Fault signal detection system and analysis positioning method

Abstract

The application provides a fault signal detection system and analysis positioning method, relates to the technical field of signal detection, and periodically applies a narrow-band acoustic disturbance with adaptive frequency selection and power at the optical cable sheath by using an injection module; a passive wireless tag node arranged along the line collects multi-dimensional data such as time stamp, amplitude, frequency, phase and local temperature in real time, and uploads the data through near-field coupling. The gathering module completes node identification association and batch aggregation, the analysis and construction module constructs an acoustic propagation path map, and the graph search algorithm is used to dynamically recover the actual laying path after temperature difference segmentation compensation. The projection and identification module maps the map to the geographic information system coordinate system, outputs the fault space position, realizes real-time and accurate optical cable fault positioning without topological prior conditions, significantly reduces manual inspection and misjudgment rate, and is suitable for long-distance communication or power optical cable scenes with multiple branches and multiple repairs.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of signal detection, in particular to a fault signal detection system and an analysis positioning method. BACKGROUND

[0002] In modern communication infrastructure, optical cables are widely used in interconnection between telecom, energy, transportation and data center due to their advantages of large transmission capacity and good stability. However, due to environmental influence, construction error, natural aging or accidental damage by third party, local fiber breakage, loose joint or sheath damage may occur in optical cable line. Once such faults occur, if they cannot be located and repaired in time, it will lead to interruption of communication service or serious performance degradation, and even cause large-scale information island or economic loss. The commonly used optical cable fault detection methods, such as optical time domain reflectometer (OTDR) test method, mostly adopt single-end detection mode, which relies on pre-drawn and updated line topology map and accurate line length information. However, in actual operation environment, due to the complexity of optical cable network layout, frequent changes of line topology and multiple repairs, as well as the accumulation of differences between original drawings and on-site layout, it is often difficult to provide real-time, accurate and up-to-date topology data, so that the traditional single-end optical test method is easily affected by multiple reflections, branch scattering or complex distribution of line, resulting in large positioning error or blind area. In addition, such methods usually require multiple comparisons and on-site surveys by manual work, which has the problems of slow response speed, large positioning error, strong dependence on manual work and high operation and maintenance cost.

[0003] Therefore, under the condition of lacking or being difficult to obtain accurate line topology information, how to realize real-time and accurate spatial positioning of optical cable fault points and significantly reduce manual dependence and misjudgment rate has become a technical problem to be solved. SUMMARY

[0004] In view of the deficiencies of the prior art, the present application provides a fault signal detection system and an analysis positioning method.

[0005] In a first aspect, the present application provides a fault signal detection system, comprising:

[0006] An injection module configured to inject a preset narrowband acoustic disturbance signal into a target optical cable;

[0007] Passive wireless tag nodes are arranged at intervals along the target optical cable, and each tag node is configured to generate and output multi-dimensional data representing the propagation characteristics of the acoustic disturbance signal when receiving the acoustic disturbance signal;

[0008] An aggregation module configured to collect the multi-dimensional data and associate the corresponding tag node identifier;

[0009] The analysis and construction module is configured to calculate the propagation difference between each of the label nodes based on the multi-dimensional data, and construct an acoustic propagation path map;

[0010] The projection and identification module is configured to map the acoustic propagation path map to a geographic information system coordinate system corresponding to the geographic coordinates of the label nodes, determine whether the propagation difference between each of the label nodes in the multi-dimensional data exceeds a preset propagation threshold, identify the label node as an abnormal propagation node if the propagation difference exceeds the preset propagation threshold, and determine the geographic information system coordinate of the abnormal propagation node as the fault space position.

[0011] As an optional implementation, the injecting a preset narrowband acoustic disturbance signal into the target optical cable comprises:

[0012] The injection module is driven to inject at least two groups of trial acoustic pulses in an increasing frequency order within a preset sweep frequency range, and the reference label node arranged at the deployment point of the injection module records the arrival amplitudes of each group of trial acoustic pulses;

[0013] A target frequency is selected according to the arrival amplitudes of each group of trial acoustic pulses, and the injection frequency corresponding to the trial acoustic pulse with the maximum amplitude and a decay rate lower than a preset threshold is determined as a target narrowband injection frequency;

[0014] Based on the arrival amplitudes measured by the reference label node in real time, the injection power is adjusted so that the amplitude of the acoustic disturbance signal at the target narrowband injection frequency at the reference label node is maintained within a preset target amplitude range;

[0015] After completing the injection power adjustment, the acoustic disturbance signal at the target narrowband injection frequency is cyclically injected into the target optical cable according to a preset injection period.

[0016] As an optional implementation, it further comprises:

[0017] Each of the passive wireless label nodes is further configured to report the local temperature data together with the identification of its previous-hop label node when outputting the multi-dimensional data; wherein the previous-hop label node refers to the label node that is adjacent to the current label node and located upstream of the current label node in the acoustic propagation path map along the propagation direction of the acoustic disturbance signal;

[0018] In response to the absence of other label nodes between the current label node and the injection module, the reference label node is regarded as the previous-hop label node of the current label node;

[0019] In response to the current label node being the farthest end label node in the acoustic propagation path, the previous-hop label node thereof is the nearest upstream label node adjacent thereto;

[0020] The analysis and construction module is configured to generate a tag pair temperature difference vector based on the local temperature data of adjacent tag nodes;

[0021] According to the temperature difference vector, an acoustic propagation path map is divided into several isothermal segments, and the temperature difference of each isothermal segment does not exceed a preset threshold; for each isothermal segment, a pre-stored temperature-sound velocity mapping data table corresponding to the optical cable sheath material and tension level is called, and the propagation difference of the segment is temperature compensated by using the mapped sound velocity parameter.

[0022] As an optional implementation, the generating and outputting of the multi-dimensional data representing the acoustic disturbance signal propagation characteristics comprises:

[0023] At the moment when the amplitude of the acoustic disturbance signal is detected to exceed a preset trigger threshold for the first time, the acoustic disturbance signal is analog-digital converted at a preset sampling rate to obtain a discrete acoustic sampling sequence;

[0024] In a preset analysis window, the discrete acoustic sampling sequence is processed, when the amplitude of the discrete acoustic sampling sequence exceeds a rising edge threshold for the first time, the moment is recorded as an arrival time stamp, a root mean square amplitude value is calculated as an arrival amplitude, a maximum energy frequency component is extracted as an arrival frequency by fast Fourier transform, and an instantaneous phase of the maximum energy frequency component is synchronously obtained;

[0025] According to a preset data frame format, a tag node identifier, a last hop tag node identifier, local temperature data, and the arrival time stamp, arrival amplitude, arrival frequency and instantaneous phase are packaged into an upstream data frame;

[0026] The upstream data frame is sent to the convergence module by using a near-field energy coupling mode.

[0027] As an optional implementation, the generating and outputting of the multi-dimensional data representing the acoustic disturbance signal propagation characteristics comprises:

[0028] Based on the arrival amplitude and the instantaneous phase, tag node data with a peak signal power to noise variance ratio below a preset signal-to-noise ratio threshold and / or an adjacent frame phase difference greater than 180° is removed to obtain an effective tag data set;

[0029] For each tag node in the effective tag data set, another tag node with the smallest and greater than zero propagation time difference is selected as an adjacent node pair to adaptively generate an adjacent edge;

[0030] A composite weight is assigned to each adjacent edge, and the composite weight is obtained by linear combination of the propagation time difference and the arrival amplitude attenuation value of the adjacent edge according to a preset weighting coefficient;

[0031] Constructing an acoustic propagation path atlas based on a graph search algorithm using the composite weight, and updating incremental information to the acoustic propagation path atlas in real time when a new effective label dataset is generated.

[0032] As an optional implementation, constructing an acoustic propagation path atlas based on a graph search algorithm using the composite weight includes:

[0033] Abstracting each label node in the effective label dataset as a directed graph vertex, taking the reference label node as a source vertex, and mapping the adjacent node pair determined by the composite weight as a directed edge from the source vertex to the propagation direction;

[0034] In the graph search process, when the composite weight of a candidate directed edge exceeds a preset pruning threshold, terminating the expansion of the corresponding edge;

[0035] Based on the pruned directed graph, using Dijkstra algorithm to calculate the shortest cumulative composite weight path from the source vertex to the remaining label nodes to obtain an acoustic propagation path result set;

[0036] For each path in the acoustic propagation path result set, outputting a path record composed of a label node sequence, a cumulative propagation time difference, and a cumulative arrival amplitude attenuation, and writing the path record into the acoustic propagation path atlas.

[0037] As an optional implementation, writing the path record into the acoustic propagation path atlas includes:

[0038] For each path in the acoustic propagation path result set, calculating a path confidence according to the cumulative composite weight of the path and the number of label nodes contained in the path;

[0039] Sorting the acoustic propagation path result set in descending order according to the path confidence, and taking the top N paths in confidence as a target path set, where N is a preset maximum output number;

[0040] For each path in the target path set, calculating an average temperature difference value of the path based on the local temperature data of the label nodes constituting the path;

[0041] Writing the target path set together with the corresponding label node sequence, cumulative propagation time difference, cumulative arrival amplitude attenuation, and average temperature difference value into the acoustic propagation path atlas.

[0042] As an optional implementation, determining whether the propagation difference between each label node in the multi-dimensional data exceeds a preset propagation threshold, and if it does, identifying the label node as an abnormal propagation node; and determining the geographic information system coordinates of the abnormal propagation node as a fault space position include:

[0043] a baseline mean μ and a standard deviation σ are calculated for a set of propagation time differences between adjacent tag nodes, and a downstream tag node is marked as a candidate abnormal node when an absolute value of a difference between a certain propagation time difference and the baseline mean is greater than kσ; wherein k is a preset abnormality coefficient;

[0044] According to the propagation direction, the candidate abnormal nodes that appear continuously are grouped, and when the number of nodes in the same group is not less than a preset continuous abnormal length, the group is confirmed as an abnormal propagation node cluster;

[0045] For each abnormal propagation node cluster, based on the geographic information system coordinates of the head and tail tag nodes constituting the cluster, the center coordinates of the abnormal propagation node cluster are calculated as the fault spatial position through linear interpolation along the cable path.

