Method and system for distributed acoustic sensing - Patents.com

Abstract

A method and system for distributed acoustic sensing in a city or metropolitan area or the like that includes a dedicated, established fiber optic communication network, including a data center, is provided. [Solution] The method and system include the steps of: (a) selecting an optical fiber cable plant having a route extending across a selected geographic area, including a bundle of optical fibers and forming part of an optical fiber communications network; (b) determining characteristics associated with the optical fibers associated with the plant and / or the selected optical fibers; (c) transmitting emitted light within the optical fiber; (d) receiving reflected light backscattered along the optical fiber; and (e) generating an alarm signal representative of an acoustic event based on the reflected light and the determined characteristics, and may be useful for detecting acoustic events near or within the selected geographic area.

Description

[Technical Field]

[0001] The present invention relates generally to a method for distributed acoustic sensing based on one or more optical fibers, and more particularly to a method for distributed acoustic detection based on one or more installed optical fiber cables. [Background technology]

[0002] Fiber optic distributed acoustic sensing can detect acoustic events in the surrounding area along the optical fiber. The acoustic events can be caused by accidents such as underground excavations near gas pipes, water pipes, or power cables, or by pedestrian and road traffic activity. Different types of events can cause different acoustic signatures in the acoustic event. Therefore, monitoring of acoustic events allows alerts to be generated for the prevention or identification of these events, or for tracking road users in the case of pedestrians and road traffic.

[0003] Deploying dedicated optical fibers for distributed acoustic sensing can make sense from a design perspective, so that the optical fiber conditions and parameters (e.g., spatial uniformity along the optical fiber, trench depth, and level of acoustic attenuation) are known or well controlled at the time of installation. However, installing dedicated optical fibers for distributed acoustic sensing can be expensive and disruptive, especially in and around urban centers.

[0004] The reference to any prior art herein is not, and should not be construed as, an acknowledgment or any form of suggestion that this prior art forms part of the common general knowledge in any jurisdiction, or that this prior art could reasonably be expected by a person skilled in the art to be relevant to and / or combined with other pieces of prior art. Summary of the Invention [Means for solving the problem]

[0005] According to a first aspect of the present disclosure, there is provided a method of distributed acoustic sensing, comprising: selecting a fiber optic cable plant having a route extending across a selected geographic region, said fiber optic cable plant including a bundle of optical fibers and forming part of an established dedicated optical fiber communications network, said fiber optic bundle including channels unused for communications or including optical fibers that are not lit; determining characteristics associated with the optical fiber and / or the selected optical fiber facility, including geographically calibrating the optical fiber for mapping between one or more locations along the length of the optical fiber and one or more corresponding locations within the geographic area; transmitting the output light to the optical fiber; receiving reflected light backscattered along the optical fiber, the reflected light including variations over time; and generating an alert signal representative of the acoustic event based on the variations and the determined characteristics; A method for distributed acoustic sensing is provided, comprising:

[0006] The optical fiber communications network may be a city or metropolitan area network. Alternatively or additionally, the optical fiber communications network may be an enterprise network. The enterprise network may include one or more data centers. When the enterprise network includes multiple data centers, they are interconnected by optical fiber facilities.

[0007] The spatially calibrating step may include generating acoustic calibration signals at or near one or more locations along the optical fiber and detecting corresponding variations at one or more locations within the geographic area, which may further include determining and recording geographic locations of the locations with respect to geographic coordinates on the surface of the Earth, detecting corresponding variations at one or more locations within the geographic area in received reflected light backscattered along the optical fiber, and determining path lengths of the optical fiber corresponding to the one or more geographic locations.

[0008] The step of spatially calibrating may further include correlating the optical fiber path length with the geographic coordinates of one or more locations to generate a look-up table correlating optical path length with geographic coordinates.

[0009] The alert generating step may include determining a location of the occurrence of the incident in a geographic area based on the corresponding variations detected based on the mapping.

[0010] The characterizing step can include acoustically calibrating the optical fiber to reduce the effects of unwanted acoustic interference. The acoustically calibrating step can include applying a spectral filter to the variations to bandpass or bandblock the unwanted acoustic interference.

[0011] The selected geographic area can include a plurality of zones corresponding to a plurality of sections of optical fiber, and the spectral filter applying step can include applying a spectral filter having bandpass or bandstop characteristics based on one of the plurality of zones or corresponding sections. Alternatively or additionally, the spectral filter applying step can include applying a spectral filter having bandpass or bandstop characteristics based on a time of day and / or a day of the week.

[0012] The characterizing step can include physically calibrating the optical fiber. The physically calibrating step can include obtaining any one or more of core characteristics, attenuation characteristics, and trench characteristics of the optical fiber. The core characteristics can include core diameter and / or numerical aperture. The attenuation characteristics can include propagation loss, existing defects, and / or splices. The trench characteristics can include burial conditions and / or cable enclosure conditions.

[0013] It will be appreciated that there are significant variations in the materials surrounding the trench and cable, including rock, gravel, concrete, sand, water, soil, clay, bitumen, or a combination of one or more of these, which can significantly change the acoustic impedance of these materials, resulting in a change in impedance between the perturbation or source of interest and the fiber optic cable.

