Radar Signal Parameter Estimation And Physical Feature Extraction Of Micro-motion Targets

Some targets, such as wareheads, planes, vehicles and lives usually perform their unique micro-motion, which are determined by their intrinsic kinetics property. With the development of the radar technology and advanced signal processing technology, the micro-motion information is got more attraction in the area of target recognition. Currently, the application of micro-doppler, the frequency modulation to the radar signal by the micro-motion targets, is a hot problem in the field of academe and industry. In this thesis, based on the background of radar target recognition, we systematically study the radar signal processing and feature extraction of micro-motion targets by combining both the theoretical analysis and the experimental certifications. In summary, the work is composed with three parts according the study sequence: establishing micro-doppler model, parameter estimation of the micro-doppler and the physical feature extraction of the micro-motion targets.Firstly, we investigate the micro-doppler model and establish the explicit relationship among the kinetic parameters, the structure parameters and the micro-doppler parameters. At the same time, two different scattering mechanisms– both the ideal scattering and the nonideal scattering, are considered in the derivation of our micro-doppler models. In the case of ideal scattering centers, three basic micro-motions– rotation, vibration and coning, are modeled into a uniform expression. Then, the frequency, time-frequency and high-order instantaneous moment characteristics of these micro-motions are studied. In the case of the nonideal scattering centers, we analyze the influence of slip-type scattering center to the micro-doppler and establish the coning micro-doppler model of the ring edge structure. Also, considering the frequency modulation of the mass translation, we establish the Doppler model of the micro-motion targets with the mass translation and then give the relationship among the parameters of the signal, the mass translation and the micro-motion.Secondly, we study the parametric estimation method and the fast estimation algorithm of the micro-doppler by using the tool of TFD (Time Frequency Distribution)-Hough transformation. According to the micro-doppler model in the ideal scattering centers situation, an estimation method of sinusoidal frequency modulation signal with three parameters is proposed, where the window length choice of the pseudo wigner-ville distribution is studied; the discrete expression and the improvement of signal noise ratio are analyzed; and then a fast estimation method is given based on time-frequency ridge-random Hough transformation. According to the coning micro-doppler model of the ring edge structure, the parametric function of the Hough transformation of the time frequency curve when partial parameters are known is proposed, based on the target recognition in the ballistic midcourse. At the same time, the fusion method of the parametric space is given to improve the estimation precision according to the rigidity and symmetry of the ring structure. For the situation of non-even scattering intensity, a parametric multi-component micro-doppler estimation method based on"CLEAN"in frequency domain is proposed, which can avoid estimating of the signal amplitude. According to the Doppler model of the micro-motion targets with the mass translation in the case of ideal scattering centers, the parameter estimation of the SinFM-PPS (Sinusoidal Frequency modulation-Polynomial Phase Signal) are proposed based on the high instantaneous moment character of the SinFM signal and the PPS signal. The algorithm uses the multilag high instantaneous moment transformation to decouple the SinFM and PPS components. Then, by using the TFD-Hough transformation of the SinFM and linear FM signals, we propose an estimation method to obtain the parameters of the micro-doppler and the mass Doppler step by step. Furthermore, we give a fast estimation algorithm for SimFM-PPS based on the time frequency ridge-random Hough transformation of the SinFM and linear FM signals.Finally, we investigate the physical feature extraction method based on the relationship among the kinetic parameters, the structure parameters and the micro-doppler parameters. In the meanwhile, two dynamic measurement experiments of the micro-motion targets are carried out. Based on the application of the target recognition of the millimeter wave seeker, an outdoor experiment to measure of the rotating target by the millimeter wave seeker was carried out. Through the experiment, the micro-doppler parameter estimation method for the ideal scattering center is verified, and the physical features of the rotating target are extracted. Based on the target recognition in the ballistic midcourse, an experimental research method of the wideband full-polarization dynamic scattering properties is proposed for the space precession target. The observability of the micro-Doppler of the precession target by microwave radar is demonstrated and also the non-ideal scattering centers caused by the normal structure of the warhead are observed.The coning micro-doppler model of the ring edge structure and the signal processing method are verified. The physical features of the space precession target are extracted. The parameter estimation of micro-doppler, the feature extraction of micro-motion targets and the experimental method of dynamic measurement of micro-motion targets, which are proposed in the thesis, will improve the theoretical architecture of radar information processing of micro-motion targets and furthermore support the target recognition application of ballistic missile defense, anti-aircraft combat and battlefield surveillance.