KF_Capstone

MS Capstone project, advised by Dr. Arnold Kim. 2025 In this project, we simulate electromagnetic wave scattering from rough surfaces using the method of fundamental solutions (MFS) to study problems in which a localized object is situated above the surface. We focus on two-dimensional problems with both sound-soft and sound-hard boundary conditions. We assess the accuracy of the MFS on flat surfaces and extend the approach to random rough surfaces. To recover target signals in the presence of strong surface reflections, we introduce a PCA-based filtering method that subtracts dominant portions of the measurements associated with surface-only scattering. We evaluate its effectiveness using numerical simulations of synthetic aperture radar (SAR)-like data. Our results demonstrate that even simple imaging strategies benefit significantly from this filtering, particularly when surface scattering dominates the response.