Fibre Bragg Grating Wavelength Shift Demodulation with Filtering and Enhancement of Spectra by Simple Nonlinear Signal Processing

A novel approach to fibre Bragg grating spectra processing is proposed. The method is based on the use of nonlinear filtration and raising the spectrum value to the second power. A combination of geometric and arithmetic mean filtering is used as nonlinear filtration. The properties of the denoising method are presented on simulation data with different signal-to-noise ratios and on experimental data. The most advantageous combination is raising the intensity of the spectrum to the square power preceded by its significant smoothing. The centroid, fast phase correlation, and cross-correlation methods are used to determine the wavelength shift in the spectrum. To obtain experimental data, measurements of the FBG in reflection mode placed in a temperature chamber with the possibility of temperature control are obtained. The simulations, measurements, and their numerical analysis confirm the usefulness of the proposed methods for demodulating the wavelength shift of FBG spectra.