An Improved Routing Architecture for Upgrading Quality of Service using Metaheuristics and Machine Learning in VANETs

Abstract

Vehicular Ad-hoc Networks (VANETs) is gaining popularity in academics and businesses, an important part of the intelligent system that benefits the environment using vehicle communication. VANET has been one of most demanding, random and sophisticated networks of the modern world from the last couple of years. The increasing popularity of VANET architecture is due to the increasing number of vehicles on the road. A VANET network is made up of multiple vehicles that communicate to each other in order to avoid congestion. The main objective of the proposed work is to enhance the route discovery process and QoS of the VANET using clustering algorithm. The proposed route discovery algorithm is supported by the clustering technique, followed by the broadcast mechanism. The clustering policy of the algorithm intakes the quality of LEACH clustering algorithm and enhances it in a moderate way.The proposed algorithm utilizes the policy of clustering of the nodes based on the distances of the nodes from the Cluster Head (CH). The working mechanism of the proposed clustering algorithm in which one node transfers the data through CHs. This clustering algorithm utilizes a novel algorithmic calculation based on the deployment region and distance awareness from the fog server. The algorithm takes the advantage of the LEACH algorithm and modifies it further. The process of route discovery results into the evaluation of the hops of the route, throughput, PDR and jitter of the route. Additionally, an integration of a trust model based on machine learning for intercommunication has been presented and compared the performance of the proposed trust model with a non-integrated clustering algorithm. newline