[0046] The fault spatial position is written into the acoustic propagation path map, and the corresponding geographic information system coordinates are output to the projection and identification module.

[0047] In a second aspect, the application provides an analysis and positioning method, comprising:

[0048] A preset narrowband acoustic disturbance signal is injected into the target optical cable;

[0049] Passive wireless tag nodes are arranged at intervals along the target optical cable, and each tag node is configured to generate and output multi-dimensional data representing the propagation characteristics of the acoustic disturbance signal when receiving the acoustic disturbance signal;

[0050] The multi-dimensional data is collected and associated with the corresponding tag node identifier;

[0051] Based on the multi-dimensional data, the propagation differences between each of the tag nodes are calculated, and an acoustic propagation path map is constructed;

[0052] The acoustic propagation path map is mapped to a geographic information system coordinate system corresponding to the geographic coordinates of the tag nodes; it is judged whether the propagation differences between each of the tag nodes in the multi-dimensional data exceed a preset propagation threshold, and if they do, the tag nodes are identified as abnormal propagation nodes; and the geographic information system coordinates of the abnormal propagation nodes are determined as the fault spatial position.

[0053] Compared with the prior art, in the absence of or inability to obtain accurate line topology drawings, the application uses passive wireless tag nodes distributed along the optical cable and injected narrowband acoustic disturbance signals to realize self-organizing detection based on acoustic propagation path; Compared with the traditional method of relying on complete topology drawings and single-end optical pulse testing, the need for manual intervention and external drawing maintenance can be greatly reduced. BRIEF DESCRIPTION OF DRAWINGS

[0054] Figure 1 A schematic diagram of a fault signal detection system provided by an embodiment of the present application is shown in FIG. 1.

[0055] Figure 2 A flowchart of a method for injecting a preset narrowband acoustic disturbance signal into a target optical cable provided by an embodiment of the present application is shown in FIG. 2.

[0056] Figure 3 A schematic diagram of differential processing provided by an embodiment of the present application is shown in FIG. 3.

[0057] Figure 4 A schematic diagram of a flow dynamic processing system based on multi-module cooperation in a smart logistics scenario provided by an embodiment of the present application is shown in FIG. 4.

[0058] Legend: 10, injection module; 20, passive wireless tag node; 30, aggregation module; 40, analysis and construction module; 50, projection and identification module. DETAILED DESCRIPTION

[0059] The technical solutions in the embodiments of the present application will be described clearly and completely below with reference to the drawings in the embodiments of the present application. Obviously, the described embodiments are only some of the embodiments of the present application, not all the embodiments.

[0060] Referring to FIG. 1, Figure 4 A schematic diagram of a fault signal detection system provided by an embodiment of the present application is shown in FIG. 1, which includes an injection module 10, passive wireless tag nodes 20, an aggregation module 30, an analysis and construction module 40, and a projection and identification module 50, wherein:

[0061] The injection module 10 is configured to inject a preset narrowband acoustic disturbance signal into a target optical cable.

[0062] The passive wireless tag nodes 20 are arranged at intervals along the target optical cable, and each tag node is configured to generate and output multi-dimensional data representing the propagation characteristics of the acoustic disturbance signal when the acoustic disturbance signal is received.

[0063] The aggregation module 30 is configured to collect the multi-dimensional data and associate the corresponding tag node identifiers.

[0064] The analysis and construction module 40 is configured to calculate the propagation differences between each of the tag nodes based on the multi-dimensional data and construct an acoustic propagation path map.

[0065] The projection and identification module 50 is configured to: map the acoustic propagation path pattern to a geographic information system coordinate system corresponding to the geographic coordinates of the tag nodes; determine whether the propagation difference between each tag node in the multi-dimensional data exceeds a preset propagation threshold, and if so, identify the tag node as an abnormal propagation node; and determine the geographic information system coordinates of the abnormal propagation node as the fault space position.

[0066] The narrow-band acoustic disturbance signal specifically refers to an acoustic excitation signal with a single center frequency or a very limited bandwidth. This signal is usually injected into the optical cable in the form of a pulse to ensure good propagation and signal recognition, thereby effectively reducing noise interference and signal loss and improving the overall detection accuracy and positioning precision of the system.

[0067] The injection module 10 can be composed of a set of piezoelectric excitation devices and a driving circuit matched therewith. By controlling the frequency, amplitude, and pulse form of acoustic vibration, a preset narrow-band acoustic disturbance signal can be applied to the optical cable outer sheath. The center frequency of the acoustic disturbance signal can be selected between several kilohertz and more than ten kilohertz, which can balance the propagability in the optical cable sheath material and will not cause serious energy attenuation due to excessively high frequency. In order to realize tunable acoustic output, the control circuit inside the injection module 10 changes the driving voltage or driving pulse width of the exciter within a certain range according to the preset sweep frequency range or the echo amplitude measured by the reference tag end.

[0068] When the injection module 10 starts to work, the piezoelectric transducer thereof is attached to or coupled with the optical cable sheath, and a series of mechanical waves are formed on the optical cable in the cable direction through short and regular pulses. Compared with traditional optical fiber flaw detection or single-end optical pulse testing, by directly applying acoustic disturbance to the optical cable sheath, the multi-layer sheath or loose sleeve structure can be penetrated more flexibly, and signal blind areas can be avoided at local jumper connections.

[0069] In some embodiments, the injection module 10 can also use the Chirp method to smoothly transition from a lower frequency to a higher frequency, thereby obtaining uniform and controllable propagation characteristics on different frequency bands of the sheath for further multi-dimensional data analysis.

[0070] In terms of mechanical coupling, the injection module 10 is detachably installed near the exposed end of the optical cable or the joint box, and a rubber clamp or a special coupling agent is used to reduce the loss of acoustic energy at the injection interface. If the laying environment of the optical cable is complex or there are interference sources around, a shielding cover or a damping component can be additionally provided on the injection module 10 to reduce the influence of environmental noise on the acoustic disturbance signal.

[0071] Since the acoustic disturbance generated by the injection module 10 can propagate across the fiber and disperse in the branching structure, combined with the passive wireless tag nodes 20 along the way, the characteristics of the optical cable transmission path can be gradually collected without prior topological information. Compared with the breakpoint detection scheme based on optical pulses, this acoustic injection mode is more adaptable to fiber splices or branching sections, avoiding complex jumpers or joint box areas that produce difficult-to-interpret echoes.

[0072] In addition, narrowband acoustic disturbance can obtain better signal-to-noise ratio in a controllable frequency band, which not only facilitates subsequent tag nodes to perform multi-dimensional parameter detection and time delay calculation, but also helps to obtain a more uniform power distribution between the near end and the far end.

[0073] In this way, whether the optical cable has branches, bends, or multiple repairs along the way, the injection module 10 can maintain effective "knocking" on downstream nodes, laying a solid data foundation for the entire fault detection system.

[0074] For example, a 10-kilometer-long optical cable needs to be quickly located. After several modifications and local repairs, there is a large deviation between the original drawings and the on-site layout, making it difficult to determine the specific location of the broken fiber or loose sleeve using conventional positioning methods that rely on optical reflection testing.

[0075] In this embodiment, a 1-meter-long exposed sheath section is reserved at one end of the optical cable as an injection point, and an injection module 10 equipped with a tunable piezoelectric exciter and control circuit is installed at this point. The exciter is set to an initial operating center frequency of 10 kHz, with a pulse width of 2 ms. Each time the pulse amplitude is quickly raised to a peak value of 80 V through a linearly rising voltage. In this way, an acoustic disturbance is formed on the optical cable, which is short and energy-concentrated.

[0076] To ensure that the far-end tag nodes still have sufficient signal amplitude, while preventing the near-end from being oversaturated, the injection module 10 is also equipped with a reference sensor to monitor the echo signal strength at a distance of about 0.5 meters from the injection point in real time. If the amplitude at this point is detected to exceed the pre-set safety value, the system will automatically shorten the pulse width or reduce the peak voltage to avoid excessive impact on the optical cable. Conversely, if the signal amplitude of the tag node reported by the far-end is lower than the 3 dB signal-to-noise ratio threshold, the pulse width is appropriately extended to 3 ms, and the voltage is adjusted to a peak value of 100 V to ensure that the acoustic disturbance can penetrate the multiple repairs or joint box sections in the middle.

[0077] The acoustic disturbance formed by the injection module 10 can effectively overcome the difficulties caused by unknown topology, providing sufficient signal strength and credibility for fault detection and positioning of downstream nodes.

[0078] For the above passive wireless tag nodes 20:

[0079] The passive wireless tag nodes 20 are fixed at intervals along the optical cable to effectively cover the entire laying section. Each tag node integrates a near-field coupling antenna, a piezoelectric micro-transducer and a low-power control circuit for capturing and analyzing the acoustic disturbance signals applied by the injection module 10. When the acoustic wave propagates along the sheath to the location of the tag node, the micro-transducer converts it into a voltage change, which is amplified and filtered internally and sent to the control circuit. Since the node itself does not require an external power supply, the required energy is obtained from external near-field coupling or spontaneous vibration, thereby achieving a passive design to simplify deployment and later maintenance.

