Research On Characteristic Analysis Of Low Probability Intercept Radar Signal

In the modern electronic warfare, the prerequisite of radar interference and anti-radiation against is accurately acquisition the enemy radar tactical and technical parameters by radar reconnaissance. The application of low probability of intercept(LPI) radar had brought new challenges to intercepting and processing the radar reconnaissance signal. Effectively interception signal and accurate analysis its intra-pulse modulation characteristic had been the core issue of radar reconnaissance. Nevertheless, limited by the technological level in the design stage, our active service radar reconnaissance equipment couldn't effectively intercept and analyze the modern LPI radar signal. So, rely on the science research project, further research on LPI radar signal interception and intra-pulse modulation characteristic analysis are carried out in this dissertation.Effective interception of LPI radar signal is the foundation to analyze its characteristic.The polarization modulation technique was widely used in modern LPI radar system. Firstly, the problem of traditional unipolar detection affect the parameter measurement accuracy was studied. We proposed the full polarization reconnaissance method and gave implementation scheme of full polarization receiver. To solve the influence of the civilian communication signals to radar reconnaissance equipment, a civilian communication signal suppression method based on polarization filtering was proposed. According to the polarization difference between the radar reconnaissance signal and civilian communications signal, the civilian communications signal was filtered in polarization domain. Theoretical analysis and simulation results verified the validity of the proposed civil communication signal suppression method.Aiming at the problem of the low recognition accuracy of LPI radar intra-pulse modulation type at low signal to noise ratio(SNR), an automatic intra-pulse modulation recognition algorithm based on hierarchical decision was proposed. Firstly, we recognize the non-modulated signal using the time-bandwidth product of intercepted signals, and extracted the characteristic parameter of signal based on the phase difference and the improved short time Fourier transform(STFT). And then, the signal modulation type recognition was achieved by hierarchical decision. Simulation results show the proposed recognition algorithm had obvious classification features, strong anti-noise ability, and high engineering values.In the research of frequency modulated signal parameter estimation, two modulation types were considered. Firstly, based on autocorrelation function and cumulative rearrangement spectrogram, the parameter extraction algorithm of sinusoidal frequency modulated continuous wave was proposed. And then, in order to solve the parameter estimation problem in low SNR, multi-scale path tracing algorithm was introduced. For reducing the computational complexity of multi-scale path tracing algorithm, the wavelet transform and power spectrum smoothing method was adopted to pre-estimate partial modulation parameter. Simulation results showed the two proposed frequency modulated signal parameter estimation method, compared to other methods, had high estimation performance.The parameter estimation and classification recognition of polyphase code signal were studied. Firstly, to solve the problem of polyphase code signal parameter estimation in low SNR, the parameter estimation based on the quadratic phase function(QPF) and fractional Fourier transform(Fr FT) was proposed. Simulation results show that the proposed method could simultaneously accumulate multiple ridge line energy, achieved the parameter estimation of polyphase code signal in low SNR, and estimation performance is not affected by coding type. Secondly, a signal modulation recognition method based on the multi-feature fusion support vector machine(SVM) was proposed. Simulation results showed that the proposed recognition method had high accuracy rate.