[0014] Seismic calibration of the surrounding medium is performed in a similar manner to the type of seismic profiling performed in oil and gas exploration, including known techniques for characterizing the acoustic impedance around well casing. In this case, the objective is not to determine the type and condition of the surrounding materials, but rather to determine the acoustic and seismic transfer functions that these materials spatially form between the fiber and the perturbation of interest. Such transfer functions allow for heterogeneous media to be taken into account, thus enabling accurate estimation of the spatial location, dynamics, and source frequency of any given perturbation around the optical fiber. Thus, three precursor calibration steps can be performed: optical calibration of the fiber itself, geospatial calibration, and seismic calibration of the surrounding medium.

[0015] The alarm signal generating step may include classifying the alarm signal into one or more classes of alarms based on the acoustic signature of the fluctuations. The selected geographic area may include a plurality of zones corresponding to a plurality of sections of optical fiber, each zone or corresponding section being associated with generating one or more selected classes of alarms. Alternatively or additionally, the selected geographic area may include a plurality of zones corresponding to a plurality of sections of optical fiber, each zone or corresponding section being associated with not generating one or more excluded classes of alarms.

[0016] One or more selected or excluded classes of alarms corresponding to each zone or section of optical fiber may relate to rail monitoring, road monitoring, and perimeter intrusion detection.

[0017] The method may further include switching the transmission of the emitted light and the reception of the reflected light to another unused channel or unlit optical fiber for communication, the other optical fiber being in another optical fiber bundle in another selected optical fiber cable plant having another route extending over another selected geographic area. The switching step may include time multiplexing the transmission of the emitted light and the reception of the reflected light to multiple optical fiber cable plants.

[0018] The method can further include bypass splicing to bypass the connection infrastructure. The connection infrastructure can also include one or more fiber transport panels (FTPs) or patch panels. The bandpass or bandstop frequency range of the spectral filter can also be adjusted, and a resulting noise level can be determined based on the adjusted frequency range.

[0019] The method may further include determining a rate of the acoustic event and, based on the determination, suppressing or enabling generation of an alert signal representative of the acoustic event.

[0020] According to a second aspect of the present disclosure, 1. A system for distributed acoustic sensing, comprising a distributed sensing unit and an optical switch, The distributed sensing unit transmitting the output light to an optical fiber; receiving reflected light backscattered along the optical fiber, the reflected light including variations over time; and generating an alert signal representative of an acoustic event based on the variation; The optical switch comprises: coupling said distributed sensing unit to a selected one of a plurality of fiber optic cable plants, each plant extending across a respective selected geographic area and including a bundle of optical fibers and forming part of a dedicated optical fiber telecommunications network, the fiber optic bundle including unused channels for communication or non-sensing illuminated optical fibers; A system for distributed acoustic sensing is provided.

[0021] The optical switch and distributed sensing units may be located in a data center that connects to an enterprise network.

[0022] Multiple fiber cable plants may be connected to or terminated at a data center.

[0023] Further aspects of the invention and further embodiments of the aspects described in the preceding paragraphs will become apparent from the following description, given by way of example and with reference to the accompanying drawings, in which: [Brief explanation of the drawings]

[0024] [Figure 1a] FIG. 1a shows an example of a system for distributed acoustic sensing. [Figure 1b] FIG. 1b shows an example of a density plot over time of the electrical signal generated by the system of FIG. 1a. [Figure 2] FIG. 2 shows an example of the disclosed method of distributed acoustic sensing. [Figure 3a] FIG. 3a is a schematic diagram of a data center cluster including several data center buildings. [Figure 3b] Figure 3b is a schematic diagram of a single data center building. [Figure 4] Figure 4 shows aerial maps of different trench conditions. [Figure 5] FIG. 5 illustrates an example of the system of FIG. 1 optically coupled to an optical switch and a plurality of fiber optic cable plants. DETAILED DESCRIPTION OF THE INVENTION

[0025] The principle of fiber optic distributed acoustic sensing relies on the occurrence of an acoustic event that causes a corresponding local perturbation in the refractive index of the optical fiber. Due to the perturbed refractive index, an optical interrogation signal transmitted along the optical fiber and then backscattered in a dispersive manner along the length of the fiber (e.g., via Rayleigh scattering or other similar scattering phenomena) exhibits fluctuations over time (e.g., intensity and / or phase) of the reflected light. The magnitude of the fluctuations is related to the severity or proximity of the acoustic event. The timing of the fluctuations along the distributed backscattering time scale is related to the location of the acoustic event.

