An Incentive Scheme Based On Coalitional Game Theory For Sparse And Dense VANET

In VANET (vehicular ad hoc network), especially the sparse VANET, because of the nonexistence of end-to-end connections, to make end-to-end messaging possible, it is essential that nodes take advantage of connection opportunities to forward messages. On the basis of VANET’s characteristics, we formulate a incentive scheme based on game theory SVIS (Incentive Scheme for Sparse VANET). The scheme can efficiently stimulate every nodes in VANET cooperate each other. All nodes forwarding messages has some limitations. It has obvious advantage in sparse VANET, however, it does not adapt to dense VANET. Because of the high mobility of nodes, vehicular scenarios in VANET are various. The density of VANET has effect on message forwarding. When VANET is sparse, there are no end-to-end connections. It is essential that nodes should take advantage of connection opportunities to forward messages. Thus, it is crucial to make sure that selfish nodes have motivations to forward messages. When VANET is dense, if all nodes are selfish and do not forward messages, messages cannot be delivered successfully. By contrast, if all nodes are stimulated to forward messages, lots of redundant packets would exist and even cause broadcast storm problem. To address these issues, we extend our scheme SVIS to take the density of VANET into consideration. An Incentive Scheme SDVIS (Incentive Scheme for Sparse and Dense VANET) which can adapt to variable vehicle density is proposed based on coalitional game theory. When VANET is sparse, SDVIS constrains the selfishness and all nodes (all vehicles in VANET) make full use of connection opportunities to forward messages. Further, the delivery rate is improved and the delay is reduced. When VANET is dense, SDVIS can reduces the number of redundant messages and improves the delivery rate. We give some rigorous proofs to show the existence of solution. At last, simulation shows the correctness and effectiveness.