Effect of source and receiver radiation characteristics in subsurface prospecting within the distorted Born approximation

Abstract

[1] Linear models are widely adopted in subsurface prospecting because of the advantages in terms of effectiveness and numerical efficiency. They make it possible to simplify the problem of inverse scattering and at the same time to obtain acceptable results in locating buried objects. In this paper, we deal with subsurface prospecting where a model linearization is performed via distorted Born approximation and considering a half-space geometry in a two-dimensional scalar problem within a multistatic/multiview configuration. In this realm, most inversion algorithms fail to take into account the radiation properties of both transmitting and receiving antennas. Our goal is thus twofold. First, we aim to provide an insight into the role played by the radiation properties of antennas on the quality of the reconstructed tomographic image of the investigated domain. This task is performed by first tackling the single-frequency case in order to gain insight into the effect of the radiation pattern using simple arguments and then examining the multifrequency case. Numerical results confirm the theoretical expectation that the quality of the reconstruction gets worse for antennas with directive radiation patterns. Second, we numerically investigate the effects of inaccurate knowledge of the antenna pattern.Citation: Soldovieri, F., R. Persico, and G. Leone (2005), Effect of source and receiver radiation characteristics in subsurface prospecting within the distorted Born approximation, Radio Sci., 40, RS3006,

Two different approaches for georadar data processing: A case study in archaeological prospecting

“…First of all, we recall that the electromagnetic field radiated by the antenna can be given under a spectral representation (Soldovieri et al, 2005); therefore, the radiated field (i.e., the incident field in our inversion) can be "physically interpreted" as the superposition of plane waves of different propagation directions. Differently in our inversion scheme we consider as incident field a plane wave with normal incidence (i.e., incidence direction along the z-axis).…”

Two-Dimensional Linear Inversion of GPR Data with a Shifting Zoom along the Observation Line

“…The problem at hand inserts into the framework of microwave tomography from GPR data, which has been largely investigated in the literature, both from the point of view of the imaging algorithms [30][31][32][33] and from the point of view of the effects of the soil and antennas on the data [34][35][36]. Applications to cases in the field have been also presented, e.g., in [37,38].…”

Linear and non linear tomographic approaches for imaging dielectric targets embedded in an inhomogeneous medium

“…in cases wherein the objects are embedded in a homogeneous medium and the probes are placed around the investigated region, some interesting approaches and results have been given in [2]- [4]. For non destructive testing for geophysical, civil engineering, military and humanitarian applications, wherein the objects are buried in a homogeneous half-space and aspect limited data are available, a number of techniques have been developed in the framework of the BA [SI- [7] as well as within the more general non-linear case [SI-[ lo]. Similar approaches have been also proposed for the case ofobjects embedded in stratified media [ll], [12].…”

“…As a consequence, the contributions related to the singular values closer to zero would be strongly corrupted in presence of even little errors on data. In order to tackle such an instability, we introduce a regularized solution based on a Truncated SVD (TSVD) expansion [14]: (7) By restricting the solution space to that spanned by the first N singular functions, such a regularized solution allows to reduce the effect of errors on data in the reconstruction [l4]. However, a reduced number of terms in the expansion (7) leads to an approximate representation o f the sought unknown.…”

“…In the general case the inverse scattering problem is a nonlinear one, but when the targets are embedded in a lossy medium a large number o f cases can be modeled by a linearized approximation such as the BA [5]- [7]. It amounts to assume the internal field E in D equal to the unperturbed field E,, ' By so doing, a considerable simplification is achieved because the overall problem reduces to the inversion of a single linear integral equation, that is:…”