Improved SAR-GMTI via optimized Crame-Rao Bound

This paper presents a theoretical study on an optimum SAR-GMTI mode that exploits the flexible programmability of modern phased array radars such as RADARSAT-2 (R2) and TerraSAR-X. One major challenge for SAR-GMTI is the considerable re-positioning error owing to large stand-off ranges. A new GMTI mode is proposed that combines the two concepts of antenna tapering with aperture switching between subsequent transmit (Tx) pulses. The suitable antenna taper is determined via global minimization of the Cramer-Rao-Bound (CRB) of a new generalized signal model in the Doppler domain. We show that the new mode outperforms existing GMTI modes on R2 albeit trading it off with some (acceptable) SNIR loss.