01661914

Component

This study is performed to investigate the feasibility of the Hilbert-Huang Transformation (HHT) technique to predict pavement layer thickness accurately from Ground Penetrating Radar (GPR) data with multiple reflections at shallow depth. These multiple reflections are often misunderstood as pavement layers due to the reasons: (a) data are not carefully analyzed; (b) lack of experience; and (c) insufficient core thicknesses needed for validation. To facilitate this study, GPR survey was conducted on the selected pavement sections with 2.0 GHz air-launched horn antenna. These pavement sections are US285, NM14, and Interstate 40 (I-40). GPR survey was conducted on different layers of the first two sections during reconstruction and shoulder of the third section which is an instrumented pavement section. The HHT is performed on GPR signal data by reconstruction of signal through Empirical Mode Decomposition (EMD) followed by a Hilbert Transformation (HT). Instantaneous amplitudes of reconstructed signal are determined using the HHT. It is repeatedly observed that instantaneous amplitudes can predict the 2-way travel time along layer interfaces even after the presence of multiple reflections. Later, layer thicknesses are predicted from the 2-way travel times and distribution of thickness is consistent. Predicted layer thicknesses are compared to the measured thicknesses. Predicted thicknesses are fairly accurate in most of the cases except some layers where a little over-prediction is observed. Finally, this method is recommended to be incorporated during the thickness prediction from GPR data.

This paper was sponsored by TRB committee AFD20 Standing Committee on Pavement Condition Evaluation.

18-05699

English

20p

2018

Transportation Research Board 97th Annual Meeting

Digital/other

Figures; Photos; References

Ground penetrating radar; Pavement layers; Thickness

Empirical Mode Decomposition; Hilbert-Huang transform

Highways; Pavements

Transportation Research Board Annual Meeting 2018 Paper #18-05699

TRIS, TRB, ATRI

Jan 8 2018 11:27AM