3-D Time-Domain SAR Imaging of a Forest Using Airborne Multibaseline Data at L- and P-Bands

摘要

In this paper, a time-domain back-projection based tomographic processing approach to a 3-D reconstruction grid is detailed, with the focusing in the third dimension being either modified versions of multilook standard beamforming, robust Capon beamforming, or multiple signal classification. The novel feature of the proposed approach compared to previous synthetic aperture radar (SAR) tomography approaches is that it allows for an approximation-free height-dependent calculation of the sample covariance matrix by exploiting the azimuth-focused data on the 3-D reconstruction grid. The method is applied to experimental multibaseline quad-pol SAR data at L- and P-bands acquired by German Aerospace Center's (DLR) E-SAR sensor: Tomographic images of a partially forested area, including a 3-D voxel plot that visualizes the very high level of detail of the tomographic image, are shown, and an analysis of the focusing performance is given for the full as well as reduced synthetic aperture in the normal direction.