Fast BOTDA (Brillouin Optical Time Domain Analysis) system and demodulation method based on spectrum and dynamic signal compressed sensing

Abstract

The invention provides a quick BOTDA (Brillouin optical time domain analysis) system based on spectrum and dynamic signal compressed sensing and a demodulation method. The demodulation method comprises the following steps: 1) carrying out Brillouin spectrum compressed sampling and reconstruction: realizing Brillouin spectrum compressed sampling through detection light frequency Gaussian random scanning; according to the observation signal and the Brillouin spectrum transform domain dictionary, Brillouin spectrum reconstruction is realized based on a Brillouin spectrum compressed sampling transform vector; 2) dynamic signal compression sampling and reconstruction: dynamic signal compression sampling is realized through a control pulse sequence and a corresponding microwave scanning sequence, and dynamic signal reconstruction is realized based on a dynamic signal compression sampling transformation vector according to a dynamic signal transformation domain dictionary; and 3) carrying out demodulation according to the reconstructed Brillouin spectrum and the reconstructed dynamic signal. By applying the technical scheme provided by the invention, the technical problem that the data compression requirement of a quick BOTDA system cannot be met in the prior art can be solved.

Description

Technical Field

[0001] The present invention relates to the field of optical technology, and in particular to a fast BOTDA system and a demodulation method based on spectrum and dynamic signal compression sensing. Background Art

[0002] Fiber optic sensing has the characteristics of anti-electromagnetic interference and corrosion resistance, and can work in extreme conditions. In addition, fiber optic materials have the advantages of small size, light weight, and easy deployment, showing excellent engineering properties. Distributed fiber optic sensing technology based on the scattering effect can simultaneously transmit and sense signals to achieve continuous spatial measurement. Its monitoring points can reach millions, and scattering-type distributed fiber optic sensing technology has also become an important development direction for fiber optic sensors. The types of scattered light signals mainly include Rayleigh scattering, Brillouin scattering, and Raman scattering, and the corresponding distributed fiber optic sensors have been widely studied. The Brillouin Optical Time Domain Analysis (BOTDA) system has the advantages of high spatial resolution, fast measurement, and high precision, and has become one of the typical representative technologies of distributed fiber optic sensing technology.

[0003] After more than ten years of technological development, the measurement time of the BOTDA system based on the frequency scanning scheme has approached its performance limit. During the measurement process, the frequency difference between the pump light and the detection light needs to be scanned. The main factors affecting the measurement time of the BOTDA system are the fiber length, the average number, the frequency switching speed and the number of frequency scanning steps. The fiber length limits the trigger interval of the pulse signal. It is necessary to ensure that there is only one pulse signal in the fiber to be measured to avoid signal overlap, and the measured Brillouin signal also requires a certain number of averages to improve the signal-to-noise ratio. For short-distance sensing, the pulse interval and the required average number can be reduced. At this time, the latter two factors (frequency switching time and frequency scanning steps) become the main limitations. The arbitrary waveform generator (AWG) can achieve fast frequency switching using high-speed digital circuits, that is, optical frequency agility technology, which effectively solves the limitation of frequency switching time on the measurement time of the BOTDA system. In 2012, researchers from Tel Aviv University in Israel first proposed the use of optical frequency agility technology to achieve rapid scanning of the detection light frequency of the BOTDA system. The frequency to be scanned is written into the memory of the arbitrary waveform generator and output once. Combined with IQ upscaling, rapid scanning of the detection light frequency is achieved. The timing synchronization of the pump pulse sequence and the optical frequency agility modulated detection light is input from both ends of the optical fiber to be tested. The length of the optical fiber to be tested is 100m. In the case of 10 averages, the system sampling rate can reach 10kHz. Therefore, the number of frequency scans of the Brillouin spectroscopy has become the main factor limiting the BOTDA system. In particular, for large temperature and large strain monitoring, the number of frequency scans needs to be further increased, resulting in an increase in the measurement time of the swept frequency BOTDA system.

