Encrypted Network Behavior Recognition Based on Dynamic Time Warping

In order to solve the problem of encrypted traffic recognition in big data environment, the method based on dynamic time warping for the encryption network behavior was proposed. The method took the encrypted Twitter traffic as the research object, and a large number of encrypted Twitter network behaviors including message, pictures and other behaviors were analyzed, and then the characteristics of the encrypted network behavior were extracted, and the specific encryption network behavior module database based on dynamic time warping were established, and then the interactive network data were collected at real time, and the dynamic time warping value between the collection data set and the module database were calculated, and the value was normalized, by comparing with the preset threshold, the classification and recognition of the encrypted network behavior was realized, the amplitude information that the method based on correlation coefficient could not take into account were overcome, at the same time, the restrictions that the acquisition data length must be consistent with the length of the sample data were avoided, the recognition accuracy was improved, the misjudgment rate was reduced, the proposed method was effective by the experiment.