A method for extracting the central frequency of a Brillouin scattering spectrum and an optical fiber strain sensing system

Abstract

The application provides a Brillouin scattering spectrum center frequency extraction method and an optical fiber strain sensing system. The Brillouin scattering spectrum center frequency extraction method comprises the following steps: acquiring discrete Brillouin scattering spectrum data; performing interpolation on the Brillouin scattering spectrum data to obtain incremental Brillouin scattering spectrum data; performing denoising processing on the incremental Brillouin scattering spectrum data; performing weighted summation on the frequencies of all data points in the incremental Brillouin scattering spectrum data with the spectral density as a weighted weight to obtain a frequency weighted sum; and acquiring the center frequency of the Brillouin scattering spectrum data based on the weight sum of the weighted weight and the frequency weighted sum. The above scheme removes the interference of noise points on the weighted calculation through the combination of interpolation, denoising and spectral density weighting, and improves the stability and accuracy of the center frequency extraction.

Description

[0001] This application is a divisional application of the patent application filed on July 21, 2025, with application number 202511003191.1 and entitled "A method for extracting the center frequency of Brillouin scattering spectrum and an optical fiber strain sensing system". Technical Field

[0002] This application relates to the field of fiber optic sensing technology, and more specifically, to a method for extracting the center frequency of Brillouin scattering spectrum and a fiber optic strain sensing system. Background Technology

[0003] Stimulated Brillouin Scattering (SBS) is a nonlinear process that occurs in optical fiber transmission. When the pump light power input to the fiber reaches the SBS threshold, the density and dielectric constant of the fiber medium undergo periodic changes due to the electrostriction effect, forming an acoustic field. When the pump light is incident on this shifted acoustic field, it produces back-propagating scattered light, i.e., Stokes light. The frequency difference between the Stokes light and the pump light is called the Brillouin frequency shift. The amount of the Brillouin frequency shift is determined not only by the material properties of the fiber medium itself but also by the temperature and stress at which the fiber is exposed.

[0004] Currently, most related technologies employ fitting algorithms (such as single-peak fitting algorithms) to obtain the center frequency of the Brillouin spectrum, thereby calculating the Brillouin frequency shift. However, in practical applications, noise is superimposed on the Brillouin gain spectrum data, and the center frequency extraction schemes in related technologies have poor noise robustness. Slight fluctuations in the data values ​​near the peak of the obtained Brillouin spectrum can lead to drastic changes in the center frequency fitting results, ultimately resulting in a deterioration in the fitting stability. Summary of the Invention

[0005] The purpose of this application is to provide a method for extracting the center frequency of Brillouin scattering spectrum and an optical fiber strain sensing system to solve the above-mentioned technical problems.

[0006] In a first aspect, embodiments of this application provide a method for extracting the center frequency of a Brillouin scattering spectrum, comprising: acquiring discrete Brillouin scattering spectrum data; interpolating the Brillouin scattering spectrum data to obtain incremental Brillouin scattering spectrum data; performing noise reduction processing on the incremental Brillouin scattering spectrum data; using the spectral density as a weighting weight, performing a weighted summation of the frequencies of all data points in the incremental Brillouin scattering spectrum data to obtain a frequency-weighted sum; and obtaining the center frequency of the Brillouin scattering spectrum data based on the weighted summation of the weighting weights and the frequency-weighted summation.

[0007] In the implementation of the above scheme, interpolation can increase the resolution of the data, making the details of the Brillouin scattering spectrum richer, which helps to more accurately locate spectral peaks, reduce frequency estimation errors caused by insufficient data points, and improve the accuracy of center frequency extraction. On the other hand, denoising can effectively remove noise points in low signal-to-noise ratio regions, reduce the influence of outliers and interference signals, and make the data purer, thus providing a more reliable basis for subsequent weighted calculations and improving the accuracy of center frequency extraction. Furthermore, using spectral density as the weighting factor can assign higher weights to points with larger amplitudes, making the calculation of the center frequency focus more on the main part of the spectrum. In the case of bimodal or multimodal conditions, even if multiple peaks exist, the calculation will be more accurate. Spectral density weighting ensures that the center frequency calculation is more focused on the location of the main spectral peaks, thus improving the stability and accuracy of center frequency extraction. Furthermore, in the denoised data, the characteristics of the effective signal are more prominent, and the amplitude distribution is more realistic. When using spectral density as the weighting factor for frequency weighted summation, it can more accurately reflect the contribution of the main spectral peaks, avoid noise interference with the results, reduce the impact of random fluctuations and outliers, and make the weighted center frequency more stable, more repeatable, and more reliable. In addition, the calculation processes for interpolation, denoising, and spectral density weighting are relatively simple, computationally intensive, and do not rely on complex mathematical models or assumptions, making them suitable for real-time processing and large-scale data processing.

[0008] In one implementation of the first aspect, the denoising process for the incremental Brillouin scattering spectrum data includes: obtaining the maximum amplitude of the Brillouin scattering spectrum data; obtaining a point selection threshold based on the maximum amplitude and a preset scaling factor; and performing denoising processing on the incremental Brillouin scattering spectrum data based on the point selection threshold.

[0009] In the implementation of the above scheme, the maximum amplitude reflects the strongest signal in the dataset. When combined with the preset scaling factor, the selection threshold can effectively distinguish between signal and noise, remove noise points in low signal-to-noise ratio areas, improve data quality, and thus improve the accuracy of center frequency extraction. On the other hand, the linear operation based on the maximum amplitude and the preset scaling factor is easy to implement, requires low computing resources, does not rely on complex mathematical models or assumptions, and is suitable for engineering and real-time applications.

[0010] In one implementation of the first aspect, the denoising process for the incremental Brillouin scattering spectrum data includes: obtaining the maximum amplitude, a first amplitude corresponding to the minimum frequency point, and a second amplitude corresponding to the maximum frequency point of the Brillouin scattering spectrum data; obtaining a selection threshold based on the first amplitude, the second amplitude, the maximum amplitude, and a preset scaling factor; wherein the selection threshold is the maximum value among the products of the first amplitude, the second amplitude, the maximum amplitude, and the preset scaling factor; and performing denoising processing on the incremental Brillouin scattering spectrum data based on the selection threshold.

