Demodulation device and method for phase-sensitive optical time domain reflectometer

Abstract

The invention relates to the field of optical fiber sensing, in particular to a demodulation device and method for a phase-sensitive optical time domain reflectometer. Comprising an optical fiber sensor, a first data acquisition processor and a second data acquisition processor, sensing signals obtained by the optical fiber sensor enter the first data acquisition processor or the second data acquisition processor in the form of two paths of electric signals; the first data acquisition processor converts the two paths of electric signals into low-speed digital signals by using a band-pass sampling mode, sequentially performs band-pass filtering and Hilbert transformation on the low-speed digital signals, performs fixed point sampling on data after the Hilbert transformation, and correspondingly demodulates the sampled data; and the second data acquisition processor converts the two paths of electric signals into corresponding high-speed digital signals in a low-pass sampling mode, and correspondingly demodulates the high-speed digital signals. According to the invention, demodulation under band-pass sampling can be carried out by using a low-cost ADC and an optimized demodulation algorithm, so that performance requirements and power consumption of a demodulation device are reduced.

Description

Technical Field

[0001] The invention relates to the field of optical fiber sensing, and in particular to a demodulation device and method for a phase-sensitive optical time domain reflectometer. Background Art

[0002] Coherent detection phase-sensitive optical time domain reflectometer (Φ-OTDR) usually uses an acousto-optic modulator (AOM) to provide the frequency difference between the detection light and the intrinsic light, and performs amplitude modulation of the continuous light output by the light source with a high extinction ratio. On the premise of meeting the bandwidth of the balanced photodetector (BPD), the higher the AOM operating frequency, the steeper the rise / fall edge of the modulated optical pulse is, thereby obtaining better spatial resolution; and a higher center frequency can effectively avoid the low-frequency noise output by the BPD and improve the signal-to-noise ratio after photoelectric conversion.

[0003] However, the selection of a higher center frequency means greater data collection pressure in analog-to-digital conversion, requiring the use of high-speed and high-precision analog-to-digital converter (ADC) chips, which causes a surge in instrument costs and makes it difficult to select ADCs with high quantization bits. In addition, high sampling rates will put pressure on subsequent signal processing, and the higher the speed requirements for signal processing, the more high-performance embedded processors or field programmable gate arrays (FPGAs) are needed. In addition, the power consumption of high-speed ADC chips and high-performance FPGAs will also be relatively high, which is a challenge for low-power application sites.

[0004] In view of this, how to overcome the defects of the prior art and solve the problem that high sampling rate has high requirements on chips and power consumption is a problem to be solved in this technical field. Summary of the invention

[0005] In view of the defects or improvement needs in the prior art, the present invention solves the problem that high sampling rate has high requirements on chips and power consumption.

[0006] The embodiment of the present invention adopts the following technical solution:

[0007] In a first aspect, the present invention provides a demodulation device for a phase-sensitive optical time domain reflectometer, specifically comprising: an optical fiber sensor, a first data acquisition processor and a second data acquisition processor; the sensing signal obtained by the optical fiber sensor enters the first data acquisition processor in the form of two electrical signals or enters the second data acquisition processor in the form of two electrical signals; the first data acquisition processor is used to convert the two electrical signals into corresponding low-speed digital signals using a bandpass sampling method, perform bandpass filtering and Hilbert transform on the low-speed digital signals in sequence, perform fixed-point sampling on the digital signals after the Hilbert transform, and perform corresponding demodulation on the fixed-point data obtained by sampling; the second data acquisition processor is used to convert the two electrical signals into corresponding high-speed digital signals using a low-pass sampling method, and perform corresponding demodulation on the high-speed digital signals.

[0008] Preferably, the sampling frequency of the first data acquisition processor meets the following conditions:

[0009] 2Fh / (m+1)≤Fs≤2Fl / m;

[0010] Wherein, Fs is the sampling frequency of the first data acquisition processor; Fl is the lower limit frequency of the electrical signal, and Fh is the upper limit frequency of the electrical signal; m is a positive integer and satisfies m≤[Fl / (Fh-Fl)], and [] indicates rounding down.

[0011] Preferably, the optical fiber sensor comprises: a narrow linewidth optical fiber laser, an optical fiber coupler, an acousto-optic modulator, a pulse generator, an optical filter, an acousto-optic modulation driver, an erbium-doped optical fiber amplifier, an optical fiber circulator, a sensing optical fiber and an integrated coherent receiver, and specifically comprises: the optical signal emitted by the narrow linewidth optical fiber laser is divided into a detection light and an intrinsic light by the optical fiber coupler; the first data acquisition processor or the second data acquisition processor is used to control the pulse generator to generate the required modulation pulse, and the acousto-optic modulation driver is used to control the acousto-optic modulator to modulate the detection light based on the modulation pulse to generate modulated light; the modulated light passes through the erbium-doped optical fiber amplifier and enters the optical fiber circulator. The optical fiber circulator transmits the modulated light to the first port of the optical fiber circulator, and then enters the sensing optical fiber through the second port of the optical fiber circulator; the Rayleigh backscattered light generated by the modulated light in the sensing optical fiber is returned through the second port of the optical fiber circulator, and the returned Rayleigh backscattered light is output to the optical filter through the third port of the optical fiber circulator to obtain the filtered returned light; the returned light and the intrinsic light are coherently mixed in the integrated coherent receiver, and the mixed optical signal is used as the sensing signal; the integrated coherent receiver includes two pairs of balanced photodetectors, which convert the sensing signal into two corresponding electrical signals through the two balanced photodetectors, and the two electrical signals are output to the data acquisition processor after transimpedance amplification.

[0012] Preferably, the first data acquisition processor includes: a low-speed analog-to-digital conversion module, a bandpass filter, a Hilbert transform module, an interval sampling module, a complex conjugate multiplication module, an inverse tangent processing module and an unwinding and adjacent phase difference merging processing module, specifically including: the low-speed analog-to-digital conversion module is used to sample two electrical signals, and send the sampled digital signals to the bandpass filter for bandpass filtering; the Hilbert transform module is used to perform Hilbert transform on the signal after bandpass filtering; the interval sampling module is used to extract fixed point data of a specified interval from the transformed data; the complex conjugate multiplication module is used to perform complex conjugate multiplication of the fixed point data and the reference frame of this period; the inverse tangent processing module is used to perform inverse tangent processing on the complex conjugate multiplication result; the unwinding and adjacent phase difference merging processing module is used to perform adjacent phase differentiation on the inverse tangent processing result, and perform phase unwinding of the data after phase differentiation, so as to complete the demodulation of the digital signal obtained by bandpass sampling.

[0013] Preferably, the first data acquisition processor also includes a block storage module, which is used to receive the fixed-point data extracted by the interval sampling module, store the first frame of fixed-point data in each period as a reference frame, and input the reference frame into the complex conjugate multiplication module.

[0014] Preferably, the second data acquisition processor includes: a high-speed analog-to-digital conversion module, a first bandpass filter, a first Hilbert transform module, a complex conjugate multiplication module, an inverse tangent processing module, an unwrapping processing module, a single-frame interval phase difference processing module and a phase difference merging and averaging module, specifically including: the high-speed analog-to-digital conversion module is used to sample two electrical signals, and send the sampled digital signals to the first bandpass filter for bandpass filtering; the first Hilbert transform module is used to perform Hilbert transform on the signal after bandpass filtering; the complex conjugate multiplication module is used to perform complex conjugate multiplication of the transformed data and the reference frame of this cycle; the inverse tangent processing module is used to perform inverse tangent processing on the result of complex conjugate multiplication; the unwrapping processing module is used to perform phase unwrapping on the result of inverse tangent processing; the single-frame interval phase difference processing module sequentially performs the difference between two adjacent unwrapped data, and sequentially outputs the differential data; the phase difference merging and averaging module is used to sequentially average the fixed point data for the data output by the single-frame interval phase difference processing module to complete the phase demodulation of the digital signal obtained by low-pass sampling.

