Distributed Acoustic Sensing Sensitivity Improvement Using MIMO Sampling and Phase Resynthesis

Abstract

Aspects of the present disclosure describe distributed fiber optic sensing (DFOS) / distributed acoustic sensing (DAS) systems, methods, and structures that exhibit sensitivity enhancement through MIMO sampling and phase recombination.

Description

[Technical Field]

[0001] The present disclosure relates generally to distributed fiber optic sensing (DFOS) systems, methods, and structures, and more particularly to distributed acoustic sensing (DAS) sensitivity enhancement using MIMO sampling and phase recombination. [Background technology]

[0002] Distributed acoustic sensing (DAS) is a fiber optic sensing technology that enables an interrogator to remotely detect vibration and acoustic signals along the entire length of an installed fiber optic cable. DAS can be advantageously applied in a wide range of applications, including infrastructure monitoring, oil and gas operations, and earthquake detection, among others. In operation, DAS utilizes the Rayleigh scattering effect in optical fibers to detect dynamic strain changes in localized fiber segments.

[0003] One important characteristic of a DAS system is its sensitivity, i.e., the lowest dynamic strain it can detect in any fiber segment. When measured over short distances, the sensitivity of a DAS system is typically on the order of a few pico-epsilons to tens of pico-epsilons per square root hertz (pε / √Hz).

[0004] Current DAS systems exhibit an interrogation repetition rate limit determined by the round-trip time required to interrogate a given fiber optic cable length. As a result, the Nyquist bandwidth of the "acoustic" sampling rate is much lower than the bandwidth of the optical signal.

[0005] Within the interrogation signal, there are a combination of noise sources, including laser phase noise and additive white Gaussian noise (AWGN), that affect system operation. Without a sufficient acoustic sampling rate, system noise outside the acoustic Nyquist band cannot be filtered out and is "folded back" into the received acoustic band, raising the phase noise floor of the recovered DAS signal. Current commercially available DAS systems have a sampling rate mismatch of over 100 times between optical and acoustic sampling, meaning that the distortion sensitivity of current DAS systems can be off by several orders of magnitude when compared to discrete-phase interferometers without dispersive operation.

[0006] Given these characteristics, improvements in DFOS sensitivity would be welcomed by the art. Summary of the Invention

[0007] An advancement in the art is provided in accordance with aspects of the present disclosure directed to DFOS systems, methods, and structures, and more particularly, to increased sensitivity in distributed acoustic sensing (DAS) using MIMO sampling and phase resynthesis.

[0008] In stark contrast to the prior art, systems, methods, and structures according to aspects of the present disclosure provide enhanced DAS sensitivity through massive MIMO sampling and phase resynthesis, mitigating excess noise folding back into the recovered DAS phase signal due to sampling rate mismatches. Advantageously, by using multiple channels along the same path for synchronized DAS interrogation, systems and methods according to the present disclosure use MIMO detection to obtain time-offset DAS results for each channel. The multi-channel DAS results can then be recombined in a DSP to produce a single-phase signal stream exhibiting higher fidelity through suppression of out-of-band laser phase noise and additive white Gaussian noise (AWGN).

[0009] As will now be shown and described, the inventive systems and methods according to this disclosure improve DAS sensitivity through MIMO interrogation and DSP-implemented channel recombination for MIMO DAS implementations.

[0010] In a first aspect, the systems and methods of the present invention employ one of two methods: time interleaving and time alignment. The time interleaving method employed in this disclosure is used to increase the acoustic sampling rate to suppress out-of-band laser phase noise and AWGN. The time alignment method accounts for the time offset between samples of different DAS MIMO channels and applies a corresponding phase shift before digitally recombining the phase signals. Advantageously, the resulting DAS output retains the native acoustic sampling rate of each channel and also has the same sensitivity improvements as the time interleaving method, such as being more robust to Rayleigh fading.

[0011] Yet another aspect of the systems and methods according to aspects of the present disclosure includes how MIMO interrogation is performed using multiple frequency channels. Advantageously, this is not limited to optical frequency implementations, and provides an elegant way to provide such functionality, as transmitter and receiver hardware can be shared between different channels. In frequency-domain DAS MIMO, interrogation pulses / codes using multiple frequencies are transmitted in a time-staggered manner, enabling higher acoustic sampling rates and noise suppression after DSP recombination. [Brief explanation of the drawings]

[0012] A more complete understanding of the present disclosure may be realized by reference to the accompanying drawings.

