Minimum-redundancy Multi-Master TomoSAR Framework for UAV-SAR 3-D Imaging

Unmanned aerial vehicle (UAV) synthetic aperture radar (SAR) is a microwave imaging technology that can work all day and all weather. It can conduct multiple observations at different spatial locations and realize three-dimensional (3-D) imaging through the tomographic SAR (TomoSAR) technique. However, the limited repeat-pass observations will lead to sparse baselines, resulting in elevation ambiguity and high sidelobes in 3-D imaging. In addition, the unavoidable spatial incoherence will introduce phase noise, which reduces the elevation estimation accuracy (EEA) and affects the imaging quality. To address these problems, this paper proposes a minimum-redundancy multi-Master (MM) TomoSAR framework for UAV-SAR 3-D imaging. It can achieve unambiguous, low-sidelobe, and high-accuracy 3-D imaging with a limited number of observations. The main contributions are summarized as follows. First, the MM-TomoSAR signal model is constructed based on the traditional TomoSAR model and the interferogram-based tomographic processing. Then, a minimum-redundancy baseline design strategy is proposed. Combined with the MM-TomoSAR model, it allows for unambiguous elevation estimation while maintaining a low sidelobe level. Finally, a 3-D imaging method combined with tomography and back projection (BP) is proposed to solve the problem of scattering information loss caused by the nonlinear processing of compressed sensing (CS). It can realize high-quality and lossless 3-D imaging. Computer simulation and UAV-SAR 3-D imaging experiment are conducted to verify the proposed method.