[0011] In implementing the above scheme, the combination of maximum amplitude, minimum frequency amplitude, and maximum frequency amplitude can more comprehensively reflect the characteristics of Brillouin scattering spectrum data, making the selected threshold more accurately reflect the true situation of the data, thereby more effectively removing noise points. In complex noise environments, the above scheme can better adapt to different noise distributions and interference conditions, improving the robustness of denoising. On the other hand, by considering the amplitudes of the minimum and maximum frequency points, misjudgments caused by abnormal amplitudes of a single frequency point can be avoided, ensuring that effective data points at the beginning and end are not mistakenly filtered out, and more meticulously preserving key details in the signal, especially for those spectral peaks with important characteristics at the beginning or end, thus improving the integrity of the data.

[0012] In one implementation of the first aspect, interpolating the Brillouin scattering spectrum data to obtain incremental Brillouin scattering spectrum data includes: Linear interpolation is performed on the Brillouin scattering spectrum data to obtain incremental Brillouin scattering spectrum data.

[0013] In the implementation of the above scheme, by uniformly or non-uniformly inserting new data points during linear interpolation, the data resolution is improved. Higher data resolution helps to more accurately depict the shape of the Brillouin scattering spectrum, especially for the more accurate characterization of features such as the position, width, and intensity of the spectral peaks, which is beneficial to improving the accuracy of center frequency extraction. On the other hand, linear interpolation is easier to implement, which is beneficial to improving the center frequency extraction efficiency of the above Brillouin scattering spectrum center frequency extraction method.

[0014] In one implementation of the first aspect, after obtaining the center frequency of the Brillouin scattering spectrum data, the method further includes: calculating the Brillouin frequency shift of the Brillouin scattering spectrum data based on the center frequency of the Brillouin scattering spectrum data.

[0015] In the implementation of the above scheme, by accurately extracting the center frequency, the Brillouin frequency shift can be calculated more accurately, which is beneficial to improving the calculation accuracy of the Brillouin frequency. On the other hand, the calculated Brillouin frequency shift can be used in scenarios such as temperature measurement, strain measurement, and fiber characteristic analysis, which is beneficial to improving the adaptability of the above Brillouin scattering spectrum center frequency extraction method.

[0016] Secondly, embodiments of this application provide an optical fiber strain sensing method, comprising: acquiring discrete Brillouin scattering spectrum data using a sensing optical fiber; wherein the sensing optical fiber is installed in the structure to be monitored; Extract the center frequency from the Brillouin scattering spectrum data; wherein the center frequency is obtained using the method provided by the first aspect or any possible implementation thereof; Based on the center frequency, strain monitoring is performed on the structure to be monitored.

[0017] Thirdly, embodiments of this application provide an optical fiber strain sensing system, comprising: a host computer, a photodetector, and a sensing optical fiber, wherein the photodetector is electrically connected to the host computer, and the sensing optical fiber is connected to the photodetector, wherein: The host computer is used to acquire discrete Brillouin scattering spectrum data; interpolate the Brillouin scattering spectrum data to obtain incremental Brillouin scattering spectrum data; denoise the incremental Brillouin scattering spectrum data; acquire the spectral density of each data point in the denoised incremental Brillouin scattering spectrum data; use the spectral density as a weight to perform a weighted sum of the frequencies of all data points in the incremental Brillouin scattering spectrum data to obtain a frequency-weighted sum; and acquire the center frequency of the Brillouin scattering spectrum data based on the total weight of the weighted weights and the frequency-weighted sum.

[0018] Fourthly, embodiments of this application provide an electronic device, including: a processor, a memory, and a communication bus, wherein the processor and the memory communicate with each other through the communication bus; the memory stores computer program instructions that can be executed by the processor, and the computer program instructions are read and executed by the processor to perform the method provided by the first aspect or any possible implementation of the first aspect or the second aspect or any possible implementation of the second aspect.

[0019] Fifthly, embodiments of this application provide a computer-readable storage medium storing computer program instructions, which, when read and executed by a processor, perform the method provided by the first aspect or any possible implementation thereof, or the second aspect or any possible implementation thereof.

[0020] Other features and advantages of this application will be set forth in the following description and will be apparent in part from the description or may be learned by practicing embodiments of this application. The objectives and other advantages of this application may be realized and obtained by means of the structures particularly pointed out in the written description, claims, and drawings. Attached Figure Description

[0021] To more clearly illustrate the technical solutions of the embodiments of this application, the accompanying drawings used in the embodiments of this application will be briefly introduced below. It should be understood that the following drawings only show some embodiments of this application and should not be regarded as a limitation of the scope. For those skilled in the art, other related drawings can be obtained based on these drawings without creative effort.

[0022] Figure 1 A flowchart illustrating the method for extracting the center frequency of Brillouin scattering spectrum provided in this application embodiment; Figure 2 A comparison chart of the fitting accuracy of the single-peak Lorentz spectrum under different noise levels in the comparative scenario one provided for the embodiments of this application, using the single-peak Lorentz fitting scheme and the linear interpolation-spectral density weighted extraction scheme; Figure 3 A comparison chart of the fitting accuracy of the single-peak Lorentz spectrum under different noise levels in the comparative scenario two provided for the embodiments of this application, using the single-peak Lorentz fitting scheme and the linear interpolation-spectral density weighted extraction scheme; Figure 4 A schematic diagram of the fitting result of fitting single-peak data using a single-peak Lorentz fitting scheme in a comparative scenario provided in the embodiments of this application; Figure 5 A schematic diagram of the fitting result of using a single-peak Lorentz fitting scheme to fit multi-peak data in the second comparative scenario provided in the embodiments of this application; Figure 6 This is a schematic diagram of the structure of an electronic device provided in an embodiment of this application. Detailed Implementation

[0023] The technical solutions of the embodiments of this application will now be described with reference to the accompanying drawings. The following embodiments are only used to more clearly illustrate the technical solutions of this application, and are therefore merely examples and should not be used to limit the scope of protection of this application.

