Evaluation of pavement layer thicknesses using GPR: A comparison between full-wave inversion and the straight-ray method

• Published in: Construction & building materials Vol. 168; pp. 91 - 104
• Main Authors: De Coster, Albéric, Van der Wielen, Audrey, Grégoire, Colette, Lambot, Sébastien
• Format: Journal Article
• Language: English
• Published: Elsevier Ltd 20.04.2018; Elsevier B.V
• Subjects: Analysis; Full-wave inversion (FWI); Ground penetrating radar; Ground-penetrating radar (GPR); Pavement characterization; Pavements; Properties; Surface reflection coefficient (SRC); Thickness estimation; Wave interference; Belgium; Full-wave inversion (FWI); Ground-penetrating radar (GPR); Thickness estimation; Surface reflection coefficient (SRC); Pavement characterization
• ISSN: 0950-0618; 1879-0526
• DOI: 10.1016/j.conbuildmat.2018.02.100

•Estimation of the electromagnetic and physical properties of pavements.•Analysis of the capabilities of the surface reflection coefficient and full-wave inversion methods.•Fast and good results provided by the surface reflection coefficient approach.•Wider range of configurations and smaller errors provided by the full-wave inversion approach.•Adequate processing steps required before inverting the data with both inversion methods. The characterization of pavement properties is essential to manage transport infrastructures. In this paper, we aim at assessing the abilities of the ground-penetrating radar (GPR) full-wave inversion and straight-ray methods to achieve this task. The first approach consists in combining a full-wave-inversion procedure with a recently developed electromagnetic model. The latter takes advantage of a closed-form solution of Maxwell’s equations to describe the antenna-medium system. The second approach resorts to the surface reflection coefficient method. We showed through numerical experiments that the straight-ray method in general provides fast and good results, but full-wave inversion applies to a wider range of model configurations and is subject to less important errors. A laboratory experiment was also conducted to take into account the effects of measurement errors as well as inherent pavement heterogeneities. The results evidence that noisy data and a lack of information can sometimes lead to an inappropriate estimation of the parameters. Nonetheless, applying an adequate processing before inverting the data made both inversion methods able to provide an estimation of the pavement thickness along the acquisition profile.