PROJECT TITLE :

An Intelligent Machine Learning Based Routing Scheme for VANET

ABSTRACT:

Developing Intelligent Transport Systems now includes Vehicular Ad-hoc Networks (VANET), which have become an interesting research topic in recent years. Because of the constant motion of vehicles traveling at varying speeds through urban environments, the topology of the network is subject to frequent shifts. When it comes to delivering data from one vehicle to another, the performance of routing protocols is the primary obstacle that must be overcome in an urban setting. In this paper, we introduce ECRDP, an Efficient Clustering Routing approach that makes use of a new clustering algorithm that is based on Density Peaks Clustering (DPC) and Particle Swarm Optimization (PSO). In the first step of the process, either the PSO algorithm is used to identify the leaders of the clusters or the DPC algorithm is used to create a new fitness function for the purpose of locating the optimal solutions. Following this, clustering is carried out utilizing the dependability of links as the determining factor between vehicles. After that, a maintenance phase is going to be proposed so that the cluster heads can be brought up to date and the vehicles can be redistributed among the clusters. Lastly, a MATLAB simulation is run on a real-life urban environment in order to test the efficacy of the proposed plan by comparing it to the simulated results. The results that were obtained demonstrate an overall improvement in stability, which was demonstrated by a decrease in the change rate by 74%; an improvement in performance, which was demonstrated by an increase in intra-cluster throughput by 34% and inter-cluster throughput by 47%; and an overall reduction in average delay by 16%.