SAR tomography via sparse representation of multiple snapshots and backscattering signals — The L1-SVD approach

TomoSAR, as one of the hot technical topics these years, gives an advanced way to use the single orbit multiple baselines SAR images. Among the many TomoSAR methods, the sparse based spectrum estimator is the most popular one. In this paper, we make use of the truncation version of multiple snapshots of compressive sensing (MCS), called L1-SVD, for retrieving information in the urban area. Compared to other sparse-based methods, the multiple scheme excavates the potential valuable information to achieve the accurate spectrums and the truncation strategy improves the computational cost. To prove the compatibility of the L1-SVD of TomoSAR in urban area, an analysis of the snapshot is made and a validation using Radarsat-2 images of a stadium are processed. Finally, with the achieved sparsity of the stadium, the rough structure is retrieved.