[0015] Preferably, the second data acquisition processor also includes a block storage module, a second bandpass filter and a second Hilbert transform module, specifically including: the block storage module is used to receive sampling data from the high-speed analog-to-digital conversion module, store the first frame of fixed-point data of this cycle as a reference frame, and output the reference frame to the second bandpass filter for bandpass filtering; the second Hilbert transform module is used to perform Hilbert transform on the reference frame after bandpass filtering, and output the reference frame after Hilbert transform to the complex conjugate multiplication module.

[0016] On the other hand, the present invention provides a demodulation method for a phase-sensitive optical time domain reflectometer, specifically: using the demodulation device of the phase-sensitive optical time domain reflectometer provided by the first aspect to complete the method, specifically including: using an optical fiber sensor to obtain a sensing signal, and converting the sensing signal into two electrical signals through an integrated coherent receiver; when the number of abnormal sensing events does not exceed a threshold, using a first data acquisition processor to sample and demodulate the two electrical signals; when the number of abnormal sensing events exceeds a threshold, using a second data acquisition processor to sample and demodulate the two electrical signals.

[0017] Preferably, the use of the first data acquisition processor to perform sampling and demodulation of two electrical signals specifically includes: inputting each sampling signal into at least two bandpass filters for filtering; wherein each bandpass filter has the same passband width and order, but different passband center frequencies; performing Hilbert transform on the filtered signal, performing fixed-point extraction based on the Hilbert transform result, using the first frame of fixed-point data in each period as the reference frame of the period, and performing complex conjugate multiplication of each frame of fixed-point data after extraction and the reference frame to obtain a set of real data and a set of imaginary data; adding all real data to obtain a first data channel, adding all imaginary data to obtain a second data channel, dividing the first data channel by the second data channel, and performing inverse tangent calculation on the quotient; performing differential processing on the result of the inverse tangent between adjacent data, performing unwrapping processing on the differentiated data, and using the phase difference information obtained by unwrapping as the demodulation result.

[0018] Preferably, the use of the second data acquisition processor to perform sampling and demodulation of the two electrical signals specifically includes: inputting each sampling signal into at least two bandpass filters for filtering, performing Hilbert transform on the filtered signal to obtain a filtered signal after Hilbert transform; wherein each bandpass filter has the same passband width and order, but different passband center frequencies; performing Hilbert transform on the filtered signal; using the first frame data of the two electrical signals in each cycle as the reference frame of the cycle, performing bandpass filtering and Hilbert transform on the reference frame to obtain a reference frame after Hilbert transform, and performing complex conjugate multiplication on each frame of the filtered signal after Hilbert transform and the reference frame after Hilbert transform to obtain a set of real data and a set of imaginary data; adding all real data to obtain the first data, adding all imaginary data to obtain the second data, dividing the first data by the second data, and performing inverse tangent calculation on the quotient; performing unwrapping processing on the result of the inverse tangent, obtaining the phase difference of adjacent data for the unwrapped data, and obtaining the average value of the phase differences of a specified number, and taking the average value of the phase differences as the demodulation result.

[0019] Compared with the prior art, the beneficial effects of the present invention are: two data acquisition processors are used simultaneously to process the sensing signals of the optical fiber sensor, the first data acquisition processor uses a low-cost ADC and FPGA to perform demodulation at a low sampling rate and real-time monitoring, so as to reduce the performance requirements and power consumption of the ADC chip and FPGA in most scenarios; the second data acquisition processor uses a high-performance ADC to meet high positioning accuracy and faster detection response in specific scenarios, so as to expand the scope of use scenarios of the device. In the preferred embodiment of the present invention, the sampling rate requirements for the ADC chip are further reduced by using bandpass sampling.

[0020] On the other hand, in the demodulation method provided by the present invention, the Φ-OTDR demodulation algorithm is executed by using a combination of data acquisition and processing components unique to the demodulation device provided by the present invention, so that FPGA resources are consumed less without affecting its demodulation effect, thereby saving system costs. BRIEF DESCRIPTION OF THE DRAWINGS

[0021] In order to more clearly illustrate the technical solutions of the embodiments of the present invention, the following briefly introduces the drawings required for use in the embodiments of the present invention. Obviously, the drawings described below 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.

[0022] Figure 1 A schematic diagram of a demodulation device of a phase-sensitive optical time-domain reflectometer provided by an embodiment of the present invention;

[0023] Figure 2A schematic diagram of the structure of an optical fiber sensor in a demodulation device provided in an embodiment of the present invention;

[0024] Figure 3 A schematic diagram of a demodulation process of a first data acquisition processor in a demodulation device provided in an embodiment of the present invention;

[0025] Figure 4 A schematic diagram of the structure of a first data acquisition processor in a demodulation device provided in an embodiment of the present invention;

[0026] Figure 5 Another structural schematic diagram of the first data acquisition processor in the demodulation device provided in an embodiment of the present invention;

[0027] Figure 6 A schematic diagram of the structure of a second data acquisition processor in a demodulation device provided in an embodiment of the present invention;

[0028] Figure 7 Another structural schematic diagram of the second data acquisition processor in the demodulation device provided in an embodiment of the present invention;

[0029] Figure 8 A schematic diagram of the structure of an inverse tangent processing module in a demodulation device provided in an embodiment of the present invention;

[0030] Fig. 9 A flow chart of a demodulation method of a phase-sensitive optical time-domain reflectometer provided by an embodiment of the present invention;

[0031] Fig.10 A flow chart of a demodulation method of a first data acquisition processor provided in an embodiment of the present invention;

[0032] Fig.11 A schematic diagram of a Hilbert transform calculation process in a demodulation method provided in an embodiment of the present invention;

[0033] Fig.12 A flow chart of a demodulation method of a second data acquisition processor provided in an embodiment of the present invention. DETAILED DESCRIPTION

[0034] In order to make the purpose, technical solution and advantages of the present invention more clearly understood, the present invention is further described in detail below in conjunction with the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are only used to explain the present invention and are not used to limit the present invention.

[0035] The present invention is an architecture of a specific functional system, so the specific embodiments mainly illustrate the functional logical relationship between the various structural modules, and do not limit the specific software and hardware implementation methods.

[0036] In addition, the technical features involved in the various embodiments of the present invention described below can be combined with each other as long as they do not conflict with each other. The present invention will be described in detail below with reference to the accompanying drawings and embodiments.

[0037] In the prior art, according to the Nyquist sampling theorem, the ADC sampling rate must be at least twice the highest frequency component of the signal to ensure that the reconstruction of the signal does not alias. According to engineering experience, in practical data acquisition processes, the ADC usually needs to ensure at least 4 to 5 times the oversampling rate to better restore the signal. Therefore, the actual ADC sampling rate fs should meet the following requirements:

[0038]

[0039] Among them: the center frequency of Φ-OTDR is f, the spatial resolution is τ, the bandwidth of the phase modulation signal is approximately expressed as B = 1 / τ; N is the oversampling multiple. In a certain actual scenario, the center frequency of Φ-OTDR is 80MHz, the pulse width is 50ns, and N is 5; in this scenario, the sampling rate fs of the ADC should be greater than or equal to 500MHz. This sampling rate must use a high-speed and high-precision ADC chip, which will cause a surge in instrument costs.

