Vehicular clustering based on future position estimation for highway scenario

Wireless communication through vehicles has been a fascinating research area for providing a variety of applications. Vehicular Ad hoc Networks has attracted attention from both industrial and research communities in recent years. It evolved with an aim for addressing comfort and safety issues of the vehicles and has received increasing interest as it integrates the capabilities of new generation wireless networks with vehicle characteristics. However the communication in VANETs faces a lot of overheads due to packet collisions. There is a need of wireless backbone architecture which gathers the nodes into the clusters from which some nodes are elected as cluster heads for handling network responsibilities in order to reduce the overheads and improve communication. Although several clustering algorithms have been proposed for VANETs, but clustering based upon prediction is still a challenge. In this paper prediction based clustering algorithm for VANETs is proposed that specifically considers vehicles moving in highways. The proposed scheme has been tested and performance evaluation results demonstrate that this scheme provides a better and stable clustering structure thereby reducing the overheads.