Incentive Strategies Based On Game Theory For Selfish Nodes In Cooperative Communication Networks

Through the method of sharing antennas among neighboring nodes to transmit signals, cooperative communication can not only form a virtual multi-antenna array that brings spatial diversity gain to improve the signal quality at the receiving terminal, but also can expand the coverage of the signal by forwarding of relay nodes. Therefore, cooperative communication will become the key technology in the future wireless communication systems to combat multi-path fading, provide high quality and high speed service. However, most of the research on cooperative communication is based on the assumption that the nodes are willing to cooperate. When the nodes belong to different authorities(users or service providers) in the network, the cooperation action will cause the consumption of nodes’ resource. And a self-interested rational user will not take part in cooperation to save its own resource. Thus, based on the assumption that the nodes are willing to cooperate in the actual network application is not realistic. It is necessary to study the mechanism of encouraging the selfish node to take part in cooperation in order to improve the work efficiency of the network, work for the same task and realize the effect of “the whole greater than the sum of the parts”. Furthermore, the problems of necessity and prerequisite of cooperation, the selection of partners, the implementation of cooperation strategy, the benefit of nodes or the entire communication network system are resolved. Consequently, it is very necessary and meaningful to establish a set of effective incentive mechanism to promote the selfish nodes to participate in the cooperation in the network, so as to make cooperative communication can be applied to the specific wireless communication scenarios.Aiming at the problem that the selfish nodes in a wireless network are unwilling to cooperate to save their own resource, an incentive mechanism based on exchange bandwidth resource is proposed. By sharing the bandwidth resources of the nodes, the purpose of establishing cooperation and increasing the benefit of the nodes is achieved. First, the cooperation bandwidth allocations among the nodes is modeled as a bargaining problem of cooperation game, and the cooperation bandwidth allocations based on the Nash bargaining solution(NBS) are obtained by the Lagrange multiplier method. Secondly, a new method of cooperation bandwidth allocations based on the Kalai-Smorodinsky bargaining solution(KSBS) is proposed, and two kinds of fairness index for evaluating the utility fairness of NBS strategy and KSBS strategy are proposed. Finally, based on the analysis of the effect of different modulation order of M-QAM on cooperation performance, a new adaptive modulation KSBS(AM-KSBS) strategy is proposed. Compared with the non-adaptive cooperation strategy, the proposed AM-KSBS method can increase the cooperation scope of the nodes and can ensure that the nodes obtain the best cooperation utility gain in cooperation.To stimulate the selfish nodes of networks to participate in cooperation, a cooperation strategy of optimal price incentive algorithm based on NBS and optimal transmit power algorithm based on particle swarm optimization(OPINBS-OTPPSO) is proposed. As a result, two key problems, i.e. how to cooperate and when to cooperate, are solved. For the first problem, relay node is paid by source node for forwarding source node’s data, then cooperation between source node and relay node can be reached. We model the optimal bid of the source node as Nash bargain, and Nash equilibrium of optimal bid which has Pareto efficient is given. Consequently, the optimal bid can guarantee that source node and relay node obtain optimal revenue. For the second problem, it is decided by a proposed cooperation condition and which can guarantee nodes’ utility achieved through cooperation is not lower than that achieved without cooperation. Then, after obtaining the bid of source node, the source node and relay node determine optimal transmit power through particle swarm optimization to maximize their revenue. Simulation results show that, compared to random price incentive mechanism, the proposed OPINBS-OTPPSO strategy can make source node and relay node obtain optimal revenue, meanwhile, the proposed OPINBS-OTPPSO strategy can ensure optimal performance of the whole system compared to the strategy that source node and relay node use constant transmit power.According to the problem that very few literatures used both non-cooperative game and cooperative game to the one scenario simultaneously to study the incentive mechanism, the both game approaches are applied to analyze the same network where the price incentive mechanism is used to stimulate the selfish nodes to take part in cooperation. First, the non-cooperative and cooperative game are adopted to analyze the problems that how much bandwidth of initiating cooperation node(ICN) forwards data through participating cooperation node(PCN) and how much bandwidth of PCN helps ICN to relay data. Although the non-cooperative game theory and cooperative game theory can achieve the same results, it can be concluded that the cooperative game theory is more suitable for the networks with central control node, while the non-cooperative game theory is more suitable for the networks without central control node. Secondly, a cooperation engine(CE) is proposed and a cooperative communication system model with CE is depicted. Finally, an algorithm based on game theory is put forwarded to realize the proposed CE. The simulation results demonstrate that, compared with the system without incentive mechanism, our proposed system with CE can significantly improve the ICN’s metric that is measured by bit-per-Joule, and increase the PCN’s revenue.