Distributed fiber optic vibration sensing method based on OFDM-NLFM time series pulse modulation

Abstract

The present invention relates to a distributed optical fiber vibration sensing method based on OFDM-NLFM time series pulse modulation, and belongs to the field of optical fiber. The present invention uses OFDM-NLFM time series frequency modulation pulse signals with a wider pulse width, which can inject more light energy into the sensing optical fiber, and uses a matching filter in the reception, and the pulse signal is compressed into a narrow pulse with a high sidelobe suppression ratio, which can not only improve the signal-to-noise ratio of the sensing signal to extend the sensing distance, but also obtain a very high spatial resolution. The use of OFDM technology can increase the repetition frequency of the detection light, obtain a high frequency response bandwidth, realize more frequency band multiplexing within the limited detector bandwidth of the receiver, and improve the spectrum efficiency. The present invention uses OFDM-NLFM, which can achieve a high frequency response bandwidth under the conditions of ultra-long sensing distance and high spatial resolution.

Description

Technical Field

[0001] The invention belongs to the field of optical fibers, and in particular relates to a distributed optical fiber vibration sensing method based on OFDM-NLFM time series pulse modulation. Background Art

[0002] Vibration is one of the most common physical phenomena in nature. Research on advanced vibration sensors for efficient and high-quality detection of vibration information (amplitude, frequency, speed and acceleration) is of great significance to ensuring the safety of human activities and improving efficiency. For example, by measuring the vibration of steel bars and concrete, researchers can effectively check the health status of important buildings such as bridges, dams and high-rise buildings and prevent collapse accidents; by measuring the generation and propagation of seismic waves in soil, rocks or glaciers, early warning of certain serious geological disasters such as landslides can be achieved, and geological structures can be inferred to explore underground resources such as oil, gas and metals. By measuring the vibration distribution of pipelines, cables and rails, illegal activities such as digging and drilling that damage pipelines, cables and rails can be effectively monitored, and the location of pipeline leaks or cable failures can be determined to speed up maintenance progress, etc.

[0003] Since the 1960s, with the rapid development of technologies such as optical fiber, optical modulators, lasers, and digital signal processing, optical fiber sensing technology has also flourished and attracted widespread attention from researchers. Optical fiber sensors are new sensors that use optical fiber as a sensing element, or only as a signal transmission carrier, or both. There are many types of optical fiber sensors that can detect physical quantities such as temperature, magnetic field, vibration, etc. in a single or multi-parameter manner.

[0004] Distributed fiber optic sensing technology is an important part of the fiber optic sensing field. Its working principle is mainly to use fiber optic sensing to receive a signal and transmit and return this signal in real time. It has the following outstanding advantages: there is no need to make sensors on the optical fiber. The sensing optical fiber integrates sensing and transmission, and can realize long-distance and large-scale sensing and networking.

[0005] Distributed fiber optic sensing systems mainly modulate the backscattered light in the sensing fiber according to changes in the external environment, causing the physical information of the backscattered light in the fiber to change, thereby realizing the perception of external parameters. In 1980, Rogers proposed to monitor the intensity of the backscattered Rayleigh light in the fiber to measure the temperature distribution in the fiber. Since then, distributed fiber optic sensing technology based on scattered light signals in the fiber has achieved unprecedented development. At present, the distributed fiber optic sensing technology based on the scattering principle at home and abroad is mainly divided into the following categories: First, the distributed fiber optic sensing technology based on Brillouin scattering uses the linear change relationship between the Brillouin frequency shift of the scattered light and temperature and stress to achieve measurement. Since the spontaneous Brillouin scattered light intensity in the optical fiber is weak, and the Brillouin frequency shift is sensitive to both temperature and strain, the distributed fiber optic sensing technology based on Brillouin scattering is not suitable for long-distance dynamic monitoring; the second is the distributed fiber optic sensing technology based on Raman scattering, which uses the linear relationship between the power ratio between Stokes light and anti-Stokes light and temperature to measure the temperature change along the optical fiber. It can be used for temperature monitoring of oil pipelines, high-voltage wires, etc. The intensity of Raman scattered light in the optical fiber is also weak, and it can only measure temperature changes. The third is the distributed fiber optic sensing technology based on Rayleigh scattering. Rayleigh scattering is caused by the uneven distribution of the optical fiber medium. There is no energy exchange. The scattered light frequency is consistent with the incident light. When the external strain changes, the refractive index of the optical fiber will change, causing the Rayleigh scattered light to undergo phase modulation. The intensity of Rayleigh scattered light is higher than Raman or Brillouin scattering, and it is more suitable for dynamic measurement.

