Bayesian Compressive Sensing Approaches for the Reconstruction of Two-Dimensional Sparse Scatterers Under TE Illuminations

In this paper, the reconstruction of sparse scatterers under multiview transverse-electric illumination is dealt with. Starting from a probabilistic formulation of the “inverse source” problem, two Bayesian compressive sensing approaches are introduced. The former is a suitable extension of the single-task method presented earlier for the transverse-magnetic scalar case, while the other exploits an innovative multitask implementation to take into account the relationships among the “contrast currents” at the different probing views. Representative numerical results are discussed to assess, also comparatively, the numerical efficiency, the accuracy, and the robustness of the proposed approaches.