Research On Detection Methods Of Constellation Modulation Wireless Covert Channels

With the continuous optimization of the coverage,delay,rate and stability of wireless communication technology,wireless communication has become one of the most important communication methods.The advantages of wireless communication,such as high flexibility of frequency band,diversity of modulation methods,natural broadcasting characteristics and wide existence of signals,have made it one of the most optimal carriers for constructing covert channel.Covert channel is to hide the secret communication behavior and parasitize the communication content in the cover object that can be accessed by both sides of the communication.The covert channel based on wireless communication signal is called wireless covert channel(WCC),which modulates secret information by adjusting the specific characteristics of wireless signal on the premise of reducing the impact on the cover communication bit error rate and signal statistical characteristics as much as possible.In the wireless covert channel,constellation modulation wireless covert channel has attracted much attention because of its advantages in channel capacity and concealment.It uses the noise tolerance of physical layer modulation method and the natural noise characteristics of wireless channel to modulate secret information into artificial noise signal and superimpose it on the cover signal to cover up the existence of covert channel.However,at present,there is little research on constellation modulation wireless covert channel detection technology,especially with the introduction of anti-detection enhancement mechanisms such as periodic rotation of secret information modulation points,model-based shape modulation and offset correlation modulation,the detection for constellation modulation wireless covert channel is facing more and more severe challenges.In view of the above problems,this dissertation conducts research on the detection methods of different types of constellation modulation wireless covert channels,from single constellation modulation wireless covert channel detection,multi-constellation joint modulation wireless covert channel detection,general detection based on image representation,multi-detection antenna joint detection and detection under Multiple Input and Multiple Output(MIMO)scenario,and the main research work and innovative results are as follows:(1)The detection method based on the quantized regression of the residual signal amplitude-phase distribution difference is proposed to address the shortcomings of the existing single constellation modulation wireless covert channel detection methods in terms of detection accuracy.Based on the typical statistical model of normal wireless channel noise,the residual signal is obtained by removing the standard constellation signal from the detecting signal,and the same intensity noise signal conforming to the typical noise distribution is generated.The amplitude and phase distribution differences between the residual signal and the same intensity noise signal are quantified respectively.The logistic regression method is used to learn the quantitative characteristics of distribution difference,and then realize the effective detection of single constellation modulation wireless covert channel.The detection results of simulation and real software radio platform show that the detection accuracy of the proposed method is better than the similar detection methods based on distribution statistical analysis.(2)Aiming at the problem that the current detection methods based on distribution statistical analysis are difficult to detect the multi-constellation joint modulation wireless covert channel with stronger concealment,a detection method is proposed based on autocorrelation analysis of balanced residual signal.This method makes use of the difference in autocorrelation between normal communication signal and covert communication signal.By analyzing the square autocorrelation coefficient of constellation points of the balanced residuals under different channel models,the hypothesis test method is used to detect the multi-constellation joint modulation covert channel.The detection results of simulation and real software radio platform show that the detection method can effectively detect the covert channel under different channel models.(3)In order to realize the general detection of the above different types of constellation modulation wireless covert channels,a detection method is proposed based on deep learning of wireless signal image representation.This method transforms the detection problem of constellation modulation wireless covert channel into the problem of image classification.After mapping the detected wireless communication signal and the calculated residual signal to the two-dimensional gray image space respectively,a two-channel convolutional neural network is designed to train the two kinds of gray images at the same time to obtain the detection model,so as to realize the effective detection of the constellation modulation wireless covert channel.The detection results of simulation and real software radio platform show that this method can detect single constellation modulation and multi-constellation joint modulation wireless covert channels.(4)Starting from the common means to improve the detection ability in wireless communication countermeasures,multi detection antennas,a constellation modulation wireless covert channel detection method is proposed based on joint analysis of multi detection antennas.The wireless signals captured at different loactions are demodulated and the residual signals are extracted respectively.The cover information demodulated by multiple detection signals under the influence of channel noise with different intensity is used to realize multi antenna synchronization,and then the correlation between the extracted residual signals is used to realize covert channel detection.Simulation results show that this method can effectively detect single constellation point modulation and multi-constellation joint modulation wireless covert channels,and the detection accuracy is significantly improved compared with the detection method based on deep learning of wireless signal image representation.(5)Aiming at the problem that the constellation modulation wireless covert channel constructed in MIMO scenario is still lack of effective detection methods,a MIMO wireless covert channel detection method is proposed based on error signal autocorrelation matrix.This method extracts the error signal autocorrelation matrix for different models of MIMO wireless covert channels,and realizes the detection of wireless covert channels in the MIMO scenario by comparing the F-norm of the error signal autocorrelation matrix of normal communication signal and covert communication signal in MIMO scenario.Simulation results show that this method can effectively detect a variety of constellation modulation wireless covert channels under different MIMO communication modes.