[0013] [Figure 1(A)] FIG. 1 is a schematic diagram illustrating a DFOS system according to an embodiment of the present disclosure.

[0014] [Figure 1(B)]FIG. 1 is a schematic diagram illustrating a coded constant amplitude DFOS system with out-of-band signal generation, according to an aspect of the present disclosure.

[0015] [Figure 2] FIG. 2 is a schematic diagram illustrating an example MIMO DAS architecture configuration with staggered interrogation signals, according to an aspect of the present disclosure.

[0016] [Figure 3] FIG. 1 is a schematic diagram illustrating an example MIMO DAS implemented using a single TX / RX / FDM configuration, according to an aspect of the present disclosure.

[0017] [Figure 4(A)] FIG. 10 illustrates an algorithm for combining multiple DAS results for out-of-band noise suppression, in accordance with an embodiment of the present disclosure, illustrating a time-interleaved method. [Figure 4(B)] FIG. 10 illustrates an algorithm for combining multiple DAS results for out-of-band noise suppression, according to an embodiment of the present disclosure, illustrating a time alignment method.

[0018] [Figure 5(A)] 10A-10C are plots illustrating improved DAS sensitivity with higher acoustic sampling rates, according to embodiments of the present disclosure, showing a low acoustic sampling rate. [Figure 5(B)] 10A-10C are plots illustrating improved DAS sensitivity with higher acoustic sampling rates, according to embodiments of the present disclosure;

[0019] [Figure 6(A)] FIG. 10 is a plot illustrating an example sampled receiver spectrum of an FDM MIMO DAS, for eight wavelengths, in accordance with an embodiment of the present disclosure. [Figure 6(B)] FIG. 10 is a plot illustrating an example sampled receiver spectrum of an FDM MIMO DAS, illustrating 16 wavelengths, in accordance with an embodiment of the present disclosure. [Figure 6(C)]FIG. 10 is a plot illustrating an example sampled receiver spectrum of an FDM MIMO DAS, illustrating 32 wavelengths, in accordance with an embodiment of the present disclosure.

[0020] [Figure 7] 10 is a plot illustrating an example average DAS sensitivity using massive FDM MIMO according to an embodiment of the present disclosure. DETAILED DESCRIPTION OF THE INVENTION

[0021] The following merely illustrates the principles of the present disclosure, and it will thus be appreciated that those skilled in the art will be able to devise various arrangements which, although not explicitly described or shown herein, embody the principles of the present disclosure and are included within its spirit and scope.

[0022] Furthermore, all examples and conditional language set forth herein are intended to be for educational purposes only to aid the reader in understanding the concepts contributed by the inventors to further the principles and techniques of the present disclosure, and should not be construed as being limited to such specifically recited examples and conditions.

[0023] Moreover, all statements herein reciting principles, aspects, and embodiments of the present disclosure, as well as specific examples thereof, are intended to encompass both structural and functional equivalents thereof. Additionally, such equivalents are intended to include both currently known equivalents as well as equivalents developed in the future, i.e., elements developed that perform the same function, regardless of structure.

[0024] Thus, for example, it will be appreciated by those skilled in the art that any block diagrams herein represent conceptual views of illustrative circuitry embodying the principles of the disclosure.

[0025] Unless otherwise specified herein, the figures comprising the drawings are not drawn to scale.

[0026] As additional background, we begin by noting that distributed fiber optic sensing (DFOS) is an important and widely used technology for detecting environmental conditions (e.g., temperature, vibration, acoustic excitation, and strain levels) anywhere along a fiber optic cable that is in turn connected to an interrogator. As known, a modern interrogator is a system that generates an input signal into the fiber, detects and analyzes the reflected / scattered, and then received signal. The signal is analyzed, and an output is generated that indicates the environmental conditions encountered along the fiber. Such received signals can result from reflections within the fiber, such as Raman backscattering, Rayleigh backscattering, and Brillion backscattering. DFOS can also use forward signals that exploit the velocity differences of multiple modes. Without loss of generality, the following discussion assumes reflected signals, but the same approach is equally applicable to forward signals.