[0004] On the other hand, the performance of frequency scanning BOTDA is subject to high data storage pressure. A single measurement of the BOTDA system needs to record the data of all fiber positions during the frequency scanning process and obtain the Brillouin gain spectrum of all fiber positions. However, during the dynamic signal measurement process, the Brillouin gain spectrum of a single measurement is a single sampling of the dynamic signal, which needs to be recorded continuously for a certain period of time to complete the dynamic signal measurement. Under high-speed sampling conditions, the BOTDA system based on frequency scanning cannot realize fast data processing and can only record all data first, which also puts forward high requirements on the storage space of the acquisition card with high sampling rate (above 100 MHz). Traditional compression technology first collects the signal data completely, and then uses algorithms to eliminate redundant information in the data to achieve data compression, and uses decompression algorithms to restore the original signal during data reconstruction. In this acquisition-compression-transmission-decompression mode, data compression technology only reduces the amount of data stored and transmitted by the post-processing equipment, and there is still huge hardware pressure for the acquisition equipment.

[0005] Compressed sensing technology directly realizes data compression during the acquisition process, reducing the memory requirements of the acquisition card. BOTDA sensing research based on compressed sensing technology has been initially applied. The Brillouin gain spectrum is represented in other sparse domains such as the cosine domain, and the Brillouin gain spectrum is reconstructed using a small number of collected samples combined with a reconstruction algorithm. In 2020, researchers from Harbin Institute of Technology proposed compressed sampling of Brillouin spectra based on principal component analysis. With a scanning step of 4MHz, distributed Brillouin fiber optic sensing can be completed by uniformly sampling 30% of the data volume at equal intervals, effectively improving the system sampling rate and reducing the amount of data storage.

[0006] However, the existing technology still cannot meet the demand for data compression in fast BOTDA systems. How to further expand the application of compressed sensing algorithms in fast BOTDA systems has high scientific research and engineering application value. Summary of the invention

[0007] The present invention aims to solve at least one of the technical problems existing in the prior art.

[0008] According to one aspect of the present invention, a demodulation method for a fast BOTDA system based on spectrum and dynamic signal compression sensing is provided, the demodulation method comprising:

[0009] 1) Perform Brillouin spectrum compression sampling and reconstruction: realize Brillouin spectrum compression sampling by Gaussian random scanning of the detection light frequency; realize Brillouin spectrum reconstruction based on Brillouin spectrum compression sampling transformation vector according to the observed signal and Brillouin spectrum transform domain dictionary;

[0010] 2) Dynamic signal compression sampling and reconstruction: Dynamic signal compression sampling is achieved by controlling the pulse sequence and its corresponding microwave scanning sequence, and dynamic signal reconstruction is achieved based on the dynamic signal compression sampling transformation vector according to the dynamic signal transformation domain dictionary;

[0011] 3) Demodulation is performed based on the reconstructed Brillouin spectrum and the reconstructed dynamic signal.

[0012] Furthermore, in step 1), according to the Brillouin spectrum characteristics, a Brillouin spectrum database with different frequency shifts is obtained by simulation, Brillouin spectrum sparsification is achieved based on the Brillouin spectrum transform domain dictionary, and Brillouin spectrum compressed sampling is achieved based on the Brillouin spectrum measurement matrix.

[0013] Furthermore, the Brillouin spectrum transform domain dictionary is obtained through the principal component analysis algorithm.

[0014] Furthermore, a Gaussian random matrix is ​​used to construct a Brillouin spectroscopy measurement matrix.

[0015] Furthermore, the Brillouin spectrum compression sampling transformation vector and the dynamic signal compression sampling transformation vector are solved by using a convex optimization algorithm, a combinatorial algorithm, a statistical optimization algorithm, a greedy algorithm or an intersection matching pursuit algorithm.

