|
Abstract Anonymous communication can effectively protect personal privacy and data security, but it is also used by criminals to
evade network censorship and engage in various illegal activities because of its concealment. The characteristics of anonymous traffic
protocols are concealed, traffic characteristics are confused, and detection of anonymous traffic is difficult. In order to strengthen
the supervision of anonymous traffic, this paper proposes an anonymous traffic detection method based on entropy estimation.
Through the analysis of the packet entropy characteristics of the network session, the entropy value of the encrypted message sequence
is used as the input feature of the classifier, and combined with the eXtreme Gradient Boosting Tree (XGBoost) algorithm,
the detection of anonymous traffic is realized. Different from the existing anonymous traffic feature detection methods based on protocol
features and traffic features, this method extracts the characteristics of the traffic itself for research. The experimental results
show that the accuracy of this method can reach 98%.
|
