Research Of Campus Instant Message User Network Behavior

In modern society, the instant message has become an important channel for communication between human. With the highest rate of use in Internet software, instant message become the mainstream internet communication tool by replacing e-mail. With the increasing users and the increasing number of service nodes, instant message traffic takes up an increasingly large share in overall network traffic. Therefore, observation and analysis of instant message user behavior have important significance to study the complex network behavior.This paper first analyze the market structure of instant message. Then the communication mode and principle of QQ, the most representative of the instant message tools, are researched.A study case in the campus LAN is built for collecting the QQ packet and analysis the QQ packet using Ethereal software. The port identification and Keep-alive trace identification to identify QQ packets are proposed in the paper. Comparison results indicate Keep-alive trace identification method can identify the QQ packets more correctly than port identification method.In the study case, operation data of the campus LAN are collected for24hours. And the QQ packets are recognized using Keep-alive identification method. Then the online user number of QQ and communication traffic data can be obtained. Therefore, the change law can also be get between the online user number and communication traffic. At last the relationship between the QQ communication traffic and overall network traffic is studied.