Iterative and single-step solutions of multi-offset ultra wide band data in the time domain

Abstract

This paper deals with the inverse scattering of ultra wide band (UWB) tomography used in reconstruction the dialectic properties of the unknown targets in 2D. The image reconstruction algorithm is based on the gradient minimization of an augmented cost function defined as the different between measured and calculated fields. The computation requires two successive steps: (i) direct and (ii) adjoint solutions. The forward-backward time stepping algorithm, implementing the finite-difference time-domain (FDTD) method with Mur's absorbing boundaries is employed in both steps. The imaging algorithm is based on non-linear optimization technique from which the single-step and iterative inversion schemes are derived. The experimental results demonstrate that the algorithms can resolve features whose sizes are comparable to the half wave-length even though scaterring data are collected only from limited view angles. These experimental evidence suggest that the technique could potentially be used to solve practical imaging problems such as detecting cancerous tumors in breast. DOI: 10.1109/IST.2009.5071658