Study on Preventing Data Collision by Enhanced Safety or Alert Message Broadcasting Strategy in Vehicular Ad-Hoc Network (VANET)

Abstract

A vehicular adhoc network (VANET) is a mobile adhoc network in which moving road vehicles are network nodes. The major problem faced in VANET is transmitting the safety message or an ALERT message between the vehicles to avoid collisions. The large number of vehicles moving in a particular area will cause the data congestion that results in a long delay on delivering the messages. Clustering the vehicles into groups is an efficient approach to broadcast the messages with a less delay and high delivering ratio. The performance of the clusters is greatly influenced by the selection of Cluster Heads (CHs). In this paper, the formation of clusters and cluster head selection is formulated as optimization problem in order to maximize the efficiency of the network energy and link quality. The formulated problem has been solved using two approaches: Density based Clustering and the cluster heads selection using Differential Evolution (DE) Algorithm. By combining these approaches the safety or Alert messages will be transmitted easily to all the vehicles by dynamically adapting transmission power and contention window size. The performance of the proposed ESMBS protocol is compared and evaluated against the existing AdvB and CBAPA protocols. The proposed scheme can control various performance metrics such as the collision, packet delivery ratio, packet delay ratio, latency, transmission time and throughput. The proposed scheme broadcasts the alert or safety messages by reducing the probability of collisions.