[0026] In one example, a unit 100 for using distributed acoustic sensing (DAS) is shown in FIG. 1a. The DAS unit 100 includes an optical time domain reflectometer (OTDR) 102. The OTDR 102 includes a light source 104 that emits an optical interrogation signal 106. The interrogation signal 106, transmitted to an optical fiber 105, may be in the form of a short optical pulse. The OTDR 102 includes a photodetector 108 configured to detect reflected light 110 and generate a corresponding electrical signal 112 having an amplitude proportional to the reflected light intensity. The DAS unit 100 also includes a processing unit 114 configured to measure variations in the electrical signal 112, either within the OTDR 102 or separately from the OTDR 102, to determine acoustic events based on measured variations 116 in intensity compared between two different times (t1 and t2). FIG. 1b shows an exemplary density plot combining the electrical signal 112 generated by the DAS unit 100 over time. The horizontal axis (labeled "Channel") represents position along the fiber, the vertical axis (labeled "Time") represents time, and the color-coded amplitude of the plot represents reflection intensity. In FIG. 1b, line-like features with relatively constant slopes are associated with moving objects (the slope indicates velocity) that cause associated acoustic events detected by the DAS unit 100. If the OTDR is phase-sensitive, phase variations in the reflected light can additionally or alternatively be measured. FIG. 1b is also offset to remove the attenuation gradient of the electrical signal 112 present in FIG. 1a. The determined acoustic events can indicate specific stationary or moving occurrences, such as excavation, excavation, traffic flow, passing trains, and pedestrian flow.

[0027] A method for distributed acoustic sensing is described herein. An arrangement of the disclosed method 200 is shown in FIG. 2. Generally, the disclosed method includes the steps of (a) selecting a fiber optic cable installation having a path extending across a selected geographic area, the fiber optic cable installation including a bundle of optical fibers and forming part of an optical fiber communications network (step 202); (b) determining characteristics associated with the optical fibers and / or the selected optical fibers (step 204); (c) transmitting an output light into the optical fiber (step 206); (d) receiving reflected light scattered along the optical fiber (step 208); and (e) generating an alarm signal indicative of an acoustic event based on the reflected light and the determined characteristics (step 210). The disclosed method may be useful for detecting acoustic events near or within a selected geographic area.

[0028] Rather than deploying dedicated optical fiber for distributed acoustic sensing, the disclosed method relies on selecting an existing optical fiber cable plant that forms part of an optical fiber communications network. For example, the optical fiber communications network may be a datacom network (e.g., to or from a data center), a telecom network (e.g., to or from a local exchange), or an enterprise network (e.g., to a large enterprise and a cloud and a data center supplier, or from a large enterprise and a cloud and a data center supplier, or between enterprises). Selecting an existing communications network for distributed acoustic sensing avoids the expense of installing dedicated cables, but may require overcoming many technical challenges. As a result, those skilled in the art would not be motivated to select an existing optical fiber communications network for distributed acoustic sensing. Furthermore, those skilled in the art would not recognize the technical challenges of using an existing optical fiber communications network for distributed acoustic sensing and would not recognize how to address these technical challenges.

[0029] Fiber optic cable installation options

[0030] In one example, the disclosed method 200 includes, in step 202, selecting an existing optical fiber plant. The selected optical fiber cable plant has a route extending across a selected geographic area. The selected geographic area may be a city or metropolitan area. In one configuration, the optical fiber cable plant includes a bundle of optical fibers, where the optical fibers are unlit for communications purposes. Thus, the unlit optical fibers can be used for distributed acoustic sensing. In another configuration, the optical fiber cable plant includes a bundle of optical fibers, where one or more of the optical fibers include time or wavelength channels that are not used for communications. For example, in an enterprise network where dense wavelength division multiplexing (DWDM) is used, only some, but not all, of the DWDM channels carry communications traffic. The remaining DWDM channels may be unused. Thus, one or more of the unused channels can be used for distributed acoustic sensing. Further references to the use of unused channels or unlit optical fibers are applicable to either configuration.

[0031] One or more factors can influence the selection of an optical fiber plant. For example, an appropriate selection is based on the installation forming part of an existing enterprise network. Unlike dedicated optical fiber with an end-to-end geometry, an enterprise network connects to multiple optical fiber cable plants, for example, through one or more data centers or hubs. A data center represents a collection of servers and storage that has the advantage of having a large amount of fiber and fiber cable. In one configuration, multiple optical fiber cable plants can span different regions of a city or metropolitan area. Therefore, selecting an enterprise network facilitates relatively large geographic coverage for distributed acoustic sensing by accessing multiple optical fiber cable terminations or connection points at or near a central location (e.g., a data center or hub).

[0032] In one configuration, the selection of optical fiber facilities can be based on selecting an optical path through multiple data centers. For example, in a hub-and-spoke configuration, a central data center is connected to multiple nearby data centers. As another example, in a data center cluster, multiple nearby data centers are interconnected. FIG. 3a shows an example data center cluster 300 including four data center buildings 302a, 302b, 302c, and 302d that communicate with each other. The data center buildings 302 within the cluster 300 share a trunked communication fiber cable 304 that enables external communication from the cluster 300 to remote service providers (e.g., carriers and content providers). Within the cluster 300, the data center buildings 302 are communicatively interconnected by an intra-data center communication fiber cable 306 in a trench that includes cross-connects between server banks in different data center buildings 302. Within each data center building 302, there is a cross-connect 308. The intra-data center fiber cable 306 represents critical infrastructure to the data center operator because it is not the service provider's responsibility, a break in the intra-data center fiber cable 306 impacts customer hosting, and lacks SDN overlay protection, similar to a carrier mesh network. The trunk fiber cable 304, the intra-data center fiber cable 306, and the cross-connect 308 each represent optical fiber assets. The selection of optical fiber equipment can be based on selecting an optical path that traverses most, if not all, of the optical fiber assets to be protected.

