Effective Media Traffic Classification Using Deep Learning
Qing Lyu, Xingjian Lu · 2019
Traffic classification (TC) is very important as it can provide useful information which can be used in the flexible management of the network. However, TC has become more and more complicated because of the emergence of various network applications and techniques. In this paper, we apply deep learning based method to the classification of four different kinds of media traffic, i.e., audio, picture, text and video. We collect traffic data from the real network environment. Multilayer Perceptron (MLP) and Convolutional Neural Network (CNN) based traffic classification methods are designed to accurately classify the target traffic into different categories. We found that MLP has very good performance in most scenarios. Moreover, specific architecture can reduce the training time of the neural network in the classification.