RF Exposure Assessment of Children's Organs using surrogate model built with electromagnetic solvers and Statistics

The stochastic dosimetry is a novel approach combining numerical electromagnetic solver and statistics to manage the influence of variations of the inputs on the outputs of heavy dosimetric numerical calculations. In this paper, a surrogate model to heavy simulations is built and used to assess the RF exposure of children's organs with a mobile located in a trouser pocket, with a device having 4 degrees of freedom for its location The surrogate model is built using a truncation of a Generalized Polynomial Chaos Expansion. The surrogate model, substituted to the FDTD method (Finite Difference in Time Domain method), is after that, with 10,000 estimations, used to characterize the statistical distribution of the organs exposure