A Stochastic FDTD Method for the Analyses of Geometrical and Media Uncertainties Based on the Conditional Probability

A stochastic finite-difference time-domain approach for analyzing geometrical and media uncertainty is described. The proposed method can convert the geometry uncertainties into media uncertainties based on the conditional probability. It can solve 1-D, 2-D, and 3-D problems. In order to improve the accuracy of the method, the improved correlation coefficients related to space division are proposed. The proposed method and improved correlation coefficients are validated with simulations of both 1-D and 3-D stochastic scattering objects. The simulation results agree well with the results of the Monte Carlo (MC) method, while the computation time of this method is only about 2% of the time consumption of the MC method.