Research On The Spinning Target Moving Imaging And Static Target Staring Imaging

High resolution imaging of spinning target and static target is a difficult problem to traditional static imaging radar (ISAR). The spinning target has a coupling of translational motion and spin motion, which leads to the sinusoidal modulation of the range and doppler of the scatters. Therefore, the RD algorithm become invalid. And the static targets have no relative motion with radar, which leads to impossibility of the synthetic aperture. Therefore, the research of imaging methods of these two kinds of targets is important.With analysis of the motion of spinning target, there are two ideas of the imaging methods.First, if the target rotates several rounds during the imaging, the RD algorithm can be used to achieve the ISAR image when the spin motion has been eliminated. The slow time re-sampling ISAR imaging technology is proposed with this idea. But if the estimate of RPM is error, the proposed method will fail. To solve this problem, the re-sampling ISAR imaging method based on Keystone transform and RID is proposed, which can obtain the clear image of target with the RPM estimation error.Secondly, the spin component can be used because the scatters with different spatial locations have different sinusoidal modulations of the HRRP. Based on the characteristic, the instantaneous HRRP polar coordinate matching method is proposed. The intra-pulse spin motion is eliminated by the time-frequency distribution instantaneous slice of the echo. And the image is obtained by the polar coordinate matching method. Considering the occlusion and undulate effects of scatters, the imaging method based on RBF network and OMP algorithm is proposed. The RBF network is used to establish a space mapping of the target imaging, which converts the imaging process to the weight matrix learning problem. The OMP algorithm is used to seek the parameters of the weight matrix with the minimal-norm criterion, which can correctly obtain the high resolution images of the target.All of the proposed imaging methods require high-accuracy RPM estimation. The narrowband echoes of spinning targets are multi-component SFM signals. Based on the period of the power spectrum of multi-component SFM signals, the AMSF is defined and used to estimate the period of the signal power spectrum. The method of AMSF can get a high RPM estimation accuracy with low SNR. The narrowband echoes are multi-component LSFM signals when the motion of the target is coupling of the translational motion and the spin motion. The estimation method based on the correlation function of Wigner-Ville distribution is proposed, which has the best energy aggregation. But the symmetric scatters will cause the estimation ambiguity of the above method. The ambiguity-resolving process based on the sinusoidal Radon transformation with the STFT-PWVD is proposed. The proposed estimation methods combined with the solution of ambiguity can effectively achieve the high-accuracy RPM estimation.Be different from the moving targets, the static targets have no relative motion with the radar. The synthetic aperture of moving imaging can not be constructed with this kind of targets. The staring imaging method based on the temporal-spatial stochastic radiation field is a brand new imaging method of static targets. The basic idea of the method is to construct the temporal-spatial stochastic radiation field. Under this radiation field, the scattered field of are stochastic modulations of the static targets by the stochastic radiation field.This method requires correlated processing by combining the received signal and the temporal-spatial stochastic radiation field. The first-order correlated method is proposed based on the first-order noncorrelation characteristic of the ideal temporal-spatial stochastic radiation field, which is simple, low computing complexity and can be real-time processed. But the ideal temporal-spatial stochastic radiation field is hard to be achieved in reality. According to this problem, the correlated processing based on the optimization problem is proposed, which uses the BP algorithm to convert the correlated problem to a linear program. This method gets better effect than the first-order correlated method under the nonideal temporal-spatial stochastic radiation field. These proposed methods can get a good image reconstruction effect respectively in ideal and nonideal temporal-spatial stochastic radiation field.