Malicious vehicle detection based on beta reputation and trust management for secure communication in smart automotive cars network

Abstract

High reliance on wireless network connectivity makes the vehicular ad hoc network (VANET) vulnerable to several kinds of cyber security threats. Malicious vehicles accessing the network can lead to hazardous situation by disseminating misleading information or data in the network or by performing cyber-attacks. It is a requirement that the information must be originated from the authentic and authorized vehicle and confidentiality must be maintained. In these circumstances, to protect the network from malicious vehicles, reputation system based on beta probability distribution with trust management model has been proposed to differentiate trustworthy vehicles from malicious vehicles. The trust model is based on adaptive neuro fuzzy inference system (ANFIS) which takes trust metrics as input to evaluate the trustworthiness of the vehicles. The simulation platform for the model is in MATLAB. Simulation results show that the vehicles need at least 80% trustworthiness to be considered as a trusted vehicle in the network.