AI‐Driven Edge Computing for Secure and Efficient Internet of Vehicles (IoV) Communication

The rapid evolution of the Internet of Vehicles (IoV) has scaled it up to provide secure and reliable communication among infrastructure, cloud platforms, and vehicles. Due to limited computation and high delay, traditional security solutions are challenged to handle real-time threats. To enhance IoV security, we propose an artificial intelligence–based edge computing with deep learning–based intrusion detection systems at edge nodes. For real-time detection of cyberattacks and preserving privacy, the system uses federated learning to distribute model training to roadside units and vehicles. There is also implementation of a zero-trust architecture as well as blockchain-based authentication for protection against Man-In-The-Middle attacks and unauthorized access. In comparison to traditional cloud-based security frameworks, experimental evaluations on IoV datasets demonstrate that the proposed framework enhances response time for threat mitigation by 38%, reduces latency by 45%, and has an intrusion detection accuracy of 97.2%. Based on studies, artificial intelligence–based edge computing has the potential to significantly enhance the efficiency and security of next-generation IoV networks toward smart and autonomous transportation systems.