Optimized clustering for data dissemination using stochastic coalition game in vehicular cyber-physical systems

Vehicular cyber-physical systems (VCPS) have emerged as one of the most powerful technologies for providing cost-effective services to the end users with minimum delay even with high mobility of the end users. As the contact time of the vehicles with the nearest access points is very less, so by integrating vehicles with the cloud environment can provide various services to the end users even with their high mobility. Keeping focus on these points, this paper proposed a new optimized strategy selection for data dissemination using a stochastic coalition game in VCPS environment. Vehicles in the coalition game are assumed as the players of the game which access a finite number of resources from the cloud. Learning automata (LA) stationed on the vehicles collect and process the information from the environment based upon pre-defined strategies. They are assumed to form a coalition based upon predictive clustering among one another using a pre-defined metric. Based upon the payoff matrix, each player executes, algorithms for cluster and leadership formation. Moreover, an algorithm for centralized supervision is also designed. At any instant in the game, only those players are allowed to make a move which are having highest payoff value so as to make a balance with respect to the moves taken by all the players in the game. The performance of the proposed scheme is evaluated using various evaluation metrics in different network scenarios. The results obtained show that the proposed scheme is better than the case where it is not applied. Specifically, there is an increase of 10–20 % in packet delivery ratio and 20 % reduction in delay in accessing various services from the cloud using the proposed scheme.