[0033] Within each data center, as shown at 302e in FIG. 3b, fiber breakout cabinets 310a and 310b are shown for receiving respective outdoor fiber optic cables 312a and 312b within fiber transport panels 314a and 314b. The disclosed method 200 may include a bypass splicing step. Bypass splicing refers to a thermal splicing step in which the sensing fiber 318 is spliced ​​at 320 to bypass connection infrastructure such as one or more fiber transport panels (FTPs) 314a and 314b or patch panels. Some of the fiber optic cables 322 extend from the breakout cabinets 310a and 310b to building customer cabinets 324. Others are optical cross-connect cables, as shown at 326.

[0034] Deploying coherent OTDRs requires a tighter tolerance for back reflections than standard telecommunications transceivers (e.g., as found in enterprise networks). The use of flat connector types, such as FC-PC, commonly found in FTPs results in undesirable levels of back reflections. The bypass splicing step provides that in a data center environment, the sensing fiber 105 in a selected cable plant used for distributed acoustic sensing (or the sensing fiber 118) is not connected to an FTP or patch panel. In one configuration, the bypass splicing continues the sensing fiber 105 or 118 from a patch lead in the DAS unit 110 to a termination unit. The disclosed method 200 may also include a step of determining whether the sensing fiber 105 or 118 passes through any connection infrastructure. According to this determination, the disclosed method 200 may include removing the portion of the sensing fiber 105 or 118 corresponding to the connection infrastructure and splicing in a bypass fiber as a replacement.

[0035] Furthermore, the selection of enterprise networks over long-haul networks is related to the observation that relatively short-reach networks do not require regeneration or amplification and therefore tend to have a relatively large number of optical fibers (i.e., a relatively large cross-section) in the bundle, increasing the likelihood of one or more unused channels or unlit optical fibers. Unused channels or unlit optical fibers may be installed as part of the optical fiber bundle to serve as spare capacity to allow for future growth in network demand. In some embodiments, all fibers are deployed, but it is also feasible that less-critical fibers may be used for applications other than communications traffic, such as distributed acoustic sensing. In contrast, long-haul networks may not be an appropriate choice because the long reach of the network requires regeneration or amplification. Therefore, long-haul networks tend to have a relatively small cross-section, reducing the likelihood of unused channels or unlit fibers, or fibers that do not play a critical role.

[0036] An example of a large-section optical fiber cable is Prysmian's (registered trademark of Prysmian Cavi E Sistemi Engergia srl) Multi Loose Tube Duct Cable, which has 216 to 624 fibers. In other examples, large cross-section can refer to an optical fiber cable with 32 to 64 optical fibers. In some configurations, the reach of a city or metropolitan area network is less than approximately 50 to 100 km. This relatively short reach is limited by attenuation and optical receiver sensitivity, without regeneration or amplification. In other configurations, this disclosure is not limited to city or metropolitan area networks, nor is it limited to a reach of 50 to 100 km. This disclosure is applicable to other communication networks (e.g., international submarine optical fiber cables) that have unused channels or disable optical fiber for communication purposes.

[0037] Alternatively or additionally, the selection of fiber optic cable locations may be based on proximity to existing infrastructure, which may include, but is not limited to, roads, railroads, water, power, electricity, telecommunications, data centers, buildings, bridges, tunnels, pedestrian access ways, rivers, harbors, lakes, docks, construction sites, industrial parks, and critical infrastructure. Alternatively or additionally, the selection of fiber optic cable locations may be based on the type of stakeholder that will be serving the alarm signal, which may include, but is not limited to, emergency and disaster management, critical infrastructure management, civil services, public administration services, law enforcement, and corporate security and asset management.

[0038] Once the fiber optic cable plant is selected, the disclosed method 200 includes optically coupling (not shown) the DAS unit 100 to the selected plant. Depending on the termination type of the optical fiber 105, the optical coupling may include splicing (if the termination type is bare fiber) and / or splicing (if the termination type is an optical connector, such as an SC or FC connector). If an acoustic event is determined, an alarm signal representative of the acoustic event may be generated. The alarm signal generation step may include classifying the alarm signal into one or more classes of alarm (e.g., excavation threat, heavy pedestrian traffic, heavy road traffic, etc.) based on the acoustic signature of the fluctuation. Some techniques for alarm classification are summarized and further referenced in, for example, "Fiber Sensing: Optical Fiber Monitors the Arterial Networks of Commerce," Laser Focus World, volume 51, issue 08, June 8, 2015 (http: / / www.laserfocusworld.com / articles / print / volume-51 / issue-08 / features / fiber-sensing-optical-fiber-monitors-the-arterial-networks-of-commerce.html). In one configuration, a geographic area is divided into multiple zones corresponding to multiple sections of optical fiber. In this configuration, each zone or corresponding section is associated with the generation of one or more selected classes of alarms (or the non-generation of one or more excluded classes of alarms). For example, each zone may be represented by a different stakeholder. If the stakeholder is a utility connection operator (e.g., for supplying gas, electricity, or water), the class of alarm selected for power generation may relate to excavation, drilling, or excavation near supply cables. Furthermore, the class of alerts excluded for generation may be associated with pedestrian traffic or road traffic, whereas if the stakeholder is a transport operator (e.g., a bus or rail operator), the class of alerts selected for generation may be associated with pedestrian traffic or road traffic.Additionally, classes of alarms excluded for generation may relate to excavations, drilling, or excavations near supply cables. One or more selected or excluded classes of alarms along the same optical fiber for different zones or sections may relate to rail monitoring, road monitoring, and perimeter intrusion detection.

