Spatial Domain Generation of Random Surface Using Savitzky-Golay Filter for Simulation of Electromagnetic Polarimetric Systems

摘要

A spatial-domain algorithm is introduced for generating both isotropic and anisotropic natural rough surface models with predetermined statistical properties using Savitzky-Golay filter. Unlike the spectral-domain methods, the proposed method does not require the calculation of two-dimensional inverse Fourier transforms which constitute a computational burden for generating large ensembles of high-resolution surfaces. A comparative analysis between the proposed spatial-domain method and one of the conventional spectral-domain methods is presented. The comparison shows that the proposed spatial-domain method results in substantially improved computational performance. The fitness of the generated rough surface to the predetermined statistical properties is verified by calculating a variogram for numerical measurement of the variance and the correlation lengths. To demonstrate the importance of generating such a rough surface for simulation of electromagnetic polarimetric systems, a rough surface with anisotropic statistical properties is placed in the near field region of two antennas of orthogonal polarizations. It is shown that the self and mutual electromagnetic coupling coefficients of the two antennas can be used for measuring the orientation of the rough surface.