迅雷协议分析及迅雷流量识别方案的研究  

Analysis of Thunder Protocol and Research of Traffic Identification Scheme

作　　者：彭骥[1] 陈运[1] 杨义先[2] 

机构地区：[1]成都信息工程学院信息安全研究所,四川成都610225 [2]北京邮电大学信息安全中心,北京100083

出　　处：《成都信息工程学院学报》2013年第6期574-579,共6页Journal of Chengdu University of Information Technology

基　　金：国家自然科学基金资助项目(60873216)

摘　　要：迅雷作为中国使用最广泛的下载软件,给网络带来沉重的负担,而迅雷流量识别作为有效控制迅雷流量的基础,对其识别方案的研究有着重要的意义。首先对迅雷的下载机制进行分析和探讨,通过实际的抓包分析,提出一种基于行为模式的迅雷流量识别方案,并给出识别流程,最后再对该识别方案进行实验验证。实验结果表明,该方案对比基于特征字的流量识别方案和基于深度流检测的流量识别方案,有更高的识别率。As the most widely used download software, Thunder brought heavy burden to the network. Thunder traf- fic identification is the important base for its effective traffic controlling, and is important for the research of traffic i- dentification scheme. Firstly, this paper analyzed and discussed mechanisms of Thunder download. Secondly, by the analysis of the actual packet capture, this paper proposed Thunder traffic identification schema based on its behavior mode, and presented the identification process. Finally, we made the experimental verification for the identification scheme. The results show that this scheme has higher recognition rate than that of the scheme based on character i- dentification or DFI.

关 键 词：计算机应用技术 信息安全 迅雷 协议分析 流量识别 识别方案 

分 类 号：TP393[自动化与计算机技术—计算机应用技术]