[0006] Distributed fiber optic sensing technology based on Rayleigh scattering mainly includes optical time domain reflectometer (OTDR) for fiber fault diagnosis, and polarization-sensitive optical time domain reflectometer (POTDR), coherent optical time domain reflectometer (COTDR), phase-sensitive optical time domain reflectometer (Φ-OTDR) and optical frequency domain reflectometer (OFDR) for distributed stress and vibration measurement.

[0007] Φ-OTDR was proposed by HF Taylor and others in 1993. The core of the system is a narrow linewidth light source, and the output laser is highly coherent, so that the Rayleigh scattered light also has strong coherence. When there is a disturbance affecting the optical fiber, the phase of the backscattered Rayleigh scattered light changes with the disturbance, and the internal self-interference of the backscattered light converts the scattered position change into light intensity change. Since the disturbance information is carried on the phase of the scattered light, and the optical phase is extremely sensitive to the deformation and stress of the optical fiber, the Φ-OTDR technology has extremely high detection sensitivity, excellent dynamic response capability and reliable positioning, and has important application value. After the emergence of Φ-OTDR technology, it has been widely studied and rapidly developed towards practical application and industrialization. The feasibility of Φ-OTDR for sensing at a spatial resolution of several meters at a distance far greater than 100km has been fully confirmed, and it is one of the current research hotspots.

[0008] Spatial resolution is the minimum spatial interval at which distributed fiber optic sensing can distinguish disturbances. How to improve spatial resolution is also one of the difficulties and hot spots in the research of distributed fiber optic sensing technology. In the Φ-OTDR system, spatial resolution is determined by pulse width. To obtain higher spatial resolution, the pulse width needs to be narrowed. However, narrowing the pulse width will lead to a decrease in injected light energy, which in turn leads to a decrease in signal-to-noise ratio (SNR) and sensing distance. There is a contradictory relationship between the spatial resolution and sensing distance of the traditional Φ-OTDR system, and it cannot meet the application requirements of high spatial resolution and long sensing distance at the same time.

[0009] In distributed fiber optic sensing systems, disturbances in the optical fiber are detected by monitoring the relative changes of the position signal. According to the Nyquist sampling theorem, a larger response bandwidth requires a higher detection pulse repetition frequency. Since the repetition frequency of the detection light is restricted by the length of the sensing fiber, the frequency response range of the system is limited by the length of the fiber. The larger the sensing distance, the smaller the achievable frequency response bandwidth. How to solve the contradiction between the sensing range and the response bandwidth is also an important issue that needs to be studied in distributed fiber optic sensing. In order to solve this problem, frequency division multiplexing (FDM), positive and negative beat signals, and periodic non-uniform sampling have been applied to distributed fiber optic sensing systems. FDM can effectively increase the scanning repetition frequency and frequency response bandwidth, but the bandwidth of the detector needs to be increased by the same multiple. Therefore, when the receiver bandwidth is fixed, the high frequency response bandwidth to be achieved under the conditions of ultra-long sensing distance and high spatial resolution is still limited.

[0010] In 2017, J. Xiong et al. proposed using positive and negative beat signals to improve the frequency response range, but this method can only double the frequency response range.

[0011] In 2015, ZYWang et al. used a time series multi-frequency (TSMF) source and injected 35 pulse trains to achieve a frequency response bandwidth of 175kHz at a sensing distance of 9.6km. In 2017, D.Iida et al. of NTT, Japan, used frequency multiplexing technology and 16 frequency components to achieve a frequency response of 80kHz at a sensing distance of 5km. FDM technology can be used to obtain multiple independent signals that are unrelated or have low correlation, thereby increasing the frequency response range. However, in multi-frequency Φ-OTDR, in order to obtain independent multi-frequency measurement signals, the frequency difference between different frequencies must meet the following conditions:

[0012]

[0013] Where f1 and f2 are the detection pulse frequencies, V g is the group velocity of light transmitted in the optical fiber, and L is the interval of differential phase detection. When the frequency difference of each detection frequency satisfies the above formula, the phases of the RBS signals obtained by each detection frequency have a low correlation. At this time, they can be regarded as different independent measurement signals and can be demodulated independently. However, due to the spectral broadening of the laser pulse and the overlap between pulses, the signal-to-noise ratio deteriorates with the increase of N and the decrease of Δf. Therefore, the number of multiplexed frequencies is limited and the spectrum efficiency is not high. In addition, this technology does not solve the restrictive relationship between spatial resolution and transmission distance in the Φ-OTDR system. The spatial resolution is still determined by the pulse width, which will affect other indicators and lead to problems such as the transmission distance cannot be too long.