[0080] After detecting the acoustic disturbance signal, the tag node will discretely collect signal amplitude and phase information at a pre-configured sampling rate and perform preliminary analysis on the sampled waveform to extract multi-dimensional features such as arrival time stamp, amplitude peak, maximum energy frequency component or instantaneous phase.

[0081] In some embodiments, the node also has a sensor for measuring ambient temperature or vibration noise installed inside, so that the tag node can not only report the transmission characteristics of the acoustic wave itself, but also provide auxiliary information about the external environment. Through the integration of these multi-dimensional data, more accurate propagation path reconstruction and fault location can be performed in the subsequent stage.

[0082] In order to realize data backhaul without affecting the normal use of the optical cable, each tag node will package the extracted multi-dimensional data into a data frame and report it to the aggregation module 30 through near-field coupling or passive radio frequency. At this time, the tag node only briefly wakes up to send data after detecting the acoustic disturbance and completing data extraction, and is in a deep sleep state at other times to ensure extremely low overall energy consumption. If no valid acoustic signal is detected within a certain excitation period, the tag node will not wake up and send data. In this way, the operating efficiency of the node passive design can be further optimized to achieve long-term online monitoring of the laying line.

[0083] In actual installation process, the passive wireless tag nodes 20 are usually packaged in a shell with waterproof and dustproof properties and are tightly attached to the surface of the optical cable sheath through bonding, tape or other fixing methods. Different interval arrangement schemes can be selected according to the site conditions, such as every 50 meters or every 100 meters.

[0084] Exemplarily, on a city optical cable to be monitored, a passive wireless tag node 20 can be installed at approximately every 50 meters to capture the acoustic disturbance signals generated by the injection module 10. If there are bends, jumps or joint boxes, the deployment of the tag nodes can be moderately encrypted according to actual needs to obtain higher resolution information along the line. After detecting the acoustic disturbance, each tag node extracts the multi-dimensional data such as the arrival time stamp, amplitude peak, instantaneous phase or maximum frequency of energy of the disturbance according to the pre-set sampling frequency and analysis window, and after a short wake-up, encapsulates the data into an uplink data frame and sends it to the aggregation module 30 through near field coupling or other feasible wireless methods.

[0085] In this exemplary scenario, the tag node itself does not rely on external power supply, but completes the work through energy coupling with the external environment (such as RF near field excitation, micro-vibration collection, etc.). In order to reduce power consumption as much as possible, the node software logic will ensure that it enters a sleep state after each reporting is completed, and will be activated again only when the next acoustic disturbance is detected. At this time, since each node will capture part or all of the characteristics of the same pulse in the same excitation period, the system can obtain distributed multi-dimensional detection data in the same time slice. Compared with traditional single-point measurement or single-end reflection test methods, this distributed monitoring can significantly improve the reconstruction and identification ability of the optical cable along the line in the subsequent analysis and construction module 40.

[0086] In the specific implementation of the passive wireless tag node 20, it can include a low-power microcontroller, a group of piezoelectric microphone arrays or micro transducers, and a small-capacity memory for temporarily storing sampling results. According to different design schemes, the tag node can choose to directly perform local digital processing on the collected acoustic waveform, or only perform partial parameter extraction (such as amplitude and time stamp), and then package into a data frame. If it is necessary to dynamically adjust the detection sensitivity or sampling rate of the tag node during operation, the aggregation module 30 can also remotely configure the tag node through near field communication to realize more fine-grained monitoring or power consumption management.

[0087] In the installation process, since each tag node must be closely coupled with the cable sheath to accurately receive acoustic disturbance, a waterproof sleeve, a fixing clamp or other stable means can be used to attach the node to the surface of the optical cable. If the line needs to be expanded or relocated later, only the corresponding number of tag nodes need to be added at the newly laid cable section or the relocated section, so that the acoustic detection capability of the system can continue to be used without the need to redraw the accurate topology or conduct a large amount of manual checking. As the injection module 10 periodically excites acoustic pulses to the optical cable, the tag nodes will synchronously monitor the acoustic disturbance arriving at their positions and complete signal acquisition and feature extraction locally, laying a foundation for subsequent propagation difference analysis and fault location. Through this scalable distributed deployment, the health status of the entire optical cable line will always be in visual and detectable management, and even after multiple branches, local repair or jumpering, the system can still maintain high fault location accuracy and real-time performance.

[0088] For the above-mentioned aggregation module 30:

[0089] The aggregation module 30 is arranged in the fault signal detection system, mainly to realize the unified collection and management of the multi-dimensional data reported by the passive wireless tag nodes 20 scattered along the optical cable. The multi-dimensional data referred to here can include, but is not limited to, the arrival time stamp, amplitude information, frequency component, phase feature, environmental temperature data and identification information of the tag node itself, etc. Since the same acoustic excitation is collected by multiple nodes at the same time, the aggregation module 30 needs to be able to distinguish different excitation periods and merge the data corresponding to each node in the same excitation period into the same data set.

[0090] To achieve this function, the aggregation module 30 can be equipped with a near-field communication or passive radio frequency receiving interface in the system structure, so that it can interact with all the scattered tag nodes. After completing local signal detection and data packaging, each tag node will send an uplink data frame carrying the identification of the tag node to the aggregation module 30; when the aggregation module 30 receives the data frame, it determines which specific tag node it comes from and which detection period or acoustic disturbance sequence the data corresponds to by reading the node identification in the data frame. At the same time, the aggregation module 30 can also determine the relative order of the data in the entire excitation period according to the time stamp and other information in the data frame, so as to aggregate the observation results of multiple nodes in time sequence or disturbance sequence.

[0091] In the process of collecting multi-dimensional data, the aggregation module 30 can perform the following operations:

[0092] In the data receiving link, the data type (such as amplitude, phase, temperature, etc.) is identified according to the data frame header or the preset format and allocated to the corresponding data buffer area;

[0093] In the data processing link, a data entry is established for each detection period, and the observation values of all nodes under the same acoustic excitation are merged into the same entry to form a multi-dimensional observation matrix. In the matrix, the identification of the tagged node (such as node ID, location code, etc.) is used as the row index, and the physical quantities such as time stamp, amplitude, and frequency are used as the column index;

[0094] In the data association link, the aggregation module 30 will ensure that each observation data can be matched with the correct node identification according to the pre-recorded "node ID-space position" mapping table or real-time allocated identification information. For the uplink data frames that arrive or delay to arrive successively in the same detection period, the aggregation module 30 can insert them into the correct detection period according to the sequence number or excitation event number in the data, so as to maintain the integrity and continuity of the data.

[0095] After the data is correctly aggregated and associated with the node identification, the aggregation module 30 can submit the corresponding data batch to the analysis and construction module 40. For example, if the multi-node observation values in a detection period include arrival time stamp, amplitude attenuation, frequency offset, etc., the aggregation module 30 will package all the node data in the same batch and mark it with "the nth acoustic excitation" or "excitation sequence ID=xxx", so that the analysis and construction module 40 knows that it is the complete observation set under the same acoustic disturbance event. The analysis and construction module 40 can calculate the time delay difference and amplitude difference of each node accordingly, construct the acoustic propagation path map, and further locate the fault point or judge the node anomaly.

[0096] In summary, the purpose of the aggregation module 30 is to collect multi-dimensional data reported by each passive wireless tag node 20 by having the same or compatible near field communication protocol, identify and analyze the node identification, detection period or acoustic excitation information carried by each data, and classify and archive them; after completing the aggregation of multi-node observation values, prepare for subsequent analysis, such as packaging the same batch of data and submitting it to the analysis and construction module 40;

[0097] For the above analysis and construction module 40:

[0098] In specific implementation, the analysis and construction module 40 is used to further reconstruct the actual laying path or connection relationship of the optical cable according to the propagation characteristics between different tag nodes after the system completes the collection and aggregation of multi-dimensional data, so as to accurately infer the possible fault location.

[0099] Its core process includes: analyzing and reading the multi-node observation data under the same excitation period from the aggregation module 30, calculating the propagation difference between each tag node, and then using these differences to construct or update an acoustic propagation path map.

[0100] When the analysis and construction module 40 obtains the multi-dimensional parameters of the tag nodes, such as the arrival time stamp, amplitude attenuation, frequency offset, and temperature, it will first classify and index these observation data according to the correspondence between the node identification and the excitation event. For example, for the same acoustic pulse excitation, the analysis and construction module 40 will read all the successfully reported tag node records within the excitation period, and calculate the time delay difference, amplitude difference, or phase difference between adjacent nodes one by one. By comparing these propagation differences, the time delay increment, energy attenuation, or frequency drift experienced by the acoustic wave when propagating from the upstream node to the downstream node can be determined. If the time delay difference of two nodes is the smallest and the amplitude attenuation is within a reasonable range, it can be inferred that the physical distance between the two nodes is relatively short, and the optical cable connection relationship is more direct.

[0101] In some embodiments, the analysis and construction module 40 will regard all nodes as vertices in the graph structure, and construct weighted edges according to the propagation differences between adjacent nodes. The weight of the weighted edge is usually obtained by comprehensively considering the time delay difference and amplitude attenuation, which can reflect not only the physical distance of the optical cable, but also the quality difference or joint loss of the optical cable.

[0102] Further, the module can use common graph search algorithms (such as shortest path search or minimum spanning tree algorithm) to reconstruct the actual path structure of the optical cable on the weighted graph. If similar propagation difference distribution is obtained under multiple excitations, the module will combine the results, continuously optimize or update the acoustic propagation path atlas. In this way, even in the absence of the original topology drawing, or the pipeline has been jumpered multiple times and the branching is complex, the system can still accurately and dynamically infer the laying relationship of the optical cable.

