Scalable And Efficient PBFT Consensus Algorithm for Secure Internet of Vehicles Networks

Abstract

The rapid expansion of Internet of Vehicles (IoV) networks necessitates robust consensus protocols that can handle dynamic vehicular dynamics while maintaining security and efficiency. Current Practical Byzantine Fault Tolerance (PBFT) algorithms are plagued by severe scalability problems when used in large-scale IoV networks because communication overhead increases exponentially with network size. This paper introduces a new PBFT consensus algorithm for IoV applications that features a unique logarithmic scaling method, defining a reverse proportionality between the network size and the percentage of nodes required for consensus. The percentage of nodes necessary for consensus diminishes logarithmically as the number of vehicles grows, allowing scalable operation in large networks. The proposed algorithm employs a smart relay selection scheme using quality parameters like distance, channel quality, collision likelihood, and received signal strength measurements. Large-scale simulations conducted by OMNeT++, VEINS, and SUMO demonstrate that our enhanced PBFT achieves a consensus time saving of 30% when compared to existing methods, achieving an average time per hop of 204ms for 200 vehicles against 365ms for normal PBFT. It is Byzantine fault-tolerant, but more scalable and thus very well suited for vehicular communication in real-time. The analysis of performance proves higher efficiency than existing methods like CreditCoin and reputation-based blockchain networks, with equal gain in performance under various vehicle densities, which forms a platform for secure and scalable IoV networks.