[0040] On the other hand, there is no dedicated application-specific integrated circuit chip (ASIC) for the digital interface of the ADC output of the Φ-OTDR system front end. FPGA is usually used as the ADC front end data interface to receive the system sampling data. At the same time, FPGA is also used for the Φ-OTDR phase demodulation algorithm. The higher the sampling rate of the front-end ADC chip or the more complex the demodulation algorithm of the Φ-OTDR, the higher the requirements for FPGA performance and logic resources. In this case, the cost of FPGA will increase significantly, which will also lead to high product costs.

[0041] Furthermore, when performing demodulation, traditional phase demodulation uses in-phase / quadrature (IQ) demodulation. IQ demodulation consumes a large amount of FPGA resources, such as multipliers and storage resources. Therefore, FPGA chips with large logic resources must be selected, which is not conducive to reducing costs and power consumption.

[0042] Embodiment 1:

[0043] like Figure 1 As shown, the demodulation device of the phase-sensitive optical time-domain reflectometer provided in this embodiment includes: an optical fiber sensor, a first data acquisition processor and a second data acquisition processor.

[0044] The sensing signal obtained by the optical fiber sensor enters the first data acquisition processor in the form of two electrical signals or enters the second data acquisition processor in the form of two electrical signals.

[0045] In order to reduce the performance requirements and power consumption of ADC chips and FPGA chips, the first data acquisition processor provides a low-cost ADC and the optimized demodulation algorithm in this embodiment, which can meet the demodulation requirements in general scenarios. On the other hand, in the actual application of optical fiber sensors, the existing sampling and calculation accuracy is usually at the centimeter level, and due to the accuracy of the optical path system or system interference, the effective accuracy of the actual test is at the meter level, and the sampling and calculation accuracy is much greater than the accuracy of the optical path system. In the usual sensing measurement scenario, no matter how high the sampling frequency is used, the sensing result finally obtained will not be able to achieve the accuracy that can be obtained theoretically at the corresponding sampling frequency due to other precision limitations, resulting in data waste. Therefore, in this embodiment, the corresponding sampling interval is also obtained according to the use accuracy of the optical fiber sensor, and the fixed point data is obtained in the sampling data of the ADC, so that the amount of data obtained by sampling is smaller on the basis of meeting the accuracy requirements, so as to further reduce the storage and calculation amount of data. The first data acquisition processor is used to convert the two-way electrical signal into a corresponding low-speed digital signal using a bandpass sampling method, perform bandpass filtering and Hilbert transform on the low-speed digital signal in turn, perform fixed-point sampling on the digital signal after the Hilbert transform, and perform corresponding demodulation on the fixed-point data obtained by sampling.

[0046] Since the sampling rate of the first data acquisition processor is low, the spatial resolution of the optical fiber sensor will be reduced. In the prior art, a low-pass sampling method is usually used, and its sampling rate is greater than 2 times the maximum frequency of the input electrical signal; in this embodiment, a bandpass sampling method is adopted, and bandpass sampling is also called undersampling, and its sampling frequency is lower than 2 times the maximum frequency of the input electrical signal, which also makes the signal-to-noise ratio of the sampling system lower than the signal-to-noise ratio of the low-pass sampling system, which will make its detection sensitivity not as high as the system sensitivity of low-pass sampling. In places where the system monitoring index is required to be high, high-performance ADC sampling is also required. Therefore, a high-speed ADC and the optimized demodulation circuit in this embodiment are used in the second data acquisition processor, and the single-frame sampling signal is not sampled at intervals to adapt to high-performance scenarios. The second data acquisition processor is used to convert the two electrical signals into corresponding high-speed digital signals using a low-pass sampling method, and demodulate the high-speed digital signals accordingly. In some scenarios, the accuracy requirements are not too high, and the second data acquisition processor can also use the same ADC as the first data acquisition processor, and only retain a relatively higher positioning accuracy by not sampling at intervals.

[0047] In actual implementation, both the first data acquisition processor and the second data acquisition processor can be implemented using the corresponding logic resources on the FPGA. The first data acquisition processor and the second data acquisition processor can be used independently in different scenarios or synchronously. In hardware design, a low-speed and low-cost ADC chip and a high-speed ADC chip, as well as corresponding peripheral devices, can be placed on an FPGA circuit board at the same time, and whether to mount the corresponding chips and peripheral devices is determined according to the actual scenario requirements.

[0048] In the actual use process of Φ-OTDR, what is concerned is the phase change rather than the carrier frequency. It is necessary to obtain the phase information of the sensing signal through complex conjugate multiplication and arctangent calculation, and then through unwrapping, the phase value is changed from a discontinuous value to a continuous value, and then the adjacent phase values are subtracted, so as to express the sensing result of the vibration information through the change amplitude of the phase difference during this period. Therefore, the first data acquisition processor and the second data acquisition processor in this embodiment can adopt the band-pass sampling method to obtain the phase information of the sensing signal with a smaller sampling frequency. After reducing the amount of data to be processed, it will not affect the calculation of the phase difference. The band-pass sampling method greatly reduces the difficulty of ADC selection and the pressure of subsequent data processing without affecting the function of the sensing system. The corresponding signal processing chips such as FPGA can also be of a general type, which greatly saves the cost of the system.

[0049] However, during band-pass sampling, the analog input bandwidth of the ADC chip needs to be greater than the maximum frequency component of the heterodyne center frequency signal, otherwise the sampled and quantized signal will be distorted, and the distorted digital signal cannot demodulate the fiber vibration phase information. The device provided in this embodiment can obtain the sampling frequency of the ADC in the first data acquisition processor according to the center frequency and bandwidth of the output electrical signal of the integrated coherent receiver, so that the sampling rate is greater than twice the signal bandwidth and there is no spectral aliasing.

[0050] When performing band-pass sampling, if the lower limit of the center frequency is Fl and the upper limit is Fh, the ADC sampling frequency Fs in the first data acquisition processor satisfies:

[0051] 2Fh / (m + 1) ≤ Fs ≤ 2Fl / m;

[0052] Wherein, Fs is the sampling frequency of the first data acquisition processor; Fl is the lower limit frequency of the electrical signal, Fh is the upper limit frequency of the electrical signal; m is a positive integer and satisfies m ≤ [Fl / (Fh - Fl)], [] represents rounding down. In particular, when Fh is an integer multiple of the bandwidth B = Fh - Fl, the minimum required sampling frequency min{Fs} = 2B; if it is not an integer multiple, let Fh = mB + Kb, where k is an integer, 0 < m < 1, then min{Fs} = 2B(1 + (m / k)).

[0053] In a certain actual scenario, the center frequency of the electrical signal that needs to be collected and demodulated is 80MHz and the bandwidth is 10kHz. When the sampling rate is more than twice the signal bandwidth, the vibration signal demodulation error rate can be stabilized below 0.3%, thereby achieving a better demodulation effect.

[0054] The lowest sampling rate of the system can be obtained through the bandpass sampling calculation formula, but in actual system operation, too low a sampling rate will lead to a low signal-to-noise ratio of the sampling system and reduce the accuracy of event positioning in the Φ-OTDR. Therefore, it is also necessary to determine the required sampling rate based on the design indicators of the system, and increase the sampling rate as much as possible while meeting the bandpass sampling requirements to improve the signal-to-noise ratio of the sampling system. In common usage scenarios, the sampling rate of the first data acquisition processor ranges from 10MSPS to 100MSPS, and the sampling rate of the second data acquisition processor ranges from 125MSPS to 2GSPS.

[0055] In the specific implementation, the optical fiber sensing system is as follows: Figure 2 As shown, the optical fiber sensor includes: a narrow linewidth fiber laser, a fiber coupler, an acousto-optic modulator, a pulse generator, an optical filter, an acousto-optic modulation driver, an erbium-doped fiber amplifier, a fiber circulator, a sensing fiber and an integrated coherent receiver.

