A Novel Approach for Malicious Node Detection in Vehicular Ad-hoc Network using Support Vector Machine

Vehicular Ad-hoc Networks (VANETs) are the most popular research area of wireless technology. It is offering various types of services such as safety and comfort for drivers as well as for passengers and requires less infrastructure than other technology. VANETs are being used in the concept of making smart cities, and smart traffic management systems all over the world. Due to vulnerability in VANETs, undesired data may attack the network traffic and broadcast false messages in the entire network which causes poor communication among the networks. False data or undesired information is termed malicious data. These malicious data can affect the performance of VANETs it increases vehicle delay, increases fuel consumption and there will be more safety threats. So, there must be a relative study, which will show the variation of network performance metrics with enhancing the number of malicious nodes in the network. This paper had focused to discuss the detection and a deep analysis of affects of malicious nodes on the network performance as throughput, average latency, and packets drop in the network. In this paper, we have discussed a novel approach to detect malicious data with Artificial Neural Networks as well as malicious nodes with Support Vector Machines in VANETs.