Experimental Validation of a Numerical Forward Model for Tunnel Detection Using Cross-Borehole Radar: Fid-Bau Portal

Experimental Validation of a Numerical Forward Model for Tunnel Detection Using Cross-Borehole Radar

Farid, Arvin / Martinez-Lorenzo, Jose A. / Alshawabkeh, Akram N. / Rappaport, Carey M.

Abstract

The goal of this research is to develop an experimentally validated two-dimensional (2D) finite difference frequency domain (FDFD) numerical forward model to study the potential of radar-based tunnel detection. Tunnel detection has become a subject of interest to the nation because of the use of tunnels by illegal immigrants, smugglers, prisoners, assailants, and terrorists. These concerns call for research to nondestructively detect, localize, and monitor tunnels. Nondestructive detection requires robust image reconstruction and inverse models, which in turn need robust forward models. Cross-well radar (CWR) modality was used for experimentation to avoid soil-air interface roughness. CWR is not a versatile field technology for political boundaries but is still applicable to monitoring the perimeter of buildings or secure sites. Multiple-depth wideband frequency-response measurements were experimentally collected in fully water-saturated sand, across PVC-cased ferrite-bead-jacketed borehole monopole antennae at a pilot-scale facility (referred to as SoilBED). The experimental results were then compared with the 2D-FDFD model. The agreement between the results of the numerical and experimental simulations was then evaluated. Results provide key diagnostic tools that can help to develop the algorithms needed for the detection of underground tunnels using radar-based methods.