Trust aware support vector machine intrusion detection and prevention system in vehicular ad hoc networks

Abstract

As a mean to improve safety and convenience on the road, Vehicular Ad Hoc Networks (VANETs) provide various advantages to passengers. However, considering that it is a wireless ad hoc type network, it is usual to see numerous security exploits present in the environment. There are prevention methods as well as responsive solutions for network intrusions; the former is known as Intrusion Detection System (IDS), which monitors and detects potential intrusions that are ongoing in the network. Active network attacks are generally designed to reduce or interrupt availability of the network. Effect of these attackers on the network can be measured by select parameters, which can in turn be used as the main lead for detecting malicious behaviors. In this paper, we propose a complete IDS in VANET using the combination of modified promiscuous mode for data collection and Support Vector Machine (SVM) for data analysis to establish a shared trust value for every vehicle on the network as Trust Aware SVM-Based IDS (TSIDS). This method ensures that the source vehicle or node as well as any intermediate network node is aware of the activity of their next hop and in case of malicious behavior or malfunction, they will respond accordingly to keep the network performance as high as possible. (C) 2018 Elsevier Ltd. All rights reserved.