| The hardware difference of Radio Frequency(RF)transmitter,as well as the interaction and action under certain environmental conditions,constitute the overall nonideal characteristics of the transmitter,giving the unique RF fingerprint(RFF)to the transmitted signal.Stable and effective RFF features are the important basis for transmitter identification and are directly and closely related to fingerprint generation mechanism and individual distinguishable mechanism.On the other hand,the rapid development of identification technology will greatly threaten the survival and working efficiency of the transmitter.The anti-identification technology is an important measure to protect the safety of the transmitter to counter the identification technology by means of interference and camouflage.The main work of this thesis includes two aspects.On the one hand,in order to extract the differences between different transmitters related to the mechanism of RFF generation and improve the accuracy of transmitter identification,this thesis studies the influence of transmitter characteristics on signals,and designs the method of RFF feature extraction.On the other hand,aiming at the anti-identification of transmitter,this thesis studies the method of camouflaging transmitter,and equivalates the non-ideal characteristics of transmitter to baseband to realize fingerprint camouflage of transmitter.The details are as follows:Firstly,the mechanism of RFF generation is studied through the characteristics of key transmitter devices,and the overall non-ideal characteristics of the transmitter are abstracted into IQ imbalance,phase noise and nonlinearity.Then the frequency domain and the time domain modeling method of phase noise are studied,as well as the Volterra series modeling method of power amplifier.The fitting of the actual power amplifier shows the correctness of the nonlinear model.Secondly,in order to accurately characterize the individual differences of transmitters and extract fingerprint features that can distinguish individual transmitters from the RFF generation mechanism.In this thesis,the extraction methods of IQ imbalance feature,nonlinear feature of spectral regrowth and Volterra coefficient,phase noise feature are studied.A RFF extraction method based on timing synchronization error,carrier synchronization error and constellation pattern is proposed.On the other hand,combined with the working nature of the transmitter,the anti-identification of the transmitter is studied,the actual non-ideal characteristics are equivalent to the baseband,and the effective camouflage of the transmitter is realized.Then,the simulation model of RF transmitter was built based on SIMULINK,whose fingerprint features were extracted.MATLAB classifier was used to classified and identified the simulated transmitter.Based on the anti-identification method proposed in this paper,the camouflaged transmitter is modeled and the same classifiers are used to complete the classification and identification.The classification recognition and misjudgment results are good,which verifies the correctness of the RFF feature algorithm and the anti-identification method in this thesis.Finally,the experimental test platform was built,and the measured transmitted signals were collected by signal acquisition equipment,and the feature extraction,individual camouflage and classification recognition of the measured signals were completed.The test results further verify the feasibility of applying the proposed method to real transmitters. |
