P2P系统中基于信誉感知的超级节点选择算法研究  

The research of the Reputation-Aware SuperNode Selection Algorithm in P2P system

作　　者：刘玉枚[1] 杨寿保[1] 陈万明[2] 郭磊涛[1] 韦冬[1] 

机构地区：[1]中国科学技术大学计算机科学与技术系,合肥230026 [2]中国科学技术大学自动化系,合肥230026

出　　处：《中国科学院研究生院学报》2008年第2期197-203,共7页Journal of the Graduate School of the Chinese Academy of Sciences

基　　金：国家自然科学基金项目(60273041和60673172);国家863计划(2002AA104560和2006AA01A110)资助

摘　　要：针对层次式P2P系统中恶意超级节点频繁离开网络导致系统不稳定甚至崩溃的问题,提出一种新的基于信誉的超级节点选择算法.建立节点基于历史交换回馈信息的信誉评估算法,以信誉评估为基础,构建信誉感知的超级节点选择算法,选择信誉高的节点为超级节点.理论分析和仿真结果表明,该算法有效地提高了系统的稳定性.In the P2P Hybrid Structure,it makes the system fluctuate,even makes the system collapse when hostile superNode leaves the system frequently.Therefore,we proposed a new Reputation-Aware SuperNode Selection Algorithm to make the system stable.It built the reputation model which was based on the history interaction feedback,and then built the reputation-aware superNode selection algorithm,selecting the high reputation node as the superNode.Simulation and analysis show that when it selected high reputation node for SuperNode,the frequency that the superNode leaves the system is reduced and the system＇s stabilization is improved.

关 键 词：对等网络 超级节点 信誉感知超级节点选择算法 信誉 恶意节点 

分 类 号：TP393[自动化与计算机技术—计算机应用技术]