A Quartic Algorithm For Squint Sar Imaging

In this paper some improvements of non-linear chirp scaling (NCS) approach are proposed and a modified algorithm based on the NCS is described. The improvements include three main aspects. A quartic polynomial model is adopted to represent SAR signal in range-Doppler domain and 2-D frequency domain to improve precision. A constant scaling factor is employed to replace the reference frequency in the NCS, which makes NCS more flexible. A second chirp scaling operation is added to the end of the NCS so as to remove scaling effect on range direction of SAR image. Those improvements make NCS applicable to process highly squinted SAR data from a large range swath at a fine resolution