[0016] According to another aspect of the present invention, a fast BOTDA system based on spectral and dynamic signal compression sensing is provided. The fast BOTDA system uses the demodulation method as described above to achieve demodulation. The fast BOTDA system includes: a first laser, a coupler, a first electro-optical modulator, a second electro-optical modulator, an arbitrary waveform generator, an amplifier, a second laser, an optical fiber to be tested, a photodetector, a data acquisition card and a data processing module. The output signal of the first laser is divided into two signals after passing through the coupler, and enters the first electro-optical modulator and the second electro-optical modulator respectively. The arbitrary waveform generator provides loading current for the first electro-optical modulator and the second electro-optical modulator respectively. The first electro-optic modulator outputs a pump pulse light signal, and the second electro-optic modulator outputs an optical frequency-agile signal to realize Brillouin spectrum compression sampling and dynamic signal compression sampling; the pump pulse light signal is amplified by the amplifier and enters the optical fiber to be tested, and the optical frequency-agile signal is shaped by the second laser and enters the optical fiber to be tested, and the signal in the optical fiber to be tested undergoes stimulated Brillouin scattering and is output to the photodetector carrying Brillouin gain spectrum information, and the data acquisition card collects data from the photodetector, and the data processing module reconstructs the Brillouin spectrum and the dynamic signal according to the data collected by the data acquisition card, and demodulates according to the reconstructed Brillouin spectrum and the reconstructed dynamic signal.

[0017] Furthermore, the fast BOTDA system also includes a first circulator. The pump pulse light signal amplified by the amplifier enters the optical fiber to be tested from the second port after passing through the first port of the first circulator. The signal in the optical fiber to be tested undergoes stimulated Brillouin scattering and is output from the third port of the first circulator to the photodetector carrying Brillouin gain spectrum information.

[0018] Furthermore, the fast BOTDA system also includes a second circulator, and the optical frequency-agile signal output after double-sideband modulation by the second electro-optical modulator enters the second laser from the second port after passing through the first port of the second circulator, and the lower sideband optical signal is injected and locked to output, and the detection light is filtered, amplified and spectrum-shaped, and then output from the third port of the second circulator.

[0019] Furthermore, the fast BOTDA system further includes an isolator, which is located between the second circulator and the optical fiber to be tested.

[0020] Furthermore, the amplifier is an erbium-doped fiber amplifier.

[0021] By applying the technical solution of the present invention, a fast BOTDA system and demodulation method based on compressed sensing of spectrum and dynamic signal are provided. The demodulation method compresses and samples and reconstructs the Brillouin spectrum and dynamic signal respectively, reduces the number of frequencies required for Brillouin spectrum measurement and the number of Brillouin spectra required for dynamic signals, and greatly reduces data storage space. Compared with the existing compressed sensing BOTDA system, the data storage space is further reduced, the requirements of the fast BOTDA system for the memory of the acquisition card are alleviated, and the ability to break through the Nyquist sampling theorem to achieve higher frequency dynamic signal sampling is achieved. Compared with the prior art, the technical solution of the present invention can solve the technical problem that the prior art cannot meet the data compression requirements of the fast BOTDA system. BRIEF DESCRIPTION OF THE DRAWINGS

[0022] The included drawings are used to provide a further understanding of the embodiments of the present invention, which constitute a part of the specification, are used to illustrate the embodiments of the present invention, and together with the text description, explain the principles of the present invention. Obviously, the drawings in the following description are only some embodiments of the present invention, and for ordinary technicians in this field, other drawings can be obtained based on these drawings without creative work.

[0023] Figure 1 A structural schematic diagram showing a schematic diagram of a compressed sensing solution process provided according to a specific embodiment of the present invention;

[0024] Figure 2 A schematic diagram of Brillouin spectrum compression sampling provided according to a specific embodiment of the present invention is shown;

[0025] Figure 3A schematic diagram of dynamic signal compression sampling provided according to a specific embodiment of the present invention is shown;

[0026] Figure 4 A schematic diagram of dynamic signal compression sampling provided according to another specific embodiment of the present invention is shown;

[0027] Figure 5 A schematic diagram of a fast BOTDA system provided according to a specific embodiment of the present invention is shown. DETAILED DESCRIPTION

[0028] It should be noted that, in the absence of conflict, the embodiments in this application and the features in the embodiments can be combined with each other. The technical solutions in the embodiments of the present invention will be clearly and completely described below in conjunction with the drawings in the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. The following description of at least one exemplary embodiment is actually only illustrative and is by no means intended to limit the present invention and its application or use. Based on the embodiments in the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0029] It should be noted that the terms used herein are only for describing specific embodiments and are not intended to limit the exemplary embodiments according to the present application. As used herein, unless the context clearly indicates otherwise, the singular form is also intended to include the plural form. In addition, it should be understood that when the terms "comprise" and / or "include" are used in this specification, it indicates the presence of features, steps, operations, devices, components and / or combinations thereof.

