PROJECT TITLE :

Inverse Scattering Using a Joint $L1-L2$ Norm-Based Regularization

ABSTRACT:

Inverse scattering problems suffer from unwell-posedness and ill-conditioning, necessitating the utilization of regularization methods to urge meaningful solutions. Commonly used regularizations are norm based, but these generate over-swish solutions. We tend to propose a regularization method using each the and norms to obtain sharp object boundaries, whereas conjointly achieving good interior reconstruction of the object permittivity. Knowledge concerning the permittivity can additionally be used as a priori information. The applicability of the tactic is demonstrated on synthetically generated knowledge for 2-dimensional (2-D) microwave imaging using the Born-iterative method (BIM). The optimization routine systematically estimates all parameters, while minimizing the cost perform. Completely different objects chosen to represent realistic options have been thought of to judge the performance. The reconstructed pictures indicate that the tactic can produce correct object localization, form identification, and good permittivity estimation.