[0103] In some optional designs, the analysis and construction module 40 can also do hierarchical processing for the branching structure: if it is detected that some nodes form a sub-path that is significantly different from the main path, it indicates that there may be an independent optical cable branch; the module can extract the branch to generate a local path atlas, and keep it associated with the main atlas. For the observation data accumulated in multiple detection periods, the module can also use time sequence superposition, by comparing the propagation difference changes in different periods, to evaluate whether the optical cable in a certain section has a persistent increase in attenuation or a certain tag node disappears (which may indicate a broken cable or a more serious damage to the sheath).

[0104] After completing the atlas construction or updating, the analysis and construction module 40 will send the key results of the acoustic propagation path atlas to the projection and identification module 50, which can include the distance or loss inference value between nodes, and the marking of some high-risk sections. After receiving, the projection and identification module 50 can map the atlas to the actual geographic coordinate system, providing visual optical cable distribution and possible fault point location for maintenance personnel.

[0105] For the above projection and identification module 50:

[0106] In the system, the projection and identification module 50 is responsible for transforming the acoustic propagation path map generated or updated by the analysis and construction module 40 into a visualization coordinate corresponding to the actual geographical location of the tag node, and determining whether the node is abnormal based on the preset propagation threshold. The implementation logic can be roughly divided into two levels: spatial coordinate mapping and threshold detection and fault labeling.

[0107] In terms of spatial coordinate mapping, the projection and identification module 50 first reads the path map provided by the analysis and construction module 40, which contains the correlation between the tag nodes and the propagation difference metric (such as time delay difference, amplitude attenuation, etc.). At the same time, in order to realize the abstract graph structure in the geographic information system (GIS), the module obtains the geographical coordinates corresponding to each tag node from the previously established "node identification-spatial coordinate" mapping table or the reference coordinate information maintained on site. The coordinates can be in the form of latitude and longitude, or in the form of plane coordinates or projection coordinates, depending on the site environment and the usage habits of the upper GIS platform.

[0108] After pulling or converting the node coordinates, the projection and identification module 50 will draw each node-to-node propagation edge in the graph to the GIS coordinate system, forming one or more line distribution diagrams. For those node sets that are determined to be independent branches or branch structures in the analysis and construction stage, the module can also distinguish them in different colors or line types on the GIS interface, making it easy for maintenance personnel to intuitively understand the current actual routing and connection relationship of the optical cable.

[0109] After completing the spatial mapping, the projection and identification module 50 will also compare the propagation difference characteristics of each node with the set threshold according to the data accumulated by each node or multiple detections within the same detection period. For example, if the time delay difference, amplitude attenuation, or frequency offset value between node A and its adjacent nodes in a certain detection is found to be significantly above the threshold, the module will mark node A as an "abnormal propagation node". This anomaly may mean that the optical cable has problems such as jacket damage, fiber breakage, or loose joints near node A. When the same out-of-limit condition appears in multiple consecutive detection periods, it indicates that the reliability of the fault at this location is higher.

[0110] Once the projection and identification module 50 determines that a certain tag node exceeds the preset propagation threshold range, it will present the node in the form of a fault marker in the GIS coordinate system and produce an alarm prompt on the display terminal or management background of the maintenance personnel.

[0111] Meanwhile, in order to better assist on-site disposal, the module can further combine the optical cable linear interpolation or the coordinates of other nodes in the periphery to refine the spatial position of the fault or estimate the center point, so as to give a more accurate "fault spatial position" coordinate. Subsequently, if integrated with an external database or a geographic pipeline information system, the module can also add the administrative district, street name or ground marker position where the fault point is located, so that the repair team can quickly and accurately locate the fault site.

[0112] Through the above technical means, the projection and recognition module 50 combines the topological information and propagation characteristic results output by the analysis and construction module 40 with the spatial coordinates of the label nodes to realize visual mapping and fault node recognition in the GIS coordinate system. With the help of the set propagation threshold and abnormality determination strategy, the module can automatically identify the suspected problem section and output its exact position in the real physical space, thereby helping the maintenance and repair team to efficiently and accurately locate the fault. In combination with other modules of the application, even in the environment lacking original topological drawings or multiple renovations, the accurate detection and rapid disposal of the optical cable fault point can still be ensured.

[0113] As an optional implementation, please refer to Figure 2 , Figure 2 A flowchart of a method for injecting a preset narrowband acoustic disturbance signal into a target optical cable provided by an embodiment of the application includes steps S201-S204, wherein:

[0114] S201: The injection module 10 is driven to inject at least two groups of trial acoustic pulses in a preset sweep frequency range in an increasing frequency order, and the arrival amplitudes of the trial acoustic pulses are recorded by the reference label node arranged at the layout point of the injection module 10;

[0115] S202: A target frequency is selected according to the arrival amplitudes of the trial acoustic pulses, and the injection frequency corresponding to the trial acoustic pulse with the maximum amplitude and a decay rate lower than a preset threshold is determined as the target narrowband injection frequency;

[0116] S203: Based on the arrival amplitudes measured by the reference label node in real time, the injection power is adjusted so that the amplitude of the acoustic disturbance signal at the target narrowband injection frequency at the reference label node is maintained within a preset target amplitude range;

[0117] S204: After completing the injection power adjustment, the acoustic disturbance signal at the target narrowband injection frequency is cyclically injected into the target optical cable according to a preset injection period.

[0118] In some scenarios, in order to find the optimal or better injection frequency and power as soon as possible at the initial excitation, so that the remote tag node can also reliably detect the acoustic disturbance signal, the injection module 10 is optionally set to a sweep mode of operation, that is, according to the pre-set sweep range, a plurality of groups of trial acoustic pulses are injected in order of increasing frequency, and the arrival amplitudes of each group of trial pulses are recorded by the reference tag node located at or immediately adjacent to the injection module 10. The reference tag node can use a transducer and control circuit similar to the downstream passive wireless tag node 20, but is deployed near the injection module 10, so that it is physically close to the injection point to measure the amplitude of the acoustic energy actually applied to the cable sheath in real time.

[0119] In this sweep process, each group of trial pulses is applied to the optical cable at a slightly different injection frequency, and the reference tag node judges the propagation effect of the frequency under the current sheath and environmental conditions according to the arrival amplitude. If the recorded amplitude of a certain group of pulses is much higher than that of other frequency points, it indicates that this frequency is more likely to be transmitted in the optical cable with lower loss; if the recorded amplitudes of certain frequency points are significantly reduced or the decay rate is too high, it indicates that the frequency band may have sheath resonance attenuation, air cavity influence or other reasons, and it is difficult to form a stable and detectable acoustic wave.

[0120] According to the arrival amplitudes of each group of trial pulses, the injection module 10 or the upper control program selects the injection frequency corresponding to the trial pulse with the maximum amplitude and a decay rate lower than a certain pre-set threshold, and determines it as the subsequent target narrowband injection frequency. The target narrowband injection frequency can remain unchanged for a certain period of time, or it can be periodically re-swept to adapt to changes in environmental conditions (such as temperature, burial depth, and sheath aging) over time.

[0121] After the target frequency is selected, the system also needs to adjust the injection power accordingly. At this time, the reference tag node will measure the arrival amplitude of the injected pulse at its location in real time, and if it is found that the amplitude is above or below the pre-set target amplitude range, the injection module 10 will accordingly increase or decrease the drive voltage of the exciter, or adjust the pulse width, so that the amplitude at the reference tag node is stably maintained within a suitable range that can ensure subsequent downstream node detection. Through this closed-loop power control, both near-end over-saturation-induced sheath damage and far-end insufficient detection signal can be avoided.

[0122] After the above sweep-selection-power adjustment steps are completed, the injection module 10 enters a periodic injection mode. According to the pre-set injection period (for example, once every 1 second or every 10 seconds), the module will repeatedly apply acoustic disturbance signals to the optical cable at the determined target narrowband injection frequency and suitable power parameters. All downstream distributed passive wireless tag nodes 20 can monitor and capture the disturbance within the corresponding period, and report multi-dimensional data to the aggregation module 30 for subsequent analysis.

[0123] For example, the preset sweep frequency range is 8 kHz-12 kHz, and the module first injects multiple sets of trial acoustic pulses with an incremental step of 0.5 kHz when starting. The reference tag node records the amplitudes of each set of pulses, and the trial frequency with the maximum amplitude and a decay rate of <10% is determined as the optimal frequency in this round. If a certain frequency stably obtains a high amplitude value in multiple trials, the system selects this frequency as the target narrowband injection frequency, and adjusts the injection power in a closed loop to maintain a target amplitude range of 80 dB at the reference tag node. Finally, the acoustic pulses of the target frequency are injected into the optical cable at an injection period of 1 Hz, providing long-term and stable fault detection support for downstream passive wireless tag nodes 20.

[0124] In this way, the system can adaptively select the optimal injection signal parameters under different optical cable materials, different temperatures or different burial depths, and improve the detection signal-to-noise ratio and availability of the remote tag nodes.

[0125] As an optional implementation, it further comprises:

[0126] Each of the passive wireless tag nodes 20 is further configured to, when outputting the multi-dimensional data, report the local temperature data together with the identification of its previous-hop tag node; wherein the previous-hop tag node refers to the tag node that is adjacent to the current tag node and located upstream of the current tag node in the acoustic propagation path map along the propagation direction of the acoustic disturbance signal.

[0127] In response to the absence of other tag nodes between the current tag node and the injection module 10, the reference tag node is regarded as the previous-hop tag node of the current tag node.

[0128] In response to the current tag node being the farthest tag node in the acoustic propagation path, its previous-hop tag node is the nearest upstream tag node adjacent thereto.

[0129] The analysis and construction module 40 is configured to generate a tag pair temperature difference vector based on the local temperature data of adjacent tag nodes.

