Frequency Beamforming-Enhanced DBIM for Limited-Aperture Quantitative Imaging

Abstract

The distorted Born iterative method (DBIM) is a technique for reconstructing a dielectric profile from scattered electric fields. The potential for effective imaging with DBIM has been demonstrated for a variety of applications. However, it is more challenging to reliably produce high-fidelity imagery when the aperture formed by the sensors does not fully surround the region of interest and when data is collected over a narrow bandwidth. To address this challenge, we propose a beamforming enhancement to DBIM for limited-aperture scenarios. The beamforming enhancement pre-focuses the data prior to performing the DBIM optimization in order to achieve a better-conditioned inversion. The enhancement is an advance on previous beamforming work for DBIM in that it focuses simultaneously across not only space but also frequency, which allows for better leveraging of range information in the limited-aperture signal. Results for simulated and experimental data demonstrate that the frequency beamforming enhancement to DBIM results in less sensitivity to the choice of the regularization parameter and produces better reconstructions when spatial and frequency information is limited.