Secure FLOATING - Scalable Federated Learning Framework for Real-time Trust in Mobility Data using Secure Multi-Party Computation and Blockchain

Explore a 14-minute conference presentation that introduces Secure-FLOATING, an innovative scalable framework designed to establish real-time trust in mobility data for Connected and Autonomous Vehicles (CAVs), micromobility devices like e-scooters and e-bikes, and smartphone users. Learn how this framework addresses the critical safety challenge of verifying trajectory data used for navigation without compromising user privacy, as malicious data could pose deadly risks, particularly to vulnerable road users. Discover how the researchers combine federated learning and blockchain technology to enable nearby nodes to coordinate and learn to trust mobility data from surrounding devices, storing this trust information through consensus on a tamper-proof distributed ledger. Understand the implementation of lightweight Secure Multi-party Computation (SMPC) with reduced message exchanges that preserves user privacy while ensuring real-time data validation. Examine the evaluation results using realistic trajectories for up to 8,000 nodes including vehicles, micromobility devices, and pedestrians in New York City, demonstrating lower delays and overhead while achieving up to 75% successful endorsement even with 50% attacker penetration. Gain insights into how this scalable approach accurately validates mobility data in real-time scenarios, presented by researchers from Western Washington University and New York University at the VehicleSec '25 conference.