带背景流的P2P流量识别技术研究  

Study on the Identification of P2P Traffic with Background Traffic

作　　者：欧阳玲[1] 宋克[2] 

机构地区：[1]中原工学院,河南郑州450052 [2]国家数字交换系统工程技术研究中心,河南郑州450005

出　　处：《计算机测量与控制》2011年第10期2562-2563,2570,共3页Computer Measurement &Control

摘　　要：针对P2P(peer-to-peer,对等体网络)应用系统中对等体主机的行为特征与P2P业务流量特征多样化、复杂化,使得单纯利用一种典型特征的P2P流量分类技术的识别精度不高的问题,提出了一种新的P2P流量多阶段识别方法;该方法根据P2P应用流量的一系列固有特征,可以从聚合网络流中识别P2P流量;通过实验表明,该方法P2P流识别精度可达99.7%,同时错误分类精度0.3%。The Diversity and complexity of Peer hosts in P2P application systems and P2P application traffic make P2P traffic identification approach based on only typical feature inaccurate. We propose a novel multi-phase identification method to reveal P2P traffic from traffic aggregation. Our method is based on a set of heuristics derived from the robust properties of P2P traffic. Experiments indicate the classification accuracy of our proposed method can reach 99. 7%, while the false positive is lower than 0.3%.

关 键 词：P2P 聚合流量 典型特征 流量分类 

分 类 号：TP393.0[自动化与计算机技术—计算机应用技术]