[0014] In 2015, Zou W et al. proposed the optical pulse compression reflectometer (OPCR) technology, which uses linear frequency modulated light as the detection pulse. The use of longer detection pulses can ensure that the receiving end signal has a higher SNR. At the same time, the receiving end can obtain a higher spatial resolution through the process of pulse compression similar to matched filtering. The use of chirp modulation pulse compression can break through the limitations of traditional Φ-OTDR. In 2016, M. Gonzalez-Herraez et al. used direct detection chirped pulse Φ-OTDR, added frequency chirp modulation to the input pulse of traditional Φ-OTDR, and then used cross-correlation to demodulate the received signal, achieving a spatial resolution of 1m at a sensing distance of 1km. Z. He et al. proposed TGD-OFDR based on chirped frequency modulated pulse compression technology, and then used optical IQ demodulation to increase the system's sensing distance to 40km at a spatial resolution of about 3.5m. In 2018, D. Chen et al. used intensity modulated chirped pulses to achieve a spatial resolution of 0.8m at a sensing distance of 9.8km. In 2019, D. Chen et al. used chirped pulses and double-ended first-order DRA to achieve a spatial resolution of 5m at a sensing distance of 108km. The sidelobe suppression ratio of the LFM pulse echo after matched filtering is very low. Excessive sidelobes will cause crosstalk between Rayleigh signals at different positions. The LFM signal has little improvement on the system's indicators. If windowing is performed to improve performance, while increasing the signal distance sidelobe suppression ratio, the impulse response of the matched filter is changed, causing a mismatch between the signal waveform and the matched filter, resulting in a 1-3dB signal-to-noise ratio loss. In addition, the windowing method will cause the signal main lobe to widen, thereby reducing the spatial resolution of the system. In order to further suppress fading noise and improve the signal-to-noise ratio, the moving average method can also be used. However, the spatial resolution will decrease after moving average.

[0015] In 2017, D. Chen et al. proposed a DAS system based on frequency division multiplexing TGD-OFDR, which uses chirped modulated pulses in five frequency bands, which not only increases the frequency response range by five times, but also has a higher spatial resolution

[11] . In 2018, J. Xiong et al. combined positive and negative beat signals, FDM technology and chirped pulse Φ-OTDR, and used four-band FDM to increase the frequency response range by eight times, reaching a repetition frequency of 9.7kHz within a sensing range of 75km. When the sensing distance is further increased and the maximum response frequency is required to be higher, the number of frequency division multiplexing pulses will be large when frequency division multiplexing is used. Considering that multi-carrier diversity is also required to suppress interference fading in the system, the number of actual frequency division multiplexing pulses will be even greater, which will greatly increase the bandwidth of the received detection signal and exceed the bandwidth range of the detector. In addition, the sidelobe suppression ratio of the LFM pulse echo after matched filtering is very low, only about 13dB. In distributed sensing, the detection signal is mainly composed of the backward Rayleigh scattered light of the pulse light, which is a continuous signal that changes with distance. Excessively high sidelobes will cause crosstalk between Rayleigh signals at different locations.

[0016] Technical problem to be solved by the present invention (purpose of the invention)

[0017] (1) According to the Nyquist sampling theorem, the highest vibration frequency that can be correctly measured is determined by the repetition frequency of the detection pulse. At ultra-long sensing distances, the pulse repetition frequency will decrease, thereby reducing the high-frequency response of the system. In order to achieve a wider frequency response range at ultra-long sensing distances, the pulse modulation method needs to be improved.

[0018] (2) When using FDM, a protection band must be reserved to avoid interference between different frequency bands. A high frequency response bandwidth requires a relatively high detector bandwidth. The high frequency response bandwidth that can be achieved under ultra-long sensing distance and high spatial resolution conditions is still limited, and the pulse modulation method needs to be improved.

[0019] (3) Since the spatial resolution of the original Φ-OTDR is mutually constrained by the transmission distance and the vibration high-frequency response, in order to achieve high spatial resolution over ultra-long transmission distances, it is necessary to decouple the relationship between the spatial resolution and the detection pulse width.

