A stochastic FDTD method for statistically varying biological tissues

THE analysis of radio wave absorption is a continuing concern for the cell phone industry due to health effects (and associated regulations) of the person using the cell phone [1,2]. The amount of allowable power absorbed has a strong impact on the design of the phone (antenna, electromagnetic interference, shielding, etc.). Cost and size are typically also conflicting tradeoffs. The analysis of these designs prior to the building of prototypes and the actual testing of the radio is critical to contain cost and design cycle time, so numerical simulations are routinely used in this industry. One of the unanswered questions with these simulations is how variation between individuals or uncertainty in measured tissue properties may impact the absorbed power. Studies of adults and children have shown that size of the head and thickness of the ear have a significant effect on absorbed power [3-5]. Other studies [6,7] have shown the non-negligible effect of head shape. Variability in tissue properties (from person to person or just because of uncertainty in the measurements) has also been shown to have a significant effect on absorbed power [8]. This paper describes a novel finite difference time domain (FDTD) method of directly calculating the expected variation in the absorbed power for statistically varying tissue properties. This method directly carries this variability through the simulation, unlike others that use multiple simulations (such as Monte Carlo) to determine this variability. This new stochastic FDTD (S-FDTD) method takes a small fraction of the time to perform the variability analysis compared with the Monte Carlo method.