Research On The Technology Of Fiber Bragg Grating Sensing Signal In Wavelength Detection

Optical fiber sensor has been used in various fields such as engineering, medicine, aerospace and military. Also, it is suitable for harsh environment and easy to reuse for its advantages of light, sensitivity and anti electromagnetic interference. Fiber sensor, especially the Fiber Bragg Grating(FBG) sensor are the most widely used. The FBG sensor detects physical parameters such as temperature and stress, by demodulating the center wavelength of the reflection spectrum.Denoising and peak detection, which the former is the precondition for the latter, are the two important components in the FBG wavelength demodulation technology. In this paper, the system noise in FBG temperature sensing system is analyzed, and an effective denoising scheme is proposed. An improved empirical mode decomposition method is proposed for the problem of single Gauss white noise interference in the signal. Based on the empirical mode decomposition(EMD) algorithm, the singular value decomposition(SVD) which determines the high and low frequency demarcation index value is used, the adaptive threshold and the convex curve threshold function are presented according to the distribution characteristics of noise and signal in different intrinsic mode function components. An improved generalized multi atom matching pursuit algorithm based on compressive sensing(CS) theory is proposed for mixed noise consisting of Gauss white noise and pink noise. According to the correlation coefficient, the degree of sparsity is estimated and the number of autom of each iteration and the number of iteration are determined. The reconstruction precision of sensor signal is improved, and the computation time complexity is greatly reduced.In this paper, based on the analysis of the asymmetry of the FBG reflection spectrum and the large calculation and much time consuming of the current peak detection algorithm, we propose a trapezoidal integration algorithm. FBG temperature calibration model is analysised by building a FBG temperature sensing measurement system, and the FBG temperature sensing signal was processed by the method proposed in this paper, then, the wavelength shift of different temperature was obtained to complete the temperature calibration to verify the proposed scheme.