[0130] According to the temperature difference vector, the acoustic propagation path map is divided into a plurality of isothermal segments, and the temperature difference of each isothermal segment does not exceed a preset threshold. For each isothermal segment, a pre-stored temperature-acoustic velocity mapping data table corresponding to the optical cable sheath material and tension level is called, and the propagation difference of the segment is temperature-compensated using the mapped acoustic velocity parameters.

[0131] In some high-temperature gradient or complex environment cable scenarios, there can be significant differences in soil temperature, burial conditions or pipeline ventilation status along the line, which will directly affect the propagation speed and attenuation coefficient of acoustic waves in the sheath. Therefore, the application further configures a temperature measurement function for each passive wireless tag node 20, and allows the node to include local temperature data and "last hop tag node identification" information when reporting multi-dimensional data. In this way, the analysis and construction module 40 can take the temperature difference between adjacent nodes into account in the calculation of the propagation difference, and perform temperature compensation or segmented correction when necessary, thereby greatly improving the matching accuracy of the actual propagation path and time delay.

[0132] In a specific implementation, each passive wireless tag node 20 is built-in or externally connected with a low-power temperature sensor (such as a digital semiconductor thermometer or a thermistor), and after detecting acoustic disturbance, the local real-time temperature data is packaged into the uplink data frame together with the time stamp, amplitude peak value and other parameters. If the system also needs to determine the relative order or topological relationship between nodes, the "last hop tag node identification" can be combined for chain marking: when a node reports data, it will tell the system which upstream node it received the acoustic wave from, i.e. who is the "last hop tag node".

[0133] In order to form a clear path chain, each passive wireless tag node 20 is assigned a globally unique ID. When node A detects acoustic disturbance in the current excitation period, it checks who is the node closer to the injection module 10 in the propagation direction, and if there is no other node within the distance, the reference tag node or the injection module 10 itself is regarded as the last hop node of A; if A itself is the farthest end, the last hop node can be traced back to the nearest upstream node adjacent to it. Through this identification mechanism, the nodes can be connected in the order of "last hop - next hop" after receiving data at the convergence module 30, and an approximate acoustic propagation chain structure is formed.

[0134] In order to facilitate temperature compensation by the analysis and construction module 40, the system records the temperature difference of each group of "adjacent nodes (A, B)", forming a "tag pair temperature difference vector". If the temperature difference between two adjacent nodes exceeds a certain preset threshold, it indicates that there is a significant temperature gradient in this section, and the acoustic wave speed deviates from the normal temperature value. The analysis and construction module 40 divides the path map into several "isothermal sections" accordingly, ensuring that the temperature difference of nodes within the same section does not exceed the threshold, so that the wave speed or attenuation coefficient can be approximately calculated as a uniform temperature interval.

[0135] To obtain more accurate acoustic propagation parameters, the module can access a pre-established "temperature-sound velocity" or "temperature-attenuation coefficient" mapping data table, which lists the sound velocity values ​​under different temperatures, sheath materials, and tension levels. The system will select or interpolate the corresponding sound velocity based on the average temperature of the segment, and then correct the time delay of all node pairs within that segment, thereby more accurately inferring the distance or attenuation between nodes. If there are tension differences in a segment (such as pipe subsidence or stretching), additional data can be used for composite queries.

[0136] For example, the temperature difference between optical cable segments in a certain urban area can reach 5-8℃, especially the temperature difference between the surface line and the deeply buried pipe gallery. Each passive wireless tag node 20 includes the local temperature when reporting data. and the previous hop label node identifier Upon receiving the data, the analysis and construction module 40 will first... By performing an interpolation calculation, if the difference is greater than 3°C, that segment of the link is considered the segment boundary, forming a new isothermal segment. The temperature-velocity mapping table v=f(T) is then used to correct the velocity of this segment. In this way, the propagation delay error caused by temperature differences can be significantly reduced, and the positioning accuracy is improved accordingly.

[0137] in, This refers to the local temperature, i.e., the ambient temperature detected by the current node (e.g., node A), which can be in degrees Celsius (°C) or other temperature scales. The temperature is the temperature of the previous hop tag node, i.e., the local temperature reported by the immediate upstream node (e.g., node B) along the direction of acoustic disturbance propagation; v=f(T) is the function / table of the relationship between temperature T and sound velocity v, representing the possible changes in the sound velocity value in the optical cable sheath under different temperature environments. This function can be in interpolation form or discrete lookup table form. For example, when the sheath material, tension level, and other conditions are fixed, several temperature points and corresponding measured or calibrated sound velocity values ​​can be stored. This serves as the identifier for the previous hop label node, used for chain tracking of which upstream node A received the acoustic disturbance from during the current excitation cycle.

[0138] For example, when performing propagation difference analysis on adjacent nodes (A, B), if it is found that... Exceeding a preset threshold This indicates a significant temperature difference between points A and B, suggesting a possible substantial change in the velocity of sound within this section of the optical cable sheath. Therefore, the analysis and construction module 40 divides this section into a new "isothermal segment" or "thermal segment" and calls the aforementioned v=f(T) to adjust the time delay or attenuation calculation for this segment. A common mapping relationship can be written as: ;

[0139] the sound velocity in the segment after correction;

[0140] the reference temperature the sound velocity calibration value at the reference temperature;

[0141] α is the sound velocity temperature coefficient, used to quantify the linear effect of temperature on sound velocity;

[0142] wherein is the average temperature of the segment (which can be derived from and the mean or more node temperature).

[0143] It can be understood that other more complex functions or lookup table methods can also be used by those skilled in the art to reflect the effects of nonlinearity or multi-parameter coupling.

[0144] In some cases, if there is an abnormal node temperature data (such as a large jump or distortion), the system can also mark it as a suspicious data point, and trace back along the last jump label node to identify whether there is a sensor failure or data reporting missing. After such upstream and downstream cooperation and segment compensation, the analysis and construction module 40 can generate an acoustic propagation path map that better fits the actual physical environment, providing a segment temperature compensated path result for the back-end projection and recognition module 50, further improving the accuracy and robustness of fault location.

[0145] In this way, not only can the cumulative error caused by traditional optical cable ranging in temperature difference environment be addressed, but also the complex laying situation with multiple jumps, branch disorder, and large depth difference can be compatible, providing more reliable and intuitive fault location basis for repair personnel.

[0146] As an optional implementation, the generating and outputting of the multi-dimensional data representing the propagation characteristics of the acoustic disturbance signal comprises:

[0147] At the moment when the amplitude of the acoustic disturbance signal is detected to exceed a preset trigger threshold for the first time, the acoustic disturbance signal is analog-digital converted at a preset sampling rate to obtain a discrete acoustic sampling sequence;

[0148] Within a preset analysis window, the discrete acoustic sampling sequence is processed, when the amplitude of the discrete acoustic sampling sequence exceeds a rising edge threshold for the first time, the moment is recorded as an arrival time stamp, a root mean square amplitude value is calculated as an arrival amplitude, a maximum energy frequency component is extracted as an arrival frequency through fast Fourier transform, and an instantaneous phase of the maximum energy frequency component is synchronously obtained;

[0149] According to the preset data frame format, the tag node identifier, the last hop tag node identifier, the local temperature data, and the arrival time stamp, the arrival amplitude, the arrival frequency, and the instantaneous phase are packaged into an uplink data frame;

[0150] The uplink data frame is sent to the convergence module 30 in a near-field energy coupling manner.

[0151] When the node captures the acoustic disturbance signal and determines that the amplitude thereof exceeds a preset trigger threshold for the first time, the tag node switches from a standby or light sleep state to a sampling working mode. In this mode, the node continuously converts the analog acoustic wave transmitted by the optical cable sheath into digital sampling data according to a preset sampling rate (for example, 10 kHz or 20 kHz).

[0152] After the storage of the sampling sequence is completed, the node performs preliminary processing on the data by using the basic operation capability of the local microcontroller. The processing logic can include time domain scanning of the amplitude change: when the amplitude of the sampling sequence exceeds a rising edge threshold at a certain point for the first time, it is considered that the arrival of the pulse is detected, and the node records the time stamp corresponding to the sampling point as the arrival time stamp.

[0153] In this way, the system can compare the relative timing of the arrival of the respective pulses between different nodes to obtain the reference time required for subsequent propagation difference analysis.

[0154] When the arrival amplitude value is calculated, in order to balance the calculation complexity and the robustness to instantaneous noise, a root mean square amplitude (RMS) operation can be performed on the data in the analysis window. The node squares and sums a small part of the waveform of the sampling sequence after the rising edge determination, and then takes the square root to obtain an amplitude index representing the overall energy level. If the amplitude is much higher than the background noise of the node, it indicates that the signal still retains a considerable degree of energy when the pulse arrives; if the amplitude is flat or does not exceed the threshold, it may indicate that the pulse has been severely attenuated during propagation.

[0155] In the case where frequency information needs to be extracted, the node can also perform a simplified frequency domain analysis on the sampling sequence, for example, performing a fast Fourier transform (FFT). In order to avoid consuming too many computing resources in the passive mode, some implementation schemes only perform a low-resolution or truncated length FFT, focus on searching for the frequency component with the maximum energy, and record the instantaneous phase of the component. This helps to subsequently determine whether the acoustic disturbance in the optical cable is additional distortion due to sheath resonance or environmental noise, and also provides more basis for the analysis and construction module 40 to identify abnormal passages.

[0156] When the node finishes processing the discrete acoustic sampling sequence, it will package the time of arrival timestamp, amplitude value (or root mean square amplitude), energy maximum frequency, and phase information, etc. key parameters, together with the node's own identification, the last hop label node identification, and the local temperature data (if the node is configured with a temperature sensor) according to the preset data frame format. In order to adapt to short distance communication or near field coupling in passive mode, the passive wireless label node 20 often uses a highly compact frame structure to record each field in a fixed order or small head key value pair. After completing the packaging, the node will activate the near field coupling transmission module to send the uplink data frame to the convergence module 30. If the node completes the reporting or confirms the packet within a limited number of times, it can quickly switch back to the sleep mode to keep the overall energy consumption at a very low level.

