Towards a Vehicle's behavior monitoring and Trust Computation for VANETs

Abstract

Vehicular Ad-hoc NETworks (VANETs) create an Intelligent Transportation System (ITS) by eradicating the accidents and traffic congestion on the roads and highways. In VANETs, there are three types of communication viz., Vehicle to Vehicle (V2V), Vehicle to Infrastructure (V2I), and Vehicle to Everything (V2X). A malicious node may communicate a fake message (road congestion, accident, etc.) to other vehicles in the network. Therefore, there is a need for detecting the genuinity of the message. The existing detection systems in literature fail due to high computational complexity, less detection accuracy, etc. In this paper, a trust and behaviour monitoring system is proposed using Neuro-fuzzy technique to differentiate the fake messages from the legitimate messages. From the experimental results, it is evident that our proposed system achieves high detection accuracy and low computational complexity.