorithm development, penetration testing, and hardware security, and he currently serves as a Lead Guest Editor for Security and Communication Networks