[0039] Acoustic Calibration

[0040] The disclosed method 200 also includes determining characteristics associated with the selected optical fiber of the optical fiber and / or facility at step 204. In one example, this determining step 204 includes acoustic calibration.

[0041] Urban or metropolitan areas where distributed acoustic sensing is performed are likely to be areas with unwanted acoustic interference. Unwanted acoustic interference can interfere with, mask, or otherwise affect the characteristics of determined acoustic events. In one configuration, to reduce the effects of unwanted acoustic interference, the disclosed method 200 includes applying spectral filtering to the detected fluctuations to reduce or remove fluctuations associated with the unwanted acoustic interference. For example, acoustic interference resulting from bus engine noise may typically range from 1 to 120 Hz due to the low rotational speed of the engine at low driving speeds. Low rotational speeds result in less pervasive acoustic frequencies compared to higher acoustic frequencies. To reduce the effects of such intrusive noise, the detected fluctuations may be spectrally filtered to remove or attenuate low-frequency fluctuations, for example, by a low-pass filter with a cutoff frequency around 100 to 150 Hz.

[0042] Additionally, the disclosed method can selectively apply spectral filtering to one or more zones of a geographic area. Different zones of a geographic area may require different or no spectral filtering. For example, away from urban zones and on highways, it may not be necessary to apply a low-pass filter, but the disclosed method may apply a band-stop filter to remove or attenuate tire noise. Alternatively or additionally, the disclosed method can selectively apply spectral filtering to variations based on time, e.g., time of day or day of the week.

[0043] To determine appropriate bandpass or bandstop frequencies for a particular zone or a particular time, step 204 may include obtaining baseline data of frequency content along the optical fiber. The baseline data may be obtained separately for each zone and / or for a particular time. The baseline data may be obtained by monitoring fluctuations over a particular duration. The frequency content in the baseline data may be based on averaging several sets of data measured over a particular duration. The particular spectral filter used based on the zone or time may be configured to have an opposite (e.g., inverted) bandpass or bandstop profile to the frequency content in the baseline data.

[0044] In one configuration, the disclosed method 200 includes adjusting a bandpass or bandstop frequency range. The disclosed method 200 may further include determining a resulting noise level based on the adjusted frequency range. This adjustment can be performed dynamically until the noise level falls below a specific threshold. Alternatively, or additionally, the adjustment may be performed recursively until the noise level falls below a specific or another threshold. The noise level may be determined based on the noise bandwidth at each calibration site. The noise bandwidth may be determined by using channel analysis and acoustic waterfall functionality in the DAS software and selecting appropriate waterfall and excitation frequencies for each calibration site. In one configuration, the noise bandwidth is determined by implementing a filter with an integration time significantly longer than the transient signal of interest. For example, the noise background may be calculated based on a time integration over tens of minutes to relate the background noise threshold to a range of positions along the fiber length (e.g., corresponding to a communication pit below ground level). The difference in the acoustic intensity of the background noise compared to the intensity of the electrical signal 112 (e.g., relative to the telecommunication pit) provides an indication of the noise bandwidth.

[0045] Spatial Calibration

[0046] Fiber optic cable plants do not extend in straight lines. Furthermore, unlike the use of dedicated fiber, where the correlation between the position along the fiber length and the corresponding location in a geographic area is known during installation, selecting existing fiber optic cable presents some uncertainty in this correlation. For example, optical fiber in a telecommunications network within an exchange may be wound on a spool to provide extra length for flexibility for repair or splicing purposes. Furthermore, the length of the optical fiber may change during repair or splicing. Therefore, it is necessary to calibrate this correlation to accurately map any detected variations in position along the fiber to the corresponding location of any acoustic event.

[0047] In one configuration, step 204 includes spatially calibrating between locations along the optical fiber and locations within the geographic area. Spatial calibration can include generating an acoustic calibration signal (e.g., a single-frequency tone at 420 Hz + / - 5 Hz selected to be different from typical noise sources in urban centers) at specific locations in the geographic area to induce variations for detection along the length of the optical fiber. By limiting the acoustic calibration signal frequency to 420 Hz + / - 5 Hz, other acoustic noise sources in urban centers can be eliminated. Eliminating other acoustic noise sources allows for detection of one strong signal corresponding to the single-frequency tone, as seen near channel 1990 in Figure 1b. As an illustrative example, an operator can travel to a specific location, e.g., a cable pit along a selected optical fiber cable plant, where the cable is typically coiled in a covered pit to allow additional cable runs to be made from that point, if needed. More than 50 meters of cable can be coiled in each pit spaced at intervals of x and y meters along the cable trench. As a result, there is a significant (up to 15% or more) discrepancy between the cable path length and the geographic path length.

