Abstract
This paper presents an efficient algorithm to analyze the electromagnetic scattering from a 3-D object buried under a 2-D rough surface. The Poggio-Miller-Chang-Harrington-Wu-Tsai integral equation is implemented in the framework of the characteristic basis function method (CBFM). To expedite the solution, a new approach which utilizes a current-based physical optics (PO) method called the modified equivalent current approximation (MECA) is used for efficient calculation of the characteristic basis functions for the dielectric rough surface. Moreover, the adaptive cross approximation (ACA) is used to accelerate the generation of the reduced matrix during which the interaction between the rough surface and the object is rigorously taken into account. The proposed approach, referred to as CBFM-MECA-ACA, is shown to be particularly efficient when deriving statistical results using the azimuthal angular averaging method. In addition, the range of validity of the proposed algorithm is investigated to demonstrate that both far- and near-field quantities can be accurately derived under appropriate constraints. Finally, the numerical results are presented to show the performance and the versatility of the CBFM-MECA-ACA algorithm.