Online and automatic identification of encryption network behaviors in big data environment

Abstract

To handle the difficulty in identifying encrypted network traffic in big data environment, a fast and online identification method for encryption network behaviors was proposed. Twitter audios, messages, videos, images, and other encrypted network behaviors were deeply studied in big data environment, and the features were extracted from a lot of encryption network behaviors, and the model database based on the correlation coefficient was established by these features, and the correlation coefficient between the network interactive data and the model database was calculated by acquiring the network interactive data at real time. The reference distance will be proposed and used to eliminate the noise of similar traffic sets; at last, the automatic and online identification of encryption network behaviors based on correlation coefficient and reference distance in big data environment were implemented by combination with the classification threshold, and the online identification rate was about 93% by this method, and the experiment results show the proposed method is applicable and effective.