Signal denoising using the Stationary Wavelet Decomposition

Eleonora Stefanutti, Fabio Bruni
Abstract:
We propose a method to denoise 1D experimental signals using wavelet transform. Noise affecting experimental signals is indeed a problem shared by many different scientific and engineering fields and a proper strategy of denoising, avoiding the loss of useful information embedded in the original signal, is often essential. This is especially true for spectroscopic data, where significant features may be hidden by noise or covered by undesired components, which are not related to the physical content of the signal. In particular, we have applied the Discrete Wavelet Transform (DWT) to infrared spectra collected at a synchrotron source, overcoming the limitations of other filtering strategies conventionally employed. The good results here obtained, and the other attempts presented in the recent literature, suggest that wavelet transform can represent a valuable tool, finding its ideal application in a number of many and diverse fields, including spectroscopy, marine science, meteorology, and engineering.
Download:
IMEKO-TC19-METROSEA-2017-21.pdf
DOI:
-
Event details
IMEKO TC:
TC19
Event name:
1st IMEKO TC19 Workshop on Metrology for the Sea
Title:

"Learning to measure sea health parameters", Special session “Metrology traceability for oceanic parameters” together with TC8 and TC12

Place:
Naples, ITALY
Time:
11 October 2017 - 13 October 2017