[0024] Unless otherwise defined, all technical and scientific terms used herein have the same meaning as commonly understood by one of ordinary skill in the art to which this application pertains; the terminology used herein is for the purpose of describing particular embodiments only and is not intended to limit the application; the terms “comprising” and “having”, and any variations thereof, in the specification, claims, and foregoing description of the drawings are intended to cover non-exclusive inclusion.

[0025] In the description of the embodiments of this application, technical terms such as "first" and "second" are used only to distinguish different objects and should not be construed as indicating or implying relative importance or implicitly specifying the number, specific order, or primary and secondary relationship of the indicated technical features. In the description of the embodiments of this application, "multiple" means two or more, unless otherwise explicitly defined.

[0026] In this document, the term "embodiment" means that a particular feature, structure, or characteristic described in connection with an embodiment may be included in at least one embodiment of this application. The appearance of this phrase in various places throughout the specification does not necessarily refer to the same embodiment, nor is it a separate or alternative embodiment mutually exclusive with other embodiments. It will be explicitly and implicitly understood by those skilled in the art that the embodiments described herein can be combined with other embodiments.

[0027] In the description of the embodiments in this application, the term "and / or" is merely a description of the relationship between related objects, indicating that three relationships can exist. For example, A and / or B can represent: A existing alone, A and B existing simultaneously, or B existing alone. Additionally, the character " / " in this document generally indicates that the preceding and following related objects have an "or" relationship.

[0028] Currently, the noise in Brillouin spectrum data mainly comes from the photoelectric conversion noise of the photodetector and random environmental noise. These noises are random and will be superimposed on the real Brillouin spectrum signal, causing the measured signal to deviate from the true value. The effects of noise on the shape of the Brillouin spectrum include: (1) amplitude fluctuation: The presence of noise will cause the amplitude of the Brillouin spectrum to fluctuate randomly at different frequency points. (2) spectral line broadening or distortion: Noise may broaden the spectral lines of the Brillouin spectrum or distort their shape, thereby masking the true spectral line shape.

[0029] In the presence of noise, the fitting results obtained by the fitting algorithm may fluctuate with changes in noise, resulting in unstable fitting results.

[0030] Based on this, embodiments of this application provide a method for extracting the center frequency of a Brillouin scattering spectrum, which extracts the center frequency by combining interpolation, denoising, and spectral density weighting. Interpolation increases data resolution, enriching the details of the Brillouin scattering spectrum, which helps to more accurately locate spectral peaks, reduce frequency estimation errors caused by insufficient data points, and improve the accuracy of center frequency extraction. Denoising effectively removes noise points in low signal-to-noise ratio regions, reduces the influence of outliers and interference signals, and makes the data purer, thus providing a more reliable basis for subsequent weighted calculations. Using spectral density as the weighting factor assigns higher weight to points with larger amplitudes, making the center frequency calculation focus more on the main part of the spectrum. In the case of bimodal or multimodal data, it also ensures that the center frequency calculation is more inclined towards the position of the main spectral peaks. On the other hand, in the denoised data, the characteristics of the effective signal are more prominent, and the amplitude distribution is more realistic. When performing frequency weighted summation with spectral density as the weighting factor, it can more accurately reflect the contribution of the main spectral peaks, avoid the interference of noise on the results, and make the center frequency obtained by weighted calculation more stable and more repeatable. Furthermore, the calculation process of interpolation, denoising, and spectral density weighting is relatively simple, with low computational load, does not depend on specific mathematical models, and is suitable for real-time processing and large-scale data processing.

[0031] Please see Figure 1 The illustrated flowchart shows a method for extracting the center frequency of a Brillouin scattering spectrum provided in an embodiment of this application. This method can be applied to electronic devices, which may include physical devices such as servers, PCs, tablets, or smartphones, or virtual devices such as virtual machines or containers. The electronic device can be a single device, a combination of multiple devices, or a cluster of a large number of devices. The aforementioned method for extracting the center frequency of a Brillouin scattering spectrum may include: Step S110: Obtain discrete Brillouin scattering spectrum data.

[0032] For example, step S110 above can be obtained using Brillouin optical time domain analysis (BOTDA) or Brillouin optical time domain reflectometer (BOTDR), specifically: The main steps for obtaining Brillouin scattering spectrum data using Brillouin Optical Time-Domain Analysis (BOTDA) can include: (1) Pump and probe light injection: First, pump light and probe light are injected into the optical fiber simultaneously. The pump light is usually generated by a high-power laser to generate an acoustic field in the optical fiber; the probe light has a slightly lower frequency than the pump light and is used to detect Brillouin gain spectrum information. (2) Acoustic field generation and interaction: When the pump light propagates in the optical fiber, it will cause periodic changes in the density and dielectric constant of the optical fiber medium under the action of electrostriction, forming an acoustic field. The probe light interacts with this acoustic field, and when the frequency of the probe light matches the Brillouin frequency shift generated by the acoustic field, the probe light will obtain Brillouin gain. (3) Photodetector signal detection: After propagation through a section of optical fiber, the returned probe light signal is detected by a photodetector. At this time, the intensity and frequency characteristics of the probe light have been affected by the Brillouin scattering effect and carry Brillouin scattering spectrum information along the optical fiber. (4) Data acquisition and processing: The electrical signals detected by the photodetector are acquired and converted into digital signals by analog-to-digital converter (ADC). Then, the acquired data are analyzed and processed using signal processing algorithms to extract discrete Brillouin scattering spectrum data.

