Prediction of the effects of soil and target properties on the antipersonnel landmine detection performance of ground-penetrating radar: A Colombian case study

The performance of ground-penetrating (GPR) radar is determined fundamentally by the soil electromagnetic (EM) properties and the target characteristics. In this paper, we predict the effects of such properties on the antipersonnel (AP) landmine detection performance of GPR in a Colombian scenario. Firstly, we use available soil geophysical information in existing pedotransfer models to calculate soil EM properties. The latter are included in a two-dimensional (2D), finite-difference time-domain (FDTD) modeling program in conjunction with the characteristics of AP landmines to calculate the buried target reflection. The approach is applied to two soils selected among Colombian mine-affected areas, and several local improvised explosive devices (IEDs) and AP landmines are modeled as targets. The signatures from such targets buried in the selected soils are predicted, considering different conditions. Finally, we show how the GPR can contribute in detecting low- and non-metallic targets in these Colombian soils. Such a system could be quite adequate for complementing humanitarian landmine detection by metal detectors.