论文部分内容阅读
当今网络环境中,新型、未知应用大量涌现、并且网络技术日新月异,这对网络流量的识别带来严重挑战.传统的基于IANA端口的应用识别方法逐渐失效,利用流行为统计特征的流量分类方法在精度和实时性上又存在先天的缺陷.本文提出了一种新型的基于数据挖掘的应用识别方法,该方法从应用会话内容中自动提取应用特征,然后根据特征匹配识别应用.通过仿真实验测试识别率、正确率及综合指标,结果表明算法是有效性,能够实现应用层流量的精确分类.
In today’s network environment, the emergence of new and unknown applications, and the ever-changing network technology, which brings serious challenges to the identification of network traffic.Traditional IANA port-based application identification methods are gradually invalid, the use of traffic behavior statistical characteristics of traffic classification method Accuracy and real-time nature.This paper presents a new method of application recognition based on data mining, which extracts the application features automatically from the content of the application session, and then identifies the applications based on the feature matching.Through the simulation experiment test to identify Rate, correct rate and comprehensive index, the result shows that the algorithm is effective and can accurately classify the application layer traffic.