[0157] In this way, distributed and multi-dimensional detection data can be obtained to provide accurate and rich observation basis for subsequent fault location and path recovery of the convergence module 30 and the analysis and construction module 40. If combined with the temperature and the last hop identification as described above, the restoration degree of the physical laying condition can be significantly improved in a multi-branch, multi-temperature zone or a wave speed variable scene, and the positioning accuracy and feasibility can be maintained at a high level in a large-scale optical cable network.

[0158] As an optional implementation, please refer to Figure 3 A flowchart of a method for constructing an acoustic propagation path map provided by the present application includes steps S301-S304, wherein:

[0159] S301: Based on the arrival amplitude and the instantaneous phase, the label node data with a signal-to-noise ratio threshold below a preset signal-to-noise ratio threshold and / or an adjacent frame phase difference greater than 180° is removed, and an effective label data set is obtained;

[0160] S302: For each label node in the effective label data set, another label node with the smallest and greater than zero propagation time difference is selected as an adjacent node pair, and an adjacent edge is adaptively generated;

[0161] S303: A composite weight is assigned to each adjacent edge, and the composite weight is obtained by linear combination of the propagation time difference and the arrival amplitude attenuation value of the adjacent edge according to a preset weighting coefficient;

[0162] S304: Using the composite weight, an acoustic propagation path map is constructed based on a graph search algorithm, and incremental information is updated to the acoustic propagation path map in real time when a new effective label data set is generated.

[0163] In specific implementations, it is difficult to completely exclude data anomalies caused by incidental noise or multipath interference by relying only on basic parameters such as arrival time stamp and amplitude. To this end, the present application adds a secondary screening and weighted graph construction process for multi-dimensional data in the analysis and construction module 40, so that in the process of constructing the acoustic propagation path map, noise that is too strong or phase conflicts can be effectively removed, and the propagation difference between nodes can be calculated based on the retained effective node data, thereby obtaining a more robust path map update result.

[0164] In specific implementations, when the aggregation module 30 integrates the multi-dimensional data of the tag nodes in the current detection period (or multiple detections), the analysis and construction module 40 will first perform data cleaning based on the arrival amplitude and instantaneous phase. To measure the signal-to-noise ratio, the system can calculate the ratio of peak signal power to noise variance. If this value is lower than a certain preset signal-to-noise ratio threshold, it means that the observation data of the tag node in the current period may be almost submerged in noise and is not suitable for subsequent path inference. At the same time, if the phase difference between adjacent frames (such as the phase difference between the main energy peaks of the same node in the same period) exceeds 180°, it can be determined that the phase information has a jump or mismatch, and it is also marked as abnormal, thereby being excluded from the effective tag data set.

[0165] After this step of screening, the analysis and construction module 40 will select another tag node with the smallest and greater than zero propagation time difference for each tag node in the remaining effective node data as the adjacent node pair, that is, adaptively generating a connection line of "upstream node - current node" or "current node - downstream node". In some environments, if a node and multiple adjacent nodes all meet the condition of similar propagation time difference, further amplitude matching, phase similarity, etc. can be used to determine the final adjacent pair. The node pair thus established can be regarded as a directed edge in the graph structure, representing the most likely and shortest path propagation direction of acoustic waves.

[0166] To integrate the time delay and attenuation or other features together, the system also assigns a composite weight to each adjacent edge. The common practice is to linearly combine the propagation time difference and the arrival amplitude attenuation value with a preset weighting coefficient, or to use a nonlinear function mapping to enhance the sensitivity to a certain feature. For example, the composite weight = γ × (propagation time difference) + β × (amplitude attenuation), where γ and β are configurable parameters. The amplitude attenuation can be estimated according to the amplitude ratio or RMS energy difference between node A and node B. If the amplitude attenuation is too large, the path quality may be poor; if the time difference is too large, it also indicates that the physical distance or medium condition is far apart.

[0167] When the analysis and construction module 40 constructs such a weighted graph, a common graph search algorithm (such as a shortest path search or a minimum spanning tree algorithm) can be called to generate an overall acoustic propagation path atlas to determine the positional relationship of each node in the chain propagation. For the graph atlas that has been constructed, incremental information can also be updated to the graph structure immediately when a new detection produces an effective tag data set. In this way, the system can maintain the dynamics and real-time nature of the path atlas in the case of continuous changes in the optical cable environment or batch addition and deletion of nodes, so as to reflect the most real current laying state and propagation relationship.

[0168] In a specific deployment, if the system detects that the adjacency relationship of some nodes frequently changes or the composite weight of a certain edge continuously and significantly increases in multiple consecutive updates, it may also indicate that the optical cable in this section has a dynamic degradation phenomenon (such as stretching, sheath water immersion, or joint loosening). At this time, the analysis and construction module 40 can further output alarm information to the projection and identification module 50 to highlight the section in the GIS coordinate system and prompt the maintenance personnel to perform pre-checking.

[0169] As an optional implementation, constructing an acoustic propagation path atlas based on a graph search algorithm using the composite weight includes:

[0170] Abstracting each tag node in the effective tag data set as a directed graph vertex, taking the reference tag node as a source vertex, and mapping the adjacent node pairs determined by the composite weight as directed edges from the source vertex to the propagation direction;

[0171] In the graph search process, when the composite weight of a candidate directed edge exceeds a preset pruning threshold, the expansion of the corresponding edge is terminated;

[0172] Based on the pruned directed graph, the Dijkstra algorithm is used to calculate the shortest cumulative composite weight path from the source vertex to the remaining tag nodes to obtain an acoustic propagation path result set;

[0173] For each path in the acoustic propagation path result set, a path record composed of a tag node sequence, a cumulative propagation time difference, and a cumulative arrival amplitude attenuation is output, and the path record is written into the acoustic propagation path atlas.

[0174] Further, in some complex optical cable networks, it is difficult to obtain a complete and stable propagation path distribution in the entire network range by relying solely on simple adjacency pairs and composite weight screening. Therefore, the present application can further introduce a graph search algorithm in the analysis and construction module 40 to perform global-level search and pruning on these weighted edges.

[0175] Specifically, each label node in the effective label dataset is first abstracted as a directed graph vertex, and if the reference label node is confirmed within the detection period, the reference label node can be regarded as the source vertex of the whole graph, so as to generate a series of directed edges downstream from the source in the propagation direction. After the system assigns the composite weight between nodes, it will terminate the processing of candidate edges with excessive weight in the graph search process based on the "preset pruning threshold", so as to avoid excessive expansion or generation of a large number of invalid branches.

[0176] When the composite weight of a candidate edge exceeds the preset pruning threshold, the analysis and construction module 40 will determine that the node pair corresponding to the edge does not constitute an actual feasible propagation path in the current scene, which may be due to excessive attenuation, excessive time delay or abnormal data. By terminating the expansion of such edges in the search in a timely manner, the system can quickly converge to those paths with lower weight and stable transmission with less computational effort. The graph search algorithm itself can adopt various ways, such as Dijkstra algorithm, to find the shortest cumulative composite weight path from the source vertex to the remaining nodes on the pruned directed graph. Each shortest path actually corresponds to a feasible and relatively low-loss, low-latency optical cable propagation chain, thereby helping the system to restore or iteratively update the acoustic propagation path atlas.

[0177] When filling the result set obtained by the shortest path search into the path atlas, the analysis and construction module 40 outputs the node sequence, cumulative propagation time difference and cumulative arrival amplitude attenuation of each path record, which reflect how the acoustic disturbance attenuates or delays on the link from the source to the downstream node. If some paths seem feasible but have abnormally large cumulative attenuation, the system can also prompt manual verification in the next step. Through this graph search and pruning strategy, the analysis and construction module 40 can integrate a large number of dispersed node observation results into interpretable and visualizable propagation links, and perform incremental update as soon as new effective label data appears in subsequent detection, so as to ensure that the atlas always reflects the real-time or near real-time state and fault risk of the optical cable laying.

[0178] For example, if the composite weight is linearly combined from the time delay difference and the amplitude attenuation, when the directed edge weight exceeds a certain threshold (for example, 10, or a certain upper limit after unitization), it will be filtered out in the search. Then use Dijkstra algorithm to find the minimum weight path from the reference tag node to each other node, if a path with relatively small cumulative time delay difference and relatively small amplitude attenuation from the source node to the target node is found, it can be explained that this group of nodes is more likely to be in the position of physical adjacent or branch connection. After completing the search, each shortest cumulative weight path can be recorded in the acoustic propagation path map, and the system saves the node order and accumulates the relative attenuation, so as to visualize and subsequent fault identification in the projection and identification module 50. If new detection period data makes part of the edge weight recalculate, the module can perform incremental correction on the existing path in the graph without starting from scratch to build.

[0179] In this way, the whole system still maintains good adaptive reconstruction ability in the case of lacking explicit topology and multiple attenuations, and provides high-quality path link basis for subsequent positioning analysis.

[0180] As an optional implementation, writing the path record into the acoustic propagation path map comprises:

[0181] For each path in the acoustic propagation path result set, calculating a path confidence according to the cumulative composite weight of the path and the number of tag nodes contained in the path;

[0182] According to the path confidence, sorting the acoustic propagation path result set in descending order, and taking the top N paths in the confidence as a target path set, where N is a preset maximum output number;

[0183] For each path in the target path set, calculating an average temperature difference value of the path based on the local temperature data of the tag nodes constituting the path;

[0184] Writing the target path set together with the corresponding tag node sequence, cumulative propagation time difference, cumulative arrival amplitude attenuation and the average temperature difference value into the acoustic propagation path map.