[0030] Unless otherwise specifically stated, the relative arrangement, numerical expressions and numerical values ​​of the parts and steps set forth in these embodiments do not limit the scope of the present invention. Meanwhile, it should be understood that, for ease of description, the sizes of the various parts shown in the accompanying drawings are not drawn according to the actual proportional relationship. The technology, method and equipment known to those of ordinary skill in the relevant art may not be discussed in detail, but in appropriate cases, the technology, method and equipment should be considered as a part of the specification. In all examples shown and discussed here, any specific value should be interpreted as being merely exemplary, rather than as a limitation. Therefore, other examples of exemplary embodiments may have different values. It should be noted that similar reference numerals and letters represent similar items in the following drawings, and therefore, once a certain item is defined in an accompanying drawing, it does not need to be further discussed in subsequent drawings.

[0031] According to a specific embodiment of the present invention, a demodulation method of a fast BOTDA system based on spectrum and dynamic signal compression sensing is provided, and the demodulation method includes:

[0032] 1) Perform Brillouin spectrum compression sampling and reconstruction: realize Brillouin spectrum compression sampling by Gaussian random scanning of the detection light frequency; realize Brillouin spectrum reconstruction based on Brillouin spectrum compression sampling transformation vector according to the observed signal and Brillouin spectrum transform domain dictionary;

[0033] 2) Dynamic signal compression sampling and reconstruction: Dynamic signal compression sampling is achieved by controlling the pulse sequence and its corresponding microwave scanning sequence, and dynamic signal reconstruction is achieved based on the dynamic signal compression sampling transformation vector according to the dynamic signal transformation domain dictionary;

[0034] 3) Demodulation is performed based on the reconstructed Brillouin spectrum and the reconstructed dynamic signal.

[0035] By applying this configuration, a demodulation method for a fast BOTDA system based on compressed sensing of spectrum and dynamic signals is provided. This demodulation method compresses and samples and reconstructs the Brillouin spectrum and dynamic signal respectively, reduces the number of frequencies required for Brillouin spectrum measurement and the number of Brillouin spectra required for dynamic signals, and greatly reduces data storage space. Compared with the existing compressed sensing BOTDA system, it further reduces data storage space, alleviates the requirements of the fast BOTDA system for acquisition card memory, and has the ability to break through the Nyquist sampling theorem to achieve higher frequency dynamic signal sampling.

[0036] The theory of compressed sensing shows that under the prior condition that the sampled signal is sparse, it is possible to break through the Nyquist sampling theorem, reduce the number of sampling points required to measure the signal, and accurately restore the original signal by solving an optimization inverse problem. Figure 1 As shown in Figure 1, since the observed signal, measurement matrix and transform domain dictionary are known, the reconstructed signal can be obtained by using the transform vector and transform domain dictionary. It can be seen that the transform domain dictionary, measurement matrix and signal reconstruction algorithm are three important components for realizing the compressed sampling of Brillouin spectroscopy and strain signals.

[0037] Furthermore, in the present invention, in the process of Brillouin spectrum compression sampling, firstly, according to the Brillouin spectrum characteristics, a Brillouin spectrum database with different frequency shifts is simulated and acquired, Brillouin spectrum sparsification is realized based on the Brillouin spectrum transform domain dictionary, and Brillouin spectrum compression sampling is realized based on the Brillouin spectrum measurement matrix.

[0038] As a specific embodiment of the present invention, a transform domain dictionary (Φ) can be obtained by a principal component analysis algorithm. Compared with other fixed transform domains such as the cosine domain, a sparse transform dictionary can be adaptively constructed according to data features to achieve Brillouin spectrum sparsification. Then, a Brillouin spectrum measurement matrix (Ψ) is constructed using a Gaussian random matrix, and Brillouin spectrum compression sampling is achieved based on the Brillouin spectrum measurement matrix to meet the requirement of low correlation between the measurement matrix and the transform domain dictionary. Figure 2 As shown, randomly selected frequencies of the Brillouin spectrum are represented by circles in the figure.

[0039] Furthermore, in the present invention, after the Brillouin spectrum compression sampling is completed, the Brillouin spectrum reconstruction is achieved based on the Brillouin spectrum compression sampling transformation vector according to the observed signal and the Brillouin spectrum transform domain dictionary.