[0033] The main steps for acquiring Brillouin scattering spectrum data using a Brillouin optical time-domain reflectometer (BOTDR) can include: (1) Pulsed light injection: A pulsed light signal is emitted into the optical fiber, which is usually generated by a short-pulse laser. (2) Generation of Brillouin scattered light: When the pulsed light propagates in the optical fiber, Brillouin scattered light is generated. The frequency of these scattered lights has a Brillouin frequency shift compared to the frequency of the incident light, and the Brillouin frequency shift is related to the temperature and strain of the optical fiber. (3) Detection of scattered light by photodetector: The returned Brillouin scattered light signal is detected using a photodetector. Since the intensity of the scattered light is relatively weak, a high-sensitivity photodetector is usually required to detect these signals. (4) Data acquisition and processing: The electrical signals detected by the photodetector are also acquired and converted from analog to digital, and discrete Brillouin scattering spectrum data is extracted through signal processing algorithms.

[0034] Step S120: Interpolate the Brillouin scattering spectrum data to obtain incremental Brillouin scattering spectrum data.

[0035] Optionally, step S120 above may include: performing linear interpolation on the Brillouin scattering spectrum data to obtain incremental Brillouin scattering spectrum data.

[0036] Understandably, in actual measurements, due to limitations in the sampling interval, some subtle spectral changes may be missed. Therefore, interpolation can be performed on the Brillouin scattering spectrum data to improve data resolution and thus enhance the accuracy of center frequency extraction. Step S120 above can employ uniform linear interpolation or non-uniform linear interpolation to interpolate the Brillouin scattering spectrum data, wherein: (1) Uniform interpolation refers to the fact that the interval between interpolation points is equal within the interpolation interval, and the data points after interpolation are evenly distributed. This facilitates frequency domain analysis and the use of certain algorithms that require data uniformity, and is relatively simple to implement. For example, three new data points are evenly inserted between two adjacent data points in the discrete Brillouin scattering spectrum data mentioned above.

[0037] (2) Non-uniform interpolation allows for unequal intervals between interpolation points within the interpolation interval, and the distribution of interpolation points can be adjusted according to the variation characteristics of the data. Non-uniform interpolation can better adapt to the variation characteristics of the data, inserting more interpolation points in areas of rapid data change to capture details, and inserting fewer interpolation points in areas of gradual data change to save computational resources. For example, generally, the discrete Brillouin scattering spectrum data mentioned above changes slowly and has larger intervals in the low-frequency range, while it changes drastically and has smaller intervals in the high-frequency range. Therefore, when performing interpolation, fewer data points can be inserted in the low-frequency range, while more data points can be inserted in the high-frequency range.

[0038] In practical applications, the choice between uniform and non-uniform interpolation can be made based on data characteristics and specific needs. For example, if the data sampling intervals are relatively uniform, or if subsequent analysis requires regular data, uniform interpolation is usually chosen. If the data variation characteristics are non-uniform, or if it is necessary to capture details in certain areas, non-uniform interpolation can be considered.

[0039] The above scheme improves the data resolution by uniformly or non-uniformly inserting new data points during linear interpolation. Higher data resolution helps to more accurately depict the shape of Brillouin scattering spectral lines, especially for the more accurate characterization of features such as the position, width, and intensity of spectral peaks, thus improving the accuracy of center frequency extraction. On the other hand, linear interpolation is easier to implement, which helps to improve the center frequency extraction efficiency of the above Brillouin scattering spectrum center frequency extraction method.

[0040] It is understandable that noise points in the Brillouin scattering spectrum are typically random and irregular, and their amplitudes may be close to or even higher than those of the effective signal points. During weighted calculations, these noise points are assigned certain weights, thus affecting the weighting results and consequently the accuracy of center frequency extraction. Based on this, the embodiments of this application provide the following solution: Before step S130 above, the above method for extracting the center frequency of the Brillouin scattering spectrum may further include: before using weighted summation of the frequencies of all data points in the incremental Brillouin scattering spectrum data, the method further includes: denoising the incremental Brillouin scattering spectrum data; using weighted summation of the frequencies of all data points in the incremental Brillouin scattering spectrum data includes: using weighted summation of the frequencies of all data points in the denoised incremental Brillouin scattering spectrum data.

[0041] The denoising process described above effectively removes noise points in low signal-to-noise ratio regions, reduces the influence of outliers and interference signals, and makes the data cleaner. This provides a more reliable basis for subsequent weighted calculations, which is beneficial for improving the data quality of incremental Brillouin scattering spectrum data and thus improving the accuracy of center frequency extraction. On the other hand, in the denoised data, the characteristics of the effective signal are more prominent, and the amplitude distribution is more realistic. When using weighted summation for frequency, it can more accurately reflect the contribution of the main spectral peaks, avoid the interference of noise on the results, and thus improve the accuracy of center frequency calculation. Furthermore, the denoised data reduces the influence of random fluctuations and outliers, making the center frequency obtained by weighted calculation more stable, more repeatable, and more reliable.

[0042] Step S130: Obtain the spectral density of each data point in the denoised incremental Brillouin scattering spectrum data; use the spectral density as the weighting weight to perform a weighted summation of the frequencies of all data points in the incremental Brillouin scattering spectrum data to obtain the frequency weighted sum.

[0043] The spectral density mentioned above refers to the amplitude spectral density of the Brillouin scattering spectrum, that is, the distribution of the square of the signal amplitude as a function of frequency. Spectral density reflects the energy distribution of the Brillouin scattering spectrum at different frequency points; points with larger amplitudes have higher spectral densities, typically corresponding to more prominent spectral features, such as spectral peaks. The method for calculating spectral density is as follows: For discrete Brillouin scattering spectrum data ,in It is a frequency point. This corresponds to the amplitude. Spectral density can be defined as the square of the amplitude, i.e. .

[0044] The above scheme uses a combination of interpolation and spectral density weighting to extract the center frequency of the Brillouin scattering spectrum. Spectral density weighting assigns higher weights to points with larger amplitudes, making the center frequency calculation more focused on the main part of the spectrum. In the case of bimodal or multimodal scattering, even with multiple peaks, spectral density weighting ensures that the center frequency calculation is more inclined towards the position of the main spectral peak, thereby improving the stability and accuracy of the center frequency extraction. On the other hand, compared with the fitting method used in related techniques to obtain the center frequency, the above scheme can quickly obtain the center frequency through simple interpolation and spectral density weighting calculation, making it suitable for real-time processing and large-scale data processing.

