Time-Domain Spectral Inversion Method for Characterization of Subsurface Layers in Ground-Penetrating-Radar (GPR) Applications

摘要

Ground-Penetrating-Radar (GPR) data analysis has been widely utilized in subsurface and geophysics applications. One of the applications turning into great importance is multilayer subsurface hydraulic parameter identification and soil water content estimations. The classic GPR techniques have shown limitations in detecting deeper soil layers and unsatisfactory accuracy in estimating the electrical properties of the layers. Spectral inversion methods have been recently identified and developed to be an effective tool to tackle these problems. In this work, the spectral inversion method is extended in time-domain and a comprehensive formulation of the algorithm along with an improved well-defined cost function is presented. The Time-Domain Spectral Inversion (TDSI) method is then applied to environmentally-relevant multilayer soil geometries and the corresponding estimated electrical properties are drawn. The results show the TDSI method considerably ameliorates the performance of the inversion in terms of simplicity, accuracy, and applicability.