[0020] (4) Since the sidelobe suppression ratio of the LFM pulse echo after matched filtering is very low, only about 13 dB, excessively high sidelobes will cause crosstalk between Rayleigh signals at different locations. It is necessary to improve the sidelobe suppression ratio of the LFM signal. Summary of the invention

[0021] 1. Technical issues to be resolved

[0022] The technical problem to be solved by the present invention is how to provide a distributed optical fiber vibration sensing method based on OFDM-NLFM time series pulse modulation to solve the shortcomings of the above-mentioned prior art.

[0023] (II) Technical solution

[0024] In order to solve the above technical problems, the present invention proposes a distributed optical fiber vibration sensing method based on OFDM-NLFM time series pulse modulation, the method comprising:

[0025] Divide a pulse period T into n intervals {T1, T2, T3, T4, ..., T n}, each interval is T / n, and then the NLFM signal x(t) is multiplied by the subcarrier sequence {f1,f2,f3,f4,...,f n}, where f i =f1+(i-1)*B, i=1,...,n, and the product {s1,s2,s3,s4,...,s n}, B is the NLFM signal bandwidth, and then the product signal s corresponding to each subcarrier i Take the pulse width t p The duration of the i In this case, we get the OFDM-NLFM electrical signal;

[0026] The OFDM-NLFM electrical signal is modulated onto the detection light to generate a time series modulated pulse detection light; the time series modulated pulse detection light is input into the sensing optical fiber, the sensing optical fiber generates a backward Rayleigh scattered light signal, and the Rayleigh scattered light signal is transmitted to the balanced photodetector;

[0027] The beat frequency signal I is obtained after coherent detection by the balanced photodetector beat , collected by ADC, processed by PC, and multiplied by the same subcarrier sequence {f1,f2,f3,f4,...,f n}, where f i =f1+(i-1)*B, i=1,...,n, and then through the matched filter, the demodulation result corresponding to each subcarrier is obtained respectively, {A 11 ,A 12 ,...,A 1k ,A 21 ,A 22 ,...,A 2k ,A 31 ,A 32 ,...,A 3k ,...,A n1 ,A n2,...,A nk}, where A ik represents the demodulation result of the kth pulse period of the ith subcarrier. i1 ,A i2 ,...,A ik} is translated forward on the time axis (i-1)*(T / n), and {A 11 ,A 12 ,...,A 1k}, then follow {A 11 ,A 21 ,A 31 ,...,A (n-1)k ,A nk} are rearranged in order to obtain n*k sampling sequences; the sampling sequences are processed to obtain the vibration positions, and the sampling points corresponding to the vibration positions on each sampling sequence are taken out in order, and then FFT is performed to obtain the vibration frequencies.

[0028] Furthermore, the NLFM signal is realized by weighting the time domain or frequency domain of the LFM signal and adding it to its frequency modulation function, thereby forming the NLFM signal.

[0029] Furthermore, the method for realizing the NLFM signal is: window function inversion method.

[0030] Furthermore, assuming that the spectrum of x(t) is X(f), the frequency characteristic of the matched filter is

[0031] H(f)=X * (f)e -jΘ(f)

[0032] Where, “*” represents complex conjugate; Θ(f) = 2πft0; t0 refers to the moment when the output signal component of the matched filter forms a peak; e -jΘ(f) is the linear phase factor.

[0033] Furthermore, the method for processing the sampling sequence includes: performing direct difference, continuous averaging and moving averaging on the sampling sequence.

[0034] Further, a super-narrow linewidth laser generates a light source, which is divided into two parts by an optical coupler. One part is used as the local oscillator light for coherent reception and output to a balanced photodetector, and the other part is the probe light output to an electro-optic modulator; an AWG generates an OFDM-NLFM electrical signal, and the electro-optic modulator modulates the electrical signal onto the probe light to generate a time-series modulated pulse probe light. An EDFA is used to amplify the time-series modulated pulse probe light, and a circulator is used to input the time-series modulated pulse probe light into a sensing optical fiber. The sensing optical fiber generates a backward Rayleigh scattering optical signal, which enters the circulator, and the circulator transmits the Rayleigh scattering optical signal to the balanced photodetector. An ADC collects the signal and delivers it to a PC for processing.

[0035] Further, the OFDM-NLFM electrical signal is composed of N detection pulses with different frequency ranges. The N detection pulses are multiplexed within a sequence period T, and the pulse repetition period is T / N. The sequence period T is greater than the maximum round-trip time of the detection pulse light passing through the ultra-long-distance sensing optical fiber, and N depends on the ratio of the maximum response frequency to be achieved to the sequence repetition frequency determined by the sensing distance.

