| Summary: | Freeman proposed a general polarimetric system model in 1991 for the characterization and calibration of fully polarimetric synthetic aperture radar (SAR) systems. The model contains three-channel imbalance parameters and four cross-talk parameters, which together describe the polarimetric distortion in a practical imaging radar more accurately than the conventional six-parameter model by including a third imbalance parameter, namely γ. This parameter is independent of the others and represents the ratio of the product of the transfer functions of the copolarized channels to the product of the transfer functions of the cross-polarized channels. Nonnegligible errors may arise if this parameter is not estimated and corrected, but since this parameter was proposed, there has been very little research into calibration methods that account for γ. In this article, based on the general polarimetric system model in the 2 × 2 matrix form, we propose a γ-factor estimation method based on an active radar calibrator (ARC) or a dihedral corner reflector. We also propose a universal iterative polarimetric calibration algorithm based on three pieces of calibration equipment, which can be all ARCs, all corner reflectors (CRs), or a combination of ARCs and CRs. The convergence of this iterative algorithm is proven mathematically. These alternatives provide a variety of practical calibration method options for polarimetric SAR calibration under different conditions. Finally, the simple calibration method based on a single piece of calibration equipment and the iterative algorithm based on three pieces of calibration equipment are verified by the polarimetric calibration experiment data from the GaoFen-3 satellite.
|
|---|
