Enhanced VANET Performance by Stable Cluster Head Selection Using Novel Stack of Algorithms

The use of VANET turns the roadway into a communal space, enhancing safety and facilitating data sharing among drivers. Data from each automobile is transmitted to the cluster head (CH) via a cluster. There are two main aspects that impact the effectiveness of a VANET network: the optimal number of the vehicle groups and the choice of the control node for stability. In this paper, we model the VANET as a hypergraph network and construct a state-of the-art modularity-based hypergraph clustering formulation by maximising tensor traces and using the modularity matrix as a feed to determine the optimal number of clusters. Automobiles were selected from each group based on how fast they were compared to others in the group, how trustworthy they were deemed by the group as a whole, how long their links lasted, and how well connected they were. All parameters are first transformed to eigenvalues, then a single CH is chosen. By adopting a fuzzy decision-making technique based on three factors, the optimal CH is selected for all vehicles in that cluster. Furthermore, the proposed method for selecting the reliable CH to improve throughput has been validated and significantly improved by comparing it to the other existing techniques.