[0036] Further, the pulse width of each detection pulse is τ, τ < T / N. The magnitude of τ determines the energy of the detection pulse light, and the pulse energy cannot be greater than the threshold of the fiber nonlinear effect; the detection pulse is a non-linear frequency modulation signal, and the frequency modulation bandwidth is Bs. The spatial resolution of the system is determined by Bs; the center frequency of each detection pulse changes in steps of f △ varies.

[0037] Further, in the PC, through the corresponding matched filter demodulation, N complex backward scattering signals are obtained from different carriers. First, the complex signals of the positive and negative frequency carriers are added using the rotating vector sum method, and then the frequency response range is enhanced using the demodulated complex signals from N detection pulses with different frequency ranges, and finally the vibration point is located and the required monitoring target is obtained.

[0038] Further, the required monitoring targets include phase and frequency.

[0039] (III) Beneficial Effects

[0040] The present invention proposes a distributed optical fiber vibration sensing method based on OFDM-NLFM time-series pulse modulation. The key of the present invention lies in: using the NLFM technology to improve technical indicators such as the signal-to-noise ratio, sidelobe suppression ratio, and spatial resolution of the Φ-OTDR signal, using the OFDM technology to improve technical indicators such as the frequency response range and spectral efficiency of the Φ-OTDR signal, and the signal design method of combining OFDM and NLFM in Φ-OTDR.

[0041] The beneficial effects brought by the technical solution of the present invention:

[0042] (1) OFDM-NLFM pulse light has a wider pulse width and a higher initial light intensity, which can increase the backscattering power at long sensing distances and extend the transmission distance.

[0043] (2) When using OFDM, the limitation of ultra-long sensing distance on frequency response bandwidth can be solved by frequency division multiplexing, and the repetition frequency of detection light can be increased to obtain a higher frequency response bandwidth.

[0044] (3) OFDM can achieve more frequency band multiplexing within the limited detector bandwidth of the receiver and improve spectrum efficiency.

[0045] (4) The sidelobe suppression ratio of the NLFM pulse echo after matched filtering is very high, which can obtain very high spatial resolution and reduce the crosstalk between Rayleigh signals at different positions, thereby improving the signal-to-noise ratio. BRIEF DESCRIPTION OF THE DRAWINGS

[0046] Figure 1 It is a schematic diagram of the overall technical solution of the present invention;

[0047] Figure 2 is the frequency modulation curve of different NLFM signals;

[0048] Figure 3 A schematic diagram of an OFDM-NLFM signal of the present invention;

[0049] Figure 4 It is a schematic diagram of the OFDM-NLFM signal processing method of the present invention;

[0050] Figure 5 It is the OFDM-NLFM time series pulse modulation signal of the present invention. DETAILED DESCRIPTION

[0051] In order to make the purpose, content and advantages of the present invention more clear, the specific implementation methods of the present invention are further described in detail below in conjunction with the drawings and examples.

[0052] Definitions of Abbreviations and Key Terms

[0053] Φ-OTDR Phase Sensitive Optical Time Domain Reflectometer

[0054] (Phase-sensitive Optical Time Domain Reflectometry)

[0055] OFDM Orthogonal Frequency Division Multiplexing

[0056] (Orthogonal Frequency Division Multiplexing)

[0057] NLFM Non-Linear Frequency Modulation

[0058] (Non-linearFrequency Modulation)

[0059] PBS Polarization Beam Splitter

[0060] (PolarizationBeam Splitter)

[0061] RBS Rayleigh Backscattering

[0062] (Rayleigh Backward Scattering)

[0063] LO intrinsic light

[0064] (Intrinsic Optical)

[0065] BPD Balanced Photodetector

[0066] (Balance Photo-detector)

[0067] ADC High-speed analog-to-digital converter

[0068] (Analog-to-digital Converter)

[0069] SNR Signal to Noise Ratio

[0070] (Signal To Noise Ratio)

[0071] EOM Electro-Optic Modulator

[0072] (Electro-Optic Modulator)

[0073] Figure 1 The system structure diagram of the scheme of the present invention. The scheme of the present invention uses an ultra-narrow linewidth laser to generate a light source. The light source is divided into two parts by an optical coupler. One part is output as a local oscillator light for coherent reception to a balanced photodetector, and the other part is a detection light output to an electro-optical modulator. The AWG generates an OFDM-NLFM electrical signal, and the electro-optical modulator modulates the electrical signal onto the detection light to generate a time series modulated pulse detection light. The EDFA is used to amplify the time series modulated pulse detection light, and the circulator is used to input the time series modulated pulse detection light into the sensing optical fiber. The sensing optical fiber generates a backward Rayleigh scattered light signal, which enters the circulator. The circulator transmits the Rayleigh scattered light signal to the balanced photodetector, and the ADC collects the signal and sends it to the PC for processing.

