Research On Airborne SAR Back Projection Imaging And Motion Compensation

Synthetic Aperture Radar(SAR)is an effective and accurate remote sensing imaging system that has been widely used in both military and civilian fields.The SAR platform based on airborne system is more flexible.The pulse compression of the linear frequency modulation signal is used to obtain high resolution of distance,and the high resolution of azimuth is obtained by the synthetic aperture,so we can have the two-dimensional high-resolution SAR image is obtained.In recent years,the process of airborne SAR echo signal has become more and more perfect,more and more high-precision or high-efficiency imaging algorithms have been emerged.Among them,the backward projection algorithm based on time domain can achieve high-precision focusing of echoes under any known track,and has been widely used in data processing of airborne SAR.However,the traditional SAR systems use the chirp signal linear frequency modulation signal to achieve the high resolution of the range,so that the side lobes of the reconstruction point is higher,and the target points with smaller interval cannot be distinguished only by the matching processing.In addition,when the airborne SAR gets data,due to the influence of atmospheric propagation or self-jitter,it will inevitably produce a track offset that cause defocusing of the reconstructed image.In the reconstructed image of the backward projection algorithm,the defocus of the motion phase error exist along the line of sight of the target to the center of the aperture,and the traditional autofocus motion compensation algorithm fails.To this end,this paper mainly studies the sidelobe suppression technology of SAR backward projection imaging and the motion compensation algorithm of reconstructed image.The main work contents are as follows:Firstly,the basic principle and key steps of the backward projection algorithm are introduced,and the advantages and disadvantages of the algorithm are pointed out.Aiming at the problem that the backward projection algorithm has a large amount of computation,a fast decomposition backward projection algorithm is introduced.For the fast decomposition backward projection algorithm,the sub-aperture division,sampling requirements,distance error and computational analysis are discussed,and finally the full resolution SAR image is obtained.In order to solve the problem that linear frequency modulation signal has higher sidelobe,it is proposed to use nonlinear FM signal for imaging.The specific process of approximate synthesis of nonlinear FM signals and pulse compression processing methods is introduced.The phase-stay principle is used as a theoretical guide to approximately get the synthesis of the nonlinear frequency modulation signal,and the match filtering is selected to achieve the pulse compression.Finally,imaging is performed by a sparse driven autofocus algorithm.In order to make more accurate estimation and compensation of existing phase errors,the existing phase gradient autofocus algorithm,rank-one phase error estimation,improved rank-one phase error estimation algorithm and sparse-driven autofocus algorithm are introduced.The advantages and disadvantages of these algorithms are compared,and a sparse imaging autofocus algorithm is proposed based on back projection algorithms.The improved phase error estimation algorithm and the sparse regularized autofocus algorithm are combined to image.The experimental results of algorithms show that the algorithm can improved accuracy of the target.