HMM Feature Extraction Combined with Perceptual Hash Matching for Network Time Covert Channel Detection
Aiming at the problems of low recognition rate and poor robustness of traditional network covert channel detection,a network covert channel detection method based on hidden Markov chain and perceptual Hash matching is proposed by using the difference between normal channel and covert channel in network traffic time series characteristics.Experiments show that the proposed method can accurately detect the network traffic of covert channels,and the overall discrimination level is between 70％～85％.Under the interference of signal noise,it can maintain high recognition rate and has good robustness.The pro-posed method takes less time than traditional spectral domain sensing feature extraction,and the complexity is low.
万方数据
维普
