A Trust-Based Malicious RSU Detection Mechanism in Edge-Enabled Vehicular Ad Hoc Networks

Edge-enabled Vehicular Ad Hoc Network (VANET) introduces real-time services and storage, computation, and communication facilities to the vehicles through Roadside Units (RSUs). Nevertheless, RSUs are often easy targets for security assaults due to their placement in an open, unprotected environment and resource-constrained nature. The malicious RSUs compromised by security attacks impose threats to human safety by impeding the operations of VANETs. Hence, an effective malevolent RSU detection mechanism is crucial for VANETs. Existing trust-based detection mechanisms assign trust scores to RSUs based on their interactions with moving vehicles where precise detection of rogue RSUs depends on the accuracy of trust scores. However, brief interaction of RSUs with the running vehicles permits inadequate time to estimate trust accurately. Besides, current works use only vehicle speed and density in beacon messages to assess trust without considering the sensor-detected data in the same messages. Nonetheless, sensor data is useful for traffic management, and neglecting them creates inaccuracy in trust estimation. In this paper, we address these limitations and propose a trust-based scheme to detect malicious RSUs that uses stable and frequent RSU-to-RSU (R2R) interaction to precisely analyze the behavior of an RSU. We also offer a mechanism to detect alteration of sensor-detected data in beacon content and incorporate this scheme in the trust calculation of RSUs. The experimental results show that the proposed solution effectively detects approximately 92% malicious RSUs, even in the presence of hostile vehicles. Moreover, integrating the proposed solution with the VANET routing protocols improves routing efficiency.