Zero-Trust Federated Learning via 6G URLLC for Vehicular Communications

The transition towards intelligent transportation systems is increasingly dependent on advancements in vehicular communications to support data-intensive tasks like Federated Learning (FL). This paper delves into the capabilities of sixth-generation (6G) Ultra-Reliable Low-Latency Communication (URLLC) in elevating the performance of FL within vehicular networks, with a focus on integrating Zero-Trust security principles. By employing real-world vehicular trajectory data from the HighD dataset within an NS-3 simulated network environment, our study rigorously evaluates the combined impact of 6G URLLC and Zero-Trust mechanisms on FL. The findings highlight not only substantial improvements in latency, achieving reductions of up to 81%-83% compared to existing FL models, but also enhancements in throughput, reliability, and model accuracy, alongside a significant increase in security compliance rate. These improvements are pivotal for FL models, promising to optimize the data exchange process, enhance overall learning efficiency, and ensure robust security against evolving cyber threats. Our research indicates that the synergistic integration of 6G URLLC with FL, fortified by Zero-Trust security, could be instrumental in the advancement of intelligent transportation systems, ensuring enhanced vehicular safety, operational efficacy, and data security.