Privacy-Preserving and Decentralized Authentication for IoV Using Federated Learning and Blockchain

Abstract

The Internet of Vehicles (IoV) is a rapidly evolving field, and it is imperative to safeguard user privacy while ensuring effective and secure authentication. Conventional centralized authentication methods are susceptible to single points of failure, privacy issues, and data breaches. This study presents an innovative decentralized authentication method utilizing federated learning and blockchain technology to address these concerns. Federated learning safeguards private data by allowing vehicles to collectively develop authentication models locally, hence eliminating the need for data transfer to centralized servers. Blockchain offers a secure, immutable ledger for documenting authentication activities, enhancing transparency and minimizing manipulation. Our privacy-preserving methodology reduces communication overhead and computational expenses while providing real-time, scalable, and tamper-resistant authentication for IoV scenarios. The proposed technique outperforms conventional methods for computational expense, communication costs, throughput, and security robustness. This methodology integrates the advantages of federated learning and blockchain technology to provide a strong solution to the increasing security demands of IoV networks.