[0048] In the geospatial calibration process, a calibration acoustic signal is generated at each successive pit location, and decimal latitude and longitude GPS coordinates are obtained for successive cable pits. Alternatively, other locations along the cable route may be used, and if the GPS coordinates are recorded or logged, the calibration acoustic signal is easily detectable. Optical fluctuations corresponding to the calibration acoustic signal are expected to be detected at specific locations along the optical fiber. Corresponding pairs of coordinates corresponding to a location within a geographic region and a location along the optical fiber at which the fluctuation is detected form geospatial calibration reference points, which in turn form part of a lookup table of the type shown below, which includes additional spatial calibration points within the geographic region along the fiber.

[0049] [Table 1]

[0050] In the table above, each point is identified by a date and timestamp, and a name or reference that is similarly associated with a GPS coordinate captured by a data logger on a suitable GPS-enabled device, such as a smartphone. The date logger transmits this information to a data center where it is mapped. The peak point of the fluctuation on the optical fiber is recorded along with the end optical distance and start optical distance if a pit is recorded, which is likely to be the presence of a fiber loop. The width reading represents the length of that loop. The type column indicates whether the reading is taken on the cable (in which case the energy peak point is relevant) or on the pit (in which case the start optical distance, end optical distance, and end optical width are relevant).

[0051] If an acoustic event is detected at a location along the fiber between two calibration points, interpolation (e.g., linear or nonlinear) can be used to estimate the location of the corresponding occurrence within the geographic area. If an acoustic event is detected at a location along the fiber beyond the first and last calibration points, extrapolation (e.g., linear or nonlinear) can be used to estimate the location of the corresponding occurrence within the geographic area. By using the above geospatial calibration method, substantial variations encountered in optical fiber networks that are not dedicated to performing location and detection functions, but rather are dedicated communications and enterprise networks of the type contemplated in this disclosure, can be reduced, and location accuracy can be significantly and substantially improved throughout the network. This calibration of optical path length to geospatial location allows for accurate queuing of personnel to a threat or cable break event in a manner that applicant understands is not possible with existing methods of determining the location of cable events.

[0052] Physical Calibration

[0053] Step 204 can further include physical calibration of the optical fiber selected for acoustic distributed sensing. Unlike dedicated fiber deployment, the disclosed method 200 generally involves the use of optical fibers with uncharacterized properties. For example, their core, attenuation, and trench characteristics are generally unknown. The resulting characteristics can be used to calibrate the detected variations.

[0054] In one configuration, step 204 includes obtaining core characteristics of the optical fiber. The core characteristics may include the core diameter and / or the numerical aperture. The core characteristics may affect the launch power of the light source, which in turn affects the intensity of the reflected light. For example, the minimum reflected intensity that can be increased by the launch power is limited by the noise floor of the photodetector. Based on the core characteristics, the launch power of the light source may be adjusted accordingly in step 206 to achieve the desired reach. Alternatively or additionally, step 204 includes obtaining attenuation characteristics of the optical fiber. The attenuation characteristics may include propagation loss per unit length, existing faults, and / or splices. The attenuation characteristics may affect the reach of distributed acoustic sensing. For example, increased propagation loss reduces the reach. Furthermore, existing faults and / or splices may cause different variations (amplification or reduction) in the reflected light compared to variations in an otherwise fault-free or splice-free fiber.

[0055] Based on the attenuation characteristics, the launch power may be adjusted accordingly in step 206 to achieve the desired range. Alternatively or additionally, step 204 may include obtaining trench characteristics of the optical fiber. Trench characteristics include characteristics affected by burial conditions and / or cable enclosure conditions. For example, the optical fiber may be enclosed in a 100 mm PVC conduit and / or buried in a cement trench, earthenware, or underground tunnel. Trench characteristics can mask or otherwise affect the acoustic signature of an acoustic event. Figure 4 shows an aerial map of the Circular Quay area in Sydney, Australia. The aerial map is overlaid with multiple sections of optical fiber (represented by at least the labels "ch1346" and "ch1384"), each corresponding to a location near the Circular Quay area (the corners of Bent and Bligh Streets, and Macquarie and Bridge Streets, respectively). Here, different sections of the optical fiber are exposed to different trench conditions, which can be obtained from the cable supplier or from acoustic measurements at a test site. Trench characteristics obtained, for example, via acoustic measurements, may be used to calibrate the detected variations in step 210 in conjunction with some of the techniques in alarm classification described above in "Fiber Sensing: Optical Fibers Monitor Commercial Arterial Networks."

[0056] Static, slow-moving, or fast-moving occurrences

[0057] As described above, the acoustic event (including the determined object) can indicate a particular stationary or moving occurrence. For example, as shown in FIG. 1b, a feature such as a line with a relatively constant slope is associated with a moving object (the slope indicating speed) that causes the associated acoustic event to be detected by the DAS unit 100. The disclosed method 200 can include determining whether the acoustic event is stationary or moving. This determination can include whether the moving acoustic event is associated with a slowly moving noise source (e.g., excavation, drilling, bore tunneling, etc.) or a fast moving noise source (e.g., vehicle, train, etc.). For example, the determination can include comparing the estimated speed of the acoustic event (e.g., based on the slope of the line) to a threshold speed value. If the estimated speed of the acoustic event is lower than the threshold speed value, the acoustic event is determined to be slowly moving or stationary; otherwise, the acoustic event is determined to be fast moving. The disclosed method 200 can further include suppressing generation of an alarm signal representing the acoustic event based on the determination. This suppression is particularly useful in avoiding false alarms in urban environments where the occurrence of benign urban activities (e.g., pedestrian walking, bus and train movement) can be much higher than the occurrence of real threats (e.g., technicians or engineers working on fiber optics, drilling, excavation and drilling).