[0040] Since the perception matrix A is A=ΨΦ, and since the dimension of the perception matrix is ​​much smaller than the dimension of the transformation vector X, the recovery of the Brillouin spectrum is a solution to an ill-conditioned equation. Generally, solving the transformation vector is a process of finding the sparsest solution. As a specific embodiment of the present invention, the main algorithms for solving the transformation vector include convex optimization algorithms, combinatorial algorithms, statistical optimization algorithms, greedy algorithms, and orthogonal matching pursuit (OMP) algorithms. Among them, the core of the greedy algorithm is to reconstruct the transformation vector by continuously iterative comparison, each time selecting the local optimal solution, with good convergence and fast reconstruction speed.

[0041] Taking the orthogonal matching pursuit (OMP) algorithm as an example to realize signal recovery, the core steps of the algorithm are as follows:

[0042] Input: compressed sampling signal Y, perception matrix A and sparsity K;

[0043] Output: estimated signal y and residual R;

[0044] Initialization: residual R0 = Y, support set F = [], transformation vector X = [], number of iterations h = 1;

[0045] Loop through steps 1-5:

[0046] Step 1: Calculate the inner product of the residual R and the column of the perception matrix A, and find the column j_h where the inner product is maximum;

[0047] Step 2: Find the corresponding column of the perception matrix A and update the support set: F = [F, A j_h ], after updating, set the corresponding column of the perception matrix to 0;

[0048] Step 3: Solve the transformation vector coefficient X by the least squares algorithm h , X h =(F T F) -1 F TY;

[0049] Step 4: Update the residual R, R = Y-FX;

[0050] Step 5: Determine whether the iteration termination condition h>K is met. If so, output X, R; otherwise, return to step 1.

[0051] Furthermore, in the invention, after the Brillouin spectrum compression sampling and reconstruction are completed, dynamic signal compression sampling is performed: dynamic signal compression sampling is achieved by controlling the pulse sequence and the corresponding microwave scanning sequence. In the present invention, it is only necessary to output a randomly sampled pulse sequence and its corresponding microwave sequence, and the microwave sequence contains randomly sampled microwave frequencies, so that dynamic signal compression sampling can be achieved.

[0052] In the process of compressed sampling and reconstruction of dynamic signals, dynamic strain generally has good sparsity in the frequency domain, and the dynamic signal transform domain dictionary is obtained by discrete cosine transform. For BOTDA sensing, the strain is demodulated by the frequency shift of the Brillouin spectrum, and each Brillouin spectrum can be used as a sampling point of the dynamic signal. Similar to the compressed sampling of the Brillouin gain spectrum, a Gaussian random matrix is ​​selected as the measurement matrix to meet the low correlation between the measurement matrix and the transform domain dictionary. The schematic diagram of compressed sampling of different strain signals is shown in the figure. Figure 3 and Figure 4 shown.

[0053] Furthermore, in the present invention, after the dynamic signal compression sampling is completed, the dynamic signal is reconstructed based on the dynamic signal compression sampling transformation vector according to the dynamic signal transformation domain dictionary.

[0054] As a specific embodiment of the present invention, a convex optimization algorithm may be used to calculate the dynamic signal compression sampling transformation vector.

[0055] By solving the convex optimization problem, the sparse coefficient x, Y = Ax, can be obtained from a small number of observations. Here, dynamic signal reconstruction is achieved by solving the linear band equality constraint minimization L1 norm problem. The mathematical expression of this problem is as follows:

[0056] min‖x‖1, st (constraint): Ax = Y

[0057] The solution of the above equation is also called the basis pursuit algorithm, which is to find the vector with the minimum L1 norm:

[0058]

[0059] The principle of this algorithm is to continuously look for the x with the smallest L1 norm to explain the sampled compressed signal Y. If a set of sufficiently sparse signals (signals with a small enough L1 norm) is found, then it is considered that the system of equations has found the most appropriate solution.

[0060] Furthermore, in the present invention, after the reconstruction of the dynamic signal is completed, demodulation is performed according to the reconstructed Brillouin spectrum and the reconstructed dynamic signal.

