基于统计特征的即时通信认证流量识别

根据网络数据流的概念及特点,在数据流的初始单位时间片内,统计其数据包个数、大小、时间等基本包属性,并分析提取出数据流的统计特征集。以特征集作为参数,利用即时通信登录认证过程的流量特征匹配函数和特征权重,可计算出特定网络数据流的识别结果百分比。识别结果高于一定阈值,则认为即时通信的登录认证流量识别成功。在中小型局域网内,该方法的识别率可以达到98%以上。

Network Traffic Identification of Instant Messenger based on Statistical Features

According to the concepts and characteristics of network flow, the basic properties of a network packe! such as number, size, time and so on could be all determined within the initial unit of time slices of a network flow, as well as the statistical flow feature sets. The statistical flow feature sets as the parameters, with the flow feature matching functions of an instant messenger＇ s login and authentication process and the relevant weights, the identification result of a specific network flow in percent could be achieved. It indicates success if the result is higher than a certain threshold. The identification rate can reach 98% or above with this method in a small or medium LAN.