Performance Improving Techniques For Synthetic Aperture Radar-ground Moving Target Indication

Synthetic aperture radar-ground moving target indication(SAR-GMTI) has been playing an essential roal for reconnaissance. As a consequence, especially in recent years, the SAR-GMTI technique has become a promising issue. The aim of this thesis is to improve the performance of the multichannel SAR-GMTI system, and the study focuses mainly on the key technological problems such as channel balancing, unambiguous velocity estimation, baseline estimation, and so on.Our works are summarized in details as follows.First, to improve the clutter suppression performance, we propose a new two-dimensional(2-D) frequency domain channel balancing algorithm. Based on a useful tool called correlation analysis, a new algorithm for solving a very general channel balancing problem is obtained, for which only the knowledge of the first two moments of the channel output signals is assumed and no other distributional assumptions are made. Furthermore, the proposed algorithm has the merits of possessing a simple implementation form and needing no iterative operation. The experiment results based on real data show that the proposed algorithm can achieve better channel calibration performance than the ordinary methods.Second, to improve the clutter suppression performance, we present a two-stage channel balancing technique. We find out that the channel errors can be classified into two categories, i.e., the so called convolutional and multiplicative errors. The convolutional errors are more suitable to be compensated in 2-D frequency domain, whereas the multiplicative errors should be compensated in image domain. By employing an already-existed conventional 2-D frequency doman calibration method as a Stage 1 balancing processor, the convolutional errors can be effectively compensated. To compensate the multiplicative errors, we introduce an image-domain median filter as a Stage 2 balancing processor. The proposed two-stage balancing technique is tested with real data obtained from a SAR-GMTI system with terribly mismatched channels, and the results show that the channel balancing performance of the proposed algorithm is good.Third, to approach the ambiguous problem of the along-track interferometry(ATI) velocity measurement, we propose a new method of estimating the unambiguous range velocity of moving targets. The basic idea is to decompose the unambiguous velocity into fractional-velocity and integer-velocity parts, and to estimate them individually. The proposed method jointly uses the ATI processing and the multilook beat frequency(MLBF) processing. The ATI processing can obtain an accurate velocity estimation but the range of unambiguous velocity is small, whereas the range of unambiguous velocity by MLBF processing is generally large enough but less accurate. As such, we use the ATI processing to estimate the fractional-velocity part, whereas the ambiguity number is estimated via jointly using the results of the ATI processing and the MLBF processing. After having estimated the ambiguity number and the fractional-velocity part, the unambiguous velocity can be accurately estimated. The proposed method is tested with real data, and the results show that the proposed method can accurately estimate the unambiguous velocity of the target.Fourth, to obtain accurate baseline estimation so that the velocity of the targets can be estimated accurately and even the targets can be relocated accurately, we propose a new method of effective baseline estimation based on co-registration and median filtering. The basic idea is to perform the baseline estimation via exploiting the slope of the ATI phase ramp in 2-D frequency domain. The proposed scheme of effective ATI-baseline estimation includes two key procedures: fine co-registration and median filtering. The proposed method is tested with real data, and the results show that the proposed method can obtain an accurate estimation of the effective baseline.Fifth, to solve the baseline estimation problem, we present a simple and robust effective baseline estimation method. The basic idea is performing the baseline estimation via a baseline search operation. In this method, the degree of coherence(DOC) is employed as a metric to judge whether the optimum baseline estimate is obtained, and the rationale behind this is that the more accurate the baseline estimate, the higher the coherence of the two channels after co-registering with the estimated baseline value. The merits of the proposed method are twofold: simple to design and robust to the Doppler centroid frequency. The effectiveness of the proposed method is tested withreal SAR data.Finally, all the works presented in this disseratation are summarized, followed by an outlook on future issues and potentials.