Scattering of waves by rough surface based on a hybrid UV/SMCG method

In simulations of 3D rough surface scattering, a fast algorithm developed to accelerate method of moments (MoM) solutions is the sparse-matrix canonical grid method (SMCG). We have also developed a multilevel UV method (Tsang, L. et al., Radio Sci., vol.39, no.5, 2004), in which fast coarse-coarse sampling is used to search the rank of each impedance block, and construct matrices U and V to compress each block based on an interpolation technique. We demonstrate a hybrid UV/SMCG technique in the 3D scattering problem of a Gaussian random rough surface with an exponential correlation function and with a Gaussian correlation function. It has been shown that surfaces with exponential correlation functions are applicable to land surfaces such as soil surfaces and the interface between snow and ground. The frequency dependencies of scattering using a single set of physical parameters are illustrated in the simulations. The numerical simulation results are compared with the small perturbation method (SPM) and the Kirchhoff approximation (KA).