A peak-finding method and system for brillouin gain spectrum

Abstract

The application discloses a peak searching method and system for Brillouin gain spectrum, which comprises the following steps: obtaining Brillouin curves under different swept carrier frequencies, signal conditioning of Brillouin gain spectrum at different time, and fitting and peak searching of the signal after signal conditioning. The application reduces the noise introduced by carrier drift, relative intensity disturbance and other factors through signal decomposition, filtering and reconstruction of the original Brillouin gain spectrum, improves the signal-to-noise ratio of the Brillouin gain spectrum signal, and improves the accuracy of Brillouin gain spectrum peak searching.

Description

Technical Field

[0001] This invention belongs to the field of peak point location determination algorithms for energy signals, and particularly relates to a peak determination method for Brillouin gain spectrum. Background Technology

[0002] Distributed fiber optic strain sensors based on Brillouin scattering are widely used in the field of fiber optic sensing. Taking a sensor based on a Brillouin optical time-domain analyzer and employing an optical frequency sweep method as an example, a pump pulse light with a constant frequency is injected into one end of the sensing fiber; a frequency-shifted continuous pulse light modulated by a carrier is injected into the other end. According to the stimulated Brillouin effect, the energy of the pulse light will be transferred to the continuous light, and the magnitude of the transferred energy is related to the Brillouin frequency shift of the fiber medium at the energy transfer point. By controlling the frequency of the continuous light, the transferred energy changes in a Lorentzian line, forming a Brillouin gain spectrum. By determining the peak position of the Brillouin gain spectrum, the strain of the sensing fiber caused by external force can be measured.

[0003] Among them, the frequency scanning interval of continuous light determines the dispersion of the Brillouin gain spectrum, thus affecting the peak finding accuracy of the Brillouin gain spectrum; by fitting the Brillouin gain spectrum and using cross-correlation operation, the peak finding accuracy can be effectively improved, breaking through the strain measurement accuracy determined by the frequency scanning interval.

[0004] Furthermore, due to wavelength temperature drift and relative loudness noise in devices such as electro-optic modulators, microwave generators, and lasers, the Brillouin gain spectrum is distorted and the signal-to-noise ratio is reduced, which makes Brillouin peak finding quite difficult.

[0005] Traditional peak-finding methods involve curve fitting of the Brillouin gain spectrum, typically using a Lorentz curve to simulate the light energy distribution. The correlation coefficient distribution is obtained by cross-correlation calculation between the fitted curve and the measured Brillouin gain spectrum. The peak position of the Brillouin gain spectrum is then determined by finding the point of maximum correlation. Based on this principle, the degree of distortion in the Brillouin gain spectrum affects the noise in the cross-correlation calculation results. Specifically, noise in the Brillouin gain spectrum increases the deviation between the maximum cross-correlation coefficient and the true Brillouin peak, thus affecting the accuracy of strain measurement by the sensor. Summary of the Invention

[0006] The technical problem solved by this invention is to overcome the shortcomings of the prior art. Based on the existing Brillouin gain spectrum peak finding method, the original Brillouin gain spectrum is decomposed into a signal, the signal noise is extracted and filtered out, and then the filtered decomposed signal is reconstructed to obtain a reconstructed Brillouin gain spectrum with only one inflection point. This reduces the impact of introduced noise on subsequent cross-correlation calculations. Then, through curve fitting and cross-correlation calculations, the position of the maximum value of the correlation coefficient is obtained, which is the peak position information of the Brillouin gain spectrum.

[0007] The objective of this invention is achieved through the following technical solution: a peak-finding method for Brillouin gain spectra, comprising:

[0008] The two-dimensional Brillouin curve S(f,t) of the "frequency-time" relationship of the sensing fiber is obtained by using a Brillouin optical time domain analyzer or a Brillouin optical time domain reflectometer.

[0009] Extracting the two-dimensional Brillouin curve S(f,t) of frequency-time, we obtain the Brillouin gain curve S acquired at time t. t (f);

[0010] For the Brillouin gain curve S t (f) Perform empirical mode decomposition to obtain the decomposed sub-functions IMF. t,i =[IMF t,1 ,…,IMF t,n ] and remaining items r t ;

[0011] For the decomposition subfunction IMF t,i The filtering process is performed to obtain the filter function IMF. t,i ′;

[0012] Through the filtering function IMF t,n The reconstructed curve R is obtained. t ;

[0013] For the reconstructed curve R t Perform Lorentz curve fitting to obtain the fitted curve F. t ;

[0014] The center wavelength f0 of the fitted curve is the peak wavelength, and S(f0,t) is the peak amplitude.