[0056] The optical signal emitted by the narrow linewidth fiber laser is divided into detection light and intrinsic light by the fiber coupler. The narrow linewidth fiber laser is used as a detection light source to emit highly coherent continuous detection light. The fiber coupler is used to split the light emitted by the narrow linewidth laser into two paths according to a preset splitting ratio, one of which is sent to the acousto-optic modulator as detection light, and the other is sent to the integrated coherent receiver as intrinsic light. In a specific scenario, the operating wavelength of the narrow linewidth fiber laser is 1550nm, the linewidth is less than 100KHz, and the optical power ratio of the detection light and the intrinsic light separated by the fiber coupler is 50%:50%.

[0057] The pulse generator is used to generate a modulated pulse, which needs to be input into the acousto-optic modulation driver module, and its control signal is generated by the first data acquisition processor or the second data acquisition processor according to actual needs. The acousto-optic modulation driver is used to generate a radio frequency signal of the modulated pulse, and the radio frequency signal is controlled by the modulated pulse input by the pulse generator. The acousto-optic modulator converts the continuous detection light into a modulated pulse detection light, and under the control of the acousto-optic modulation driver, the detection light is frequency-shifted and modulated according to the frequency shift amount. In a specific implementation, the first data acquisition processor or the second data acquisition processor is used to control the pulse generator to generate the required modulation pulse, and the acousto-optic modulation driver is used to control the acousto-optic modulator to modulate the detection light based on the modulation pulse to generate modulated light.

[0058] The modulated light enters the first port of the optical fiber circulator through the erbium-doped optical fiber amplifier, and then enters the sensing optical fiber through the second port of the optical fiber circulator.

[0059] The Rayleigh backscattered light generated by the modulated light in the sensing optical fiber is returned from the second port of the optical fiber circulator, and the returned Rayleigh backscattered light is output from the third port of the optical fiber circulator to the optical filter to obtain the filtered returned light. The optical filter filters the Rayleigh backscattered light returned from the optical fiber circulator to filter out wavelengths other than the operating wavelength of the narrow linewidth fiber laser.

[0060] The return light and the intrinsic light are coherently mixed in the integrated coherent receiver, and the mixed optical signal is used as the sensing signal.

[0061] In this embodiment, the optical fiber coupler and balanced photodetector in the prior art are not used for data collection, but an integrated coherent receiver (ICR) is used. The integrated coherent receiver includes two pairs of balanced photodetectors, which convert the sensing signal into two corresponding electrical signals through the two balanced photodetectors, and output the two electrical signals to the data acquisition processor after transimpedance amplification. In a specific implementation, the Rayleigh backscattered light returned by the optical fiber circulator passes through an optical filter to generate return light, and the return light beats with the intrinsic light through an optical mixer in the ICR, and outputs two electrical signals through two balanced photodetectors, and then the first data acquisition processor and the second data acquisition processor collect the two electrical signals separately or synchronously. In actual implementation, the center frequency of the output signal of the integrated coherent receiver is the same as the radio frequency frequency of the acousto-optic modulation driver, and the bandwidth of the output signal of the integrated coherent receiver is the same as the inverse of the pulse width of the pulse generator.

[0062] In actual implementation, the data acquisition processor is used to sample the electrical signal output by the integrated coherent receiver, and then obtain the phase change information of the Rayleigh backscattered light after a series of signal processing such as bandpass filtering, Hilbert transform, data sampling, inverse tangent calculation, unwrapping processing and phase difference calculation on the sampled digital signal. In the device provided in this embodiment, the two data acquisition processors also improve the Φ-OTDR phase demodulation algorithm through a specific algorithm structure, and use fewer resources to obtain the same or better phase demodulation effect.

[0063] In actual implementation, when the optical fiber sensor is a Φ-OTDR optical fiber sensor, the monitoring result can be expressed as follows:

[0064]

[0065] When polarization diversity reception is used, the two electrical signals can be described as follows:

[0066]

[0067] in, is the phase of the sensor signal in the x direction when the disturbance occurs, and The sensing signal is y The phase of the directional component; w c is the carrier angular frequency, θ x is the initial phase of the x-direction component of the intrinsic light pulse, θ y The initial phase of the y-direction component of the intrinsic light pulse.

[0068] In general scenarios, it can be considered Since the two electrical signals come from the same pulse signal, θ x (t) = θ y (t), let Since the collected signal is essentially generated by the superposition of multiple frequency band signals, the signal strength of signals in different frequency bands will be different in space. Therefore, in the device provided in this embodiment, the first data acquisition and processing module and the second data acquisition and processing module both use bandpass filters to perform bandpass filtering on the original signal, and merge the data of all frequency bands into one channel for phase calculation after being processed by corresponding algorithms to reduce signal fading.

[0069] Assuming that there are multiple independent frequency bands symmetrical with the center frequency within the bandwidth of the original signal, interference fading and polarization fading can be effectively eliminated by comprehensive processing of the signals of these independent frequency bands, thereby ensuring the accuracy of phase demodulation.

[0070] The first data acquisition processor and the second data acquisition processor may both include a plurality of bandpass filters, but each bandpass filter has the same passband width and order but different center frequencies, so as to filter different independent frequency bands respectively to obtain filtered signals after bandpass filtering. In a specific implementation, different convolution parameters may be set for the bandpass filters so that the bandpass filters can process independent frequency bands with different center frequencies.

[0071] In a specific scenario, three independent frequency bands with the same frequency range and different center frequencies are randomly selected. The center frequencies of each independent frequency band are: 1 =w c0 -Δw c , w 2 =w c0 and w 3 =w c0 +Δw c When performing bandpass filtering, three bandpass filters with different convolution parameters can be used to complete the bandpass filtering of each independent frequency band.

[0072] The scattering curves of the three sets of filtered signals can be expressed as:

[0073]

[0074]

[0075]

[0076]

[0077]

[0078]

[0079] The above scattering curve can be described as follows using polar coordinates:

[0080]

[0081] It can be seen from the above expression that when phase demodulation is performed, the signal phases between the optical pulses in the sensing signal are independent of each other. Therefore, before phase demodulation is performed, the first frame in each time period is used as a reference frame, and the phase values ​​of other frames in this time period are obtained. The signal phase of the current sensing signal is reset to zero according to the phase values ​​of other frames and the phase value of the reference frame in this time period, thereby completing the phase correction of the current frame signal. In actual implementation, the reference frame signal and the signal of each frame can be complex conjugate multiplied at the corresponding sampling position to complete the phase correction of the current frame signal.

[0082] Assume that a certain period range is [t 0 ,t 0 +T],t 0 The complex conjugate curve expression of the moment is as follows:

[0083]

[0084] Then t 0 The phase zeroing at the moment is obtained by multiplying the reference frame with the complex conjugate of the scattering curve of the filtered signal at this moment, and then integrating the complex conjugate multiplication results on all independent frequency bands. The phase of the reference frame, that is, t 0 The phase at a moment can be described as:

[0085]

[0086] Similarly, [t 0 ,t 0 The phase change at any time during the period of t 0 The initial phase at time is used as a reference to solve:

[0087]

[0088] Since the phase calculated above is a rotation phase, its phase variation range is [-π,π], and there is phase winding. It is also necessary to unwind the phase calculated above:

[0089]

[0090] The phases calculated above are all spatially distributed phases at a certain moment. In order to accurately describe the change in phase, it is necessary to perform interval differentiation in space:

[0091]

[0092] The above formula is the spatial phase distribution at time t, where n 0 is the spatial difference interval.

[0093] In order to reduce the output data volume of phase demodulation, the demodulation results can be spatially merged. The general principle of merging is that the phases within a pulse width are merged into one point:

[0094]

[0095] The above formula is the spatial phase distribution at time t after the final merging. Among them, n is the unit of spatial merging length, and L is the maximum integer multiple of the spatial merging length.

