Dynamic Clustering Mechanism to Avoid Congestion Control in Vehicular Ad Hoc Networks Based on Node Density

Abstract

Due to the rapid growth in wireless communications, vehicular ad-hoc networks (VANETs) face many challenges over wireless communication networks. Nowadays VANETs have become a major research and they have incredible resources to improve road traffic efficiency, safety, convenience and comfort to both drivers and passengers etc. Congestion occurs when the vehicles are in the dense part while the network node is carrying more data than it can handle. The proposed DBDC: density based dynamic clustering devices determine node density of the precise location in a lane and provide a proactive solution for congestion. To trigger the congestion control process in the cluster, we use average vehicle density threshold which is calculated using trained dataset. If the node density is greater than the threshold value, the dynamic clustering process is triggered. Our simulation result indicates that the overall performance of our scheme is better than multi agent dynamic clustering and vehicular weighted clustering algorithm in terms of average cluster number, cluster head duration and average cluster member duration.