[0027] Figure 1(A) is a schematic diagram of a generalized prior art DFOS system. As will be appreciated, modern DFOS systems include an interrogator that periodically generates optical pulses (or any coded signal) and injects them into an optical fiber. The injected optical pulse signal is transmitted along the optical fiber.

[0028] At locations along the fiber, a small portion of the signal is reflected back to the interrogator. The reflected signal carries information that the interrogator uses to detect, for example, changes in power level indicative of mechanical vibrations. Although not shown in detail, the interrogator can include a coded DFOS system that can employ a coherent receiver configuration known in the art, such as that shown in Figure 1(B).

[0029] The reflected signal is converted to the electrical domain and processed within the interrogator. Based on the time of pulse injection and the time the signal is detected, the interrogator can determine which location on the fiber the signal is coming from and sense the activity at each location on the fiber.

[0030] Those skilled in the art will understand and appreciate that by implementing signal coding on the interrogation signal, more optical power can be transmitted into the fiber, which can advantageously improve the signal-to-noise ratio (SNR) of Rayleigh scattering-based systems (e.g., distributed acoustic sensing, or DAS) and Brillouin scattering-based systems (e.g., Brillouin optical time-domain reflectometry, or BOTDR).

[0031] As currently implemented in many modern applications, a dedicated fiber is allocated to the DFOS system in the fiber optic cable, physically separated from existing optical communication signals carried on different fibers. However, given the exponentially increasing bandwidth demand, it is becoming very difficult to economically operate and maintain optical fiber solely for DFOS operations. As a result, there is growing interest in integrating communication and sensing systems onto a common fiber that is part of a larger multi-fiber cable.

[0032] Operationally, we assume that DFOS systems are Rayleigh scattering-based systems (e.g., distributed acoustic sensing, or DAS) and Brillouin scattering-based systems (e.g., Brillouin optical time-domain reflectometry, or BOTDR) with coding implementations. With such coding designs, these systems are more likely to be integrated with fiber communication systems due to their lower operating power and greater sensitivity to the response time of optical amplifiers.

[0033] With this discussion in mind, it is again noted that systems, methods, and structures according to aspects of the present disclosure advantageously mitigate the extraneous noise folding back into the recovered DAS phase signal due to the aforementioned sampling rate mismatch. By using multiple channels along the same path for DAS interrogation while synchronously interrogating, the systems and methods of the present disclosure use MIMO detection to obtain time-offset DAS results for each channel. The multi-channel DAS results can then be recombined in a DSP to produce a single-phase signal stream exhibiting higher fidelity due to the suppression of out-of-band laser phase noise and AWGN. While the use of multiple channels in DAS operation to increase the effective acoustic sampling rate for sensing interrogation has previously been proposed, the focus of this disclosure is the implementation of MIMO interrogation in a DSP and methods for increasing DAS sensitivity through its channel recombination method.

[0034] As previously mentioned, the systems and methods of the present invention use one of two methods: time interleaving and time alignment. While the prior art uses time interleaving to increase the effective sampling rate of multi-channel DASs, the present disclosure employs this method to increase the acoustic sampling rate and suppress out-of-band laser phase noise and AWGN. The time alignment method accounts for the time offset between samples of different DAS MIMO channels, and applies a corresponding phase shift before digitally recombining the phase signals. The resulting DAS output achieves the same sensitivity improvement as the time interleaving method while retaining the native acoustic sampling rate of each channel. Furthermore, the time alignment method is more robust to Rayleigh fading.

[0035] Yet another feature of the systems and methods disclosed herein relates to MIMO interrogation performed using multiple frequency channels. This is not limited to optical frequency implementations, but proves to be one of the simplest ways to achieve the functionality, as transmitter and receiver hardware can be shared between different channels. In frequency domain DAS MIMO, the transmission of interrogation pulses / codes using multiple frequencies is staggered in time, allowing for higher acoustic sampling rates and noise suppression after DSP recombination.

[0036] As noted again herein, one of the key advantages of the inventive MIMO DAS technology according to this disclosure is the resulting increase in the acoustic sampling rate of the DAS system for a given fiber cable path, while suppressing out-of-band noise to improve DAS sensitivity.

