Share Email Print

Proceedings Paper

Radial-basis function network for the approximation of quasi-distributed FBG sensor spectra with distorted peaks
Author(s): Aleksander S. Paterno; Lucia Valeria R. Arruda; Hypolito Jose Kalinowski
Format Member Price Non-Member Price
PDF $17.00 $21.00

Paper Abstract

This paper describes the use of a neural network, specifically a Radial-Basis function network, to approximate spectra of the signal reflected by a fibre Bragg grating sensor. This approximation will help the interpretation of the data acquired from the sensor when it is supposed to work in a quasi-distributed way, but its reflected spectrum has non-uniformities which would cause the misinterpretation of the data if quasi-distributed demodulation techniques are used. Results using an emulated double-peaked spectrum from a fibre Bragg grating sensor show that the common practice of fitting with a gaussian curve and then finding its peak or directly finding the maximum of the raw spectrum would cause a larger error if compared to finding the peak of an approximated spectrum using the RBF network.

Paper Details

Date Published: 23 May 2005
PDF: 4 pages
Proc. SPIE 5855, 17th International Conference on Optical Fibre Sensors, (23 May 2005); doi: 10.1117/12.623814
Show Author Affiliations
Aleksander S. Paterno, Ctr. Federal de Educacao Tecnologica do Parana (Brazil)
Lucia Valeria R. Arruda, Ctr. Federal de Educacao Tecnologica do Parana (Brazil)
Hypolito Jose Kalinowski, Ctr. Federal de Educacao Tecnologica do Parana (Brazil)

Published in SPIE Proceedings Vol. 5855:
17th International Conference on Optical Fibre Sensors
Marc Voet; Reinhardt Willsch; Wolfgang Ecke; Julian Jones; Brian Culshaw, Editor(s)

© SPIE. Terms of Use
Back to Top
Sign in to read the full article
Create a free SPIE account to get access to
premium articles and original research
Forgot your username?