Compressed Sensing SAR Imaging Based On Inverse Matched Filtering

Synthetic Aperture Radar(SAR)uses a small antenna to synthesize an equivalent long antenna by moving the small antenna along the platform.It increases the imaging resolution without increasing the actual aperture of antenna.This is a historic break-through for radar imaging.SAR imaging has been widely used in both military and civil field.With the increase of the size of imaging scene,the amount of data has grown sig-nificantly.It brings much pressure to the hardware system.However,the actual scene is usually sparse or has some kind of structure,thus it is possible to solve the problem under Compressed Sensing(CS)theory.Compressed Sensing theory is used to decrease the sample rate with the help of information redundancy.CS theory has proved that sparse optimization algorithm could recover the original signal from undersampled data as long as certain conditions are satisfied.Compared with conventional information theory,CS combines the sample and compression procedure to conduct undersample directly.In this way the amount of data are decreased and the pressure of storage and transmission is reduced.CS SAR imaging has drawn great attention in recent years.In this paper,CS SAR imaging algorithm is studied.We use traditional matched filtering to optimize the reconstruction algorithm,the time and space complexity is thus reduced.In the introduction part,we first introduce the history of SAR,and the problems it is facing.Then CS theory is briefly introduced.CS and 1-bit CS have both improved the performance of SAR imaging.We point out some problems in CS SAR imaging,and will address these problems in the rest of this paper.In the second chapter,CS theory and sparse reconstruction algorithms are intro-duced.We analyze the preconditions of CS,and studied several CS models and corre-sponding reconstruction algorithms.Some problems of CS SAR imaging are pointed out.In the third chapter,considering the high computational complexity of CS SAR imaging algorithm,we proposed an optimized method using omega-K algorithm.The proposed algorithm reduced computational complexity based on approximated substi-tution.Quantitative analysis of computational complexity is shown.The feasibility of the proposed algorithm is proved,and detailed procedures are derived.Experimental results validate the effectiveness of the proposed algorithm.In the fourth chapter,we discuss the problem of low performance under low SNRin the last chapter.A novel I-bit CS SAR imaging algorithm with low computational complexity is proposed.It not only improves the performance of the reconstruction algorithm under low SNR condition,but also reduces the pressure of ADC.We analyze the feasibility of combining omega-K and 1-bit CS SAR imaging.Detailed steps are derived.The optimized algorithm decreases time and space complexity of conventional 1-bit CS SAR imaging algorithm.Experimental results prove the effectiveness of the proposed algorithm.