Investigation on improved clustering techniques for vehicular ad hoc networks

Abstract

VANET is an emerging technology used to provide an intelligent transport system. In VANET, every vehicle communicates to other nodes in a dynamic environment. The dynamic topology nature of VANET leads to poor network stability and link quality. Clustering is a node grouping technique introduced to handle these difficulties. Cluster head selection is an important task in clustering for data aggregation and spectrum management. To meet the clustering challenges of VANET, a new clustering technique is proposed in this work. The main objective of the proposed clustering technique is to increase the packet delivery ratio of the network and to effectively handle the frequent topological changes and higher vehicle mobility of the network. The contributions of this works are as follows: 1)Proposed a Middle-Order based vehicle clustering-based cluster construction model.2) Proposed an optimized cluster head selection method by considering multi-objective parameters 3) Proposed an optimized secure and stable clustering for VANET. The cluster head selection process is considered as an optimization problem and solved using red deer optimization (RDO). RDO is a nature-inspired meta-heuristic algorithm inspired by the matting behaviour of red deer. The performance metrics used in this work are defined as below. The proposed methods are analyzed in terms of Cluster stability, Average packet delay, packet loss probability, Re- clustering Ratio and Attacker detection rate. The proposed methods resulted in a better performance when compared to other standard clustering protocols. newline