Modeling measurements for quantitative imaging subsurface targets

TL;DR

This paper introduces a simplified measurement model for ground-penetrating synthetic aperture radar that captures essential information for subsurface imaging, validated through simulations and data processing techniques.

Contribution

The paper develops and validates a simplified measurement model for subsurface radar imaging that accounts for space/angle limitations and first-order interactions.

Findings

Modeled data images are nearly identical to full simulation images.

Principal component analysis effectively removes ground bounce signals.

The model provides a theoretical framework for understanding measurement limitations.

Abstract

We study ground-penetrating synthetic aperture radar measurements of scattering by targets located below a rough air-soil interface. By considering the inherent space/angle limitations of this imaging modality, we introduce a simplified model for measurements. This model assumes (i) first-order interactions between the target and the air-soil interface, (ii) scattering by the target below a flat air-soil interface, and (iii) a point target model. Using the method of fundamental solutions to simulate two-dimensional simulations of scalar waves for the direct scattering problem, we systematically study each of these data modeling assumptions. To test and validate these assumptions, we apply principal component analysis to approximately remove ground bounce signals from measurements and then apply Kirchhoff migration on that processed data to produce images. We show that images using this modeled data are nearly identical to those that use simulated measurements from the full direct scattering problem. In that way, we show that this model contains the essential information contained in measurements. Consequently, it provides a theoretical framework for understanding how inherent space/angle limitations affect subsurface imaging systems.