Title: Reconstructing a point source from diffusion fluxes to narrow windows in three dimensions

Abstract: We develop a computational approach to locate the source of a steady-state gradient of diffusing particles from the fluxes through narrow windows distributed either on the boundary of a three dimensional half-space or on a sphere. This approach is based on solving the mixed boundary stationary diffusion equation with the Neumann-Green's function and matched asymptotic. We compute the probability fluxes and develop a highly efficient analytical-Brownian numerical scheme. This scheme accelerates the simulation time by avoiding the explicit computation of Brownian trajectories in the infinite domain. Our derived analytical formulas agree with the results obtained from the fast numerical simulation scheme. Using the analytical representation of the particle fluxes, we show how to reconstruct the location of the point source. Furthermore, we investigate the uncertainty in the source reconstruction due to additive fluctuations present in the fluxes. We also study the influence of various window configurations (cluster vs uniform distributions) on recovering the source position.