Two-dimensional FDTD inverse-scattering scheme for determination of near-surface material properties

Abstract

Numerical inverse-scattering studies found in the literature are based on either frequency- or time-domain approaches. Frequency-domain algorithms can result in a very large number of unknown parameters. In contrast. time-domain approaches can exploit causality to limit the region of inversion, potentially reducing the number of unknowns. The most relevant approach to our work is the one-dimensional (1D) FDTD formulation of an inverse-scattering scheme for remote sensing of inhomogeneous lossy layered media (Umashankar, K.R. et al., 1994). A layer-stripping procedure was used to recover the conductivity and permittivity profiles simultaneously, assuming a normally incident plane wave pulse and zero noise present on the received pulse. Our work extends that technique to the case of a two-dimensional (2D) layered lossy half-space impulsively excited at its surface by an infinitely long monopole. The new algorithm permits simultaneous estimation of the electrical permittivity, /spl epsiv//sub r/, and conductivity, /spl sigma/, of a surface layer by generating a search trajectory in the (/spl epsiv//sub r/, /spl sigma/) parameter space of the layer. The robustness of the algorithm is tested in the presence of Gaussian noise. We illustrate our new algorithm by the measurement of the dielectric parameters and skin thickness of the female human breast. Monitoring of near-surface breast properties may be useful for the prediagnosis of underlying tissue pathologies.