부산대학교 도서관

Polarimetric calibration of the space-borne low-frequency SAR instruments rely more and more often on the use of PARCs. At lower frequencies the polarimetric calibration has to be performed in the presence of ionospheric distortions, in first line Faraday rotation (FR). The consideration of imperfect PARCs and FR in the Polarimetric calibration problem dramatically increases the dimensionality of the inversion problem making accurate inversion problematic. The numerical algorithm developed in this study is able to overcome two major challenges in the calibration of polarimetric SAR: First to resolve the ambiguity between the cross-talk induced LOS rotation and FR by estimating FR independently from the data itself; Second to deal with a non-negligible additive noise contribution by constraining the high-dimensional inversion problem using channel imbalance ratios estimated from distributed targets. The evaluation of the improvement follows.