Blockchain-Enabled Trust Management for Secure Data Exchange in Internet of Vehicles

Abstract

The Internet of Vehicles (IoV) enables continuous communication between vehicles and infrastructure, but its open and dynamic nature exposes it to severe security and trust challenges. Existing blockchain-based trust management solutions address these issues only partially, suffering from passive-only detection, static trust update mechanisms, absence of RSU monitoring, monolithic blockchain designs, and lack of fairness toward honest vehicles. This memoir proposes ADBT-IoV (Adaptive Dual-Blockchain Trust Framework for Secure Data Exchange in Internet of Vehicles), a unified trust management framework implemented as Solidity smart contracts on an Ethereum blockchain. ADBT-IoV integrates cryptographic vehicle authentication, cross-verification-based message confidence scoring, active probing for hidden malicious behaviour detection, an adaptive AIMD trust update with recidivism memory, RSU anomaly monitoring via the Trust Outlier Factor (TOF) with automatic handover, graduated vehicle management through grey and black lists with a rehabilitation mechanism, broadcast priority proportional to trust, and a Soft Actor-Critic (SAC) deep reinforcement learning agent for dynamic parameter optimization. Experimental evaluation on a Ganache local blockchain demonstrates that the SAC agent achieves 99.2% accuracy, 98.6% detection rate, and 0.6% false positive rate. Smart contract benchmarks show write latency averaging 61.13 ms and read throughput reaching 182.48 TPS, with CPU consumption below 0.01%, confirming real-time suitability for resource-constrained IoV environments. A comparative analysis shows that ADBT-IoV is the only solution simultaneously addressing all identified limitations of existing approaches