[0074] (1) Generation of NLFM Signal

[0075] The modulation frequency of the nonlinear frequency modulation NLFM signal changes nonlinearly with time, and the modulation frequency curve is shown in the figure below. Figure 2 The figure shows the frequency modulation curves of four different NLFM signals, each using a different window function design.

[0076] The theoretical basis of NLFM signal is the phase lingering principle. There are many NLFM signal models, and the generation of the signal is more complicated than that of the LFM signal. There is no precise signal design method, and approximate methods are used in practice. At present, there are two main methods for realizing NLFM signal: ① adding the weighted time domain or frequency domain of the LFM signal to its frequency modulation function to form an NLFM signal; ② the more classic window function inversion method. The above two NLFM signal realization methods are both existing, and the present invention adopts the second method. The specific signal generation process is as follows:

[0077] The NLFM signal is expressed as

[0078] x(t)=a(t)exp(jθ(t))

[0079] Where, “a(t)” represents the amplitude of the signal, which is generally a constant, and “θ(t)” represents the phase function of the signal;

[0080] Assuming that the spectrum of x(t) is X(f), the frequency characteristic of its matched filter should be

[0081] H(f)=X * (f)e -jΘ(f)

[0082] Where, “*” represents complex conjugate; Θ(f) = 2πft0; t0 refers to the moment when the output signal component of the matched filter forms a peak; e -jΘ(f) is the linear phase factor. At this time, the spectrum Y(f) of the output y(t) of the matched filter is

[0083]

[0084] For a certain frequency domain window function W(f), let

[0085] |X(f)| 2 =W(f)

[0086] According to the phase lingering principle, the relationship between the signal spectrum and the frequency modulation slope is:

[0087]

[0088] In the formula, T(f) is the group delay function of the signal; f(t) is the frequency modulation function, and they are inverse functions of each other. Assuming that the spectrum of the signal after pulse compression is selected as W(f), then Then, the group delay function can be obtained by integrating W(f), that is,

[0089]

[0090] Where K is a constant coefficient. When the NLFM signal bandwidth is required to be B and the time width is t p Sometimes, there are

[0091]

[0092] According to the relationship between T(f) and f(t), we can get

[0093] f(t)=T -1 (f)

[0094] So, the phase function of the signal is

[0095]

[0096] At this point, the generation of the NLFM signal x(t) can be completed.

[0097] (2) Generation of OFDM-NLFM Signals

[0098] First, a pulse period T is evenly divided into n intervals {T1, T2, T3, T4, ..., T n}, each interval has a duration of T / n, and then the NLFM signal x(t) generated in step (1) is multiplied by the subcarrier sequence {f1,f2,f3,f4,...,f n}, where f i =f1+(i-1)*B, i=1,...,n, and the product {s1,s2,s3,s4,...,s n}, B is the NLFM signal bandwidth, and then the product signal s corresponding to each subcarrier i Take the pulse width t p The duration of the i In this case, we get the OFDM-NLFM signal, such as Figure 3 shown.

[0099] (3) Demodulation of OFDM-NLFM Signals

[0100] The beat frequency signal I is obtained after coherent detection by the balanced photodetector beat , collected by ADC, processed by PC, and multiplied by the same subcarrier sequence {f1,f2,f3,f4,...,fn} where f n = f1+(n - 1)*B, and then through the matched filter designed in (1), the demodulation results corresponding to each sub - carrier are obtained, {A 11 , A 12 ,..., A 1k , A 21 , A 22 ,..., A 2k , A 31 , A 32 ,..., A 3k ,..., A n1 , A n2 ,..., A nk}, where A ik represents the demodulation result of the k - th pulse period of the i - th sub - carrier. Shift {A i1 , A i2 ,..., A ik} forward by (i - 1)*(T / n) on the time axis to align with {A 11 , A 12 ,..., A 1k}, and then rearrange them in the order of {A 11 , A 21 , A 31 ,..., A (n-1)k , A nk} to obtain n*k sampling sequences. Process the data of the sampling sequences by methods such as direct difference, continuous average, and moving average to obtain the vibration position, and then sequentially extract the sampling points corresponding to the vibration position on each sampling sequence, and then perform FFT to obtain the vibration frequency. As Figure 4 shown.