[0015] The Brillouin gain curve S t (f) The decomposition subfunction IMF after empirical mode decomposition t,i and remainder r t Represented as:

[0016]

[0017] The decomposition subfunction IMF t,i The filtering process is performed to obtain the filter function IMF. t,i ′, including: decomposition of subfunctions IMF t,n Perform filtering processing and set the filtering threshold Th. t,i Filtering function IMF t,i ′ is represented as:

[0018]

[0019] Through the filtering function IMF t,n The reconstructed curve R is obtained. t , represented as:

[0020]

[0021] For the reconstructed curve R t Perform Lorentz curve fitting to obtain the fitted curve F. t Represented as:

[0022]

[0023] Where A is the amplitude coefficient of the fitted curve, and σ is the width coefficient of the fitted curve.

[0024] A peak-finding system for Brillouin gain spectrum includes: an acquisition module, an extraction module, a decomposition module, a filtering module, a reconstruction module, and a fitting module;

[0025] The acquisition module uses a Brillouin optical time domain analyzer or a Brillouin optical time domain reflectometer to obtain the two-dimensional Brillouin curve S(f,t) of the "frequency-time" of the sensing fiber.

[0026] The extraction module extracts the two-dimensional Brillouin curve S(f,t) of frequency-time to obtain the Brillouin gain curve S acquired at time t. t (f);

[0027] The decomposition module affects the Brillouin gain curve S. t (f) Perform empirical mode decomposition to obtain the decomposed sub-functions IMF. t,i =[IMF t,1 ,…,IMF t,n ] and remaining items r t ;

[0028] The filtering module decomposes the sub-function IMF t,i The filtering process is performed to obtain the filter function IMF. t,i ′;

[0029] The reconstruction module uses the filtering function IMF t,n The reconstructed curve R is obtained. t ;

[0030] The fitting module for the reconstructed curve R t Perform Lorentz curve fitting to obtain the fitted curve F. t The center wavelength f0 of the fitted curve is the peak wavelength, and S(f0,t) is the peak amplitude.

[0031] The Brillouin gain curve S t (f) The decomposition subfunction IMF after empirical mode decompositiont,i and remainder r t Represented as:

[0032]

[0033] The decomposition subfunction IMF t,i The filtering process is performed to obtain the filter function IMF. t,i ′, including: decomposition of subfunctions IMF t,n Perform filtering processing and set the filtering threshold Th. t,i Filtering function IMF t,i ′ is represented as:

[0034]

[0035] Through the filtering function IMF t,n The reconstructed curve R is obtained. t , represented as:

[0036]

[0037] For the reconstructed curve R t Perform Lorentz curve fitting to obtain the fitted curve F. t Represented as:

[0038]

[0039] Where A is the amplitude coefficient of the fitted curve, and σ is the width coefficient of the fitted curve.

[0040] Compared with existing technologies, this invention has the following advantages: Existing technologies typically employ direct peak finding methods, i.e., without signal conditioning processes, which cannot effectively overcome the gain spectrum noise introduced by factors such as temperature affecting optoelectronic devices. In contrast, this invention uses a signal filtering method to filter and denoise the Brillouin gain spectrum signal, overcoming the relative intensity noise introduced by devices such as lasers and electro-optic modulators. Existing filtering techniques generally employ low-pass filtering, band-pass filtering, etc., which require converting the signal into a frequency domain signal. This leads to problems such as signal information distortion and poor filtering effect in the Brillouin gain spectrum, resulting in a decrease in peak finding accuracy. In contrast, this invention uses an empirical mode decomposition method to decompose the signal in the time domain, resulting in shorter signal processing time. This improves peak finding accuracy while minimizing the loss of true signal characteristics. Attached Figure Description

[0041] Figure 1 This is a flowchart of the peak-finding method for Brillouin gain spectrum provided in an embodiment of the present invention;

[0042] Figure 2 This is a schematic diagram comparing the original Brillouin gain spectrum and the reconstructed Brillouin gain spectrum provided in an embodiment of the present invention. Detailed Implementation

[0043] Exemplary embodiments of the present disclosure will now be described in more detail with reference to the accompanying drawings. While exemplary embodiments of the present disclosure are shown in the drawings, it should be understood that the present disclosure may be implemented in various forms and should not be limited to the embodiments set forth herein. Rather, these embodiments are provided to enable a more thorough understanding of the present disclosure and to fully convey the scope of the disclosure to those skilled in the art. It should be noted that, unless otherwise specified, the embodiments and features described herein can be combined with each other. The present invention will now be described in detail with reference to the accompanying drawings and embodiments.

