Resolution enhanced SAR tomography: A nonparametric iterative adaptive approach

Abstract

The ground-volume separation of radar scattering plays an important role in the analysis of forested scenes. For this purpose, the data covariance matrix of multi-polarimetric (MP) multi-baseline (MB) SAR surveys can be represented thru a sum of two Kronecker products composed of the data covariance matrices and polarimetric signatures that correspond to the ground and canopy scattering mechanisms (SMs), respectively. The sum of Kronecker products (SKP) decomposition allows the use of different tomographic SAR focusing methods on the ground and canopy structural components separately, nevertheless, the main drawback of this technique relates to the rank-deficiencies of the resultant data covariance matrices, which restrict the usage of the adaptive beamforming techniques, requiring more advanced beamforming methods, such as compressed sensing (CS). This paper proposes a modification of the nonparametric iterative adaptive approach for amplitude and phase estimation (IAA-APES), which applied to MP-MB SAR data, serves as an alternative to the SKP-based techniques for ground-volume reconstruction, which main advantage relates precisely to the non-need of the SKP decomposition technique as a pre-processing step.