[0101] The waveform generator generates an OFDM - NLFM time - series signal as the pulse modulation signal. This sequence consists of N detection pulses with different frequency ranges. The N detection pulses are multiplexed within a sequence period T, and the pulse repetition period is T / N. The sequence period T is greater than the maximum round - trip time of the detection pulse light passing through the ultra - long - distance sensing optical fiber. N depends on the ratio of the maximum response frequency to be achieved and the sequence repetition frequency determined by the sensing distance. The pulse width of each detection pulse is τ (τ < T / N), and the magnitude of τ determines the energy of the detection pulse light. The pulse energy cannot be greater than the threshold of the fiber nonlinear effect. The detection pulse is a non - linear frequency - modulated signal, and the frequency - modulation bandwidth is Bs. The spatial resolution of the system is determined by Bs. The center frequency of each detection pulse changes in steps of f △ , Figure 5 is the time - frequency curve of the OFDM - NLFM signal.

[0102] AWG generates OFDM-NLFM electrical signals, and the electro-optic modulator modulates the electrical signals onto the detection light generated by the ultra-narrow linewidth laser to generate time-series modulated pulse detection light. EDFA is used to amplify the optical signal, and the circulator is used to input the time-series modulated pulse detection light into the sensing fiber. The sensing fiber generates a backward Rayleigh scattered light signal, which enters the circulator. The circulator transmits the Rayleigh scattered light signal to the balanced photodetector, and the ADC collects the signal and sends it to the PC for processing.

[0103] By demodulating with the corresponding matched filter in the PC, N complex backscattered signals can be obtained from different carriers. The demodulated complex signals from N detection pulses with different frequency ranges can be used to increase the frequency response range, and ultimately the vibration point can be located and the required monitoring targets such as phase and frequency can be obtained.

[0104] The present invention proposes a distributed optical fiber vibration sensing method based on OFDM-NLFM time series pulse modulation. The key points of the present invention are: using NLFM technology to improve technical indicators such as the signal-to-noise ratio, sidelobe suppression ratio and spatial resolution of Φ-OTDR signals, using OFDM technology to improve technical indicators such as the frequency response range and spectrum efficiency of Φ-OTDR signals, and a signal design method combining OFDM with NLFM in Φ-OTDR.

[0105] Beneficial effects brought by the technical solution of the present invention:

[0106] (1) OFDM-NLFM pulse light has a wider pulse width and a higher initial light intensity, which can increase the backscattering power at long sensing distances and extend the transmission distance.

[0107] (2) When using OFDM, the limitation of ultra-long sensing distance on frequency response bandwidth can be solved by frequency division multiplexing, and the repetition frequency of detection light can be increased to obtain a higher frequency response bandwidth.

[0108] (3) OFDM can achieve more frequency band multiplexing within the limited detector bandwidth of the receiver and improve spectrum efficiency.

[0109] (4) The sidelobe suppression ratio of the NLFM pulse echo after matched filtering is very high, which can obtain very high spatial resolution and reduce the crosstalk between Rayleigh signals at different positions, thereby improving the signal-to-noise ratio.

[0110] The above is only a preferred embodiment of the present invention. It should be pointed out that for ordinary technicians in this technical field, several improvements and modifications can be made without departing from the technical principles of the present invention. These improvements and modifications should also be regarded as the scope of protection of the present invention.

Claims

1. A distributed optical fiber vibration sensing method based on OFDM-NLFM time series pulse modulation, characterized in that: The method includes: Divide a pulse period T into n intervals {T1, T2, T3, T4, ..., T n }, each interval is T / n, and then the NLFM signal x(t) is multiplied by the subcarrier sequence {f1,f2,f3,f4,...,f n }, where f i =f1+(i-1)*B, i=1,...,n, and the product {s1,s2,s3,s4,...,s n }, B is the NLFM signal bandwidth, and then the product signal s corresponding to each subcarrier i Take the pulse width t p The duration of the i In this case, we get the OFDM-NLFM electrical signal; The OFDM-NLFM electrical signal is modulated onto the detection light to generate a time series modulated pulse detection light; the time series modulated pulse detection light is input into the sensing optical fiber, the sensing optical fiber generates a backward Rayleigh scattered light signal, and the Rayleigh scattered light signal is transmitted to the balanced photodetector; The beat frequency signal I is obtained after coherent detection by the balanced photodetector beat , collected by ADC, processed by PC, and multiplied by the same subcarrier sequence {f1,f2,f3,f4,...,f n }, where f i =f1+(i-1)*B, i=1,...,n, and then through the matched filter, the demodulation result corresponding to each subcarrier is obtained respectively, {A 11 ,A 12 ,…,A 1k ,A 21 ,A 22 ,…,A 2k ,A 31 ,A 32 ,…,A 3k ,…,A n1 ,A n2 ,…,A nk }, where A ik represents the demodulation result of the kth pulse period of the ith subcarrier. i1 ,A i2 ,…,A ik } is translated forward on the time axis (i-1)*(T / n), and {A 11 ,A 12 ,…,A 1k }, and then follow {A 11 ,A 21 ,A 31 ,…,A (n-1)k ,A nk } are rearranged in order to obtain n*k sampling sequences; the sampling sequences are processed to obtain the vibration positions, and the sampling points corresponding to the vibration positions on each sampling sequence are taken out in order, and then FFT is performed to obtain the vibration frequencies.