[0058] For example, in the context of telecommunications infrastructure, detection of a stationary or slowly moving noise source indicates that the noise source is likely a threat, while detection of a fast-moving noise source indicates that the noise source is unlikely to be a threat. Threat likelihood may be graded (e.g., on a scale of low, medium, and high, or on a scale of 1 to 10). The threshold grade at which a threat alert is generated (or not) may be adjustable, e.g., dynamically, based on the use case context. Without such suppression, a large number of false alarms may be generated, to the point of invalidating the disclosed method. In one configuration, the threshold velocity value and / or threshold slope may be adjusted to reduce the number of false alarms.

[0059] Switching

[0060] To increase the total length of existing fiber optic cable that can be monitored from one DAS unit 100, in one configuration the disclosed method further includes switching the transmission of outgoing light and reception of reflected light to another unused channel or unlit optical fiber for communication, the other optical fiber being in another bundle of optical fibers in another selected optical fiber cable plant having another route extending over another selected geographic area.

[0061] FIG. 5 schematically illustrates an example of a DAS unit 100 optically coupled to an optical switch 500, e.g., via optical fiber, to form a distributed acoustic sensing system 400. The optical switch 500 may be located in a data center or hub that connects to multiple optical fiber plants (502a, 502b, 502c, 502d, and 502e). The DAS unit 100 may be co-located with the optical switch 500 or may be located separately from the optical switch 500. When the DAS unit 100 is co-located with the optical switch 500 in a data center or hub, the DAS unit 100 is in close proximity to an ecosystem of optical communication networks (e.g., enterprise networks, cloud provider networks, IP transit provider networks, Internet service provider networks, and telecommunications carrier networks, including regional, metropolitan, and long-haul networks). Thus, the DAS unit 100 in the data center or hub can be configured to selectively access one or more different communication network types for distributed acoustic sensing. As discussed above, in one example, the enterprise network option offers the advantage of a relatively large optical fiber cross-section.

[0062] The optical switch 500 is configured to couple light between the DAS unit 100 and any one of a plurality of optical fiber plants (502a, 502b, 502c, 502d, and 502e). The plurality of optical fiber plants 502 together span a larger geographic area than would be spanned by any one of the fiber plants 502 alone. In one configuration, the optical switch 500 time-multiplexes the plurality of optical fiber plants. For example, the transmission of outgoing light and the reception of reflected light are cycled through the plurality of optical fiber plants and switched to the next plant at regular intervals.

[0063] In some circumstances, different installations may detect the same acoustic event to increase the spatial accuracy of locating the acoustic event. For example, an acoustic event may be detected by both installations 502b and 502c, but not by installations 502a, 502d, and 502e. Such a detection indicates that the corresponding occurrence is located in the geographic area between 502b and 502c. If only installation 502b is queried, the detection may present uncertainty as to where the occurrence is located (e.g., between installations 502a and 502b, or between installations 502b and 502c). If the acoustic event generates seismic waves, the waves may propagate across multiple cables. In this case, the epicenter may be triangulated based on measurements of the propagation direction and time-of-flight calculations. Now, the configuration of the present disclosure will be described; it will be apparent to those skilled in the art that the described configuration has the following advantages:

[0064] The expense of deploying dedicated optical fibers for distributed acoustic sensing is avoided.

[0065] The ability to measure assets or acoustic events in specific locations where it would not otherwise be possible to locate a dedicated sensor system because the property or land is owned by another party.

[0066] The calibration step adapts the existing installation to imperfect or non-ideal characteristics that would not otherwise be present in a dedicated fiber optic deployment.

[0067] When switching is used, the system can be easily scaled to extend the geographic area of ​​interest or the total length of cable to be monitored. Additionally, switching can increase the spatial accuracy of identifying locations that appear outside the cable.

[0068] It will be understood that the invention disclosed and defined herein extends to all alternative combinations of two or more of the individual features mentioned or apparent from the text or drawings. For example, any one or more calibration steps may be used separately or together. All of these different combinations constitute various alternatives of the present disclosure.

