A P2P Traffic Identification Method Based on VFDT

摘要：

We analyzed the memory limitation problem of traffic identification arithmetic when faced the large and fast stream data, and extracted the traffic attribute based on the P2P working mechanism. Using VFDT method to identify the P2P traffic can scan the traffic data only once relying on the Hoeffding Restriction, the method reduce the complexity of algorithm on the part of timing and memory and ensure the identification correction rate. The experiment shows the method can get good performance.

关键词： P2P Application Traffic Identification Machine Learning

作者: Meifeng Sun Yun Zhang Jingtao Chen Tingting Shi

作者单位: Information Engineering College, Yangzhou University Yangzhou, P.R. China

会议类型: 国际会议

会议名称: 2010 International Conference on Intelligent Computation Technology and Automation(2010 智能计算技术与自动化国际会议 ICICTA 2010)

会议地点: 长沙

会议语种:英文

页码: 281-284

在线出版日期: 2010-05-11（万方平台首次上网日期，不代表论文的发表时间）