A Decentralized Reputation Management Model for Enhanced IoV Networks With 5G Broadcast Services

Abstract

This paper investigates the challenges of data trust sharing faced by Internet of Vehicles (IoV) network with 5G broadcast services. Particularly we develop a decentralized IoV reputation management model with spatiotemporal feature perception fusion (RMM-STFP) based on blockchain. The proposed reputation evaluation method evaluates the reputation value of nodes from the two aspects of time continuity and spatial transitivity and thus improves the identification accuracy of malicious nodes. To further accelerate the dissemination of reputation data, we have constructed a blockchain-based management storage system, where PBFT consensus scheme combines reputation and Bayesian inference. Finally, numerical results are given to justify the superiority of our proposed scheme. When proportion of malicious nodes reaches 45%, the accuracy of our proposed method is 94.5%, and the suppression rate of malicious messages is 83%. Moreover, compared with the traditional PBFT consensus scheme, the consensus delay and communication overhead are reduced by 87.57% and 78.45%, respectively, and the transaction throughput is increased by 70.65%.