Abstract
The paper discusses two methods for addressing finite frequency selective surface (FSS) problem. The first of these deals with a relatively small-size FSS, which is treated via the spectral-Galerkin method that accounts for all of the interactions between the elements in a rigorous way. However, this method is not well-suited for larger surfaces as it places a heavy burden on the CPU time and memory requirements, both of which become unmanageably large very rapidly with increasing size. To circumvent this difficulty, we employ a second scheme based on the plane wave spectral decomposition approach, that enables us to treat the finite FSS problem as though it were doubly periodic and infinite.