A P2P Network Traffic Classification Method Using SVM

Accurate identification and classification of network traffic according to application type is an important element of many network management tasks. In this paper, a P2P network traffic classification method using SVM classifier is proposed. By this method, the P2P network traffic can been classified according to application types with statistical characteristics of network traffic. This paper mainly introduces the network traffic classification problem on four application types of P2P, namely, BitTorrent, PPLive, Skype, MSN. The classification framework based on the SVM is introduced. The methods to gain the traffic samples and construct classifier are described also. The experimental results show the validity of the proposed methods and the average classification precise rate was high.