[0185] In partial fault detection tasks, a single Dijkstra algorithm may generate a large number of shortest cumulative composite weight paths, for example, in a multi-branch structure or a densely distributed node optical cable network, there may be several similar weight feasible paths between each terminal node and the source vertex. In order to facilitate subsequent interpretation and focus, the present application introduces a path confidence calculation mechanism in the analysis and construction module 40, sorts all paths in descending order, and then according to the preset maximum output number N, extracts several representative or high-quality paths, and records their average temperature difference and other additional information, and writes them into the final acoustic propagation path atlas.

[0186] After completing the Dijkstra search for the pruned directed graph, the system obtains one or more shortest cumulative composite weight paths from the source vertex (usually the reference label node or the location of the injection module 10) to the remaining label nodes. The analysis and construction module 40 organizes these paths into an acoustic propagation path result set, and then calculates a path confidence for each path in the set according to the cumulative composite weight of the path and the number of label nodes contained in the path.

[0187] Among them, the path confidence can be defined as a value represented by a floating point number or other fraction between 0 and 1, or by a linear / non-linear function to give a higher confidence score to a path with "smaller weight and moderate number of nodes". For example, let the confidence be inversely proportional to 1 / (cumulative weight x node number), but other modification coefficients can also be added on this basis.

[0188] After the system completes the calculation of all path confidences, it sorts the entire path result set in descending order according to the confidence, and generates a target path set from the top N paths. Here, N can be a constant set in the system configuration, or it can be dynamically adjusted according to the actual scene to balance the needs of "retaining multiple candidate paths to avoid missing detection" and "avoiding too many results". For a larger N value, the system can retain enough candidate routes when there are multiple branches or multiple approximately same weight paths; if the N value is small, it is more conducive to focusing on high-confidence paths in the interface or downstream processing link.

[0189] After the target path set is selected, the system can traverse all the label nodes within each path, extract the local temperature data reported by these nodes, and then calculate the average temperature difference value of the path. If the number of nodes in the path is a, the system can average the difference between the temperatures of adjacent nodes, or compare the arithmetic average of the single node temperature with the reference temperature, or make more detailed segmented analysis to obtain a value that comprehensively represents the temperature gradient experienced by the path. When the same path is detected multiple times and the average temperature difference value tends to be consistent or changes regularly, it indicates that the temperature conditions of the path in the physical space are relatively stable; if the temperature difference fluctuates sharply in a short time, it may indicate that the environmental conditions have changed (such as pipe damage, or a section of the pipeline being laid from the ground to an elevated position, etc.).

[0190] The system will eventually write the target path set together with the corresponding label node sequence, cumulative propagation time difference, cumulative amplitude attenuation, and the above-mentioned average temperature difference value, etc. key information into the acoustic propagation path map. This not only records the specific node sequence passed by each high-confidence path, but also provides additional temperature and attenuation data, which facilitates accurate positioning of potential fault points in the subsequent projection and identification module 50. If the maintenance personnel want to view the full-range attenuation curve and temperature curve of a specific path from the source vertex to the end node, the system can also generate visual reports or trend charts based on these records, further improving the efficiency of fault diagnosis and operation decision-making.

[0191] In this way, for urban optical cable networks with different pipe burial depths, significant ground heat island effects, or frequent seasonal temperature changes, the adaptability and insight of the system in complex environments can be improved, and the subsequent processing burden caused by missed detection, false positives, or redundant paths can be reduced.

[0192] As an optional implementation, determining whether the propagation difference between each label node in the multi-dimensional data exceeds a preset propagation threshold, and if it exceeds, identifying the label node as an abnormal propagation node; and determining the geographic information system coordinates of the abnormal propagation node as the fault space position comprises:

[0193] Calculating the baseline mean μ and standard deviation σ of the set of propagation time differences between adjacent label nodes, and when the absolute value of the difference between a certain propagation time difference and the baseline mean is greater than kσ, marking the corresponding downstream label node as a candidate abnormal node; wherein k is a preset abnormal coefficient;

[0194] Grouping the continuously appearing candidate abnormal nodes according to the propagation direction, and when the number of nodes in the same group is not less than a preset continuous abnormal length, confirming the group as an abnormal propagation node cluster;

[0195] For each cluster of abnormal propagation nodes, based on the geographic information system coordinates of the head and tail tag nodes constituting the cluster, the center coordinates of the cluster of abnormal propagation nodes are calculated as the fault space position by linear interpolation along the cable path.

[0196] The fault space position is written into the acoustic propagation path map, and the corresponding geographic information system coordinates are output to the projection and identification module 50.

[0197] In some pipe network or extremely complex application scenarios, simply performing threshold judgment on a shortest path or adjacent node pair may not be able to timely identify some scattered and distributed implicit fault areas along the line. Therefore, the application adds a set of baseline mean-standard deviation statistical method in the projection and identification module 50 or the analysis and construction module 40, which is used to evaluate the propagation time difference between adjacent tag nodes as a whole, and is supplemented by node clustering and coordinate interpolation means to determine the specific fault center position in the GIS coordinate system.

[0198] When the system obtains the set of propagation time differences of tag node pairs in a detection period, it will connect the two-by-two adjacent relationship of the nodes according to the previously determined acoustic propagation path map, extract the propagation time difference list between the node pairs, and calculate the baseline mean μ and the standard deviation σ of the list. The baseline mean μ can be understood as the overall central tendency of the delay of the entire network in the normal transmission state, and the standard deviation σ reflects the random fluctuation or distribution dispersion of the delay in different sections. If the absolute value of the deviation of the propagation time difference of a pair of nodes from μ exceeds kσ, where k is a pre-configured abnormal coefficient (usually between 2 and 3), it can be determined that this adjacent relationship has extremely abnormal delay signs in the current period. The system will mark the corresponding downstream tag node as a candidate abnormal node to indicate that there may be local sheath damage, joint loosening or other physical damage.

[0199] Through this abnormality identification method based on statistical distribution, the system can not only capture abnormal values relative to the average delay, but also take into account the dispersion of the entire network. If some delays themselves have a relatively high fluctuation amplitude in the entire pipeline, the range of the mean plus multiple σ will be relatively loose, avoiding false positives for nodes that are inherently highly curved or have severe scattering; on the contrary, for pipelines with relatively small overall fluctuations and relatively stable distribution, even a large deviation can be quickly detected by the kσ threshold.

[0200] After the determination of the candidate abnormal nodes is completed, the system further performs a continuity detection on the candidate abnormal nodes according to the acoustic disturbance propagation direction. If it is found that a plurality of nodes continuously appear abnormal markers in adjacent positions (referring to adjacent in the physical propagation order, rather than adjacent in the ID), and maintain the abnormal state in the same detection period or multiple detection periods, the continuously appearing candidate abnormal nodes can be grouped. When the number of nodes in the same group is not less than the "preset continuous abnormal length" (for example, 2-3 or more), it indicates that the optical cable is extremely likely to have a concentrated failure risk at the branch or section level, and the system can further define the group of nodes as an "abnormal propagation node cluster".

[0201] Each abnormal propagation node cluster includes a head node and a tail node, and the system can calculate a central coordinate as a representative value of the fault space position based on the positions of the two extreme nodes in the GIS coordinate system through linear interpolation or more precise curve interpolation of the optical cable path. If it is known that the nodes distributed in the cluster are not strictly linear in the physical space, curve approximation or segmented interpolation can also be applied to fit the arrangement of multiple nodes in the GIS coordinate system, and the center of the curve or the center of gravity is selected as the final interpolation result. Since the fault usually has a sheet shape or a large range, such aggregated positioning can help maintenance personnel quickly lock a suspicious section, rather than just a point.

[0202] After the positioning of the abnormal propagation node cluster is completed, the system writes the fault space position information generated by the interpolation into the acoustic propagation path map, and outputs the corresponding GIS coordinates to the projection and identification module 50. The projection and identification module 50 can mark the section or the fault center position on the geographic information system with highlights, warning icons or different colors to remind the on-site operation and maintenance personnel to pay attention to the possible occurrence of continuous or multi-point faults. When the fault is checked or repaired, the system can also verify it in the next detection period, and once the propagation time delay returns to normal, the system automatically removes the area from the abnormal cluster or updates the state, so as to maintain the real-time accuracy of the network path and fault distribution information.

[0203] In this way, both scattered and isolated hop point interference and potential large-scale or continuous faults in the network can be quickly judged and pinpoint interpolated, thereby greatly shortening the repair decision and troubleshooting time, and providing more adaptive intelligent detection capability for complex laying environments of large-scale, multi-branch pipelines.

[0204] Based on the same inventive concept, the embodiments of the present application also provide an analysis and positioning method corresponding to the fault signal detection system. Since the principle of the method in the embodiments of the present application solves the problem similar to the above-mentioned fault signal detection system of the embodiments of the present application, the implementation of the method can be referred to the implementation of the system, and the repeated parts will not be described herein.

[0205] Referring to Figure 1 FIG. 1 is a flowchart of an analysis positioning method provided by an embodiment of the present application, including steps S101-S105, wherein:

[0206] S101: injecting a preset narrowband acoustic disturbance signal into a target optical cable;

[0207] S102: setting passive wireless tag nodes at intervals along the target optical cable, each tag node being configured to generate and output multi-dimensional data representing propagation characteristics of the acoustic disturbance signal when receiving the acoustic disturbance signal;

[0208] S103: collecting the multi-dimensional data and associating corresponding tag node identifiers;

[0209] S104: calculating propagation differences between the tag nodes based on the multi-dimensional data, and constructing an acoustic propagation path atlas;

[0210] S105: mapping the acoustic propagation path atlas to a geographic information system coordinate system corresponding to geographic coordinates of the tag nodes; determining whether the propagation differences between the tag nodes in the multi-dimensional data exceed a preset propagation threshold, and if so, identifying the tag nodes as abnormal propagation nodes; and determining geographic information system coordinates of the abnormal propagation nodes as fault space positions.

