The Research On A Method Of P2P Traffic Detection Based On Flow Features

P2P application played an important role in the development of broadband services in early times,but in recent years,P2P service has affected the traditional internet services (such as WWW, E-Mail, FTP, and others) with the development and especially the increasingly extensive application of P2P services.It has a negative impact on the development of the internet. The consumption of network bandwidth of P2P applications is increasing and the traditional services occupy lower proportion of network bandwidth because of the great ability to seize bandwidth of P2P application. So today P2P traffic detection technology has become a hot research.This thesis firstly looks at the principle of the P2P traffic detection and several typical P2P traffic detection methods. Then the paper investigate the data mining method in the detection of the P2P traffic, in which mainly focus on the research of the application flow feature selection.At last,the paper analyses the flow feature data on the data mining platform and get a perfect result.Paper includes the writer's main work in aspects below:1.Provides an introduction of the key P2P technology, and analyses the principle of the P2P traffic detection methods and their features and problems in the detection process.2.Analyses the significant features of P2P traffic, design and realize a P2P traffic eigenvalue analysis system to analyse several common web application flow characteristics.3.Introduces the data mining technology, demonstrates the feasibility and necessity of the data mining technology in the P2P detection.4.Calculates the value of the features using the data mining technology and test the P2P traffic detection accuracy on data mining platform based on the value of the features having been calculated.5.From a practical application,analyses the prospect of P2P traffic detection technology and the key point of next step research.