[0045] The above scheme can employ at least the following denoising methods to denoise the incremental Brillouin scattering spectrum data: The first noise reduction method: noise reduction based on filtering techniques; Denoising of incremental Brillouin scattering spectrum data can be performed using linear or nonlinear filtering methods, where: Linear filtering includes methods such as moving average filtering and Gaussian filtering. Moving average filtering smooths data by calculating the average of adjacent data points, making it suitable for smoothing processes and reducing random noise. Gaussian filtering, on the other hand, uses a weighted average based on the shape of a Gaussian function to effectively remove Gaussian noise while preserving as much of the signal's main characteristics as possible.

[0046] Nonlinear filtering includes median filtering and bilateral filtering. Median filtering removes noise by replacing data points with the median of adjacent data points, and it is effective at removing outliers such as impulse noise. Bilateral filtering considers both the neighborhood information of the data points and the similarity of their gray values, effectively removing noise while preserving important features such as signal edges.

[0047] The second denoising method: model-based denoising; Denoising of incremental Brillouin scattering spectrum data can be performed using autoregressive models or wavelet transform models, where: Autoregressive (AR) model: This model assumes that a signal can be represented as a linear combination of signals from several past moments plus a white noise term. By fitting the model parameters, the original signal can be estimated from a noisy signal, making it suitable for denoising signals with autocorrelation.

[0048] Wavelet transform: Wavelet transform possesses excellent time-frequency localization properties, allowing signals to be decomposed into different scales and locations for analysis. Denoising is achieved by processing the wavelet coefficients using an appropriate threshold function. This method effectively removes noise while preserving important signal features such as abrupt changes and spikes, making it particularly suitable for denoising non-stationary signals.

[0049] The third denoising method: Denoising based on machine learning and deep learning; Denoising autoencoders (DAEs) or generative adversarial networks (GANs) can be used to denoise incremental Brillouin scattering spectral data, where: Denoising autoencoders (DAEs) are neural network-based denoising methods. They encode and decode input data by constructing an autoencoder network, introducing noise during the encoding process. This allows the network to learn how to reconstruct the original data from the noisy data, thus achieving denoising. This method performs well on complex datasets, but requires a large amount of training data and computational resources.

[0050] Generative Adversarial Networks (GANs): Through adversarial training between a generator and a discriminator, the generator learns to generate realistic signals, while the discriminator learns to distinguish between real and generated signals. During training, the generator gradually learns to generate clean signals from noisy data, thus achieving denoising. GANs have achieved significant results in areas such as image denoising, but their application in Brillouin scattering spectrum denoising is still in the exploratory stage.

[0051] The fourth method: using a selected threshold for noise reduction; Optionally, the above-mentioned denoising process for the incremental Brillouin scattering spectrum data includes: obtaining the maximum amplitude of the Brillouin scattering spectrum data; obtaining a selection threshold based on the maximum amplitude and a preset scaling factor; and denoising the incremental Brillouin scattering spectrum data based on the selection threshold.

[0052] For example, the aforementioned preset scaling factor can be an empirical choice or determined based on the noise statistical characteristics in fiber optic sensing. For instance, choosing 0.2 can effectively filter out most noise points while retaining the main spectral peak data without being overly stringent. Alternatively, a method for denoising incremental Brillouin scattering spectrum data based on a selected point threshold can be as follows: remove data points with amplitudes not exceeding the selected point threshold from the incremental Brillouin scattering spectrum data, and retain data points with amplitudes exceeding the selected point threshold.

[0053] The maximum amplitude value in the above scheme reflects the strongest signal in the dataset. When combined with the preset scaling factor, the selection threshold can effectively distinguish between signal and noise, remove noise points in low signal-to-noise ratio areas, improve data quality, and thus improve the accuracy of center frequency extraction. On the other hand, the linear operation based on the maximum amplitude value and the preset scaling factor is easy to implement, requires low computing resources, does not rely on complex mathematical models or assumptions, and is suitable for engineering and real-time applications.

[0054] Optionally, the above-mentioned denoising process for the incremental Brillouin scattering spectrum data includes: obtaining the maximum amplitude, the first amplitude corresponding to the minimum frequency point, and the second amplitude corresponding to the maximum frequency point of the Brillouin scattering spectrum data; obtaining a selection threshold based on the first amplitude, the second amplitude, the maximum amplitude, and a preset scaling factor; wherein the selection threshold is the maximum value among the products of the first amplitude, the second amplitude, the maximum amplitude, and the preset scaling factor; and performing denoising processing on the incremental Brillouin scattering spectrum data based on the selection threshold. For example, this implementation involves calculating the measured discrete Brillouin scattering spectrum data. The maximum value is denoted as Calculate the discrete Brillouin scattering spectrum data corresponding to the minimum and maximum frequency points. The amplitudes are denoted as follows: and ;calculate The maximum value in the range is used to obtain the selection threshold, denoted as . .

[0055] The above will be introduced separately below. , and Its role in determining the point selection threshold: (1) : This is the maximum amplitude value in the entire Brillouin gain spectrum, representing the strongest part of the signal. By taking 0.2 times it, a threshold related to the overall signal strength can be obtained, ensuring that the selected data points have a certain degree of significance overall.

[0056] (2) Amplitude corresponding to the minimum frequency point The amplitude at the minimum frequency point in the Brillouin gain spectrum. This reflects the signal strength at the beginning of the spectrum. In some cases, there may be effective spectral peaks or important signal features at the beginning. By... As one of the references for the selection threshold, it can avoid misjudging the data points at the beginning due to noise or other abnormal conditions, which may cause the amplitude of the data points to be abnormally low or high, thereby ensuring that the valid data points at the beginning can be correctly retained.

