A Novel Triple Cluster based Routing Protocol (TCRP) for VANETs

Vehicular Ad-hoc Network (VANETs) is a collection of intelligent vehicles aims to get control over transportation problems and reduce accidental ratio. Clustering refers to the classification of vehicles in groups in order to minimize the traffic overhead by selecting a responsible node called Cluster Head. In this paper, we introduce a new Triple cluster based routing protocol (TCRP) for cluster formation & Vehicle Head (VH) selection by using modified K-Means & Floyd-Warshall algorithm. By assuming truncated normal distribution, the confidence interval of vehicles velocities is defined. The modified K-means divides the vehicles into three clusters within their velocities confidence range. Floyd-Warshall algorithm calculates the shortest distance for all pair of vehicles of VANETs. A vehicle having smallest average distance to the rest of vehicles & having least variance in velocity will be selected as Vehicle Head (VH). Our analysis and simulation results show that the proposed TCRP makes the cluster shape consistent and avoid reselection of VH in further rounds, thus forming quite stable groups of vehicles.