Share Email Print

Proceedings Paper

Reconstruction of non-uniform fiber Bragg grating parameters using a combination of RBF and MLP neural networks
Author(s): Arash Yazdanpanah-Goharrizi; Ali Rostami
Format Member Price Non-Member Price
PDF $17.00 $21.00

Paper Abstract

A combination of radial basis function (RBF) and multilayer perceptron neural networks (MLPNN) is employed for extracting the parameters of a non-uniform fiber Bragg grating (FBG) from its reflection spectra. The identification process is carried out in two stages; First, the type of the nonuniformity (Gaussian, Apodized, Chirped, or mixed) is identified by using RBF. Then, the parameters of the identified grating are extracted via MLPNN. In contrast to conventional reconstruction methods, which usually require a priori knowledge of the type of grating, here the grating type of nonuniformity is automatically identified. The required number of neurons for obtaining accurate results is 10 for MLPNN and 312 for RBFNN, and therefore the network can be trained fast. The proposed method is applied to different test cases and accurate results are obtained within a very short computation time.

Paper Details

Date Published: 8 September 2006
PDF: 7 pages
Proc. SPIE 6343, Photonics North 2006, 63431P (8 September 2006); doi: 10.1117/12.707779
Show Author Affiliations
Arash Yazdanpanah-Goharrizi, Univ. of Tabriz (Iran)
Ali Rostami, Univ. of Tabriz (Iran)

Published in SPIE Proceedings Vol. 6343:
Photonics North 2006
Pierre Mathieu, 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?