[0096] In the demodulation algorithm corresponding to the above data processing process, the processing process of each frame of data is as follows: Figure 3 shown. Figure 3 In the process, after sampling the two electrical signals to obtain digital signals, each processing process of digital signal demodulation is as follows: Processing 1, bandpass filtering is performed on each frequency band respectively, Figure 3 In the process, three bandpass filters with different convolution parameters are used to complete bandpass filtering; in process 2, the filtered signal is subjected to Hilbert transform; in process 3, the reference frame is complex conjugated; in process 4, the complex conjugate result of the reference frame is complex conjugated with the filtered signal after Hilbert transform; in process 5, the real part and the imaginary part of each path of the complex conjugate multiplication result are added separately; in process 6, the positive and negative tangents are calculated for the addition result to obtain the angle representation of the phase; in process 7, two adjacent phase values ​​are differentiated, and the differential result is phase unwrapped, and the differential result is used as the demodulation result.

[0097] Based on the demodulation algorithm, the corresponding data processing modules in the first data acquisition processor and the second data acquisition processor can be selected and designed. The following provides a set of available data processing module implementation methods for each data acquisition processor. In actual implementation, corresponding software and hardware design can be carried out according to actual needs.

[0098] like Figure 4 As shown, the first data acquisition processor includes: a low-speed analog-to-digital conversion module, a bandpass filter, a Hilbert transform module, an interval sampling module, a complex conjugate multiplication module, an inverse tangent processing module, and an unwinding and adjacent phase difference merging processing module, each of which is connected in sequence and used to perform corresponding data processing processes in sequence. The low-speed analog-to-digital conversion module is used to sample two electrical signals, and send the sampled digital signals to the bandpass filter for bandpass filtering; the Hilbert transform module is used to perform Hilbert transform on the signal after bandpass filtering; the interval sampling module is used to extract fixed point data of a specified interval from the transformed data; the complex conjugate multiplication module is used to perform complex conjugate multiplication of the fixed point data and the reference frame of this cycle; the inverse tangent processing module is used to perform inverse tangent processing on the complex conjugate multiplication result; the unwinding and adjacent phase difference merging processing module is used to perform adjacent phase differentiation on the inverse tangent processing result, and perform phase unwinding of the data after phase differentiation, so as to complete the demodulation of the digital signal obtained by bandpass sampling.

[0099] In actual implementation, when digital signal processing is performed, the order of the bandpass filter remains unchanged, and its corresponding filter coefficient can be dynamically configured according to the actual scene requirements. When digital signal processing is performed, the interval sampling module performs spatial sampling on the result processed by the Hill transform module, and the number of sampling intervals can be dynamically configured.

[0100] In this embodiment, when using the reference frame for deflection correction, the first frame of data in this cycle can be used as the reference frame. In the first data acquisition processor, the data has been extracted at intervals, so the amount of fixed point data in the first frame is not large. Therefore, the block random access memory (BRAM) module in the FPGA chip can be directly used to store the reference frame and output it to the complex conjugate multiplication module to complete the complex conjugate multiplication of the reference frame and each frame of data without using an additional storage module such as an external DDR. Figure 5 As shown, the block storage module is used to receive the fixed point data extracted by the interval sampling module, store the first frame of fixed point data of each cycle as a reference frame, and input the reference frame to the complex conjugate multiplication module

[0101] Similarly, although the second data acquisition processor uses a high-speed ADC, it can also use a demodulation algorithm under a bandpass sampling mode and the phase demodulation algorithm provided in this embodiment. On the other hand, in order to ensure data accuracy, the second data acquisition processing module does not perform interval point extraction. In order to reduce the amount of data processing after unwrapping, it is also necessary to evenly group the unwrapped data and calculate the average phase difference of each group of data, and use the average phase difference as the result of phase demodulation. Figure 6As shown, the second data acquisition processor includes: a high-speed analog-to-digital conversion module, a first bandpass filter, a first Hilbert transform module, a complex conjugate multiplication module, an inverse tangent processing module, an unwrapping processing module, a single-frame interval phase difference processing module and a phase difference merging and averaging module, wherein the high-speed analog-to-digital conversion module is used to sample two electrical signals and send the sampled digital signals to the first bandpass filter for bandpass filtering; the first Hilbert transform module is used to perform Hilbert transform on the signal after bandpass filtering; the complex conjugate multiplication module is used to perform complex conjugate multiplication on the transformed data and the reference frame of this cycle; the inverse tangent processing module is used to perform inverse tangent processing on the complex conjugate multiplication result; the single-frame interval phase difference processing module sequentially performs the difference between two adjacent unwrapped data and sequentially outputs the differential data; the phase difference merging and averaging module is used to sequentially average the fixed point data for the data output by the single-frame interval phase difference processing module to complete the phase demodulation of the digital signal obtained by low-pass sampling.

[0102] When using high-speed ADC to collect data for processing, in order to ensure positioning accuracy, fixed-point data is not extracted. When using reference frames for correction, the number of reference frames will also be relatively large. At the same time, after the Hilbert transform, the bit width of the data will increase, resulting in a larger amount of data. Therefore, when storing the reference frame, the original sampling value is directly stored to obtain the reference frame, rather than the reference frame after the Hilbert transform, in order to reduce the space occupied by data storage, and then the reference frame is bandpass filtered and Hilbert transformed separately to obtain the transformed data. Therefore, if Figure 7 As shown, the block storage module is used to receive the sampling data of the high-speed analog-to-digital conversion module, store the first frame of fixed-point data of this cycle as a reference frame, and output the reference frame to the second bandpass filter for bandpass filtering; the second Hilbert transform module is used to perform Hilbert transform on the reference frame after bandpass filtering, and output the reference frame after Hilbert transform to the complex conjugate multiplication module. In actual implementation, the first bandpass filter module and the second bandpass filter module need to use the same set of convolution parameters, and the first Hilbert transform module and the second Hilbert transform module also need to use the same set of convolution parameters.

[0103] In the first data acquisition processor and the second data acquisition processor, if Figure 8 As shown, the inverse tangent processing module can use a similar structure, including real and imaginary part addition components, real and imaginary part normalization components, inverse tangent calculation components and inverse tangent calculation result processing components. Each component is connected in sequence to perform corresponding data processing processes in sequence to complete the inverse tangent processing in the demodulation process.

[0104] In a certain actual scenario, the center frequency of the electrical signal is 80MHz and the bandwidth is 5MHz. If the low-pass sampling method with a sampling rate of 250MHz is used for sampling, and a conventional I / Q demodulation algorithm is used, the logic resource of the FPGA is about 350K; while the bandpass sampling at a sampling rate of 90MHz provided in this embodiment is used, and the demodulation algorithm provided in this embodiment is used, the resource consumption of the FPGA is about 120K.

[0105] The demodulation device of the phase-sensitive optical time-domain reflectometer provided in this embodiment uses a low-speed ADC and reduces the sampling rate of the acquisition system according to a bandpass sampling method, so as to reduce the cost and power consumption of the ADC chip and reduce the data processing burden of the FPGA; a lower sampling rate corresponds to a lower FPGA processing clock, which is conducive to the development of FPGA layout and routing and the achievement of timing convergence; at the same time, an optimized demodulation algorithm is used to design the corresponding data processing module, so that the FPGA consumes less resources during demodulation, which is more conducive to the FPGA timing convergence and power consumption reduction.

[0106] Embodiment 2:

[0107] Based on the demodulation device of the phase-sensitive optical time-domain reflectometer provided in the above embodiment 1, the present invention further provides a demodulation method of the phase-sensitive optical time-domain reflectometer that can be used to implement the above method.

