Skip to content
Open access

Blockchain Intelligence-Based Trust Management System for Internet of Vehicles

Nov 2026 · IEEE Internet of Things Journal · Vol 13, pp. 48785-48803 · 0 citations · 44 references

Abstract

The Internet of Vehicles (IoV) extends automotive systems through vehicle-to-everything (V2X) communication, creating a collaborative ecosystem to enhance transportation efficiency and road safety. However, the decentralized and open vehicular network is vulnerable to internal malicious participants that disseminate falsified messages or refuse to contribute local resources. While blockchain (BC) and artificial intelligence (AI) have been extensively employed in trust management systems (TMSs), existing solutions lack cross-feature correlation analysis, continuous learning capabilities for emerging threats, and efficient BC consensus mechanisms. This article proposes a novel BC intelligence (BI)-based TMS (BIT) to address these critical challenges. Specifically, BIT establishes a multidimensional feature set based on interfeature physical consistency to empower a deep learning (DL)-based trust evaluation (TE) model, enhancing detection robustness in high-adversary contexts. Furthermore, a trust-driven BC-based federated learning (BFL) framework is designed to facilitate the continuous learning of the TE model using newly collected vehicular data. Local model contributions are subsequently quantified to determine node rewards or penalties by evaluating model quality, training efficiency, and historical reliability. Finally, trust scores and model weights are secured on a consortium BC via an efficient proof-of-model-contribution (PoMC) consensus, which repurposes local training efforts to improve throughput and reduce latency. Simulation results demonstrate that BIT sustainably learns from emerging attack patterns, achieves superior detection accuracy in high-density adversarial scenarios, and maintains secure and efficient storage, outperforming state-of-the-art TMSs.

Read PDF

We use cookies to run the site and, with your consent, for analytics and to show ads. See our Cookie Policy.