Optimization of vehicular node clustering process using evolutionary algorithms

Abstract

Numerous methods are available for vehicular node clustering. These vehicular nodes in vehicular ad hoc Network (VANETs) move very fast. These nodes, on the basis of different parameters, are grouped together in same group called cluster. Elementary reason for creating vehicular cluster is to enable vehicles to communicate with each other. Cluster Heads (CHs) are selected in each cluster for managing the whole cluster. CH is liable for inter and intra cluster communication. Life time of each cluster is directly linked with the performance of network, moreover, the less number of clusters tend to more efficient network due the reduced communication overhead. Gray Wolf Optimization (GWO) is a technique based on the social behavior of gray wolves, a novel algorithm, for clustering in VANET, based on this technique is proposed in this paper. This algorithm delivers optimized solution for smooth and robust communication in the VANETs. The vital parameters which are kept under consideration in the proposed algorithm are grid size, load balance factor (LBF), the speed of the nodes, directions and transmission range. The ICGWO is compared by the tested method with the well-known meta-heuristics, Comprehensive Learning Particle Swarm Optimization (CLPSO) and Multi-Objective Particle Swarm Optimization (MOPSO) for VANETs. Experiments are performed by the varying the key parameters of the ICGWO. This is done for measuring the effectiveness of the proposed algorithm. The graphical results show that the proposed methodology is providing the optimized results as compare to competitors.