Array SAR 3D Imaging Method Based On Compressed Sensing

Different from other 3D SAR(Circular SAR,Tomograpy SAR)can only work under the side-looking mode, linear array SAR(LASAR) 3D-imaging system can also work under downward-looking mode, thus overcome the shade effect and image geometric distortion which other 3D SAR has. The conventional LASAR 3D-imaging method is based on classical signal processing theory, which demands a large number of echoes and results a low-resolution image because of high side-lobe and wide main-lobe. While 3D-imaging method based on Compressive Sensing theory points out that with a sparse signal, we can surmount Nyquist sampling rate, meaning we can reconstruct a sparse signal with under-sampling rate. In the 3D scene, it is usually to find the target sparse, happens to satisfy the premise of the theory of Compressive Sensing. Applying CS theory to LASAR 3D-imaging, we can reduce the sampling requirements of radar systems, thus reduces the amount of echo data, makes signal transmittion and processing more convenient. Besides the LASAR 3D-imaging method based on CS theory is not restricted to Rayleigh limit and can efficiently suppress the side-lobe. But the LASAR 3D-imaging method based on CS theory has a large amount of data and lacks design criterion of a sparse array, to solve problems above, this thesis do:1. Research on the principle of LASAR sparse imaging method.Introduct several LASAR working modes, conduct the distance history of radar echo of each mode. Two conventional 3D imaging methods are presented and the shortcomings of both are discussed. Compressive sensing theory is briefly stated, conduct the matrix forms of echo data and LASAR measurement matix, then construst the linear measurement models under array plane and the whole 3-D space respectively.2. Research a fast LASAR imaging method based on CS. To overcome the large amount of echo data problem brought by conventional LASAR imaging method, this thesis combines the coupling property of LASAR echo and sparse property of target, via the idea of echo block and scene block,proposes a new fast LASAR imaging method based on CS.3. Research a linear array antenna distribution optimization method for LASARsparse imaging. Conduct the relationship between the measurement matrix coherence and the LASAR system ambiguity function. To minimize the correlation coefficient, via the relationship between correlation coefficient and LASAR system ambiguity function, conduct a linear array antenna distribution optimization model for LASAR, using the model, propose a linear array antenna distribution optimization method for LASAR sparse imaging.