High Resolution Deconvolution of Ultrasonic Traces

Impedance Profiling for Bond Strength Evaluation

“…These were: frequency domain Wiener filtering, autoregressive spectral extrapolation (ASE) and Ll-norm spike extraction [7], developed by Barrodale Computing Services of Victoria, Canada. Figure 2 shows the calculated impulse responses for a rubber / araldite/ aluminium sandwich obtained from synthetic pulse-echo data, using three different deconvolution methods.…”

Further improvement of temporal resolution of seismic data by autoregressive (AR) spectral extrapolation

“…This admission is also favorable for the seismic data. Furthermore, frequency window should be selected where the SNR is high so that the extrapolated spectrum will be almost flat (Miyashita et al, 1985;Zala et al, 1988). The reflectivity sequence of layered media consisting of a number of spikes has approximately flat spectrum.…”

“…The AR technique has been used to predict the missing low and high frequencies and consequently to recover acoustic impedance by inversion of seismic reflection data (Lines and Clayton, 1977;Oldenburg et al, 1983;Walker and Ulrych, 1983). However, the combination of Wiener filtering and autoregressive spectral extrapolation have also been used to improve temporal resolution in analyzing ultrasonic signals which are also used in the discrimination of closely spaced reflectors as well as the detection of the discontinuities in coarse-grained materials, such as austenitic steel welds (Miyashita et al, 1985;Zala et al, 1988;Sin and Chen, 1992;Honarvar et al, 2004). All these studies have shown that Wiener filtering followed by autoregressive spectral extrapolation can produce very good results at a reasonable computational cost.…”

Preliminary Research on the Nonlinear Ultrasonic Detection of the Porosity of Porous Material Based on Dynamic Wavelet Fingerprint Technology

“…* corresponds to the conjugate. ψ*false(false(t−bfalse)/afalse) is formed via dilation and translation from the wavelet function [19,20]. The relationship between the scale a and the frequency f in the wavelet transform is:f=faa×T where T is the sampling period of the signal, and f a is the center frequency of the wavelet function.…”