[0108] like Fig. 9 As shown, the specific steps of the demodulation method provided by the embodiment of the present invention are as follows:

[0109] Step 101: Use an optical fiber sensor to obtain a sensing signal, and convert the sensing signal into two electrical signals through an integrated coherent receiver.

[0110] The integrated coherent receiver of the optical fiber sensor will beat the return light with the intrinsic light, and then output two electrical signals through two balanced photodetectors. Therefore, both the first data acquisition processor and the second data acquisition processor will sample two electrical signals. In actual implementation, only one of the data acquisition processors can be selected for signal demodulation as needed, or the two data acquisition processors can be switched according to the situation to meet the requirements of power consumption and accuracy at the same time.

[0111] Step 102: When the number of abnormal sensing events does not exceed the threshold, the first data acquisition processor is used to sample and demodulate the two-channel electrical signals.

[0112] In general use scenarios of fiber optic sensing, long-term continuous sensing monitoring is usually required, and there may not be a stable power supply. Therefore, it is necessary to use a method with lower power consumption for monitoring to reduce the demand for power consumption. On the other hand, in the long-term continuous monitoring process, in most cases, the sensing optical fiber will not experience abnormal sensing events such as vibration, breakage, torsion, etc., and there is no need to use high-precision calculations to accurately locate abnormal sensing events. Therefore, in ordinary scenarios, the first data acquisition processor is used to perform daily monitoring through a small amount of data extracted from fixed points to reduce the power consumption of sampling and demodulation.

[0113] Step 103: When the number of abnormal sensing events exceeds a threshold, a second data acquisition processor is used to sample and demodulate two channels of electrical signals.

[0114] When abnormal sensing events such as vibration, breakage, and torsion occur multiple times on the sensing optical fiber, it indicates that the abnormal sensing event is not an accidental event or a misjudgment, and the abnormal sensing event needs to be located to solve the problem caused by the abnormal monitoring object corresponding to the abnormal sensing event. At this time, in order to improve the positioning accuracy, a second data acquisition processor with higher accuracy needs to be used for sampling and demodulation. After the abnormal sensing event is processed, it can be switched back to the first data acquisition processor to continue daily monitoring. In actual implementation, the switching can be completed manually according to specific needs, or it can be automatically completed according to the preset abnormal sensing event number threshold.

[0115] After steps 101 to 104 provided in this embodiment, the demodulation device in embodiment 1 can be used to complete sampling and demodulation in different ways in different scenarios, thereby reducing the performance requirements and power consumption of the FPGA.

[0116] like Fig.10 As shown, when the first data acquisition processor in the demodulation device is used, the specific steps of the demodulation method provided by the embodiment of the present invention are as follows:

[0117] Step 201: Input each sampling signal into at least two bandpass filters for filtering.

[0118] The first data acquisition processor includes a plurality of bandpass filters, each of which has the same passband width and order, but different passband center frequencies. In general scenarios, three bandpass filters are usually used to convert one electrical signal into three filtered signals, and the other electrical signal is subjected to the same bandpass filtering process.

[0119] Step 202: Perform Hilbert transform on the filtered signal, perform fixed point extraction based on the Hilbert transform result, use the first frame of fixed point data in each period as the reference frame of the period, and perform complex conjugate multiplication of each frame of fixed point data after extraction and the reference frame to obtain a set of real data and a set of imaginary data.

[0120] The Hilbert transform module performs Hilbert transform on the filtered data and rotates the phase of the filtered data by 90 degrees to reduce the I / Q demodulation calculation cost. In the specific implementation, the Hilbert transform process is as follows: Fig.11 shown.

[0121] In the first data acquisition processor, the interval sampling module is first used to perform interval sampling on the data after Hilbert transformation, and then the complex conjugate multiplication module is used to perform complex conjugate multiplication on the extracted fixed point data. In order to ensure data correspondence, the data of the reference frame also uses the extracted fixed point data. In a specific implementation, the first frame data obtained by sampling each period can be used as the reference frame of the period, and the reference frame and the current frame are complex conjugate multiplied to complete the phase correction of the current frame.

[0122] After complex conjugate multiplication, each filtered signal will obtain the corresponding real part and imaginary part, and the angle between the real part and the imaginary part is the phase information.

[0123] Step 203: Add all the real data to obtain the first data path, add all the imaginary data to obtain the second data path, divide the first data path by the second data path, and perform an inverse tangent calculation on the quotient.

[0124] For the real and imaginary data obtained by complex conjugate multiplication, use the real and imaginary addition components to add the real data into one data path and the imaginary data into one data path. Then divide the added real part by the imaginary part, use the inverse tangent calculation component to perform inverse tangent calculation on the result, and use the inverse tangent calculation result processing component to process the inverse tangent calculation result.

[0125] Step 204: perform differential processing on adjacent data of the inverse tangent result, perform deconvolution processing on the differential data, and use the phase difference information obtained by deconvolution as the demodulation result.

[0126] In general scenarios, when using phase difference information, in order to avoid sudden noise interference, the unwrapped data is usually divided into uniform data segments, each data segment contains the same amount of data, and then the phase difference average of the data in adjacent data segments is calculated, and the phase difference average of two adjacent data segments is used as the phase difference. However, in the first data acquisition processor, since the interval point extraction has been performed, each fixed point data actually corresponds to the characteristics of a data segment, and accordingly, after unwrapping, the data difference between two adjacent data actually corresponds to the phase difference between two adjacent data segments. Therefore, in the first data acquisition processor, there is no need to perform the average value calculation, but directly perform the difference on the extracted fixed point data, and then use the unwrapping and adjacent phase difference merging processing module to unwrap the difference result to obtain the required phase difference data.

[0127] After step 201 to step 204 provided in this embodiment, the demodulation device in embodiment 1 can be used to perform bandpass sampling, and demodulation is completed by a new demodulation algorithm, thereby reducing the performance requirements and power consumption of the FPGA.

[0128] like Fig.12 As shown, when the second data acquisition processor in the demodulation device is used, the specific steps of the demodulation method provided by the embodiment of the present invention are as follows:

[0129] Step 301: Input each sampling signal into at least two bandpass filters for filtering, and perform Hilbert transform on the filtered signal to obtain a filtered signal after Hilbert transform.

[0130] The second data processing module will also sample two electrical signals, and then perform bandpass filtering and Hilbert transform on the two electrical signals in parallel. In a specific implementation, three bandpass filters can be used, each of which has the same passband width and order, but different passband center frequencies, and each electrical signal is divided into three filtered signals through the three bandpass filters.

[0131] Step 302: The first frame data of the two electrical signals in each cycle is used as the reference frame of the cycle, the reference frame is bandpass filtered and Hilbert transformed to obtain the reference frame after Hilbert transformation, and each frame of filtered signal after Hilbert transformation is complex conjugate multiplied with the reference frame after Hilbert transformation to obtain a set of real data and a set of imaginary data.

[0132] In the second data acquisition processor, fixed point extraction is not performed, so the amount of data in each frame is large, and the Hilbert transform will further increase the space occupied by the data. Therefore, for the data of the reference frame, the Hilbert transform is not performed first, and it is directly stored in the block storage module, and then the reference frame is bandpass filtered and Hilbert transformed separately. After the filter signal and the reference frame signal are Hilbert transformed respectively, the complex conjugate multiplication module is used to perform complex conjugate multiplication on the filter signal and the reference frame.

[0133] After complex conjugate multiplication, each filtered signal will obtain the corresponding real part and imaginary part, thereby obtaining the required phase information.

[0134] Step 303: Add all the real data to obtain the first data path, add all the imaginary data to obtain the second data path, divide the first data path by the second data path, and perform an inverse tangent calculation on the quotient.

