Random gaussian rough surfaces for full-wave electromagnetic simulations

Abstract

In the paper, we present a straightforward procedure to utilize random rough surfaces with Gaussian probability distribution in commercial full-wave electromagnetic simulation software CST Microwave Studio. Random, two dimensional (2D) matrices with desired size and statistical parameters such as correlation length and root mean square height of the profile are generated in MATLAB®. In order to obtain desired surface parameters, we employ a Gaussian random 2D matrix generator and subsequent filtering with Gaussian 2D filter. A macro in Visual Basic for Application is then developed to convert these matrices into continuous surfaces in CST MWS. Radiation patterns of horizontally and vertically polarized half-wave dipoles are finally evaluated above these surfaces and compared to analytical models.