Bayesian-based model for a reputation system in vehicular networks

Abstract

Vehicular ad hoc networks (VANETs) are a cost-effective technology to enhance driving safety and traffic efficiency. In such promising networks, security is of prime concern because an attack by a malicious vehicle might have disastrous impact leading to loss of life. Reputation trust management has been proposed in the recent years as a novel way to tackle some of those not yet solved threats in VANETs. In this paper, we propose a robust distributed reputation model based on Bayesian filter. The model allows nodes to establish profiles (e.g., malicious, honest) on their neighbors and to detect malicious behaviors (e.g., black hole, gray hole). The simulation results proved that intentionally dropping packets in VANETs can be fully detected, with our proposed Bayesian filter, with high level of accuracy.