[0044] Figure 1 This is a flowchart of a peak-finding method for Brillouin gain spectrum provided in an embodiment of the present invention. Figure 1 As shown, the peak finding method for Brillouin gain spectrum is characterized by including: acquiring the Brillouin scattering curve, extracting the Brillouin gain spectrum curve, decomposing the curve, filtering the decomposed signal, obtaining the reconstructed signal through the filtered signal, performing curve fitting on the reconstructed signal, obtaining the center frequency of the fitted curve and determining it as the peak frequency of the Brillouin gain spectrum.

[0045] Figure 2 This is a schematic diagram of the original Brillouin gain spectrum for the peak-finding method of Brillouin gain spectrum provided in an embodiment of the present invention. Figure 2 As shown, the scanning start frequency of the Brillouin scattering curve is 10,700 GHz, the scanning interval is 5 MHz, and the number of scanning points is 60.

[0046] This embodiment provides a peak-finding method for Brillouin gain spectrum, which includes the following steps:

[0047] Step 1: Use a Brillouin optical time domain analyzer or a Brillouin optical time domain reflectometer to obtain the two-dimensional Brillouin curve S(f,t) of the "frequency-time" of the sensing fiber.

[0048] Step 2: Extract the two-dimensional Brillouin curve S(f,t) of "frequency-time" to obtain the Brillouin gain curve S acquired at time t. t (f);

[0049] Step 3: Analyze the Brillouin gain curve S t (f) Perform empirical mode decomposition to obtain the decomposed sub-functions IMF. t,i =[IMF t,1 ,…,IMF t,n ] and remaining items r t ;

[0050]

[0051] Step 4: Set the filter threshold Th t,i For the decomposition subfunction IMF t,i The filtering process is performed to obtain the filter function IMF. t,i ′;

[0052]

[0053] Step 5: Apply the filter function IMF t,n The reconstructed curve R is obtained. t ;

[0054]

[0055] Step Six: Reconstruct the curve R t By performing curve fitting on the Lorentz curve and selecting the optimal amplitude coefficient A and width coefficient σ, the fitted curve F is obtained. t ;

[0056]

[0057] Step 7: The center wavelength f0 of the fitted curve is the peak wavelength, and S(f0,t) is the peak amplitude.

[0058] The present invention also relates to a peak finding system for Brillouin gain spectrum, comprising: an acquisition module, an extraction module, a decomposition module, a filtering module, a reconstruction module, and a fitting module;

[0059] The acquisition module uses a Brillouin optical time domain analyzer or a Brillouin optical time domain reflectometer to obtain the two-dimensional Brillouin curve S(f,t) of the "frequency-time" of the sensing fiber.

[0060] The extraction module extracts the two-dimensional Brillouin curve S(f,t) of frequency-time to obtain the Brillouin gain curve S acquired at time t. t (f);

[0061] The decomposition module affects the Brillouin gain curve S. t (f) Perform empirical mode decomposition to obtain the decomposed sub-functions IMF. t,i =[IMF t,1 ,…,IMF t,n ] and remaining items r t ;

[0062] The filtering module decomposes the sub-function IMF t,i The filtering process is performed to obtain the filter function IMF. t,i ′;

[0063] The reconstruction module uses the filtering function IMF t,n The reconstructed curve R is obtained. t ;

[0064] The fitting module for the reconstructed curve R t Perform Lorentz curve fitting to obtain the fitted curve F. t The center wavelength f0 of the fitted curve is the peak wavelength, and S(f0,t) is the peak amplitude.

[0065] This invention reduces temperature drift and relative intensity noise introduced by sensor devices by decomposing, filtering and reconstructing the original Brillouin gain spectrum curve, thereby improving the accuracy of curve fitting and cross-correlation calculation, and improving the accuracy of Brillouin gain spectrum peak finding.

[0066] Although the present invention has been disclosed above with reference to preferred embodiments, it is not intended to limit the present invention. Any person skilled in the art can make possible changes and modifications to the technical solutions of the present invention by utilizing the methods and techniques disclosed above without departing from the spirit and scope of the present invention. Therefore, any simple modifications, equivalent changes and alterations made to the above embodiments based on the technical essence of the present invention without departing from the content of the technical solutions of the present invention shall fall within the protection scope of the technical solutions of the present invention.

