Three-dimensional Target Feature Extraction Based On The Curve Of The Synthetic Aperture Radar (clsar)

At present, multi-parameter and multi-mode are the trend of modern microwave remote sensing system. This kind of integral design can reduce the cost and make the SAR system to be more adaptable. The Curvilinear Synthetic Aperture Radar (CLSAR) is a relative new microwave observation technology of Spotlight type. It can be used as a new mode of microwave sensor system.The basic theory of the traditional FFT-based radar imaging method is analyzed systematically in this paper. The mathematic model, theoretical analysis and numerical simulation are presented in both 1-D matching filter imaging method and 2-D polar format algorithm (PFA) of Spotlight SAR, respectively. In addition, through the analysis of the 2-D Fourier transform of the curvilinear apertures, it is proved that the 3-D images .obtained via using FFT with CLSAR suffer from severe high sidelobes, and are of little practical use.Because of the high sidelobes of curvilinear aperture, RELAX algorithm, a parameter estimation algorithm based on nonlinear least-squares (NLS) criterion, is introduced to solve the radar target feature extraction problem. At first, the RELAX algorithm is applied to the 1-D high resolution radar and 2-D spotlight SAR, numerical examples are presented and the results show that the RELAX algorithm has it's validities and advantages. Based on all these preparative work, this paper present the detailed steps of the RELAX algorithm for 3-D target feature extraction with CLSAR, numerical examples are presented respectively, the results of different shape curvilinear apertures are compared. The numerical results show that it is feasible to use CLSAR for 3-D radar target feature extraction.