Claims

1. 1. A distributed acoustic sensing method, comprising: selecting a fiber optic cable plant having a route extending across a selected geographic area, the fiber optic cable plant including a bundle of optical fibers, the optical fibers forming part of an established dedicated optical fiber communications network, the fiber optic bundle including: (1) a first subset including at least one optical fiber or channel for carrying communications traffic; and (2) a second subset including at least one optical fiber or channel that is not being used or lit for communications; using the second subset of the at least one optical fiber or channel of the optical fiber bundle for distributed acoustic sensing, determining characteristics associated with at least a portion of the optical fiber bundles within the selected fiber optic cable plant, including performing spatial calibration by geographically mapping between one or more locations along the length of the optical fiber bundles and one or more corresponding locations within the selected geographic area; acoustically and / or seismically calibrating a medium located between a perturbation source and the optical fiber by determining a transfer function produced by the medium; transmitting the output light through the second subset of at least one optical fiber or channel using an optical time domain reflectometer (OTDR); and using the OTDR to receive reflected light backscattered along the second subset of optical fibers or channels, the reflected light comprising variations over time; generating an alert signal representative of an acoustic event based on the variation and the determined characteristic; The step of determining the characteristics includes physically calibrating the second subset of at least one optical fiber or channel by obtaining one or more of core characteristics, including at least one of core diameter and numerical aperture, trench characteristics, including at least one of burial conditions and cable enclosure conditions, and attenuation characteristics, including at least one of propagation loss, existing defects, and splices, of the second subset of at least one optical fiber or channel. nothing, Distributed acoustic sensing methods.

2. 10. The distributed acoustic sensing method of claim 1, wherein the optical fiber communications network is an established city or metropolitan area network or enterprise network that includes multiple data centers interconnected by optical fiber facilities.

3. 2. The method of distributed acoustic sensing of claim 1, comprising determining whether the acoustic event is stationary, slowly moving, or fast moving by comparing an estimated speed of the acoustic event to a threshold speed value, wherein the estimated speed of the acoustic event is determined based on a gradient of a footprint produced by the acoustic event over a period of time.

4. The method of claim 3 , further comprising suppressing generation of the alarm signal representative of the acoustic event based on the determination.

5. The geographical mapping may further include: generating an acoustic calibration signal at or near one or more locations along the second subset of at least one optical fiber or channel; determining and logging the geographic location of said location with respect to geographic coordinates on the surface of the Earth; detecting corresponding variations in received reflected light backscattered along said second subset of at least one optical fiber or channel at one or more locations within the geographic area; determining a path length of the second subset of at least one optical fiber or channel corresponding to one or more of the geographic locations; The method of distributed acoustic sensing according to claim 1 .

6. 6. The distributed acoustic sensing method of claim 5, wherein the geographic mapping further comprises correlating the path lengths of the second subset of at least one optical fiber or channel with the geographic coordinates of the one or more locations to generate a lookup table correlating the path lengths with geographic coordinates.

7. 2. The method of distributed acoustic sensing of claim 1, wherein generating the alarm signal comprises determining a location of an occurrence of an incident in the selected geographic area based on corresponding variations detected based on the geographic mapping.

8. 2. The method of distributed acoustic sensing of claim 1, wherein the step of determining the characteristics includes acoustically calibrating the second subset of at least one optical fiber or channel to reduce the effects of undesired acoustic interference.

9. The method of distributed acoustic sensing of claim 8 , wherein the acoustically calibrating step includes applying a spectral filter to the variations to band-pass or band-block undesired acoustic interference.

10. the selected geographic area includes a plurality of zones corresponding to a plurality of sections of the second subset of at least one optical fiber or channel; 10. The distributed acoustic sensing method of claim 9, wherein applying the spectral filter comprises applying the spectral filter having band-pass or band-stop characteristics based on one of the plurality of zones or corresponding sections.

11. 10. The method of claim 9, wherein applying the spectral filter comprises applying a spectral filter having band-pass or band-stop characteristics based on time of day and / or day of week.

12. The method of claim 1 , wherein generating the alarm signal comprises classifying the alarm signal into one or more classes of alarms based on an acoustic signature of the variation.

13. 13. The distributed acoustic sensing method of claim 12, wherein the selected geographic area includes a plurality of zones corresponding to a plurality of sections of the second subset of at least one optical fiber or channel, and each zone or corresponding section is associated with the generation of one or more selected classes of alerts.

14. 13. The distributed acoustic sensing method of claim 12, wherein the selected geographic area includes a plurality of zones corresponding to a plurality of sections of the second subset of at least one optical fiber or channel, each zone or corresponding section being associated with non-generation of alarms of one or more excluded classes.

15. 14. The distributed acoustic sensing method of claim 13, wherein the one or more selected or excluded classes of alarms corresponding to each zone or section of a second subset of at least one optical fiber or channel relate to rail monitoring, road monitoring, and perimeter intrusion detection.

16. connecting the distributed acoustic sensing units to a plurality of fiber cable plants via optical switches; 2. The distributed acoustic sensing method of claim 1, wherein a fiber optic cable plant is selected from the plurality of fiber cable plants by the optical switch, and the alarm signal is generated by the distributed acoustic sensing unit based on the variation and the determined characteristic.

17. 10. The distributed acoustic sensing method of claim 1, wherein the second subset of at least one optical fiber or channel is used to acoustically detect acoustic events that are excavations, traffic flow, trains passing by near or within a selected geographic area, and pedestrian flow.

18. 2. The distributed acoustic sensing method of claim 1, wherein determining the transfer function produced by the medium comprises determining variations in acoustic impedance of the medium such that at least one of spatial location, dynamics, and frequency of the perturbation source can be determined taking into account variations in acoustic impedance of the medium.

19. 19. The method of distributed acoustic sensing of any one of claims 1 to 18, wherein the established dedicated optical fiber communication network includes at least one submarine optical fiber for submarine communication purposes that is unused or unlit.

Images