Algorithms for GEO SAR Imaging Processing

GEO SAR imaging processing faces serious difficulties because of the negative influence caused by high orbital altitude, curved trajectory, long synthetic aperture time (SAT) and large scene size, including complex slant range history, failure of “stop-and-go” assumption, two-dimensional (2D) spatially variant slant range model coefficients, etc. To solve these problems and achieve GEO SAR imaging, an accurate echo signal model, including the slant range model, the 2D spectrum and the spatially variant model of slant range coefficients based on the curved trajectory and Taylor series which take the error of “stop-and-go” assumption into account are first introduced in detail. Then, the time domain algorithm which can be used in any conditions is discussed detailedly, including its procedure flows and computational load. To improve the processing efficiency, the frequency domain algorithm is addressed. In this chapter, the difficulties of GEO SAR imaging in frequency domain are first analyzed in detail; then, the azimuth compensation which consists of a time domain compensation and a frequency domain compensation is derived to reduce the azimuth variance of the focus parameters and unfold the folded azimuth spectrum at certain specific orbital positions; after the azimuth compensation, a 2D nonlinear chirp scaling algorithm (NCSA) is introduced to finally obtain GEO SAR images; besides, the accuracy of the frequency algorithm, including its geometric distortion and available azimuth swath is discussed. At last, we summarize this chapter and give the conclusion.