Amelioration of the Analysis of the Ultrasonic Measurement Using the Wavelet Method

Abstract

The experimental ultrasound data is always trained by measurement errors, internal and external noise. For this reason, it is always necessary to apply some signal processing operations, such as filtering, the Fourier transform, the Laplace transform, the wavelet transform, and the statistical and intelligent methods to extract relevant information desired in case of instabilities of measurements. The wavelet method is a method that has proven its importance in scientific research, especially in noise filtering, signal analysis, superimposed signal separation and data compression. It also allows you to view the variation of signals as a function of time and frequency at the same time. In this article, we will apply the wavelet transform on experimental ultrasound data to minimize noise, separate the different superimposed signals and extract the desired information, and better analyse the experimental data from an ultrasonic wave measurement. Applying the ultrasonic prism method on fresh concrete samples results in reflected ultrasound signals, which are rich in information about the properties and different characteristics of the sample under test. These measurements are sometimes unstable, noisy, and contains several interesting information about the testing sample overlapping each other. These signals are processed and analysed by the wavelet method in order to improve the exploitation of the ultrasonic measurements. The results obtained are discussed and compared with those published in literature.