Deconvolution from instrumental devices and source effect in acoustic experiments

Abstract

Two methods to deconvolve experimental data from the distortions introduced by instrumental devices or source effects are presented. Considering a total acquistion system (emissionreception line, amplifier, pre-amplifier) as a global experimental filter, we can define its properties (module and phase) experimentally from the generation of a family of source signals dilated in time. The estimation of this filter allows the deconvolution of the recorded output signal. The first approach is based on the simple reconstruction formula of the continuous wavelet transform (CWT). The second method is based on the construction of a normalized family of a finite number of specific filters, independent of the frequency range used. In both cases, experimental results in an acoustic tank are presented. We show that after deconvolution, the source signal is correctly reconstructed from the recorded output signal and the global instrumental filter. Index Terms-Acoustic experiment, continuous wavelet transform, deconvolution, diffraction tomography experiment, family of adapted and normalized filters, heterogeneous rock, simple reconstruction formula of the CWT, transmission experiment, ultrasonic data.

Wavelet modeling of signals for non-destructive testing of concretes

“…An ultrasonic non-destructive test (NDT) uses the attenuation and the time-delay of ultrasonic waves to detect discontinuities in the detected structures. In order to enhance the spatial resolution of ultrasonic signals, a wide variety of methods have been proposed, which include time-frequency joint analysis methods [1,2], split spectrum processing [3], deconvolution of instrument response [4], damage localization by enhanced imaging [5][6][7], etc. To improve the resolution of the traditional ultrasonic B-scan image, the method of synthesized aperture focusing technology (SAFT) [7] is proposed, which simulates the effect of linear array transducers by numerically superimposing the received signals at adjacent apertures.…”

3-D seismic endoscopy: multiscale analysis and dynamic azimuthal filtering of borehole waves

“…These preliminary processes were successfully tested on experimental data in an acoustic laboratory (Valero et al 2001). When the recorded signal needs to be deconvolved from the instrumental devices and/or the source effect, the methodology developed in Saracco & Tchamitchian (1990) and Valero et al (2002) is applied. This study encountered the well‐known problem of imaging around the well, because of the large amplitude difference between the borehole waves and the far‐field reflected waves, of interest.…”