Cross layer plausibility-based trust assessment for Sybil attack detection in Intelligent Transportation Systems

The growing number of smart connected vehicles has increased dependency on vehicular information for autonomous decision-making. Despite the benefits, the open nature of Intelligent Transportation Systems (ITS) makes them vulnerable to attacks by malicious entities. Sybil attacks are one type of attack that can compromise ITS security. This article presents a trust-based cross-layer approach for identifying and countering various Sybil attack variants. The approach enables adaptive trust assessment by continuously updating a node’s trust. It uses fuzzy logic to combine application-layer kinematic semantics with physical-layer signal metrics to assess spatial plausibility and enhance security. The proposed approach requires no roadside infrastructure, no add-on roadside sensing hardware, and no resource-intensive training, making it suitable for real-time operation, with execution times well within the 1 Hz beacon interval. Experimental results indicate improvements of 2.28%, 14.67%, and 18.22% in detection accuracy for DosRandomSybil, DataReplaySybil, and DoSDisruptiveSybil attacks, respectively, compared to single-layer approaches, even at high attacker densities (60%). The approach offers a scalable, lightweight, and effective defense against Sybil attacks, thereby improving the security and trustworthiness of connected vehicular systems.