| Moving target detection and parameter estimation are important application in radar signal processing. Multi-channel synthetic aperture radar(SAR) system based moving target detection method can employ spatial degrees-of-freedom to realize effectively clutter suppress, thus achieves ground moving target detection. Therefore, multi-channel SAR system has found wide application in airspace surveillance. However, the quantity of radar sampling data increases greatly with more channels, which results in great pressure for data processing of radar system. Nevertheless, compress sensing theory can achieve the signal recovery without distortion under sparse conditions and the distribution of moving target is sparse in space, thus the research of SAR moving target detection method in sparse sampled system is of great significance for increasing airspace surveillance ability of radar system. In this thesis, sparse reconstruction based GMTI method is studied for multi-channel SAR system. The main contributions of this thesis are as follows.Sparse reconstruction algorithms usually have a large computational complexity, thus it is hard to be employed in real system. In this thesis, a fast moving target detection method for dual channel SAR is proposed by introducing null space tuning(NST) algorithm. This method combines dual channel data to suppress the clutter and employ NST algorithm to detect the moving target, thus can improve the accuracy and robustness of the target detection and reduce the computational complexity.Moving target should be signed in the high resolution SAR image. While moving targets are sparse in space, real scene is not, thus it is hard to realize moving target indication and high resolution SAR imaging at the same time. For this problem, a method is given in order to realize the target detection and imaging under the condition of lower sampling rate. In the method, alternating direction method(ADM) is employed to deal with the full sampling channel for SAR imaging, and then utilizing the acquired SAR image to suppress the clutter in the sparse sampled channel, finally, null space tuning algorithm is applied to realize the ground moving target indication. Simulation results demonstrate the excellent GMTI performance of the method.Since the sparse reconstruction error is non-isotropic for different channels, a moving target detection method is proposed for sparse sampled multi-channel SAR system, which greatly reduces the effect of reconstruction error on target detection. On this basis, a radial velocity estimation method is proposed. In our method, sparse reconstruction algorithms are used to perform SAR imaging for each channel. Then joint amplitude and phase ATI approach is utilized to realize moving target indication. Finally, a high accuracy radial velocity estimation method is achieved through array DOA method. |
