Polarization Influence on Algorithms of TFBG Sensors Data Analysis for Bending Application

Abstract

In this article we deal with the influence of polarization on determining the bending radius in TFBG (tilted fiber Bragg grating) sensors. The spectrum of TFBG structures changes under the influence of many factors at the same time. In the case of bending radius measurements, additional factors may be the temperature and polarization state of the introduced light. So far, only the cladding mode envelope algorithm has been used to determine the bending radius. An interesting issue seems to be establishing the influence of cross-sensitivity of the spectrum on changes in polarization during bending measurements. In addition to the envelope algorithm, algorithms for spectral length and average deviation from the local mean were examined. As a result of the analysis of experimental data, it was found that the level of polarization’s influence on the result may be significant in determining the bending radius. Reducing the influence of polarization can be achieved by using algorithms providing wavelength parameters. Additionally, in order to reduce the influence of polarization, we proposed the use of the PLS (partial least squares regression) algorithm for the processed spectrum.