Andrew D. Strange

Application of ground penetrating radar technology for near-surface interface determination in coal mining

The use of ground penetrating radar (GPR) for detecting near-surface interfaces is a scenario of special interest to the underground coal mining industry. The problem is difficult to solve in practice because the radar echo is often dominated by unwanted components such as antenna crosstalk and ringing, ground-bounce effects, clutter, and severe attenuation. These nuisance components are also highly sensitive to subtle variations in ground conditions, rendering the application of standard signal pre-processing techniques largely ineffective in the unsupervised case. As a solution to this problem, we develop a novel algorithm which utilizes a pattern recognition-based approach using features derived from the bispectrum of the radar data. We show that, unlike traditional second order correlation based methods such as matched filtering which fail in known conditions, the new method reliably allows the determination of layer interfaces using GPR to be extended to the near surface region.

Coal seam thickness estimation using GPR and higher order statistics - the near-surface case

A novel pattern recognition-based approach to detect near-surface interfaces using ground penetrating radar (GPR) has been reported in [1]. The approach was used to successfully detect interfaces within 5 cm of the ground surface. This technique has been adapted for the important task of layer thickness estimation in the near-surface range. This is inherently a difficult problem to solve in practice because the radar echo is often dominated by unwanted components such as antenna crosstalk and ring-down, ground reflection effects and clutter. Features derived from the bispectrum and a nearest-neighbour classifier have been utilized for this processing task. It is shown that unlike traditional second order correlation based methods such as matched filtering which can fail in known conditions, layer thickness estimation using this approach can be reliably extended to the near-surface region.