Stable Clustering Algorithm for Routing Establishment in Vehicular Ad-Hoc Networks

With regard to the complex and varied urban scenes in Vehicular Ad-Hoc Network, such as the vehicle nodes with fast speed, unstable links and frequent changes in network topology, this paper proposed a stable clustering algorithm to establish routing for VANET. In this algorithm, clustering was first formed according to the Similar Neighbor Node Table. Then on the basis of the Highest Connectivity Algorithm, parameters such as position and speed of the node were introduced to calculate the Selection Priority which were used to produce the preferred and the alternative cluster head node. The introduction of alternative cluster head nodes improved the stability of clustering to a certain extent. Finally, the clustering was established and maintained for six special scenarios. Compared with the traditional clustering algorithm in VANET, it had the characteristics of lower end-to-end delay, higher packet delivery rate and lower change rate of preferred cluster head nodes, which greatly improved the communication quality of VANET.