Implementation of evolutionary algorithms in vehicular ad-hoc network for cluster optimization

Abstract

Among many other sub-types, one sub-type of ad hoc network is Vehicular ad hoc Network (VANET). VANET can be further categorized in sub-domains like Vehicle to Infrastructure (V2I), Vehicle to Vehicle (V2V), Vehicle to pedestrian or other equipment (V2X) and Hybrid (V2V+V2I+V2X). V2V communication is the primary focus of this paper. Different methodologies are available in the literature for optimization of V2V communication. Clustering is one of them, in clustering vehicles, the same vicinity are grouped together for efficient communication. Different evolutionary algorithms for clustering already have been implemented to route information among nodes. Two evolutionary algorithms are applied for optimizing communication among the vehicles and the clustering problem in the VANETs. The bio inspired evolutionary algorithms are Comprehensive Learning Particle Swarm Optimization (CLPSO) and Multi-Objective Particle Swarm Optimization (MOPSO). After implementation, comparison for the mentioned algorithms is used to depict the results. The experimental results show that CLPSO is providing better results than MOPSO.