[0061] The present invention firstly performs compression sampling and reconstruction on the Brillouin spectrum, reduces the number of required frequency sweeps, improves the system sampling rate and reduces the data storage space, and on this basis, performs compression sampling and reconstruction on the dynamic signal, further reduces the data storage space, alleviates the requirement of the fast BOTDA system on the memory of the acquisition card, or effectively increases the recording time of the fast BOTDA system for the dynamic signal under the same hardware conditions.

[0062] like Figure 5 As shown, according to another aspect of the present invention, a fast BOTDA system based on spectral and dynamic signal compression sensing is provided, the fast BOTDA system adopts the demodulation method as described above to realize demodulation, the fast BOTDA system comprises: a first laser, a coupler, a first electro-optical modulator, a second electro-optical modulator, an arbitrary waveform generator, an amplifier, a second laser, an optical fiber to be tested, a photodetector, a data acquisition card and a data processing module, the output signal of the first laser is divided into two signals after passing through the coupler, and enters the first electro-optical modulator and the second electro-optical modulator respectively, the arbitrary waveform generator provides the first electro-optical modulator and the second electro-optical modulator with amplification. The first electro-optic modulator outputs a pump pulse light signal, and the second electro-optic modulator outputs an optical frequency-agile signal to realize Brillouin spectrum compression sampling and dynamic signal compression sampling; the pump pulse light signal is amplified by the amplifier and enters the optical fiber to be tested, and the optical frequency-agile signal is shaped by the second laser and enters the optical fiber to be tested, and the signal in the optical fiber to be tested undergoes stimulated Brillouin scattering, and is output to the photodetector carrying Brillouin gain spectrum information, and the data acquisition card collects data from the photodetector, and the data processing module reconstructs the Brillouin spectrum and the dynamic signal according to the data collected by the data acquisition card, and demodulates according to the reconstructed Brillouin spectrum and the reconstructed dynamic signal.

[0063] In the fast BOTDA system of the present invention, each pulse and its corresponding microwave signal can obtain the Brillouin gain signal at all positions, and the compressed sampling of the Brillouin spectrum is realized by the output of the frequency scanning signal of the arbitrary waveform generator. Each group of pulse sequences and their corresponding microwave scanning sequences can obtain the Brillouin spectra of all positions at a certain moment, which can be regarded as a sampling point for a dynamic signal, and the compressed sampling of the dynamic signal is realized by controlling the pulse sequence and its corresponding microwave scanning sequence.

[0064] Furthermore, in the present invention, in order to realize the detection and collection of the signal carrying the Brillouin gain spectrum information in the optical fiber to be tested, the configurable fast BOTDA system also includes a first circulator. The pump pulse light signal after being amplified by the amplifier enters the optical fiber to be tested from the second port after passing through the first port of the first circulator. The signal in the optical fiber to be tested undergoes stimulated Brillouin scattering and is output from the third port of the first circulator to the photodetector carrying the Brillouin gain spectrum information.

[0065] Furthermore, in the present invention, in order to realize the shaping of the optical agile frequency conversion signal by the second laser, the configurable fast BOTDA system also includes a second circulator. The optical agile frequency conversion signal output after double-sideband modulation by the second electro-optical modulator enters the second laser from the second port after passing through the first port of the second circulator. The lower sideband optical signal is injected and locked to the output, and the detection light is filtered, amplified and spectrum shaped before being output from the third port of the second circulator.

[0066] Furthermore, in the present invention, in order to prevent the high-power pump pulse light amplified by the amplifier from entering the second laser and the second electro-optic modulator, the configurable fast BOTDA system further includes an isolator, which is located between the second circulator and the optical fiber to be tested.

[0067] As a specific embodiment of the present invention, the amplifier can be an erbium-doped fiber amplifier. After the pump pulse light is amplified by the erbium-doped fiber amplifier, the peak power is hundreds of milliwatts.

[0068] Both the electro-optic modulator 1 and the electro-optic modulator 2 work at the lowest output working point. The pulse width of the pump pulse light affects the spatial resolution of the system. For example, a pulse width of 20 ns means a spatial resolution of 2 m for the system.