[0037] 2 is a schematic diagram illustrating an example MIMO DAS architecture configuration with staggered interrogation signals, according to an embodiment of the present disclosure. Compared to a single-channel DAS, this illustrated scheme interrogates multiple channels, each with T frame transmit an interrogation pulse / sequence in the original channel interrogation period of T and receive a Rayleigh signal. However, the interrogation signal on a different channel is frame Therefore, after MIMO detection and processing, the DAS signal information of the different channels can be combined to produce a phase signal output with a low phase noise floor and suppressed out-of-band noise.

[0038] As shown schematically in the figure, there are several ways to achieve MIMO DAS within a fiber optic cable. Perhaps the simplest is via FDM, which uses multiple frequency / wavelength channels, requiring only a single fiber and allowing the optical / electronic hardware to be shared between multiple channels.

[0039] It can also be done via spatial division multiplexing (SDM) by using different fibers, different cores of a multicore fiber (MCF), or different modes of a multimode fiber (MMF). SDM typically requires multiple sets of optical / electronic hardware because channels cannot be shared. Spatial multiplexing and demultiplexing is also required when using MCF or MMF.

[0040] FIG. 3 is a schematic diagram illustrating an exemplary MIMO DAS implemented using a single TX / RX / FDM configuration, according to an embodiment of the present disclosure. In this figure, an exemplary FDM DAS setup is shown to illustrate that a MIMO DAS can be implemented using a single transmitter and receiver setup. This setup exhibits significant differences in the generation of interrogation signals compared to the prior art. According to an embodiment of the present disclosure, the inventive scheme utilizes time staggering between different frequency channels to increase the effective acoustic sampling rate. The T of each separated channel is frame During the original frame period, the interrogation pulses or sequences of each channel are intentionally offset in time. Thus, in an N-channel MIMO DAS system, the time offset between adjacent channels is T frame / N, and the effective acoustic sampling rate increases by a factor of N.

[0041] Advantageously, this particular implementation of FDM DAS allows for the joint generation of MIMO interrogation signals for all channels using a single transmitter. First, the sensing laser is modulated by an in-phase / quadrature modulator (IQM) to generate multiple frequencies sequentially before an acousto-optic modulator (AOM) generates optical pulses at each frequency. The modulation signal to the modulator can be generated from a common DAC or arbitrary waveform generator. Note that this configuration can also be used to generate sequences / codes for each channel if desired, so the implementation is not limited to pulse interrogation alone.

[0042] At the receiver, the Rayleigh reflected signals from each channel can be detected by a single coherent receiver without demultiplexing. In DSP operation, each FDM channel is first filtered and digitally downconverted to baseband before being passed to the main DAS algorithm for parallel processing. For every position (defined by the spatial resolution of the system) in each frequency channel, a differential phase value is generated by the algorithm.

[0043] As mentioned above, the inventive system and method according to the present disclosure employs two methods for combining DAS phase results from different channels of a MIMO DAS into a single stream, which are shown in Figures 4(A) and 4(B). Figures 4(A) and 4(B) illustrate an algorithm for combining multiple DAS results for out-of-band noise suppression according to an embodiment of the present disclosure, where Figure 4(A) illustrates the time-interleaved method and Figure 4(B) illustrates the time-aligned method.

[0044] Note that both algorithms deal with the combination of multiple phase tributaries at the same fiber location. As mentioned previously, the first method is time-interleaving, where N channels, each with a native sampling rate of 1 / T, are spliced ​​together to form a higher sampling rate of N / T. Before time-interleaving, the phase vectors from each channel are rotated to prevent the introduction of additional interleaving noise. This can be done with an LPF-based combining algorithm.

[0045] The second method, by time alignment, deals with the timing offset of the individual channels and sums them together so that each contributing sample is time-aligned. To align each channel, perform an FFT and IFFT on each channel,

number

[0046] 5(A) and 5(B) are plots illustrating the improvement in DAS sensitivity with a higher acoustic sampling rate, according to embodiments of the present disclosure. FIG. 5(A) is a low acoustic sampling rate, and FIG. 5(B) is a high acoustic sampling rate. These figures illustrate the basic principles that explain how a higher acoustic sampling rate can improve the distortion sensitivity of a DAS. In a DAS interrogation system, both laser phase noise and additive white Gaussian noise (AWGN) contribute to the final DAS phase noise, limiting sensitivity performance.

