Leaky cable intelligent anomaly detection system and method

Abstract

The application relates to the technical field of communication equipment, in particular to a leakage cable intelligent abnormality detection system and method. The system comprises an acquisition module, which acquires historical signal strength data and position information; a construction module, which constructs a probability distribution model of terminal signal strength at each geographical position point and determines a normal signal strength reference interval; a selection module, which generates a position distribution probability and selects a reference position; an acquisition module, which acquires signal strength reported by multiple terminals in a current detection period in real time to form a current signal strength observation data set; an association module, which associates actual signal strength with a reference position and a position distribution probability of the reference position to form association data; a comparison module, which compares the actual signal strength with a normal signal strength reference interval of the reference position and calculates a deviation degree; a calculation module, which calculates a weighted average deviation degree; and a judgment module, which issues an abnormality alarm when the weighted average deviation degree exceeds a deviation threshold.

Description

Technical Field

[0001] This application relates to the field of communication equipment technology, specifically to an intelligent anomaly detection system and method for leaky cables. Background Technology

[0002] Leaky feeder cables, as special radio frequency devices that combine the functions of transmission lines and distributed antennas, are widely used in enclosed or semi-enclosed environments such as subways, tunnels, and mines to provide continuous and stable signal coverage for wireless communication systems. However, due to long-term exposure to complex electromagnetic environments and physical stress conditions, they are susceptible to external interference, mechanical damage, loose connections, or aging, leading to abnormal radiation characteristics and consequently causing communication interruptions, signal attenuation, or coverage blind spots. Traditional maintenance methods mainly rely on periodic manual inspections or simple alarm mechanisms based on fixed thresholds, which are difficult to identify early or localized performance degradation in a timely and accurate manner, and lack the ability to adaptively learn from historical operating states, resulting in high false alarm rates and low levels of intelligence. Therefore, it is necessary to research more intelligent anomaly detection and early warning technologies to reduce the need for manual inspections. Summary of the Invention

[0003] This specification describes a leaky cable intelligent anomaly detection system and method through several embodiments.

[0004] Firstly, embodiments of this specification provide an intelligent anomaly detection system for leaky cables, comprising:

[0005] The acquisition module acquires signal strength data and corresponding location information reported by multiple dedicated terminals along the target leaky cable during its historical normal operation.

[0006] The construction module, based on the historical signal strength data and location information, constructs a probability distribution model of the terminal signal strength at each geographical location point, and determines the normal signal strength reference interval corresponding to each location point;

[0007] Select the module to generate the terminal's location distribution probability, and select several locations whose location distribution probability is higher than the preset reference probability value as reference locations;

[0008] The acquisition module collects signal strength data reported by multiple terminals in real time during the current detection period, forming a current signal strength observation dataset.

[0009] The association module associates the actual signal strength in the current signal strength observation dataset with a reference location and the location distribution probability of the reference location to form association data. The association data includes the actual signal strength, the reference location, and the corresponding location distribution probability.

[0010] The comparison module compares the actual signal strength with the normal signal strength reference range at the reference location, calculates the deviation, and calculates the confidence level based on the corresponding location distribution probability.

[0011] The calculation module uses the confidence level as the weight to calculate the weighted average deviation of the current signal strength observation dataset;

[0012] The judgment module determines that there is an abnormality in the leaky cable when the weighted average deviation exceeds a preset deviation threshold and issues an abnormality alarm.

[0013] Secondly, this specification provides an intelligent anomaly detection method for leaky cables, comprising the following steps:

[0014] Acquire signal strength data and corresponding location information reported by multiple dedicated terminals along the target leaky cable during its historical normal operation.

[0015] Based on the historical signal strength data and location information, a probability distribution model of the terminal signal strength at each geographical location is constructed, and a reference interval for normal signal strength corresponding to each location is determined.

[0016] Generate the location distribution probability of the terminal, and select several locations whose location distribution probability is higher than the preset reference probability value as reference locations;

[0017] During the current detection period, the signal strength reported by multiple terminals is collected in real time to form the current signal strength observation dataset.

[0018] The actual signal strength in the current signal strength observation dataset is associated with a reference location and the location distribution probability of the reference location to form associated data. The associated data includes the actual signal strength, the reference location and the corresponding location distribution probability.

