SVD-Based Joint Sparsity in SAR Tomography for Urban Mapping

With the advancement of Synthetic Aperture Radar (SAR) technology, the transition from 2D imaging to 3D tomographic imaging has become a significant trend in the remote sensing community, particularly offering extensive application prospects in the extraction of urban structural details and topographical mapping. Conventional TomoSAR imaging approaches cannot utilize the scene's prior structural information with independently dealing with every pixel. In this study, we proposed a singular value decomposition (SVD) based joint sparsity algorithm of TomoSAR imaging using the multiple measurement vectors (MMV) from adjacent pixels. In particular, the proposed algorithm can enhance the utilization of spatial structure. To promote the computational efficiency, we employ the SVD technique to reduce the dimensionality of high-dimensional SAR data, effectively extracting primary feature information while removing noise and redundant components. Building on this dimensionality reduction framework, the alternating direction method of multipliers (ADMM) optimization method is utilized to construct an efficient and precise method for solving 2D sparse models. Experimental results using both simulation data and measured data demonstrate that the proposed algorithm outperforms most of the state-of-the-art (SOTA) algorithms.