ABSTRACT
Ac conductivity time series have been used in the past to study various phenomena in materials. Wavelet analysis is a processing tool, whose accuracy contributes towards the characterization of geomaterials with respect to their physical properties. Specifically in this work wavelet analysis of ac conductivity time series is used in order to identify probable structural imperfections in marble samples and differentiate between two different groups of samples (uncompressed – uniaxially compressed). For the analysis in the frequency domain the Continuous Wavelet Transform has been used instead of Fourier Transform as the former is preferable in case of non-stationary signals. The 2nd derivative of the Gaussian probability density function, known as “Mexican Hat” and the 2nd Daubechies wavelet were tested against the 10th Daubechies wavelet so as to select the most suitable as mother wavelet. By calculating the wavelet spectra, distinguishable differences were identified in the spectral analysis of both compressed and uncompressed samples.
