基于用户主动过滤的P2P文件污染防御  

Fighting Against P2P File Pollution by Encouraging User to Delete Malicious Content Actively

作　　者：黄志华[1] 陆松年[1] 张爱新[2] 李琳[1] 

机构地区：[1]上海交通大学电子工程系,上海200240 [2]上海交通大学信息安全工程学院,上海200240

出　　处：《小型微型计算机系统》2014年第10期2325-2330,共6页Journal of Chinese Computer Systems

基　　金：国家"九七三"重点基础研究发展计划项目(2010CB731403;2010CB731406)资助

摘　　要：文件污染是P2P文件共享系统面临的主要安全威胁之一,现有的文件污染防御机制包括基于反馈的声誉机制和基于用户行为的防御机制,前者面临用户合作度不高和反馈攻击等问题,后者不提供分级服务,无法激励用户过滤污染.本文提出一种激励用户主动过滤的文件污染防御机制,不需要依赖用户反馈.系统客户端自动跟踪下载节点过滤下载文件的行为,生成下载节点对服务节点的信任,该信任决定了下载节点将来获得服务的质量.仿真结果表明,本文机制能为好节点提供高效稳定的下载性能,同时惩罚自私节点和懒惰节点.这种分级服务能有效激励用户过滤污染文件.File pollution is one of several key security issues in Peer-to-Peer （P2P } file sharing systems. Existing defense mechanisms include reputation systems based on user feedback and defense mechanisms based on user sharing behaviors. However, the former suf- fers from shortage of user cooperation and feedback attacks. The latter cannot encourage user to filter the polluted files since there is no service differentiation. In this paper, we propose a defense mechanism named actively filtering based defense mechanism （ AFDM }, in which the requirement of user feedback is removed. P2P system may automatically monitor user behavior of filtering the files down- loaded from service peers and compute the trust placed by a user on the service peer. According to the trust value, user will obtain dif- ferential services in the future. Simulating results show that AFDM can provide robust efficient download performance for good peers while punish selfish and lazy peers in various environments. This can encourage user to delete the polluted files actively.

关 键 词：P2P文件共享 文件污染 用户行为 主动过滤 

分 类 号：TP393[自动化与计算机技术—计算机应用技术]