Joint Multi-hop Clustering and Routing in VANETs using Array of Doubly Linked List

This paper addresses the problems of multi-hop clustering, Cluster Head (CH) selection, and routing in vehicular ad-hoc networks. We propose an efficient algorithm called Joint Multi-hop Clustering and Routing (JMCR) to improve the performance of the network. JMCR uses only local information to cluster vehicles in such a way that not only reduces the total number of clusters, but also maximizes the stability of clusters. It considers both relative speed and Euclidean distance factors to select an appropriate CH for each cluster. In order to update the routing database of each node, an array of doubly linked list is used in which the creation of different routing paths starts from the boundary nodes and continues to CH. Compatibility with reactive clustering and the ability to dynamically follow the network conditions are another advantages of the proposed intra-cluster routing method. Simulation results conducted in NS2 confirm the efficiency of the proposed method in terms of end-to-end delay, overhead, and packet delivery rate.