[0069] According to preliminary experimental results, the traditional fast BOTDA measurement scans from 10.7GHz to 11.2GHz with a step length of 4MHz, requiring a total of 126 frequencies. The Brillouin spectrum compression sampling only requires 31.8% of the data volume of the traditional BOTDA system to achieve the compression reconstruction of the Brillouin spectrum, increasing the system sampling rate by 3.3 times, compressing the data volume by 68.2%, and further only requiring 50% of the pulse sequence to achieve the compression sampling of dynamic signals. Therefore, using the proposed simultaneous compression sampling of the spectrum and dynamic signal, the fast BOTDA data volume can be compressed by 84.1%, the dynamic signal sampling rate is increased by 3.3 times, and it has the ability to break through the Nyquist sampling theorem to achieve higher frequency dynamic signal sampling.

[0070] The fast BOTDA system and demodulation method based on spectral and dynamic signal compression sensing innovatively proposed by the present invention utilizes the compressed sampling and reconstruction of Brillouin spectra to reduce the number of frequencies required to be scanned, improve the system sampling rate and reduce the data storage space; utilizes the compressed sampling and reconstruction of dynamic signals to reduce the number of required pulse sequences and their corresponding frequency sweep sequences, further reduce the data storage space, and have the ability to demodulate high-frequency signals. Compared with the existing compressed sensing BOTDA system, it further reduces the data storage space, alleviates the requirements of the fast BOTDA system for the memory of the acquisition card, and has the ability to break through the Nyquist sampling theorem to achieve higher frequency dynamic signal sampling.

[0071] Compared with the prior art, the beneficial effects of the present invention include:

[0072] 1) Greatly compress data storage space, and compress and reconstruct Brillouin spectra and dynamic signals at the same time, reduce the number of frequencies required for Brillouin spectrum measurement and the number of Brillouin spectra required for dynamic signals, and greatly reduce data storage space;

[0073] 2) Improve the system sampling rate and have the ability to break through the limitations of the Nyquist sampling theorem to achieve high-frequency vibration signal measurement. Brillouin spectroscopy compression sensing can effectively reduce the measurement time of the BOTDA system and improve the system sampling rate. Random compression sensing of dynamic signals can break through the Nyquist sampling theorem and have the ability to measure high-frequency vibration signals.

[0074] 3) Easy to apply and promote. The proposed algorithm can be directly used in the existing BOTDA hardware system. Only by changing the scanning frequency, pulse sequence and its corresponding scanning sequence output, the compressed sensing of Brillouin spectrum and dynamic signal can be realized.

[0075] In summary, the present invention provides a fast BOTDA system and demodulation method based on compressed sensing of spectra and dynamic signals. The demodulation method compresses and samples and reconstructs the Brillouin spectrum and dynamic signal, respectively, reduces the number of frequencies required for Brillouin spectrum measurement and the number of Brillouin spectra required for dynamic signals, and greatly reduces data storage space. Compared with the existing compressed sensing BOTDA system, the data storage space is further reduced, the requirements of the fast BOTDA system for the memory of the acquisition card are alleviated, and the ability to break through the Nyquist sampling theorem to achieve higher frequency dynamic signal sampling is achieved. Compared with the prior art, the technical solution of the present invention can solve the technical problem that the prior art cannot meet the data compression requirements of the fast BOTDA system.

[0076] In addition, it should be noted that the use of terms such as "first" and "second" to limit components is only for the convenience of distinguishing the corresponding components. If not otherwise stated, the above terms have no special meaning and therefore cannot be understood as limiting the scope of protection of the present invention.

[0077] The above description is only a preferred embodiment of the present invention and is not intended to limit the present invention. For those skilled in the art, the present invention may have various modifications and variations. Any modification, equivalent replacement, improvement, etc. made within the spirit and principle of the present invention shall be included in the protection scope of the present invention.

Claims

1. A demodulation method for a fast BOTDA system based on spectral and dynamic signal compressed sensing, characterized in that: The demodulation method comprises: 1) Perform Brillouin spectrum compression sampling and reconstruction: realize Brillouin spectrum compression sampling by Gaussian random scanning of the detection light frequency; realize Brillouin spectrum reconstruction based on Brillouin spectrum compression sampling transformation vector according to the observed signal and Brillouin spectrum transform domain dictionary; 2) Dynamic signal compression sampling and reconstruction: Dynamic signal compression sampling is achieved by controlling the pulse sequence and its corresponding microwave scanning sequence, and dynamic signal reconstruction is achieved based on the dynamic signal compression sampling transformation vector according to the dynamic signal transformation domain dictionary; 3) Demodulation is performed based on the reconstructed Brillouin spectrum and the reconstructed dynamic signal.

