Stochastic FDTD for Analysis of Statistical Variation in Electromagnetic Fields

Abstract
This paper describes a new stochastic finite difference time domain (S-FDTD) method for calculating the variance in the electromagnetic fields caused by variability or uncertainty in the electrical properties of the materials in the model. Details of the 1D derivation using truncated Taylor series approximations are given. The S-FDTD analysis is then compared to Monte Carlo analysis for a 1D bioelectromagnetic example. The accuracy of the method is controlled by the approximations for the cross correlations of the fields and electrical properties. In this paper, we are able to bound the variance using cross correlations of 1 (overestimate) and the reflection coefficient (underestimate).