P2P中一种改进的可信模型  

An Improved Trust Model in P2P

作　　者：郑光勇[1] 任晓慧[1] 李肯立[1] 

机构地区：[1]湖南大学计算机与通信学院,湖南长沙410082

出　　处：《计算机工程与科学》2008年第3期7-10,27,共5页Computer Engineering & Science

基　　金：国家自然科学基金资助项目(60273075);教育部重点资助项目(05128)

摘　　要：对基于贝叶斯网络的可信模型中的资源搜索算法和可信度计算方法进行重新设计,本文提出一种改进的P2P可信模型。改进后的资源搜索算法使节点接收到的冗余包数量减少,并提高了系统的可靠性。在此基础上,针对P2P网络的匿名性,提出了一种新的运用服务次数和对评价值取对数方式进行可信度计算和更新的方法。此方法可以有效地阻止恶意节点对正常节点和可信节点的攻击,同时还可防止恶意节点间通过小集团合作方式来提高各自节点的可信度。实验结果表明,本模型较现有模型在抑制P2P网络中恶意节点的活动方面具有更好的效果,且增加了系统服务可靠性.In this paper, we propose an improved trust model based on the analysis of the Bayesian network-based trust model. It can reduce the redundancy packages received by the peers and enhance the system reliability through improving the resource search algorithm. Meanwhile we improve the trust computing and updating method, which applies logarithm to the evaluation and counting of services to compute the trust. And it prevents malicious peers to attack the normal peers and trust peers efficiently. Meanwhile it restrains the malicious peers from increasing their trust respectively by their group collaboration. The sketch maps of the trust value illustrate that the new model controls the behavior of malicious peers in P2P more efficiently.

关 键 词：P2P 可信度 名誉度 恶意节点 

分 类 号：TP393.08[自动化与计算机技术—计算机应用技术]