Abstract:

The selfsimilarity of peertopeer (P2P) traffic is studied based on fractal method. Different from former researches, the traffic of single protocol is concerned. Two popular P2P applications are tested and showen that the application layer traffic tends to be selfsimilar. The selfsimilarity is more stable under behavior scale than under time scale. The selfsimilarity of the P2P traffic is applied to P2P traffic identification. A new traffic identification algorithm is proposed based on the fractal dimension and positivity of the network traffic. Experimentss show that the performance of the proposed algorithm is better than the existed in terms of accuracy especially for encrypted traffic.

Key words: peer to peer, network traffic, selfsimilarity, fractal, fractal dimension