Numerical and Experimental Analysis of Multistatic GPR Systems for Subsurface Inspection

Ground-Penetrating Radar (GPR) is a non-invasive sensing technology [1] that exploits the propagation of electromagnetic pulses to investigate opaque media, such as soil, sand, ice, concrete, asphalt, and many others. GPR enables the detection and characterization of dielectric anomalies arising from interfaces, voids, cracks, moisture ingress, reinforcement corrosion, and variations in layer thickness. Owing to its non-destructive nature, GPR has become a key tool for the structural health monitoring of critical infrastructures (e.g. bridges, tunnels, roads, railways, and buildings) thereby contributing to the sustainability, safety, and resilience of the built environment.

Although GPR is a well-established technology employed in a wide range of operational contexts, its performance can be significantly degraded by the presence of noise and clutter, especially when operating in contactless mode and in complex scenarios, f.i. when mounted on mobile vehicles, unmanned aerial platforms, or robotic systems [2]. To overcome these limitations, advanced measurement configurations employing multiple transmitting and receiving antennas have recently been proposed [3]. At the state of art, multistatic radar technology represents a promising solution for mitigating signal disturbances and enhancing subsurface imaging capabilities [4]. However, this technology entails a substantial increase in data volume and computational complexity, thus requiring the development of efficient and robust signal processing and image reconstruction strategies. Within this framework, a key challenge lies in the identification of suitable measurement setups that achieve an optimal trade-off between imaging performance and computational cost.

In this contribution, a performance assessment of different multistatic antenna configurations operating in a three-dimensional free-space scenario and consisting of a single transmitting antenna and multiple receiving antennas is considered. First, a numerical analysis will be conducted to assess the imaging capabilities of the system. Then, experimental results obtained under controlled laboratory conditions will be presented to validate the proposed imaging approach and identify the configurations that provide the best compromise between reconstruction quality and computational cost.

References

• Daniels, David J., ed. Ground Penetrating Radar. Vol. 1. Iet, 2004.
• Catapano I., Gennarelli G., Ludeno G., Noviello C., Esposito G., Soldovieri F., "Contactless Ground Penetrating Radar Imaging: State of the art, challenges, and microwave tomography-based data processing," in IEEE Geoscience and Remote Sensing Magazine, vol. 10, no. 1, pp. 251-273, March 2022, doi: 10.1109/MGRS.2021.3082170.
• Masoodi, M.; Gennarelli, G.; Noviello, C.; Catapano, I.; Soldovieri, F. Performance Assessment of Multistatic/Multi-Frequency 3D GPR Imaging by Linear Microwave Tomography. Sensors 2025, 25, 6467.
• Noviello, C.; Braca, P.; Maresca, S. Chapter 5—Radar Networks. In Photonics for Radar Networks and Electronic Warfare Systems; SciTech Publishing, Inc.: Raleigh, NC, USA, 2019; p. 111.