Multistatic TomoSAR 3-D Imaging Technique via Matrix Completion for Structured Targets

Multistatic 3-D synthetic aperture radar (3D SAR) has shown significant potential in rapid 3-D imaging. Compared to traditional multipass or array 3-D imaging systems, it achieves high-resolution imaging in a single pass. However, due to the introduction of multiple radar systems, decoherence factors such as multichannel and synchronization cause serious degradation in data quality, posing challenges for accurate reconstruction. To address this issue, this article proposes a data recovery algorithm based on matrix completion (MC) for 3-D imaging of structured targets. The structural characteristics of architectural targets introduce a low-rank property into the data, stemming from the inherent correlation among adjacent pixels. Utilizing the principle of MC, combined with the sparsity of scattering points in elevation, a low-rank and sparse joint completion model is established. Furthermore, the Truncated Schatten-p Norm and Sparse Regularizer-Alternating Direction Method of Multipliers (TSPN-ADMM) algorithm is adopted for solving. Additionally, considering that this recovery method reconstructs the 2-D complex image, a Filter-MC processing framework is proposed to further enhance the performance. Finally, both simulation and real data verify the effectiveness of the proposed recovery method and framework.