[0019] The actual signal strength is compared with the normal signal strength reference range at the reference location to calculate the deviation, and the confidence level is calculated based on the corresponding location distribution probability.

[0020] Using the confidence level as the weight, the weighted average deviation of the current signal strength observation dataset is calculated;

[0021] When the weighted average deviation exceeds a preset deviation threshold, the leaky cable is determined to be abnormal, and an abnormality alarm is issued.

[0022] Thirdly, embodiments of this specification provide an electronic device, including a processor and a memory;

[0023] The processor is connected to the memory;

[0024] The memory is used to store executable program code;

[0025] The processor runs a program corresponding to the executable program code stored in the memory to perform the method described in any of the above aspects.

[0026] Fourthly, embodiments of this specification provide a computer-readable storage medium having a computer program stored thereon, which, when executed by a processor, implements the methods described in any of the above aspects.

[0027] Fifthly, embodiments of this specification provide a computer program product, including a computer program that, when executed by a processor, implements the methods described in any of the above aspects.

[0028] The beneficial effects of the technical solutions provided in some embodiments of this specification include at least the following:

[0029] In several embodiments of this specification, the provided intelligent anomaly detection system and method for leaky cables constructs a probability distribution model of signal strength for each geographical location and determines a normal reference interval accordingly. This effectively characterizes the reasonable fluctuation range of the signal under normal operating conditions, avoiding misjudgments caused by environmental noise or individual terminal differences. By combining the historical location distribution probability of the terminal to screen high-confidence reference locations, and using these as confidence weights to weight and fuse the current observation deviation, the detection results are more focused on data from high-frequency, stable observation areas, avoiding the interference caused by random anomalies in low-probability areas. When the weighted average deviation exceeds a preset threshold, a timely warning is issued, indicating potential early performance degradation. Although communication interruption has not yet occurred, the deviation from normal statistical patterns helps improve the reliability of leaky cable maintenance. The system enables the detection of the leaky cable's condition between two manual inspections, helping to extend the interval between manual inspections and reduce maintenance costs.

[0030] Other features and advantages of various embodiments of this specification will be further revealed in the following detailed description and accompanying drawings. Attached Figure Description

[0031] To more clearly illustrate the technical solutions in the embodiments of this specification, the accompanying drawings used in the embodiments will be briefly introduced below. Obviously, the drawings described below are only some embodiments of this specification. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0032] Figure 1 This is a schematic diagram of intelligent anomaly detection for leaky cables provided in this manual.

[0033] Figure 2 This is a schematic diagram of the intelligent anomaly detection method for leaky cables provided in this manual.

[0034] Figure 3 This is a schematic diagram of the method for constructing a probability distribution model provided in this specification.

[0035] Figure 4 This is a flowchart illustrating the method for selecting a reference location as provided in this manual.

[0036] Figure 5 This is a schematic diagram of the method for calculating the weighted average deviation provided in this manual.

[0037] Figure 6 This is a schematic diagram of the intelligent anomaly detection system for leaky cables provided in this manual.

[0038] Figure 7 This is a schematic diagram of the electronic device provided in this manual.

[0039] Among them: 1. Leaky cable, 2. Reference position, 100. Acquisition module, 200. Construction module, 300. Selection module, 400. Acquisition module, 500. Association module, 600. Comparison module, 700. Calculation module, 800. Judgment module, 1100. Electronic equipment, 1101. Processor, 1102. Communication bus, 1103. User interface, 1104. Network interface, 1105. Memory. Detailed Implementation

[0040] The technical solutions of the embodiments of this specification will be explained and described below with reference to the accompanying drawings. However, the following embodiments are only preferred embodiments of this specification and not all of them. Other embodiments obtained by those skilled in the art based on the embodiments in the implementation methods without creative effort are all within the protection scope of this specification.

[0041] The terms "first," "second," "third," etc., in the description, claims, and accompanying drawings are used to distinguish different objects, not to describe a specific order. Furthermore, the terms "comprising" and "having," and any variations thereof, are intended to cover non-exclusive inclusion. For example, a process, method, system, product, or apparatus that includes a series of steps or units is not limited to the listed steps or units, but may optionally include steps or units not listed, or may optionally include other steps or units inherent to such processes, methods, products, or apparatus.

