一种新的基于博弈论的P2P激励机制  

Novel incentive based on game theory in P2P networks

作　　者：徐海湄[1] 郑相全[2] 齐守青[2] 聂晓文[1] 

机构地区：[1]电子科技大学计算机科学与工程学院,成都610054 [2]解放军重庆通信学院,重庆400035

出　　处：《计算机应用研究》2008年第9期2787-2789,共3页Application Research of Computers

基　　金：国家自然科学基金资助项目(10577007);重庆市重点自然科学基金资助项目(CSTC;2007ab2017)

摘　　要：针对P2P系统中的搭便车和公共悲剧问题,提出了一种基于博弈论的激励机制。每个节点从个人理性出发,在最大化个人收益的同时根据请求者的贡献分配带宽,并利用动态规划方法求出了最优解,实现了有限资源的帕累托配置,达到了社会收益最大化的目的。仿真实验表明,该激励机制促进了社会收益的快速增长,达到了激励节点参与资源共享的目的。In order to solve free rider and tragedy of the commons problems in peer-to-peer（P2P） systems, this paper proposed a novel incentive based on game theory. While trying to maximize its own utility subjected to individual rationality, every peer allocated bandwidth resources efficiently according to competing peers＇ contribution values to maximize the social utility. The simulation result shows that the incentive increased the social utility of the whole P2P system rapidly and incentive every peer to share resources effectively.

关 键 词：不完全信息静态博弈 帕累托 社会收益 个人收益 动态规划 

分 类 号：TP393[自动化与计算机技术—计算机应用技术]