[0057] (3) Amplitude corresponding to the maximum frequency point Similarly, the amplitude at the maximum frequency point This reflects the signal strength at the end of the spectrum. The end of the spectrum may also contain valid spectral peaks or important signal features. As one of the references for the selection threshold, it can avoid misjudgment due to abnormally low or high amplitude of data points at the end of the spectrum caused by noise or other abnormalities, and ensure that the effective data points at the end can be correctly retained.

[0058] In addition, the above-mentioned point selection threshold It should have at least the following functions: (1) Filter out noise points (such as baseline fluctuations) in areas with low signal-to-noise ratio: Noise points in low signal-to-noise ratio regions typically have small amplitudes, often below the selected point threshold. By filtering out values ​​with amplitude greater than The data points can effectively remove the interference of noise points, improve the purity of the data, and make the subsequent center frequency extraction more accurate.

[0059] (2) Retain all spectral peaks that are significantly higher than the noise level (regardless of the number of peaks): Significant spectral peaks typically have high amplitudes, which often exceed the point selection threshold. Regardless of whether the spectral peaks are single-peaked, double-peaked, or multi-peaked, as long as their amplitudes are significantly higher than the noise, they can be preserved. This ensures that the dataset for subsequent processing contains all the important spectral peak information, providing a guarantee for accurate extraction of the center frequency.

[0060] The above scheme combines the maximum amplitude, minimum frequency amplitude, and maximum frequency amplitude, which can more comprehensively reflect the characteristics of Brillouin scattering spectrum data, making the selected threshold more accurately reflect the true situation of the data, thereby more effectively removing noise points. In complex noise environments, the above scheme can better adapt to different noise distributions and interference conditions, improving the robustness of denoising. On the other hand, by considering the amplitudes of the minimum and maximum frequencies, misjudgments caused by abnormal amplitudes of a single frequency point can be avoided, ensuring that effective data points at the beginning and end are not mistakenly filtered out, and more meticulously preserving key details in the signal, especially for those spectral peaks with important characteristics at the beginning or end, thus improving the integrity of the data.

[0061] Step S140: Obtain the center frequency of the Brillouin scattering spectrum data based on the weighted sum of the weighted weights and the frequency weighted sum.

[0062] Optionally, after step S140 above, the above Brillouin scattering spectrum center frequency extraction method may further include: calculating the Brillouin frequency shift of the Brillouin scattering spectrum data based on the center frequency of the Brillouin scattering spectrum data.

[0063] The Brillouin shift mentioned above refers to the frequency difference between the scattered light (Stokes light) and the incident pump light during stimulated Brillouin scattering. The Brillouin shift is a key parameter in fiber optic sensing technologies such as Brillouin Optical Time Domain Analysis (BOTDA) and Brillouin Optical Time Domain Reflectometer (BOTDR). These technologies utilize the linear relationship between the Brillouin shift and temperature and strain, and by analyzing the Brillouin gain spectrum curve along the fiber, they obtain temperature and strain information at various locations along the fiber, enabling health monitoring of large infrastructure such as communication cables, buildings, oil pipelines, power lines, and railways.

[0064] The above scheme can more accurately calculate the Brillouin frequency shift by precisely extracting the center frequency, which is beneficial to improving the calculation accuracy of the Brillouin frequency. On the other hand, the calculated Brillouin frequency shift can be used in scenarios such as temperature measurement, strain measurement, and fiber characteristic analysis, which is beneficial to improving the adaptability of the above-mentioned Brillouin scattering spectrum center frequency extraction method.

[0065] The following provides the specific application steps of the above Brillouin scattering spectrum center frequency extraction method in a certain application scenario. In this application scenario, interpolation, denoising processing based on point selection thresholds, and spectral density weighted calculation are performed sequentially: Step 1: Obtain the measured discrete Brillouin scattering spectrum data ,in , The number of scan points. The initial frequency is and the sweep step size is . ; Step 2: Calculate the measured discrete Brillouin scattering spectrum data The maximum value is denoted as Calculate the discrete Brillouin scattering spectrum data corresponding to the minimum and maximum frequency points. The amplitudes are denoted as follows: and ; Step 3: Calculation The maximum value in the range is used to obtain the selection threshold, denoted as . ; Selecting a threshold point can effectively remove the influence of noise and non-primary spectral peaks. In the case of bimodal or multimodal peaks, it can highlight the characteristics of the primary spectral peaks, allowing subsequent center frequency calculations to focus more on the main part of the true Brillouin gain spectrum. For example, in a bimodal spectrum, if one peak has a low amplitude and may be affected by noise, setting an appropriate threshold point can help. This allows us to exclude or reduce its influence, thereby focusing on the dominant spectral peak; Step 4: Convert the discrete Brillouin scattering spectrum data Following the principle of linear interpolation, three new data points are uniformly inserted into adjacent frequency sweep points to obtain the linearly interpolated Brillouin scattering spectrum data. The new step size is The total number of scan points is Assume the frequency and amplitude of adjacent data points are... and The frequency and amplitude of the three newly added data points based on linear interpolation are... , , ; By uniformly inserting three new data points into adjacent frequency sweep points through linear interpolation, the originally discrete Brillouin scattering spectrum data becomes more refined, increasing the number of data points and thus improving the data resolution. When dealing with bimodal or multimodal spectra, it can more precisely depict the shape of the spectral lines, avoiding the omission of key spectral features due to insufficient data points, and providing richer information for subsequent center frequency extraction. Step 5: The linearly interpolated Brillouin scattering spectrum data Medium amplitude greater than Select data points and build a new dataset Simultaneously calculate the number of data points in the dataset, denoted as . ; Step Six: Process the selected dataset The center frequency was extracted by performing a spectral density weighted calculation.

[0066]

[0067] in, The center frequency; The number of data points; For frequency point spectral density; For frequency point The amplitude.

[0068] The weighted average frequency is calculated using spectral density (the square of the amplitude) as the weight, assigning higher weights to points with larger amplitudes. In the case of bimodal or multimodal frequencies, even with multiple peaks, the calculation of the center frequency is more biased towards the location of the dominant spectral peak. This is because the dominant spectral peak usually has a larger amplitude and a higher spectral density, thus dominating the weighted calculation and improving the stability and accuracy of the center frequency extraction.

