IPEK trust model for vehicle networks hits 0% false revocations

IPEK, short for Intelligent Priority-Aware Event-Based Trust with Asymmetric Knowledge, comes from İpek Abasıkeleş Turgut and addresses a critical blind spot in vehicular network security: attackers who build trust gradually by reporting correctly on low-priority events, then inject false data during safety-critical situations. Existing VANET trust models treat all events equally, allowing this 'good-citizen-then-betray' strategy to go undetected. IPEK introduces event-aware and location-aware attack models that capture these strategic adversaries, then pairs them with an asymmetric local trust mechanism: penalties scale with event and location severity, while rewards follow an asymptotic model, making it exponentially harder for malicious nodes to recover reputation after an exploit.

For global trust fusion, IPEK adapts Dempster-Shafer Theory using Yager's combination rule, which assigns conflicting evidence to uncertainty rather than forcing premature decisions. This is combined with sequential source-reliability ordering and an asymmetric risk accentuation mechanism. Experiments in OMNeT++, Veins, and SUMO benchmarked IPEK against MDT and TCEMD under attacker densities ranging from 15% to 35%. Results show IPEK maintains a 0% False Positive Rate—no honest vehicle was ever wrongly revoked—while sustaining recall above 75% and F1-scores exceeding 0.86. For safety-critical vehicle-to-everything (V2X) communication, this means networks can filter malicious inputs during accidents or traffic hazards without penalizing legitimate drivers, a major step toward deployable resilient VANETs.