Underground Cavity Detection through Group Dispersion of a GPR Signal

We present a novel method for underground cavity detection using cross-borehole ground-penetrating radar (GPR). Modeling the soil-cavity system as a filter allows us to estimate the group dispersion of the GPR signal induced by propagation through a cavity. We adopt a hypothesis-testing approach based on the observed group dispersion and propose a method which allows us to detect cavities with high probability. The proposed method is validated on both simulated and real measurements. Geologic clutter that causes conventional methods to fail has negligible impact on the efficacy of the proposed method, and the proposed method may be combined with conventional methods to provide a more robust and effective cavity-detection solution.

Automated summary

We propose a novel method for underground cavity detection using cross-borehole ground-penetrating radar (GPR). The soil-cavity system is modeled as a filter, allowing estimation of the GPR signal's group dispersion induced by propagation through a cavity. Based on the observed group dispersion, we employ a hypothesis-testing approach and propose a method that can detect cavities with high probability. The effectiveness of the proposed method is validated through both simulated and real measurements, and it is not significantly affected by geologic clutter. Furthermore, the proposed method can be combined with conventional methods to offer a more robust and effective cavity-detection solution.