Encrypted Traffic Classification Based on a Convolutional Neural Network

Abstract To resolve the issues of low accuracy, weak universality, and easy invasion privacy in traditional encryption traffic classification methods, an method based on a convolutional neural network is offered. Firstly, according to packet size time message net traffic, original transformed into two-dimensional picture avoid relying payload violate privacy, then model embedded. The Inception module performs feature fusion improve accuracy. Finally, average pooling layer convolution are used replace fully connected layer, increasing calculation speed avoiding overfitting. Experimental results show that algorithm achieves accuracy more than 95% for application tasks.