[0047] For ease of discussion, we focus on the short-distance DAS operating region, where laser phase noise is the dominant noise source. As shown in Figure 5(B), at frequencies below 10-100 Hz, the laser phase noise becomes very high due to the limitations of the mechanical and thermal isolation of the laser module. The laser phase noise gradually decreases to a stable level corresponding to the laser's intrinsic linewidth.

[0048] Note that laser phase noise is inherent to the DAS interrogation operation and can only be partially filtered by the receiver hardware LPF and digital filters in the DSP. The bandwidth of these filters is typically designed according to the spatial resolution requirements of the DAS and is therefore much wider than the acoustic sampling bandwidth, which is limited by the interrogation round-trip time. Therefore, even with a relatively high repetition rate at short distances, there is a lot of phase noise outside the Nyquist bandwidth that cannot be adequately filtered, and it folds back into the DAS's acoustic bandwidth, significantly raising the phase noise floor. For example, a repetition rate of ~10 kHz (for a 10 km distance) and signal filtering of ~10 MHz (for a 10 m spatial resolution) results in an approximately 30 dB increase in noise level due to an insufficient acoustic sampling rate.

[0049] The inventive systems and methods of the present disclosure allow the acoustic sampling rate to be effectively increased by MIMO DAS interrogation. As shown in FIG. 5(B), a higher sampling rate substantially reduces the amount of noise present outside the Nyquist band, resulting in less aliased noise and therefore a reduced in-band phase noise floor. Therefore, it should be noted that the implementation of MIMO DAS according to the present disclosure does not aim to achieve a wider DAS bandwidth for high-frequency signal detection and identification, but rather aims to suppress in-band phase noise through a higher acoustic sampling rate.

[0050] If 100x MIMO is implemented on the aforementioned DAS system at a repetition rate of 10 kHz, the phase noise can theoretically be suppressed by 20 dB.

[0051] We can also see how the noise floor is suppressed by examining the example operations shown in Figures 4(A) and 4(B). For the time-interleaved method, the detected DAS phase can be expressed as:

number

[0052] The time interleaving result is (k=m·N+n-1)

number

[0053] As the final result shows, the distorted signal

number

number

number

[0054] In the time-aligned method, the operation of the DSP can be described as follows:

number

[0055] The final time-interleaved result is

number

[0056] The first two terms of the result

number

number

number

[0057] From the above equation, if a single-channel DAS is dominated by out-of-band laser phase noise and AWGN, using an N-channel MIMO-DAS can suppress the noise level by N times through the time alignment method.

[0058] To confirm the phase noise suppression using MIMO-DAS, we conducted several experiments based on the described concept. A baseline single-channel DAS using a commercially available sensing laser was

number

number

[0059] Single-sideband (SSB) signal modulation and detection is used to avoid I / Q imbalance penalties from the transmitter and receiver. The DAS signals for each channel are processed separately in a DSP before being combined using the two proposed methods.

[0060] The spectrogram, phase plot, and phase spectrum were determined for the FDM DAS results using the time alignment method at eight frequencies. The output DAS results are at the same 20 kHz sampling rate. From these results, the stretched signal at 1.267 kHz and 0.16 rad can be clearly seen, allowing the PZT location to be identified. Examining the phase spectrum, it is clear that for the eight-frequency system:

number

[0061] In another experiment, a time-interleaving method was used to construct the final DAS results. With time interleaving, the final acoustic sampling rate was 160 kHz. Due to the presence of broader band noise in the result plots (which can be removed with further digital filtering), the phase plots contain more noise compared to the time-aligned experiment, but the stretcher position and applied signal can also be obtained. Examination of the phase spectrum revealed that the noise level was similar to that of the time-aligned method. However, it should be noted that this method adds additional interleaving noise to the spectrogram.