[0069] In addition, this application also uses two sets of control scenarios to demonstrate the effectiveness of the above-mentioned Brillouin scattering spectrum center frequency extraction method, specifically: Compare with Scenario 1: Construct a Brillouin gain spectrum with an amplitude of 100, a full width at half maximum (FWHM) of 40MHz, and a center frequency of 100MHz, and superimpose multiplicative noise. Please refer to [link / reference]. Figure 2 , Figure 2This paper presents a comparison of the fitting accuracy (stability) of the single-peak Lorentz fitting scheme and the aforementioned linear interpolation-spectral density weighted extraction scheme under different noise levels. Figure 2 It can be seen that, under the same noise level, the above linear interpolation-spectral density weighted extraction scheme has higher measurement accuracy and stronger noise resistance compared to the single-peak Lorentz fitting scheme.

[0070] Compare with Scenario 2: Construct a Brillouin spectrum, with center frequencies corresponding to 10, 20, and 25, forming a trimodal distribution. Please refer to [link / reference]. Figure 3 , Figure 3 This paper presents a comparison of the fitting accuracy (stability) of the single-peak Lorentz fitting scheme and the aforementioned linear interpolation-spectral density weighted extraction scheme for single-peak Lorentz spectra under different noise levels. Figure 3 It can be seen that, under the same noise level, the above linear interpolation-spectral density weighted extraction scheme has higher measurement accuracy and stronger noise resistance compared to the single-peak Lorentz fitting scheme.

[0071] Additionally, please see Figure 4 and Figure 5 , Figure 4 The fitting results of fitting the unimodal data using the unimodal Lorentz fitting scheme are shown. Figure 5 The results of fitting multi-peak data using a unimodal Lorentz fitting scheme are shown. Figure 4 and Figure 5 It can be seen that the fitting schemes in related technologies are highly dependent on mathematical models, and will fail when the models do not match. Figure 5 The abnormal situation shown is as follows. The Brillouin scattering spectrum center frequency extraction method provided in this application directly processes the measurement data through interpolation, weighting, and other methods, without relying on a specific mathematical model. This method focuses more on the characteristics and processing of the data itself, and can better adapt to Brillouin spectra of different shapes.

[0072] Based on the same inventive concept, embodiments of this application also provide an optical fiber strain sensing method, including: Discrete Brillouin scattering spectrum data are acquired using a sensing fiber; wherein the sensing fiber is installed in the structure to be monitored. Extract the center frequency from the Brillouin scattering spectrum data; wherein the center frequency is obtained by the above-described Brillouin scattering spectrum center frequency extraction method. Strain monitoring is performed on the structure to be monitored based on the center frequency.

[0073] It is understood that the method of using center frequency for strain monitoring is a relatively mature technology in this field. For its specific implementation, please refer to the relevant technology. The embodiments in this application will not be described in detail.

[0074] Based on the same inventive concept, this application also provides an optical fiber strain sensing system, including a host computer, a photodetector, and a sensing optical fiber. The photodetector is electrically connected to the host computer, and the sensing optical fiber is connected to the photodetector, wherein: The host computer is used to acquire discrete Brillouin scattering spectrum data; interpolate the Brillouin scattering spectrum data to obtain incremental Brillouin scattering spectrum data; denoise the incremental Brillouin scattering spectrum data; acquire the spectral density of each data point in the denoised incremental Brillouin scattering spectrum data; use the spectral density as a weight to perform a weighted sum of the frequencies of all data points in the incremental Brillouin scattering spectrum data to obtain a frequency-weighted sum; and acquire the center frequency of the Brillouin scattering spectrum data based on the weighted sum of the weighted weights and the frequency-weighted sum.

[0075] Figure 6 This is a schematic diagram of an electronic device provided in an embodiment of this application. (Refer to...) Figure 6 The electronic device 200 includes a processor 210, a memory 220, and a communication interface 230. These components are interconnected and communicate with each other via a communication bus 240 and / or other forms of connection mechanism (not shown).

[0076] The memory 220 includes one or more (only one is shown in the figure), which may be, but is not limited to, Random Access Memory (RAM), Read Only Memory (ROM), Programmable Read-Only Memory (PROM), Erasable Programmable Read-Only Memory (EPROM), Electrically Erasable Programmable Read-Only Memory (EEPROM), etc. The processor 210 and other possible components may access the memory 220 to read and / or write data therein.

[0077] Processor 210 includes one or more (only one is shown in the figure), which can be an integrated circuit chip with signal processing capabilities. The processor 210 described above can be a general-purpose processor, including a central processing unit (CPU), a microcontroller unit (MCU), a network processor (NP), or other conventional processors; it can also be a special-purpose processor, including a digital signal processor (DSP), an application-specific integrated circuit (ASIC), a field-programmable gate array (FPGA), or other programmable logic devices, discrete gate or transistor logic devices, or discrete hardware components.

[0078] The communication interface 230 includes one or more (only one is shown in the figure) and can be used to communicate directly or indirectly with other devices to exchange data. For example, the communication interface 230 can be an Ethernet interface; it can be a mobile communication network interface, such as an interface for 3G, 4G, or 5G networks; or it can be other types of interfaces with data transmission and reception functions.

[0079] One or more computer program instructions may be stored in memory 220. Processor 210 may read and run these computer program instructions to implement the Brillouin scattering spectrum center frequency extraction method provided in the embodiments of this application and other desired functions.

[0080] Understandable. Figure 6 The structure shown is for illustrative purposes only; the electronic device 200 may also include more than [other components]. Figure 6 The more or fewer components shown, or having the same Figure 6 The different configurations shown. Figure 6 The components shown can be implemented using hardware, software, or a combination thereof. For example, electronic device 200 can be a single server (or other device with computing power), a combination of multiple servers, a cluster of a large number of servers, etc., and can be either a physical device or a virtual device.

