Skip to main content
U.S. Department of Energy
Office of Scientific and Technical Information

Wavelet transform techniques and signal analysis

Conference ·
OSTI ID:6690930
;  [1];  [2]
  1. Oak Ridge National Lab., TN (United States) Tennessee Univ., Knoxville, TN (United States). Dept. of Nuclear Engineering
  2. Universidad Politecnica de Madrid (Spain). Facultad de Informatica

Traditionally, the most widely used signal analysis tool is the Fourier transform which, by producing power spectral densities (PSDs), allows time dependent signals to be studied in the frequency domain. However, the Fourier transform is global -- it extends over the entire time domain -- which makes it ill-suited to study nonstationary signals which exhibit local temporal changes in the signal's frequency content. To analyze nonstationary signals, the family of transforms commonly designated as short-time Fourier transforms (STFTs), capable of identifying temporally localized changes in the signal's frequency content, were developed by employing window functions to isolate temporal regions of the signal. For example, the Gabor STFT uses a Gaussian window. However, the applicability of STFTs is limited by various inadequacies. The Wavelet transform (NW), recently developed by Grossman and Morlet and explored in depth by Daubechies (2) and Mallat, remedies the inadequacies of STFTs. Like the Fourier transform, the WT can be implemented as a discrete transform (DWT) or as a continuous (integral) transform (CWT). This paper briefly illustrates some of the potential applications of the wavelet transform algorithms to signal analysis.

Research Organization:
Oak Ridge National Lab., TN (United States)
Sponsoring Organization:
DOE; USDOE, Washington, DC (United States)
DOE Contract Number:
AC05-84OR21400
OSTI ID:
6690930
Report Number(s):
CONF-930601-4; ON: DE93007919
Country of Publication:
United States
Language:
English