[0211] Those skilled in the art can appreciate that the units and algorithm steps of the examples described in conjunction with the embodiments disclosed in the present application can be implemented in electronic hardware or a combination of computer software and electronic hardware. Whether the functions are implemented in hardware or software depends on the specific application and design constraints of the technical solution. A person skilled in the art can use different methods to implement the described functions for each specific application, but such implementation should not be considered beyond the scope of the present application.

Claims

1. A fault signal detection system, characterized by, The method comprises the following steps: an injection module configured to inject a preset narrowband acoustic disturbance signal into a target optical cable; passive wireless tag nodes arranged at intervals along the target optical cable, each tag node being configured to generate and output multidimensional data representing the propagation characteristics of the acoustic disturbance signal when the acoustic disturbance signal is received; a convergence module configured to collect the multidimensional data and associate corresponding tag node identifiers; an analysis and construction module configured to calculate the propagation differences between the tag nodes based on the multidimensional data and construct an acoustic propagation path map; a projection and identification module configured to map the acoustic propagation path map to a geographic information system coordinate system corresponding to the geographic coordinates of the tag nodes; determine whether the propagation differences between the tag nodes in the multidimensional data exceed a preset propagation threshold; if so, identify the tag nodes as abnormal propagation nodes; and determine the geographic information system coordinates of the abnormal propagation nodes as the fault spatial position. The method further comprises the following steps: driving the injection module to inject at least two groups of probe acoustic pulses in an increasing frequency order within a preset sweep frequency range, and recording the arrival amplitudes of each group of probe acoustic pulses by a reference tag node arranged at the deployment point of the injection module; selecting a target frequency according to the arrival amplitudes of each group of probe acoustic pulses, and determining the injection frequency of the probe acoustic pulse with the maximum amplitude and the decay rate lower than the preset threshold as the target narrowband injection frequency; adjusting the injection power based on the real-time measured arrival amplitudes of the reference tag node, so that the amplitude of the acoustic disturbance signal at the target narrowband injection frequency at the reference tag node remains within a preset target amplitude range; after completing the injection power adjustment, cyclically injecting the acoustic disturbance signal at the target narrowband injection frequency into the target optical cable according to a preset injection period.

2. The fault signal detection system of claim 1, wherein The method further comprises the following steps: each passive wireless tag node is further configured to report local temperature data together with the identifier of its previous hop tag node when outputting the multidimensional data; wherein the previous hop tag node refers to the tag node that is adjacent to the current tag node and located upstream of the current tag node in the acoustic propagation path map; in response to the absence of other tag nodes between the current tag node and the injection module, regarding the reference tag node as the previous hop tag node of the current tag node; in response to the current tag node being the farthest end tag node in the acoustic propagation path, the previous hop tag node thereof being the nearest upstream tag node adjacent thereto; the analysis and construction module is configured to generate a tag pair temperature difference vector based on the local temperature data of adjacent tag nodes; divide the acoustic propagation path map into several isothermal segments according to the temperature difference vector, and the temperature difference of each isothermal segment does not exceed a preset threshold; for each isothermal segment, call a pre-stored temperature-acoustic velocity mapping data table corresponding to the optical cable sheath material and tension level, and use the mapped acoustic velocity parameter to perform temperature compensation on the propagation difference of the segment.

3. The fault signal detection system of claim 2, wherein The method further comprises the following steps: analog-digital conversion is performed on the acoustic disturbance signal according to a preset sampling rate to obtain a discrete acoustic sampling sequence, starting from a moment when the acoustic disturbance signal amplitude is detected to exceed a preset trigger threshold for the first time; within a preset analysis window, the discrete acoustic sampling sequence is processed, when the amplitude of the discrete acoustic sampling sequence exceeds a rising edge threshold for the first time, a time when this happens is recorded as an arrival time stamp, a root mean square amplitude value is calculated as an arrival amplitude, a maximum energy frequency component is extracted as an arrival frequency through a fast Fourier transform, and an instantaneous phase of the maximum energy frequency component is synchronously obtained; a label node identifier, a last-hop label node identifier, local temperature data, and the arrival time stamp, arrival amplitude, arrival frequency, and instantaneous phase are packaged into an uplink data frame according to a preset data frame format; the uplink data frame is sent to the convergence module by using a near-field energy coupling mode.

4. The fault signal detection system of claim 3, wherein the propagation difference between each label node is calculated based on the multi-dimensional data, and an acoustic propagation path atlas is constructed, including: based on the arrival amplitude and the instantaneous phase, label node data with a peak signal power to noise mean square ratio lower than a preset signal-to-noise ratio threshold and / or an adjacent frame phase difference greater than 180° is removed, and an effective label data set is obtained; for each label node in the effective label data set, another label node with the smallest and greater than zero propagation time difference is selected as an adjacent node pair, and an adjacent edge is adaptively generated; a composite weight is assigned to each adjacent edge, which is obtained by linear combination of the propagation time difference and the arrival amplitude attenuation value of the adjacent edge according to a preset weighting coefficient; an acoustic propagation path atlas is constructed based on a graph search algorithm using the composite weight, and incremental information is updated to the acoustic propagation path atlas in real time when a new effective label data set is generated.

5. The fault signal detection system of claim 4, wherein, an acoustic propagation path atlas is constructed based on a graph search algorithm using the composite weight, including: each label node in the effective label data set is abstracted as a directed graph vertex, the reference label node is taken as a source vertex, and adjacent node pairs determined by the composite weight are mapped as directed edges from the source vertex to the propagation direction; in the graph search process, when the composite weight of a candidate directed edge exceeds a preset pruning threshold, the expansion of the corresponding edge is terminated; based on the pruned directed graph, a Dijkstra algorithm is used to calculate the shortest cumulative composite weight path from the source vertex to the remaining label nodes to obtain an acoustic propagation path result set; for each path in the acoustic propagation path result set, a path record composed of a label node sequence, a cumulative propagation time difference, and a cumulative arrival amplitude attenuation is output, and the path record is written into the acoustic propagation path atlas.

6. The fault signal detection system of claim 5, wherein, writing the path record into the acoustic propagation path atlas includes: for each path in the acoustic propagation path result set, a path confidence is calculated according to the cumulative composite weight of the path and the number of label nodes contained in the path; the acoustic propagation path result set is sorted in descending order according to the path confidence, and the top N paths in confidence are taken as a target path set, where N is a preset maximum output number; For each path in the target path set, an average temperature difference value of the path is calculated based on local temperature data of label nodes constituting the path; The target path set is written into the acoustic propagation path atlas together with corresponding label node sequence, accumulated propagation time difference, accumulated amplitude decay and the average temperature difference value.

7. The fault signal detection system of claim 6, wherein It is judged whether the propagation difference between each label node in the multi-dimensional data exceeds a preset propagation threshold, and if it exceeds, the label node is identified as an abnormal propagation node; And the geographic information system coordinates of the abnormal propagation node are determined as the fault space position, including: The baseline mean μ and standard deviation σ of the propagation time difference set between adjacent label nodes are calculated, and when the absolute value of the difference between a certain propagation time difference and the baseline mean is greater than kσ, the corresponding downstream label node is marked as a candidate abnormal node; wherein k is a preset abnormal coefficient; According to the propagation direction, the continuously appearing candidate abnormal nodes are grouped, and when the number of nodes in the same group is not less than a preset continuous abnormal length, the group is confirmed as an abnormal propagation node cluster; For each abnormal propagation node cluster, based on the geographic information system coordinates of the head and tail label nodes constituting the cluster, the center coordinates of the abnormal propagation node cluster are calculated as the fault space position through linear interpolation along the optical cable path; The fault space position is written into the acoustic propagation path atlas, and the corresponding geographic information system coordinates are output to the projection and identification module.

8. An analysis positioning method characterized by, It includes: A preset narrowband acoustic disturbance signal is injected into a target optical cable; Passive wireless label nodes are arranged at intervals along the target optical cable, and each label node is configured to generate and output multi-dimensional data representing the propagation characteristics of the acoustic disturbance signal when receiving the acoustic disturbance signal; The multi-dimensional data is collected and associated with the corresponding label node identification; Based on the multi-dimensional data, the propagation difference between each label node is calculated, and an acoustic propagation path atlas is constructed; The acoustic propagation path atlas is mapped to a geographic information system coordinate system corresponding to the geographic coordinates of the label nodes; it is judged whether the propagation difference between each label node in the multi-dimensional data exceeds a preset propagation threshold, and if it exceeds, the label node is identified as an abnormal propagation node; And the geographic information system coordinates of the abnormal propagation node are determined as the fault space position; The preset narrowband acoustic disturbance signal is injected into the target optical cable, including: The driving injection module respectively injects at least two groups of trial acoustic pulses in an increasing frequency order within a preset frequency sweep range, and the reference label node arranged at the layout point of the injection module records the arrival amplitude of each group of trial acoustic pulses; According to the arrival amplitudes of each group of trial acoustic pulses, a target frequency is selected, and the injection frequency corresponding to the trial acoustic pulse with the maximum amplitude and the decay rate lower than the preset threshold is determined as the target narrowband injection frequency; Based on the arrival amplitude measured by the reference label node in real time, the injection power is adjusted so that the amplitude of the acoustic disturbance signal at the target narrowband injection frequency at the reference label node remains within a preset target amplitude interval; After completing the injection power adjustment, the acoustic disturbance signal of the target narrowband injection frequency is cyclically injected into the target optical cable according to a preset injection period.

Images