[0135] For the real and imaginary data obtained by complex conjugate multiplication, use the real and imaginary addition components to add the real data into one data path and the imaginary data into one data path. Then divide the added real part by the imaginary part, use the inverse tangent calculation component to perform inverse tangent calculation on the result, and use the inverse tangent calculation result processing component to process the inverse tangent calculation result.

[0136] Step 304: perform deconvolution processing on the result of the arc tangent, obtain the phase difference of adjacent data for the deconvolution data, and obtain the average value of the phase differences of a specified number, and use the average value of the phase differences as the demodulation result.

[0137] In the second data acquisition processor, interval point extraction is not performed. After the deconvolution processing module is used to deconvolve the result of the inverse tangent, it is also necessary to use the single-frame interval phase difference processing module to divide the data into uniform data segments, and use the phase difference merging and averaging module to calculate the average phase difference of adjacent data segments, and use the difference between adjacent data segments as phase difference information.

[0138] After steps 301 to 304 provided in this embodiment, the demodulation device in embodiment 1 can be used to complete demodulation through a new demodulation algorithm under low-pass sampling, and the performance requirements and power consumption of the FPGA can be reduced through the optimized demodulation algorithm.

[0139] Embodiment 3:

[0140] Based on the demodulation device of the phase-sensitive optical time-domain reflectometer provided in Example 1 and the demodulation method of the phase-sensitive optical time-domain reflectometer provided in Example 2, in some specific implementations, they can be implemented through the specific implementations in this embodiment. It can be understood that the specific implementations provided in this embodiment are only used to illustrate the specific implementation process of the method in Example 1 in certain specific scenarios, and are not intended to limit the scope of protection.

[0141] In a specific implementation scenario, the bandpass sampling result can meet the demodulation requirement, and the first data acquisition processor is used for demodulation.

[0142] According to the current optical path system, the frequency of the RF of the acousto-optic modulation driver and the pulse width of the pulse generator, the center frequency and bandwidth of the input signal of the acquisition card can be obtained. Among them, the center frequency of the RF signal output by the integrated coherent receiver is the frequency of the RF of the acousto-optic modulation driver, and its bandwidth is the inverse of the pulse width of the pulse generator. In this scenario, the frequency of the RF of the acousto-optic modulation driver is 80MHz, and the pulse width of the pulse generator is 200ns. Then the center frequency of the input signal in the first data acquisition processor is 80MHz, and its bandwidth is 5MHz.

[0143] According to the center frequency and bandwidth of the input signal, the center frequency and bandwidth of the integrated coherent receiver output signal are obtained through theoretical calculation. According to the theoretical calculation results, the RF frequency of the acousto-optic modulation driver and the pulse width of the pulse generator are configured, and the center frequency and bandwidth of the integrated coherent receiver output signal are actually tested. If the actual test results do not match the theoretical calculations, the hardware system and software parameter configurations need to be checked.

[0144] When the center frequency and bandwidth of the output signal of the integrated coherent receiver in the actual test are consistent with the theory, the appropriate sampling rate is calculated according to the bandpass sampling formula. Considering the problem that the low sampling rate of the ADC chip leads to large quantization noise in the sampling system, and the high FPGA processing clock leads to large resource consumption and difficulty in timing convergence, the sampling frequency of the bandpass sampling is obtained by comprehensive consideration. In this scenario, the center frequency of the input signal is 80MHz and the bandwidth is 5MHz. Taking comprehensive consideration, the sampling rate can be set to 90MHz.

[0145] After setting the sampling rate, the FPGA in the first data acquisition processor is used to perform algorithm processing on the sampled digital signal, and its processing flow is the same as the demodulation method in Example 2. Figure 5 As shown, according to the flow of data inside the FPGA, the data mainly passes through a bandpass filter, a Hilbert transform module, an interval sampling module, a complex conjugate multiplication module, an inverse tangent calculation module, and a phase unwrapping and adjacent phase differential merging processing module.

[0146] In the interval sampling module, the number of intervals is equal to the number of merged points. The number of merged points determines the positioning accuracy of the vibration event. The positioning accuracy is equal to the spatial resolution of the sampling system multiplied by the number of merged points.

[0147] After completing the phase unwrapping and adjacent phase differential merging processing, the demodulated phase change information can be obtained. The phase change information can then be sent to an external processing system through a transmission interface, such as a Gigabit Ethernet port, a 10 Gigabit Ethernet port, or PCIE, for further pattern recognition processing.

[0148] In another specific implementation scenario, the bandpass sampling result cannot meet the demodulation requirement, and low-pass sampling is required, and a second data acquisition processor is used for demodulation.

[0149] In this scenario, two high-performance ADC sampling channels are used for synchronous sampling to obtain two digital signals of polarization state X and polarization state Y. The FPGA in the second data acquisition processor is used to perform algorithmic processing on the sampled digital signals, and the two sampled digital signals are synchronously bandpass filtered, Hilbert transformed, and complex conjugate multiplied. All corresponding real numbers obtained after complex conjugate multiplication are added as one channel I, and all corresponding imaginary data are added as one channel Q, and then the arctangent value of I / Q, arctan(I / Q), is solved. Figure 7 As shown, according to the flow direction of data inside the FPGA, it mainly passes through a bandpass filter, a Hilbert transform module, a complex conjugate multiplication module, an inverse tangent calculation module, a deconvolution module, a single frame interval phase difference processing module and a phase difference merging and averaging module.

[0150] In the single-frame interval phase difference processing module, the input current frame is divided into two paths, one path is used as a reference frame through a shift register with a configurable delay parameter (i.e., the number of data in each data segment), and the other path is a normal path, while counting the number of data in a single frame. When the count value reaches the interval difference value, the phase difference between the two adjacent data segments can be obtained by subtracting the data of the normal path from the data output by the shift register, and this phase difference is sent to the phase difference merging and averaging module.

[0151] After the results of the single-frame interval phase difference processing module are processed according to the configurable number of points, the phase difference merging and averaging module accumulates the processed data in segments according to the configured number of points, and then divides it by the configured number of points to obtain the average value of the phase difference between two adjacent data segments. An average value is output for every configured number of points, and this average value is the phase change information between the two data segments after demodulation.

[0152] After completing the data processing of the phase difference merging and averaging module, the demodulated phase change information can be obtained. The phase change information can then be sent to an external processing system through a transmission interface, such as a Gigabit Ethernet port, a 10 Gigabit Ethernet port, or PCIE, for further pattern recognition processing.

[0153] The above description is only a preferred embodiment of the present invention and is not intended to limit the present invention. Any modifications, equivalent substitutions and improvements made within the spirit and principles of the present invention should be included in the protection scope of the present invention.

Claims

1. A demodulation device for a phase-sensitive optical time-domain reflectometer, characterized in that: include: An optical fiber sensor, a first data acquisition processor and a second data acquisition processor; The sensing signal obtained by the optical fiber sensor enters the first data acquisition processor in the form of two electrical signals or enters the second data acquisition processor in the form of two electrical signals; The first data acquisition processor is used to convert the two-way electrical signals into corresponding low-speed digital signals by using a bandpass sampling method, perform bandpass filtering and Hilbert transform on the low-speed digital signals in sequence, perform fixed-point sampling on the digital signals after the Hilbert transform, and perform corresponding demodulation on the fixed-point data obtained by sampling; The second data acquisition processor is used to convert the two-way electrical signals into corresponding high-speed digital signals by using a low-pass sampling method, and to perform corresponding demodulation on the high-speed digital signals.

