Traffic Classification Approach Based on Support Vector Machine and Statistic Signature

As network traffic is dramatically increasing, classification of application traffic becomes important for the effective use of network resources. Classification of network traffic using port-based or payload-based analysis is becoming increasingly difficult because of many peer-to-peer (P2P) applications using dynamic port numbers, masquerading techniques, and encryption. An alternative approach is to classify traffic by exploiting the distinctive characteristics of applications. In this paper, we propose a classification method of application traffic using statistic signatures based on SVM (Support Vector Machine). The statistic signatures, defined as a directional sequence of packet size in a flow, are collected for each application, and applications are classified by SVM mechanism.