Claims

1. A peak-finding method for Brillouin gain spectra, characterized in that, include: Two-dimensional Brillouin curves of the "frequency-time" relationship of the sensing fiber are obtained using a Brillouin optical time-domain analyzer or a Brillouin optical time-domain reflectometer. S ( f , t ); Two-dimensional Brillouin curves of frequency-time S ( f , t Extract and obtain t Brillouin gain curves acquired at different times S t ( f ); Brillouin gain curve S t ( f Perform empirical mode decomposition to obtain decomposed sub-functions. IMF t,i =[ IMF t,1 ,…, IMF t,n ] and remaining items r t ; Decomposition of subfunctions IMF t,i Perform filtering to obtain the filter function. IMF t,i ′; Through the filtering function IMF t,i ' Obtain the reconstructed curve R t ; For the reconstructed curve R t Perform Lorentz curve fitting to obtain the fitted curve. F t ; Center wavelength of the fitted curve f 0 represents the peak wavelength. S ( f 0, t () represents the peak amplitude.

2. The peak-finding method for Brillouin gain spectrum according to claim 1, characterized in that: The Brillouin gain curve S t ( f The decomposition sub-functions after empirical mode decomposition IMF t,i and remainder r t Represented as: 。 3. The peak-finding method for Brillouin gain spectrum according to claim 1, characterized in that: The decomposition subfunction IMF t,i Perform filtering to obtain the filter function. IMF t,i ′, including: decomposition of subfunctions IMF t,i Perform filtering and set the filter threshold. Th t,i Filtering function IMF t,i ′ is represented as: 。 4. The peak-finding method for Brillouin gain spectrum according to claim 3, characterized in that: Through the filtering function IMF t,i ' Obtain the reconstructed curve R t , represented as: 。 5. The peak-finding method for Brillouin gain spectrum according to claim 4, characterized in that: For the reconstructed curve R t Perform Lorentz curve fitting to obtain the fitted curve. F t Represented as: Where A is the amplitude coefficient of the fitted curve, and σ is the width coefficient of the fitted curve.

6. A peak-finding system for Brillouin gain spectroscopy, characterized in that, include: The module includes: acquisition module, extraction module, decomposition module, filtering module, reconstruction module, and fitting module. The acquisition module uses a Brillouin optical time domain analyzer or a Brillouin optical time domain reflectometer to obtain the two-dimensional Brillouin curve of the sensing fiber in terms of frequency-time. S ( f , t ); The extraction module extracts two-dimensional Brillouin curves of frequency-time. S ( f , t Extract and obtain t Brillouin gain curves acquired at different times S t ( f ); Decomposition module for Brillouin gain curve S t ( f Perform empirical mode decomposition to obtain decomposed sub-functions. IMF t,i =[ IMF t,1 ,…, IMF t,n ] and remaining items r t ; The filtering module decomposes the sub-functions. IMF t,i Perform filtering to obtain the filter function. IMF t,i ′; The reconstruction module uses a filtering function. IMF t,i ' Obtain the reconstructed curve R t ; The fitting module for the reconstructed curve R t Perform Lorentz curve fitting to obtain the fitted curve. F t The center wavelength of the fitted curve f 0 represents the peak wavelength. S ( f 0, t () represents the peak amplitude.

7. The peak-finding system for Brillouin gain spectrum according to claim 6, characterized in that: The Brillouin gain curve S t ( f The decomposition sub-functions after empirical mode decomposition IMF t,i and remainder r t Represented as: 。 8. The peak-finding system for Brillouin gain spectrum according to claim 6, characterized in that: The decomposition subfunction IMF t,i Perform filtering to obtain the filter function. IMF t,i ′, including: decomposition of subfunctions IMF t,i Perform filtering and set the filter threshold. Th t,i Filtering function IMF t,i ′ is represented as: 。 9. The peak-finding system for Brillouin gain spectrum according to claim 8, characterized in that: Through the filtering function IMF t,i ' Obtain the reconstructed curve R t , represented as: 。 10. The peak-finding system for Brillouin gain spectrum according to claim 9, characterized in that: For the reconstructed curve R t Perform Lorentz curve fitting to obtain the fitted curve. F t Represented as: ； Where A is the amplitude coefficient of the fitted curve, and σ is the width coefficient of the fitted curve.

Images