A novel P2P traffic classification approach using back propagation neural network

To meet the requirements of the network activities and take into account P2P traffic classification challenges, a promising method is to use Machine Learning (ML) techniques and identify network applications based on flow features. We present a novel P2P traffic identification approach using back propagation neural network. It is demonstrated by simulation results that our approach can identify popular P2P applications with very high accuracy, low overheads and robustness. Experiment results clearly illustrate that this approach can be competent for classifying P2P traffic which can learn unknown traffic with minimum manual intervention.