Reliable multicast as a Bayesian coalition game for a non-stationary environment in vehicular ad hoc networks: a learning automata-based approach

Abstract

Vehicular ad hoc networks (VANETs) are offering lot of services for the benefits of community of users. But, due to the dynamic nature of VANETs, it is a challenging task to perform reliable multicast. To address this issue, this paper proposes a new approach called reliable multicasting in non-stationary environment as a Bayesian coalition game using learning automata (RMBCG-LA) for VANETs. A new metric, probabilistic reliability index (PRI) is computed by each player. A coalition among the players of the game is formed using Bayesian network with a threshold in each coalition is based upon the conditional probability. For each action performed by the automaton, its action is rewarded or penalised by the non-stationary environment in which it is operates. The performance of the proposed scheme is evaluated in comparison with the well-known existing schemes. The results obtained show that our proposed scheme is better than the other schemes of its category.