A scatter search based heuristic for reliable clustering in vehicular ad hoc networks

Abstract

Achieving a safe, comfort and autonomous driving in vehicular ad hoc networks (VANET) is the great interest of a large number of researchers and car manufacturers. Despite the variety of the proposed approaches and the development of communications technologies, there are no typical solutions. Indeed, several recent studies prove the practical advantages of heuristic method to solve various problems of optimization. Therefore, we used in this paper a Hybrid Scatter Tabu Search (HSTS) based heuristic approach to assign cluster members (CMs) to convenient cluster heads (CHs). We addressed in this work the cluster formation phase in our Weighted K-medoids Clustering Algorithm (WKCA) proposed recently. The main objective is to derive new solutions from the combination of previous one, including the network coverage as a special criterion. To achieve this objective and to locate the global minimum, we integrate the tabu search in the inner process of the scatter search. To the best of our knowledge, there is no study that uses the scatter search to perform clustering in VANET. By simulation, results show that our scheme improves the network stability in term of several metrics compared with prior approaches.