[0062] FIG. 7 is a plot illustrating an example average DAS sensitivity using massive FDM MIMO according to an embodiment of the present disclosure. This figure compares the performance of MIMO-DAS using two different combining methods. The time-aligned method was found to achieve better performance than the time-interleaved method. The performance difference is likely due to Rayleigh fading. Because each frequency channel has a different Rayleigh response, there will always be channels with SNRs significantly worse than average. With time-interleaving, a single bad channel can affect overall performance, making the combined result more susceptible to fading. On the other hand, the time-aligned method uses phase averaging across all channels to suppress phase noise, and averaging can also mitigate fading. Sampling each frequency channel at 20 kHz (5 km test spur) achieved a record low signal-to-noise ratio using 32 channels and a total TX / RX bandwidth of ~2 GHz.

number

[0063] It should be noted that the two combining methods of the present invention can also be used together sequentially. Thus, instead of time-interleaving or aligning all channels, time alignment can be performed on adjacent channels first to mitigate fading and improve phase noise performance, and then time interleaving can be performed later to detect higher-frequency acoustic signals. This results in a more gradual trade-off between phase noise performance and output acoustic bandwidth. Finally, when AWGN is the dominant noise source, massive MIMO DAS can also be used in combination with long-range coded DAS. As a result, the interrogation pulse shown in Figure 3 can be replaced with a code sequence to improve the optical SNR of individual DAS channels.

[0064] At this point, while the present disclosure has been presented using some specific examples, those skilled in the art will recognize that the present teachings are not so limited. Accordingly, the present disclosure should be limited only by the scope of the appended claims.

Claims

1. 1. A method for distributed fiber optic sensing / distributed acoustic sensing (DFOS / DAS) using MIMO sampling and phase combining, comprising:

1. A multi-channel DFOS / DAS system comprising: an optical fiber sensor cable; a multi-channel DFOS / DAS interrogator in optical communication with the fiber optic sensor cable; an analyzer configured to analyze the multi-channel DFOS / DAS sensing data received by the multi-channel DFOS / DAS interrogator; providing a multi-channel DFOS / DAS system having operating the multi-channel DFOS / DAS interrogator to interrogate the fiber optic sensor cable on a plurality of channels and receive Rayleigh reflection signals from the fiber optic sensor cable on the plurality of channels; combining the received Rayleigh reflected signals of the plurality of channels to generate a phase signal output exhibiting phase noise characteristics including suppression of out-of-band laser phase noise and additive white Gaussian noise (AWGN); The received Rayleigh reflected signal is received during a channel interrogation period of T frame ; the interrogation signals generated by the interrogator are shifted in time for each of the channels by a fraction of T frame ; the received Rayleigh reflected signals of the plurality of channels are combined into the phase signal output in at least one of a time interleaved and a time aligned manner; The time interleaving method involves splicing N channels, each having an individual sampling rate of 1 / T, to form a sampling rate of N / T; The time alignment method adjusts the timing offset of each individual channel by performing a fast Fourier transform and an inverse Fourier transform on each channel and then combining them so that the contributing samples are aligned in time.

2. The method described in claim 1, wherein the phase vectors from each channel are rotated before performing the time interleaving method.

3. The method of claim 1 , wherein the time interleaving method is performed after the time alignment method is performed on adjacent channels to detect an acoustic signal.

4. 10. The method of claim 1, wherein the interrogation signal of the multi-channel DFOS / DAS interrogator is launched into a single fiber of the fiber optic sensor cable.

5. The method of claim 4 , wherein the plurality of channels comprises a plurality of frequency division multiplexed frequency / wavelength channels.

6. 10. The method of claim 1, wherein the multi-channel DFOS / DAS interrogator interrogation signal is launched into multiple fibers within the same cable.

7. 10. The method of claim 1, wherein the multi-channel DFOS / DAS interrogator interrogation signals are launched into multiple cores of the same multi-core optical sensor cable.

8. 10. The method of claim 1, wherein the interrogation signals of the multi-channel DFOS / DAS interrogator are launched as multiple modes of the same multimode fiber.

9. The time offset between adjacent channels is T frame 2. The method of claim 1, wherein N is the number of channels in the multi-channel DFOS / DAS MIMO operation.

10. 10. The method of claim 9, wherein the multi-channel DFOS / DAS interrogator comprises a single transmitter-receiver configuration (single transmit / receive / frequency division multiplexing TX / RX FDM).

Images