网络流量识别方法研究  

Study on Network Traffic Identification Method

作　　者：梁伟[1,2] 李晗[2] 

机构地区：[1]常熟理工学院计算机科学与技术学院,湖南常熟215500 [2]南京邮电大学信息网络技术研究所,江苏南京210003

出　　处：《通信技术》2008年第11期88-90,共3页Communications Technology

基　　金：江苏省科技技术项目:"新一代通信网络业务安全主动监控技术研究"(编号:BG2007045);江苏省科技技术项目:"基于智能流感知的自组织覆盖网络自适应监控机制研究"(编号:BK2007603)。

摘　　要：随着P2P和多媒体流量的发展,原有的流量识别方法越来越显现出其不足。为识别这些流量,需要识别效率更高的识别方法。文中提出了一种混合流量识别方法,此方法将特征识别和会话行为映射方法相结合,进行精确的流量识别。接着给出了这种识别过程的流程图,对识别过程进行了说明。对包的识别是基于优先级的特征识别,以提高识别效率。通过实验,选取四种应用Monkey3,eDonkey2000,MSN messenger,BitTorrent流量,将这些单个流量和混合流量的识别结果进行了比较。由实验结果可得出该方法对混合流量识别率比单个流量识别率高。With the development of P2P and multimedia traffic, the existing application traffic identification method is becoming more and more unsuitabe. In order to identify this traffic, a identification method with high accuracy is needed. Signature identification and session behavior mapping method are combined for accurate traffic identification. The paper proposes a priority-based signature identification to improve the accuracy of identification. In experiment, 4 applications included onkey3, eDonkey2000, MSN messenger and BitTorrent are chosen their identification results are compared. The experiment indicates that this combined identification method is more efficient for identifying mixture traffic than for identifying single traffic.

关 键 词：流量识别 净荷特征识别方法 流量统计特征识别方法 

分 类 号：TP393.06[自动化与计算机技术—计算机应用技术]