A Study Of Network Covert Channel Detection Based On Deep Learning

Computer covert channel which is regarded as an important area of information security research gets more and more attention.Concerning covert channels,researchers focus on the construction,analysis,detection and processing of network covert channels.The detection of network covert channels is an important issue in this field.With the development of the network,more and more complicated covert channels have been constructed.In order to detect these covert channels,researchers proposed various detection algorithms during the research process.However,these detection algorithms have many problems.Many algorithms can only detect certain specific covert channels.The speed and accuracy of detection need to be improved.The desired result cannot be obtained during the test.Deep learning algorithm is the new direction of machine learning algorithm,and great progress has been made in recent research.Through the use of deep learning algorithm can greatly enhance the efficiency of the algorithm.Based on the deep learning algorithm,this paper proposes a new idea of network covert channel detection,which is to detect the covert channel through the detection of dangerous information.By comparing the difference between normal information and dangerous information,a covert channel detection algorithm is designed.In this detection algorithm,a recurrent neural network(RNN)is used to detect the covert channel.By designing the interior of the recurrent network,the performance of the covert channel detection algorithm is improved.On the one hand,this algorithm can detect more complex covert channels,which greatly improves the detection methods that can only detect several types of covert channels.On the other hand,due to the use of a RNN,the accuracy of detection can be greatly improved.By optimizing this detection method,the better results on the evaluation indicators can be acquired.In order to test the effectiveness of the algorithm,the algorithm is used to detect the algorithm using a data set with complex network threats constituting the network covert channel.Compared with common machine learning algorithms,the proposed algorithm can show better results in some aspects,thus verifying the practicability of the new method.