摘要
为识别网络中大量的P2P流,为网络管理提供技术支持,提出一种基于行为特征加权的P2P流识别模型。该方法不需深度报文检测(DPI),采用深度流行为检测(DFI),统计P2P应用的行为特征,并通过行为特征对P2P流识别的有效性的贡献进行分级,特征按所属级别被赋予不同的权值,通过特征匹配和加权,识别P2P流。仿真实验结果表明,该方法能够有效识别P2P业务流。
For identifying large numbers of P2P flows in the Internet and giving technical support to the Internet, a new model for identifying P2P flows is developed using the weight of the sum of characteristics. This model dose not need deep packets inspection （DPI）, but only depends on the deep flow inspection （DFI）, classifies P2P behavior characteristics based on their contribution to the validity of identification and endows with the weight. It uses characteristic matching and sum of weight to identify P2P flow. The experiment results show that this method exhibits good performance for identifying P2P flow.
出处
《计算机工程与设计》
CSCD
北大核心
2009年第20期4805-4807,F0003,共4页
Computer Engineering and Design
基金
国家863高技术研究发展计划基金项目(2007AA01z2a1)
