Buried-Object Time-Reversal Imaging Using UWB Near-Ground Scattered Fields

Abstract

In this paper, the problem of time-reversal imaging (TRI) of buried dielectric and metallic targets under Gaussian rough surfaces is studied. For the scattering problem, a parallel finite-difference time-domain technique is used. Using a multistatic scattering matrix of a TR operator at a frequency range of 0.5-2.5 GHz, ultrawideband (UWB) TR multiple signal classification images are calculated. First, to reduce the clutter influence, near-ground UWB scattered fields are used. It is shown that employing received signals near the ground instead of those at the transmitter height improves the signal-to-clutter power ratio by at least 10 dB at 1 GHz. Second, a time-gated (TG) TRI algorithm is proposed to better detect and localize buried targets in the presence of clutter. Third, once the TG TR images are obtained, best single-frequency TR images are calculated, yielding even better target localization.