Covert Channel Research Based On Concealed Information Storage Distribution

As the problem of network security becomes more and more prominent,covert channel,one of the important network hazards,has gradually become a hot academic research.According to the secret information carrier,the covert channel is divided into time-type and storage-type covert channel,in addition to multi-link covert channel and length concealment channel.In this paper,we propose a concealment method based on covert information storage to solve the problem of the difference of existing length concealment channel and legitimate channel in the statistical characteristics and channel entropy.Firstly,the covert information in different encoding modes is converted into binary bit stream,the bit string of specific length is sequentially taken in the bit stream,we estimated the probability that different value of the bit string in the bit stream,the reason of the different probability and it's affection on length concealment channel.And then considering this factor into the construction of the length concealment channel,a new length concealment channel construction method is designed.In order to solve the problem of channel entropy increase,the paper designed an effective entropy adjustment mechanism,adjust the entropy of the channel to the legal range.And the classical reference length covert channel NTNCC is realized.In this paper,the threshold of statistical characteristics and entropy feature of the channel are obtained by characteristic parameters of the NTNCC and the Legitimate channel.In the end,the concealment of the covert channel model designed in this paper is detected by two methods:statistical characteristics and entropy detection.The detection results of Legitimate channel,NTNCC and the covert channel designed by this paper are compared.Experimental results show that,compared with the traditional reference length concealment channel NTNCC,the length concealment channel method designed in this paper is more close to the Legitimate channel in statistical characteristics and entropy,and with better concealment.