A multi-peak demodulation algorithm based on Ht-Exg fiber grating sensing network

Abstract

Aiming at the poor accuracy of traditional single peak detection algorithm in fiber grating demodulation, this paper proposes a symmetry corrected Gaussian nonlinear fitting algorithm based on Hilbert transform (HT-EXG) for multi peak detection. The algorithm uses Hilbert transform to segment the input reflection spectrum and uses exponential modified Gaussian nonlinear fitting algorithm to correct and optimize its peak position, so as to achieve accurate positioning of the reflection spectrum. The experimental results show that the proposed HT-EXG algorithm improves the accuracy of FBG peak demodulation and the stability of the system. The average error of peak detection is less than 3 pm, and the system stability is better than 0.5 pm. It provides an accurate demodulation method for FBG sensor network peak detection.