Time delay and surface roughness estimation by subspace algorithms for pavement survey by radar

Abstract

In civil engineering, ground penetrating radar is widely used for road pavement surveys. In contrast to the existing literature, the influence of interface roughness (surface and interlayer roughness of stratified media) is accounted for within the scope of the data processing of radar signals. The rigorous electromagnetic method PILE (propagation inside layer expansion) provides the simulated data. The observed frequency variations of the radar magnitude introduce some shape distortion on the radar wavelet. An adaptation of the root-MUSIC algorithm is proposed on the basis of the work in [4]. As a result, it is allowed to jointly estimate the time delay and the interface roughness.

Advanced Inversion Techniques for Ground Penetrating Radar

“…For example, in [62] an efficient algorithm for the computation of the time domain reflection coefficient in the transverse magnetic (TM) case has been developed in order to better characterize the surface reflections. In [63], [64] subspace methods and learning-based strategies have been used for the estimation of time delays, permittivity and roughness parameters within pavement structures. Estimation of specific soil parameters (e.g., moisture and clay content) have also been addressed in [65].…”

Estimation of time delay and interface roughness by GPR using modified MUSIC

“…For decimetre-scale GPR wavelengths (in the air), this assumption can be held, but for an ultra-wide band radar, this is no longer suitable. The influence of interface roughness and heterogeneous of medium must then be analysed [13,18,19]. In this paper, only the interface roughness is discussed.…”

“…In [19,20], this kind of work has already been carried out, but the frequency behaviour coming from the roughness has been simply approximated by an exponential function (using a curve fitting method). In this situation, the high resolution methods can easily be applied for parameters estimation (time delays and roughness).…”

“…In order to reduce the errors coming from the curve fitting, we propose a modified MUSIC algorithm which can take into account several possible frequency behaviours, more adaptable for ultra-wideband GPR. Like in [19,20], we also focus on the estimation of time delays and interface roughness. Unlike the methods in [19,20], the proposed method allows estimating the time delays without knowing the frequency behaviour from roughness.…”

“…Like in [19,20], we also focus on the estimation of time delays and interface roughness. Unlike the methods in [19,20], the proposed method allows estimating the time delays without knowing the frequency behaviour from roughness. Then, the roughness parameter is estimated by MLE [22] with the estimated time delays from the modified MUSIC algorithm.…”