[0081] This application also provides a computer-readable storage medium storing computer program instructions. These instructions are read and executed by a computer processor to perform the Brillouin scattering spectrum center frequency extraction method provided in this application. For example, the computer-readable storage medium can be implemented as follows: Figure 6The memory 220 in the electronic device 200.

[0082] In the embodiments provided in this application, it should be understood that the disclosed apparatus and methods can be implemented in other ways. The apparatus embodiments described above are merely illustrative. For example, the division of units is only a logical functional division, and in actual implementation, there may be other division methods. Furthermore, multiple units or components may be combined or integrated into another system, or some features may be ignored or not executed. Additionally, the displayed or discussed mutual couplings, direct couplings, or communication connections may be through some communication interfaces; indirect couplings or communication connections between devices or units may be electrical, mechanical, or other forms.

[0083] Furthermore, the units described as separate components may or may not be physically separate. The components shown as units may or may not be physical units; that is, they may be located in one place or distributed across multiple network units. Some or all of the units can be selected to achieve the purpose of this embodiment according to actual needs.

[0084] Furthermore, the functional modules in the various embodiments of this application can be integrated together to form an independent part, or each module can exist independently, or two or more modules can be integrated to form an independent part.

[0085] The above description is merely an embodiment of this application and is not intended to limit the scope of protection of this application. Various modifications and variations can be made to this application by those skilled in the art. Any modifications, equivalent substitutions, improvements, etc., made within the spirit and principles of this application should be included within the scope of protection of this application.

Claims

1. A method for extracting the center frequency of a Brillouin scattering spectrum, characterized in that, The method includes: Obtain discrete Brillouin scattering spectrum data; Interpolate the Brillouin scattering spectrum data to obtain incremental Brillouin scattering spectrum data; The incremental Brillouin scattering spectrum data is then denoised. Obtain the spectral density of each data point in the denoised incremental Brillouin scattering spectrum data; Using the spectral density as the weighting factor, the frequencies of all data points in the incremental Brillouin scattering spectrum data are weighted and summed to obtain the frequency-weighted sum. The center frequency of the Brillouin scattering spectrum data is obtained based on the sum of the weighted weights and the frequency weighted sum.

2. The method for extracting the center frequency of Brillouin scattering spectrum according to claim 1, characterized in that, The denoising process for the incremental Brillouin scattering spectrum data includes: Obtain the maximum amplitude of the Brillouin scattering spectrum data; Based on the maximum amplitude and the preset proportional coefficient, the selection threshold is obtained; Based on the selected point threshold, the incremental Brillouin scattering spectrum data is denoised.

3. The method for extracting the center frequency of Brillouin scattering spectrum according to claim 2, characterized in that, The denoising process for the incremental Brillouin scattering spectrum data includes: Obtain the maximum amplitude, the first amplitude corresponding to the minimum frequency point, and the second amplitude corresponding to the maximum frequency point of the Brillouin scattering spectrum data; A selection threshold is obtained based on the first amplitude, the second amplitude, the maximum amplitude, and a preset scaling factor; wherein the selection threshold is the maximum value among the products of the first amplitude, the second amplitude, the maximum amplitude, and the preset scaling factor; Based on the selected point threshold, the incremental Brillouin scattering spectrum data is denoised.

4. The method for extracting the center frequency of Brillouin scattering spectrum according to any one of claims 1 to 3, characterized in that, The step of interpolating the Brillouin scattering spectrum data to obtain incremental Brillouin scattering spectrum data includes: Linear interpolation is performed on the Brillouin scattering spectrum data to obtain incremental Brillouin scattering spectrum data.

5. The method for extracting the center frequency of Brillouin scattering spectrum according to any one of claims 1 to 3, characterized in that, After acquiring the center frequency of the Brillouin scattering spectrum data, the method further includes: The Brillouin frequency shift of the Brillouin scattering spectrum data is calculated based on the center frequency of the Brillouin scattering spectrum data.

6. The method for extracting the center frequency of Brillouin scattering spectrum according to claim 1, characterized in that, The denoising process for the incremental Brillouin scattering spectrum data includes: The incremental Brillouin scattering spectrum data can be denoised using linear or nonlinear filtering; alternatively, an autoregressive model or wavelet transform model can be used; or, a denoising autoencoder (DAE) or a generative adversarial network (GAN) can be used.

7. The method for extracting the center frequency of Brillouin scattering spectrum according to claim 1, characterized in that, The spectral density is calculated as follows: for discrete Brillouin scattering spectrum data ,in It is a frequency point. This corresponds to the amplitude, and the spectral density is defined as the square of the amplitude, i.e. .

8. A fiber optic strain sensing method, characterized in that, The method includes: Discrete Brillouin scattering spectrum data are acquired using a sensing optical fiber; wherein the sensing optical fiber is installed in the structure to be monitored. The center frequency is extracted from the Brillouin scattering spectrum data; wherein the center frequency is obtained using the Brillouin scattering spectrum center frequency extraction method as described in any one of claims 1 to 7; Based on the center frequency, strain monitoring is performed on the structure to be monitored.

9. A fiber optic strain sensing system, characterized in that, include: The system comprises a host computer, a photodetector, and a sensing fiber, wherein the photodetector is electrically connected to the host computer, and the sensing fiber is connected to the photodetector, wherein: The host computer is used to acquire discrete Brillouin scattering spectrum data; interpolate the Brillouin scattering spectrum data to obtain incremental Brillouin scattering spectrum data; denoise the incremental Brillouin scattering spectrum data; acquire the spectral density of each data point in the denoised incremental Brillouin scattering spectrum data; use the spectral density as a weight to perform a weighted sum of the frequencies of all data points in the incremental Brillouin scattering spectrum data to obtain a frequency-weighted sum; and acquire the center frequency of the Brillouin scattering spectrum data based on the total weight of the weighted weights and the frequency-weighted sum.

10. An electronic device, characterized in that, include: A processor, a memory, and a communication bus, wherein the processor and the memory communicate with each other via the communication bus; The memory stores program instructions that can be executed by the processor, and the processor can execute the method as described in any one of claims 1 to 7 by calling the program instructions.

Images