2. The demodulation method of a fast BOTDA system based on spectrum and dynamic signal compression sensing according to claim 1, characterized in that: In step 1), according to the Brillouin spectrum characteristics, a Brillouin spectrum database with different frequency shifts is simulated and acquired, Brillouin spectrum sparsification is realized based on the Brillouin spectrum transform domain dictionary, and Brillouin spectrum compression sampling is realized based on the Brillouin spectrum measurement matrix.

3. The demodulation method of a fast BOTDA system based on spectrum and dynamic signal compression sensing according to claim 2, characterized in that: The Brillouin spectrum transform domain dictionary is obtained through the principal component analysis algorithm.

4. The demodulation method of a fast BOTDA system based on spectrum and dynamic signal compression sensing according to claim 2, characterized in that: The Brillouin spectroscopy measurement matrix is ​​constructed using Gaussian random matrices.

5. The demodulation method of a fast BOTDA system based on spectrum and dynamic signal compression sensing according to claim 4, characterized in that: The Brillouin spectrum compression sampling transformation vector and the dynamic signal compression sampling transformation vector are solved by using a convex optimization algorithm, a combinatorial algorithm, a statistical optimization algorithm, a greedy algorithm or an intersection matching pursuit algorithm.

6. A fast BOTDA system based on spectral and dynamic signal compressed sensing, characterized in that: The fast BOTDA system uses the demodulation method according to any one of claims 1 to 5 to realize demodulation. The fast BOTDA system includes: a first laser, a coupler, a first electro-optical modulator, a second electro-optical modulator, an arbitrary waveform generator, an amplifier, a second laser, an optical fiber to be tested, a photodetector, a data acquisition card and a data processing module. The output signal of the first laser is divided into two signals after passing through the coupler, and enters the first electro-optical modulator and the second electro-optical modulator respectively. The arbitrary waveform generator provides loading electrical signals for the first electro-optical modulator and the second electro-optical modulator respectively. The first electro-optical modulator outputs a pump The second electro-optic modulator outputs an optical frequency-agile signal to realize Brillouin spectrum compression sampling and dynamic signal compression sampling; the pump pulse light signal is amplified by the amplifier and enters the optical fiber to be tested, and the optical frequency-agile signal is shaped by the second laser and enters the optical fiber to be tested. The signal in the optical fiber to be tested undergoes stimulated Brillouin scattering and is output to the photodetector carrying the Brillouin gain spectrum information. The data acquisition card collects data from the photodetector, and the data processing module reconstructs the Brillouin spectrum and the dynamic signal according to the data collected by the data acquisition card, and demodulates according to the reconstructed Brillouin spectrum and the reconstructed dynamic signal.

7. The fast BOTDA system based on spectrum and dynamic signal compression sensing according to claim 6, characterized in that: The fast BOTDA system also includes a first circulator. The pump pulse light signal amplified by the amplifier enters the optical fiber to be tested from the second port after passing through the first port of the first circulator. The signal in the optical fiber to be tested undergoes stimulated Brillouin scattering and is output from the third port of the first circulator to the photodetector carrying Brillouin gain spectrum information.

8. The fast BOTDA system based on spectrum and dynamic signal compressed sensing according to claim 6 or 7, characterized in that: The fast BOTDA system also includes a second circulator. The optical frequency-agile signal output after double-sideband modulation by the second electro-optical modulator enters the second laser from the second port after passing through the first port of the second circulator. The lower sideband optical signal is injected and locked to output, and the detection light is filtered, amplified and spectrum-shaped before being output from the third port of the second circulator.

9. The fast BOTDA system based on spectrum and dynamic signal compression sensing according to claim 8, characterized in that: The fast BOTDA system further comprises an isolator, which is located between the second circulator and the optical fiber to be tested.

10. The fast BOTDA system based on spectral and dynamic signal compressed sensing according to any one of claims 6 to 9, characterized in that: The amplifier is an erbium-doped fiber amplifier.

Images