| The combination of Tilted Fiber Bragg grating(TFBG)and Surface Plasmon Resonance(SPR)technology provides a new type of optical fiber surface refractive index sensing technology in recent years,with high sensitivity,real-time monitoring,fast response speed,etc,which has important applications in biomedicine,drug safety,environmental monitoring and other fields.As a TFBG-SPR sensing technology with high refractive index sensitivity,efficient analysis and accurate demodulation of spectral information is the key to improve the performance of the sensing system.Since the current TFBG-SPR sensor spectrum is a complex superposition of the SPR absorption peak and the comb spectrum of the TFBG high-order cladding modes,which brings great obstacles to the demodulation of the sensor information of the TFBG-SPR spectrum.Aiming at the key problems of inaccurate acquisition and demodulation of SPR wavelength and difficult to apply to trace biochemical sample detection for TFBG-SPR complex spectrum,this paper designs an automatic demodulation algorithm that integrates compressed sensing spectral reconstruction and SPR wavelength solution.The main work of the thesis is as follows:1.Design a demodulation algorithm to automatically solve the SPR wavelength,using the least square method,Fourier fitting and other mathematical principles to perform filtering,peak finding,fitting and other processing operations on the TFBG-SPR spectrum.The algorithm includes a filter module,a baseline processing module,a smoothing module,a peak finding module,and a fitting evaluation module.By adjusting the sequence and parameters of each module,SPR wavelength demodulation is achieved.2.In view of the problems of excessive spectral noise and complex peak information,the algorithm optimization work has been carried out,including correction of parameters,abandoning of unreasonable extreme points,and improvement of peak finding accuracy.For some problem data points in the SPR trend chart,the problem data is returned to the data point fitting curve to obtain an SPR wavelength change trend curve that is more in line with the objective law.Using the GUI interface design function in MATLAB,the demodulation algorithm is designed as an interactive graphical interface.Select the input spectrum data in the data path,you can quickly get the SPR wavelength drift trend point diagram before and after the data correction,the SPR wavelength at each time point and the overall wavelength drift of the SPR.Finally,a TFBG-SPR demodulation algorithm that can demodulate real-time signals is designed.3.Combining the principle of compressed sensing,via different sampling rates,we use sparse transformation methods and reconstruction algorithms to reconstruct the TFBG-SPR spectrum.After multiple reconstruction and restoration experiments,fast Fourier transform is selected to represent the signal sparsely.The sampling rate of under-sampling is 25%,and the Co Sa MP reconstruction algorithm is used to reconstruct and restore the spectrum.4.The experimental data measured by the TFBG-SPR refractive index sensor is used to reconstruct the spectrum,and the combine with the demodulation algorithm for analysis and verification to obtain linear demodulation data.This result also verified the spectral reconstruction effect and the stability of the demodulation algorithm.This paper proposes a spectrum demodulation system with arbitrary wavelength tunable laser as the core and hardware compressed sensing technology to achieve high precision,low cost,high performance and compact structure.In this paper,compressed sensing technology is used to reconstruct the TFBG-SPR spectrum,combined with the self-designed SPR wavelength demodulation algorithm,an automatic demodulation algorithm integrating compressed sensing spectral reconstruction and solving SPR wavelength is designed,and the refractive index is obtained by demodulation.The sensitivity is 544.64nm/RIU,and the minimum wavelength change of SPR is 0.05nm. |
PDF Full Text Request