2. The distributed optical fiber vibration sensing method based on OFDM-NLFM time series pulse modulation as claimed in claim 1, characterized in that: The NLFM signal is realized by weighting the time domain or frequency domain of the LFM signal and adding it to its frequency modulation function to form the NLFM signal.

3. The distributed optical fiber vibration sensing method based on OFDM-NLFM time series pulse modulation as claimed in claim 1, characterized in that: The implementation method of NLFM signal is: window function inversion method.

4. The distributed optical fiber vibration sensing method based on OFDM-NLFM time series pulse modulation as claimed in claim 1, characterized in that: Assuming that the spectrum of x(t) is X(f), the frequency characteristic of the matched filter is H(f)=X * (f)e -jΘ(f) Wherein, "*" indicates complex conjugate; Θ(f) = 2πft0; t0 refers to the moment when the output signal component of the matched filter forms a peak; e -jΘ(f) is the linear phase factor.

5. The distributed optical fiber vibration sensing method based on OFDM-NLFM time series pulse modulation as claimed in claim 1, characterized in that: The method for processing the sampling sequence includes: performing direct difference, continuous averaging and moving averaging on the sampling sequence.

6. The distributed optical fiber vibration sensing method based on OFDM-NLFM time series pulse modulation according to any one of claims 1 to 5, characterized in that: The ultra-narrow linewidth laser generates a light source, which is divided into two parts by an optical coupler. One part is output as the local oscillator light for coherent reception to a balanced photodetector, and the other part is the detection light output to an electro-optical modulator. AWG generates OFDM-NLFM electrical signals, and the electro-optical modulator modulates the electrical signals onto the detection light to generate time series modulated pulse detection light. EDFA is used to amplify the time series modulated pulse detection light. The circulator is used to input the time series modulated pulse detection light into the sensing optical fiber. The sensing optical fiber generates a backward Rayleigh scattered light signal and enters the circulator. The circulator transmits the Rayleigh scattered light signal to the balanced photodetector. The ADC collects the signal and sends it to the PC for processing.

7. The distributed optical fiber vibration sensing method based on OFDM-NLFM time series pulse modulation as claimed in claim 6, characterized in that: The OFDM-NLFM electrical signal consists of N detection pulses with different frequency ranges. The N detection pulses are multiplexed within a sequence period T. The pulse repetition period is T / N. The sequence period T is greater than the maximum round-trip time of the detection pulse light through the ultra-long-distance sensing optical fiber. N depends on the ratio of the maximum response frequency to be achieved and the sequence repetition frequency determined by the sensing distance.

8. The distributed optical fiber vibration sensing method based on OFDM-NLFM time series pulse modulation as claimed in claim 7, characterized in that: The pulse width of each detection pulse is τ, τ<T / N, the size of τ determines the energy of the detection pulse light, and the pulse energy cannot be greater than the threshold of the fiber nonlinear effect; the detection pulse is a nonlinear frequency modulation signal, the bandwidth of the frequency modulation is Bs, and the spatial resolution of the system is determined by Bs; the central frequency of each detection pulse is in steps of f Δ change.

9. The distributed optical fiber vibration sensing method based on OFDM-NLFM time series pulse modulation as claimed in claim 8, characterized in that: In the PC, N complex backscattered signals are obtained from different carriers through corresponding matched filter demodulation. The frequency response range is improved by using the demodulated complex signals from N detection pulses with different frequency ranges, and finally the vibration point is located and the required monitoring target is obtained.

10. The distributed optical fiber vibration sensing method based on OFDM-NLFM time series pulse modulation as claimed in claim 9, characterized in that: The required monitoring targets include phase and frequency.

Images