A REFINED TWO-DIMENSIONAL NONLINEAR CHIRP SCALING ALGORITHM FOR GEOSYNCHRONOUS EARTH ORBIT SAR

Taking into account long signal propagation time, curved orbit and \near-far-near" slant range histories at apogee, a reflned slant range model (RSRM) is presented for geosynchronous earth orbit synthetic aperture radar (GEO SAR) in this paper. Additional linear component and high order components are introduced into straight orbit assumption (SOA) model to describe relative motion during long signal propagation and curved orbit respectively. And the special slant range histories at apogee are considered through adding terms changing with the sign of Doppler rate. Then, based on RSRM under an ideal acquisition and ignoring nonideal factors (such as depolarization and attenuation efiects), a reflned two-dimensional nonlinear chirp scaling algorithm (RTNCSA) is proposed. Space-variant range cell migration (RCM) caused by range-variant efiective velocities is corrected by reflned range nonlinear chirp scaling algorithm, and the variable Doppler parameters in azimuth direction are equalized through reflned azimuth nonlinear chirp scaling algorithm. Finally, RSRM is verifled by 600-second direct signal received by a stationary receiver on a tall building from BeiDou navigation satellite, and RTNCSA is validated through simulated point array targets with resolution of 5m and scene size of 150km.