Compressive Sensing Techniques for the Detection of Surface Waves

Abstract

Understanding the scattering physics of specific objects is important for their design and use. Imaging techniques may give this insight, provided that they can distinguish different scattering mechanisms. The use of compressive sensing (CS) in imaging allows to use different scattering models. In this paper, two models for surface wave (SW) scattering are described and applied to a canonical test case. We find that CS is effective in extracting the reflection of surface waves from the total radar echo. However, the standard l1-regularization of the CS optimization problem requires careful balancing of the different scattering models. It is shown that the Bayesian approach to solve the CS optimization problem does not suffer from this problem.