A Classification Method Of Encrypted Traffic Based On Deep Neural Network

As user network security awareness increases,a large number of applications use encrypted tunnels to transmit data,such as VPN,SSH,and so on.The types of traffic in the network become more diverse.The current network traffic contains not only traditional digital,image,streaming media,but also a variety of encrypted traffic.Due to the lack of understanding of encrypted traffic and effective identification methods,encrypted traffic increases the difficulty of network management and poses a great threat to network security.The classification of encrypted traffic helps to understand the composition of network traffic,which is beneficial to network management and network security.With the widespread use of network traffic encryption technology,traditional traffic classification methods have gradually failed.This paper analyzes the traffic encrypted and transmitted by VPN and explores its classification method.By extracting the time series characteristics of encrypted traffic,this paper uses the classification model of deep neural network to classify the traffic of seven different categories in the encrypted traffic,and compares it with the commonly used naive Bayesian classification algorithm.At the same time,this paper also studies the batch size that affects the training of deep neural network models,and discusses the effect of batch size on the classification of deep neural network models.Experiments show that the classification ability of encrypted traffic classification model based on deep neural network is much better than the naive Bayesian method.During training,the batch size has different effects on the deep neural network model.When the batch size is 40,the deep neural network model has the best classification ability.