The Algorithm Research On The Identification And Parameter Estimation To The Pulse Compression Radar Signals

Pulse compression radar with low probability of intercept brings a new challenge to the modern ARM seeker and other passive surveillance receivers. We take BPSK signals, and LFM signals as main objects. And the recognition of modulation types and their corresponding further processing were studied, including the parameter estimation for PSK signals, detection and parameter estimation for frequency modulation signals. Additionally, some hardware experiments were done for the processing of pulse compression radar signals, which supplied necessary parameters for sorting, intercepting, matching and tracking of pulse compression radar signals in ARM seeker.Blind signal process (BSP) are shown to dispose the simultaneous arriving signals. Firstly, a radar signal sorting algorithm based on the Fast Independent component analysis (Fast ICA) is given in the dissertation. This sorting algorithm can separate different modulated radar signals efficiently. At the same time, the method of soring and recognition can merger in the same hardware. A new method of denoising of the signals was proposed based on ICA. The simulation results show that:The ICA method can gain higher SNR, and the denoising of the signals can apply successfully after this method.To realize the radar signal identification, an algorithm based on the signal frequency spectrum is in trduced to identify the LFM and BPSK radar signal. Based on the identification, a method of cumulative average of the signals in the discrete domain was proposed to estimate the parameter of the signals. The definition and the principle of the method is given in the paper, for BPSK and LFM radar signals, the compter simulation results show the effectiveness and fast of the algorithm. The algorithm was compared to the instantaneous autocorrelation. The test results show that:the method can adapt a lower SNR, and it is suitable for engineering. Adaptive denoising method based on DWT (Discrete Wavelet Transform) provided a feasible solution for radar signal filtering. However, DWT doesn't have the characteristic of translation invariance, so it would bring bigger reconstruction error and impact the filtering effect if different wavelets are used to reconstruct signals. The lifting methods of SWT (static wavelet transform) was introduced. The new methods can enhance SNR of the output, whose calculation amount is a little more than traditional method. A new recognition algorithm is proposed based on the lifting stationary wavelet transform (SWT) and the Fractional Fourier transform (FRFT). The algorithm can enhance the precise positioning capability of FRFT. The eclectic denoisingmethod for soft and hard thresh-olds is based on the nonlinear relationship of the wavelet coefficients before and after filtering. Firstly, the algorithm findsout themaximum pointsof each corresponding fractional, and converts the curve modulus composed by these maximums to wavelet domain through the lifting SWT. And the low-frequency parts of the noise are eliminated by the eclectic denoising method to estimate the probable position of the low-frequency parts of the signals. Then, the high-frequency coefficientsof each layer are filtered to eliminate the noise outside of the band, and second filtering to the coefficients ismade by the eclectic denoising method. Finally, signals are reconstructed and each parameter is recognized by the precise positioning of the curve peak. Simulation results show the validity of the algorithm.The LFM radar signal parameter estimation method based on digital channelized receiver is proposed, the hardware block diagram of the system is given in the paper, the software and the hardware test are introduced emphatically, at last the error of experimental results are analyzed.