Generating synthetic sample data for the detection of explosive ordnances with GPR (Counter-IED2)


Since some time ground-penetrating radar (GPR) is being used for detecting landmines, unexploded ordnances, as well as improvised explosive devices (IEDs). Because the latter is often made of non-metallic materials, metal-detectors fail and often vehicle-mounted GPR antenna arrays are introduced. In principle, GPR can distinguish between any materials, given that the contrast of their electrical and dielectric properties is great enough. However, with certain soil conditions, GPR may fail as well and miss buried objects. Experience has shown that the signatures of target objects vary strongly with the ground. The project is financed by the German Ministry of Defence (BMVg) and aims to study the various influences of the ground on the detection of explosive ordnances with GPR.

Measurements & numeric simulations

The construction of test sites for studying the performance of GPR sensors is costly and requires a lot of time and the tests only cover a limited range of soils and target objects. The current project aims to improve this step by simulating different scenarios and thereby generate synthetic sample data. Since a couple of years, numerical simulations are used to study certain aspects of GPR-aided search for buried explosive ordnances. With modern soft- and hardware is it possible to use big 3D models with such a fine resolution as to allow the correct modeling of intricate geometries of GPR antennas. Herewith all the factors that have an essential impact on the GPR performance, such as the geometry of the target objects, the frequency-dependency of the electric soil properties, the heterogeneity of the soil and the coupling of the antenna to the ground, can be considered. Figure 2 shows a cut through a numerical 3D model of a 400 MHz GPR antenna. The synthetic sample data is evaluated and optimized via experimental data from the test sites. Lastly, several scenarios shall be modeled, which can effectively be used for the systematic study for the detectability of buried explosive ordnances as well as for the purpose of training personnel.

Current publications


Project lead

Dr. Jan Igel

+49 511 643-2770

Project scientists

Sam Stadler

Stephan Schennen


Federal ministry of defense (BMVg)


15.12.2017 - 30.09.2018


Research Area GPR

Predecessor projects