Probabilistic B-Tree Based Clustering Algorithm For Vanet With Data Aggregation

Vehicular ad hoc network (VANET) is a kind of ad hoc network, where the wireless communication has been established between the moving vehicles. Recently, the clustering scheme is suggested as an effective solution to handle the fast topology changes of vehicular ad hoc networks. However, the stability of the existing clustering approaches shows poor performances due to highly dynamic scenario of VANET. Thus this paper proposes a probabilistic B-Tree based multi-hop clustering scheme for VANET. A probabilistic density function is computed based on the velocity, speed and acceleration of the vehicles in order to select the cluster head (Designated Node (DN)) and the Backup Designated Node (BDN). The clustering will be performed using the direction of vehicles. A B-tree has been constructed for each cluster and each node will keep and maintain the entire topology structure. Once the DN failed, the BDN will be the DN and new BDN will be selected and the Btree will be rearranged, where the proposed scheme enables the faster convergence. Furthermore the data aggregation will be performed at the designated nodes to reduce data transmission and that makes the effective bandwidth utilization. The NS2 simulation has been used to evaluate the performance of the proposed scheme