2. The demodulation device of phase-sensitive optical time-domain reflectometry according to claim 1, characterized in that: The sampling frequency of the first data acquisition processor meets the following conditions: 2Fh / (m+1)≤Fs≤2Fl / m; Wherein, Fs is the sampling frequency of the first data acquisition processor; Fl is the lower limit frequency of the electrical signal, and Fh is the upper limit frequency of the electrical signal; m is a positive integer and satisfies m≤[Fl / (Fh-Fl)], and [] indicates rounding down.

3. The demodulation device of phase-sensitive optical time-domain reflectometry according to claim 1, characterized in that: The optical fiber sensor comprises: a narrow linewidth optical fiber laser, an optical fiber coupler, an acousto-optic modulator, a pulse generator, an optical filter, an acousto-optic modulation driver, an erbium-doped optical fiber amplifier, an optical fiber circulator, a sensing optical fiber and an integrated coherent receiver, specifically comprising: The optical signal emitted by the narrow linewidth fiber laser is divided into detection light and intrinsic light through the fiber coupler; The first data acquisition processor or the second data acquisition processor is used to control the pulse generator to generate the required modulation pulse, and the acousto-optic modulation driver is used to control the acousto-optic modulator to modulate the detection light based on the modulation pulse to generate modulated light; The modulated light enters the first port of the optical fiber circulator through the erbium-doped optical fiber amplifier, and then enters the sensing optical fiber through the second port of the optical fiber circulator; The Rayleigh backscattered light generated by the modulated light in the sensing optical fiber is returned from the second port of the optical fiber circulator, and the returned Rayleigh backscattered light is output to the optical filter from the third port of the optical fiber circulator to obtain the filtered returned light; The return light and the intrinsic light are coherently mixed in the integrated coherent receiver, and the mixed optical signal is used as the sensing signal; The integrated coherent receiver includes two pairs of balanced photodetectors, which convert the sensing signals into two corresponding electrical signals through the two balanced photodetectors, and then output the two electrical signals to the data acquisition processor after transimpedance amplification.

4. The demodulation device of phase-sensitive optical time-domain reflectometry according to claim 1, characterized in that: The first data acquisition processor includes: a low-speed analog-to-digital conversion module, a bandpass filter, a Hilbert transform module, an interval sampling module, a complex conjugate multiplication module, an inverse tangent processing module and a deconvolution and adjacent phase difference merging processing module, specifically including: The low-speed analog-to-digital conversion module is used to sample two electrical signals and send the sampled digital signals to a bandpass filter for bandpass filtering; the Hilbert transform module is used to perform Hilbert transform on the signal after bandpass filtering; the interval sampling module is used to extract fixed-point data of a specified interval from the data after Hilbert transform; the complex conjugate multiplication module is used to perform complex conjugate multiplication on the fixed-point data and the reference frame of this cycle; the inverse tangent processing module is used to perform inverse tangent processing on the complex conjugate multiplication result; the unwrapping and adjacent phase difference merging processing module is used to perform adjacent phase differentiation on the inverse tangent processing result, and perform phase unwrapping of the data after phase differentiation, so as to complete the demodulation of the digital signal obtained by bandpass sampling.

5. The demodulation device of phase-sensitive optical time-domain reflectometry according to claim 4, characterized in that: The first data acquisition processor also includes a block storage module, which is used to receive the fixed-point data extracted by the interval sampling module, store the first frame of fixed-point data in each cycle as a reference frame, and input the reference frame to the complex conjugate multiplication module.

6. The demodulation device of phase-sensitive optical time-domain reflectometry according to claim 1, characterized in that: The second data acquisition processor includes: a high-speed analog-to-digital conversion module, a first bandpass filter, a first Hilbert transform module, a complex conjugate multiplication module, an inverse tangent processing module, an unwrapping processing module, a single frame interval phase difference processing module and a phase difference merging and averaging module, specifically including: The high-speed analog-to-digital conversion module is used to sample two electrical signals, and send the sampled digital signals to the first bandpass filter for bandpass filtering; the first Hilbert transform module is used to perform Hilbert transform on the signal after bandpass filtering; the complex conjugate multiplication module is used to perform complex conjugate multiplication on the transformed data and the reference frame of the current cycle; the inverse tangent processing module is used to perform inverse tangent processing on the result of the complex conjugate multiplication; the unwrapping processing module is used to perform phase unwrapping on the result of the inverse tangent processing; the single-frame interval phase difference processing module sequentially performs the difference between two adjacent unwrapped data, and sequentially outputs the differential data; the phase difference merging and averaging module is used to sequentially calculate the average value of the fixed point data for the data output by the single-frame interval phase difference processing module, so as to complete the phase demodulation of the digital signal obtained by low-pass sampling.

7. The demodulation device of phase-sensitive optical time-domain reflectometry according to claim 6, characterized in that: The second data acquisition processor further includes a block storage module, a second bandpass filter and a second Hilbert transform module, specifically including: The block storage module is used to receive sampling data from the high-speed analog-to-digital conversion module, store the first frame of fixed-point data in this cycle as a reference frame, and output the reference frame to the second bandpass filter for bandpass filtering; the second Hilbert transform module is used to perform Hilbert transform on the reference frame after bandpass filtering, and output the reference frame after Hilbert transform to the complex conjugate multiplication module.

8. A demodulation method for a phase-sensitive optical time-domain reflectometer, characterized in that: The method is completed using the demodulation device of the phase-sensitive optical time-domain reflectometer according to any one of claims 1 to 7, specifically comprising: The sensing signal is obtained by using an optical fiber sensor, and the sensing signal is converted into two electrical signals by an integrated coherent receiver; When the number of abnormal sensing events does not exceed the threshold, the first data acquisition processor is used to sample and demodulate the two-channel electrical signals; When the number of abnormal sensing events exceeds a threshold, a second data acquisition processor is used to sample and demodulate two channels of electrical signals.

9. The demodulation method of phase-sensitive optical time-domain reflectometry according to claim 8, characterized in that: The sampling and demodulation of the two-channel electrical signals using the first data acquisition processor specifically includes: Input each sampling signal into at least two bandpass filters for filtering; wherein each bandpass filter has the same passband width and order, but different passband center frequencies; Perform Hilbert transform on the filtered signal, perform fixed point extraction on the result after Hilbert transform, use the first frame of fixed point data of each cycle as the reference frame of the cycle, perform complex conjugate multiplication on each frame of fixed point data after extraction and the reference frame to obtain a set of real part data and a set of imaginary part data; Add all the real data to obtain the first data path, add all the imaginary data to obtain the second data path, divide the first data path by the second data path, and perform an inverse tangent calculation on the quotient; The result of the arc tangent is used to differentiate adjacent data, and the differentiated data is unwrapped, and the phase difference information obtained by unwrapping is used as the demodulation result.

10. The demodulation method of phase-sensitive optical time-domain reflectometry according to claim 8, characterized in that: The sampling and demodulation of the two-channel electrical signals using the second data acquisition processor specifically includes: Input each sampling signal into at least two bandpass filters for filtering, and perform Hilbert transform on the filtered signal to obtain a filtered signal after Hilbert transform; wherein each bandpass filter has the same passband width and order, but different center frequencies of the passband; The first frame data of the two electrical signals in each cycle is used as the reference frame of the cycle, the reference frame is subjected to bandpass filtering and Hilbert transformation to obtain a reference frame after Hilbert transformation, and each frame of filtered signal after Hilbert transformation is subjected to complex conjugate multiplication with the reference frame after Hilbert transformation to obtain a set of real data and a set of imaginary data; Add all the real data to obtain the first data path, add all the imaginary data to obtain the second data path, divide the first data path by the second data path, and perform an inverse tangent calculation on the quotient; The result of the arc tangent is unwrapped, and for the unwrapped data, the phase difference of adjacent data is obtained, and the average value of the phase differences of a specified number is obtained, and the average value of the phase differences is used as the demodulation result.

Images