Cluster Formation And Cluster Head Selection Approach For Vehicle Ad-hoc Networks Using K-Mean And Floyd-Warshall Techniques

Vehicular ad-hoc Networks(VANETs)plays a vital role in Intelligent Transportation Systems(ITS).They transmit essential infrastructural information about roads and traffic conditions along with safety messages and entertainment packages for passengers.In broader context,VANETs provide vehicles with a possibility to connect with each other for different purposes e.g.,to exchange safety messages,route congestion and under-construction path information etc.Literally,VANETs is the explicit sub-type of Mobile ad-hoc Networks(MANETs)in which procedures are carried out by utilizing nodes with considerable dynamics,hence,have principle safety applications in urban traffic environments.Since such networks are comprised of massive-scale traffic data,therefore,they function by clustering their domain vehicles by employing clustering algorithms and by specifying the leading node or Cluster Head(CH)as per systematized protocol.In VANETs,the CH is held responsible for collecting all the useful information about its respective surroundings by adopting a set of procedures.This research study devises an integrated scheme by coupling two well-known clustering algorithms,K-Mean and Floyd-Warshall algorithm(KMFWA).The integrated KMFWA scheme entails the division of vehicular points followed by their grouping using K-mean,while all pairs of shortest distances corresponding to each vehicle in the cluster are then computed by means of FloydWarshall algorithm(FWA).In KMFW,the FWA centralizes the vehicular node with the least average distance as compared to its neighboring nodes and then declares this centralized node as CH for its respective cluster.FWA overall selects a centralized vehicle as a CH,hence its stability time will improve significantly.We presented a novel Cluster formation and CH Selection approach for VANETs using KMFW.The proposed algorithms had two parts that is,providing divisions parts for vehicles group using K-Means and then in second part,calculating all pair shortest path for every vehicles within the cluster using FWA.Criteria for CH selection in FWA is,the node in the cluster having small average distance to all other vehicle in the cluster.We showed through simulations that our proposed scheme outperforms the contemporary algorithms in terms of mean Peak Signal to Noise Ratio(PSNR),transmission range,average connectivity and average duration of CH.Result shows that PSNR with varying number of vehicles.Intuitively,increase in the number of vehicles increases the signal quality.With greater number of vehicles,the signal would incur lesser path loss.In terms of mean PSNR,our proposed KMFW scheme has better performance in comparison with Clustering Algorithm in Vehicular ad-Hoc Networks(VWCA)and Cluster-Based Traffic Information Generalization(CTIG).After certain number of vehicles per kilometer,250 in this case,the mean PSNR will remain the same.However,for lesser number of vehicles the proposed KMFW performs better than VWCA and CTIG.