基于重复博弈和惩戒机制的P2P协作激励信誉模型

Collaboration Incenting Reputation Model Based on Repeated Game Theory and Punishment Mechanism in P2P Networks

现有的信任模型在计算节点信任度方面存在时间复杂度和通信报文量均比较高的问题,而且没有考虑到理性节点比率和节点耐心程度对信任模型的影响.在研究已有模型的基础上引入惩戒机制,建立一种基于重复博弈理论和惩戒机制的P2P网络信誉模型,以达到激励节点协作的目的.通过分析P2P网络中的不协作均衡,描述了基于惩戒思想的信誉机制;利用重复博弈理论分析了此机制下节点的重复交易行为,提出并证明了考虑理性节点比率和节点耐心程度的协作均衡定理;最后得出了一个协作条件,在此条件下,节点将慑于惩戒机制而采取协作策略.实验结果表明,该模型在促进节点协作方面是有效的.

To deal with the shortcomings in high time complexity and network overhead in calculating nodes＇ trust degree,as well as to take into account the impact of rational nodes ratio and nodes patience degree on the trust model,the punishment mechanism is utilized to establish the P2P trust model based on the repeated game theory.The trust mechanism is described and the repeated trade actions are analyzed by using repeated game theory.The collaborative equilibrium theorem considering the rational nodes ratio and nodes patience degree is proposed and proved,from which a collaboration condition is concluded and the nodes should adopt the collaboration strategy for fear of being punished under this condition.The experimental results show the effectiveness of this trust model in promoting the nodes collaboration.