Rough Set Based Fuzzy Scheme for Clustering and Cluster Head Selection in VANET

Abstract

In Vehicular Ad hoc Network (VANET), clustering helps vehicles to communicate with other vehicles and to the nearby Road Side Unit (RSU). Conventional clustering methods follow precise clustering which degrades the stability of cluster formation. To stabilize the formation of clusters, the vehicles in the boundary region of more than one cluster should be uniquely added to the proper cluster. Fuzzy set representation of clusters makes this possible by assigning a membership value to all the vehicles and supports the formation of clusters based on this membership value. Since the cluster lifetime is very minimal in vehicular network, fuzzy based clustering is too descriptive to interpret the clustering results. In this paper, the rough set based fuzzy clustering is employed for formation of clusters in a VANET. Using this scheme, a vehicle in the transmission range of more than one cluster namely, the boundary vehicles are assigned with a membership value. Based on the fuzzy rule base, the vehicles are assigned to the appropriate cluster. Theoretical analysis and experimental results show that rough set based fuzzy scheme obtains 10 % to 20 % more average cluster lifetime and 20 % to 25 % more cluster head lifetime when compared to existing approaches.Index Terms-Clustering, fuzzy sets, rough sets, vehicular ad hoc network.

Modeling and Analysis of New Hybrid Clustering Technique for Vehicular Ad Hoc Network

“…The low tolerance level problem was solved in [9] by proposing a neighborhood rough-sets model to create variable-precision neighborhood approximation sets and positive regions, so the rough set becomes a dynamic model. The dynamic rough set model with the fuzzy scheme was used in [10] to make a more efficient rough set clustering system. This scheme enhances the rough set and implements it in a VANET dataset as a CH selection model.…”