Internet of vehicles (IoV) allows real-time connectivity between automobiles, roadside infrastructure, and edge computer nodes for intelligent transport systems. In dynamic scenarios, secure message delivery, cooperative routing, and attack-resilient vehicle coordination need trust relationships. Dynamic topology change, malicious node insertion, and dynamic vehicle interactions make static scoring or machine learning trust management approaches fail. This impairs safe and adaptive trust assessment in decentralized IoV environments. An Adaptive Blockchain-Oriented Trust Management architecture with Proximal Policy Optimization (ABTM-PPO) may solve these difficulties. Vehicle data streams are processed by dynamic trust state encoder (DTSE) to capture temporal interaction patterns, node behavioral fingerprints, and contextual communication needs. To avoid reputation manipulation and fraudulent trust escalation, a blockchain-validated reputation consensus module (BRCM) maintains trust data in an immutable distributed ledger. PPO-driven adaptive trust optimization engines (PATOEs) adapt reward-guided reinforcement learning trust rules to network states and hostile situations. Trust assessment baselines are outperformed by the proposed technique in diverse traffic volumes and coordinated assaults. Malicious node detection is 97.1% and trust classification 98.4%. In high-mobility, packed communication intervals, it cuts decision reaction time by 21.6% while maintaining performance. These results show that the trust-aware security management system for next-generation intelligent vehicular networks is scalable and robust.
Shamili et al. (Fri,) studied this question.