[0042] In the following description, terms such as “inner,” “outer,” “upper,” “lower,” “left,” and “right” are used only to facilitate the description of the embodiments and to simplify the description, and are not intended to indicate or imply that the device or element referred to must have a specific orientation, or be constructed and operated in a specific orientation, and therefore should not be construed as a limitation of this specification.

[0043] All data involved in this application are information and data authorized by the user or fully authorized by all parties, and the collection, use and processing of the relevant data comply with the relevant laws, regulations and standards of the relevant countries and regions.

[0044] Before introducing the technical solutions described in this manual, the application scenarios and related technologies of the technical solutions will be introduced.

[0045] Leaky Coaxial Cable (LCC) is a specially designed coaxial cable with periodic slots or holes in its outer conductor, allowing electromagnetic waves to radiate and receive continuously along the cable's longitudinal direction, thus functioning as both a signal transmission and distributed antenna. LCC solves the problem of limited coverage in narrow, enclosed spaces caused by traditional base station antennas and is widely used for signal extension in wireless communication systems in tunnels, subways, mines, underground utility tunnels, and other similar scenarios. For example, it provides continuous vehicle-to-ground communication coverage for GSM-R, LTE-M, and 5G-R in rail transit and high-speed rail tunnels. It ensures personnel positioning, voice communication, and data transmission in mines and underground engineering projects. It also provides deep coverage supplementation in complex indoor structures such as airports and convention centers within large buildings.

[0046] Common fault types in leaky cables include: physical damage such as squeezing, bending, and cutting, which alter radiation characteristics or exacerbate signal attenuation; joint / connector failures such as loosening, oxidation, and water ingress, leading to increased VSWR and insertion loss; aging and degradation, where dielectric performance deteriorates after long-term use, resulting in decreased radiation efficiency; external interference or shielding, such as metal structure obstruction or construction covering, causing sudden drops in localized signal strength; and installation defects such as improper hanging spacing or excessively small bending radius, which affect radiation uniformity. These faults often do not immediately cause communication interruptions, but they manifest as abnormal fluctuations in signal strength or regional attenuation along the cable, exhibiting a concealed and gradual nature, requiring early identification through inspection activities.

[0047] Currently, maintenance personnel often carry portable field strength meters on foot or in vehicles to measure the field strength, but this method is inefficient, costly, and time-consuming. While using dedicated testing equipment (such as OTDRs and spectrum analyzers) can accurately locate breakpoints or standing wave anomalies, it requires interrupting services or periodic shutdowns for testing, making real-time monitoring difficult.

[0048] Therefore, it is necessary to continue researching fault detection technology for leaky cable 1, reduce the cost of fault detection for leaky cable 1, and ensure the normal operation of leaky cable 1. This specification provides an intelligent anomaly detection system and method for leaky cable 1. Please refer to the appendix. Figure 1 By utilizing signal strength data and corresponding location information reported by multiple dedicated terminals under historical normal operating conditions, a signal strength probability distribution model based on geographical location is constructed, and a signal strength reference interval reflecting the normal fluctuation range is determined for each location point.

[0049] Building upon this, the method further analyzes historical location data to generate the spatial location distribution probability of the terminal, and selects high-confidence areas with location distribution probabilities higher than preset reference probability values ​​as reference locations 2. During the current detection cycle, signal strength observation data from multiple terminals are collected in real time, and these actual observations are associated with geographically adjacent or matching reference locations 2, forming an associated dataset containing actual signal strength, reference locations 2, and their location distribution probabilities. By comparing the actual signal strength with the normal reference interval of the corresponding reference location 2, the degree of deviation is calculated, and the deviation is weighted by combining the historical occurrence probability (i.e., confidence level) of that location, ultimately obtaining the weighted average deviation of the entire observation dataset.

[0050] When the weighted average deviation exceeds a preset deviation threshold, it is determined that the leaky cable 1 may have a performance abnormality, thereby triggering an alarm mechanism to prompt maintenance personnel to intervene and inspect in a timely manner. By integrating spatial statistical modeling, location reliability assessment, and dynamic weighted decision-making, the ability to identify early and weak anomalies is effectively improved, avoiding the problems of traditional fixed threshold methods being susceptible to noise interference and having poor adaptability, thus achieving more intelligent and reliable health status monitoring of the leaky cable 1.

[0051] This includes signal strength data and corresponding location information reported by multiple dedicated terminals during normal historical operation. The information reported by the dedicated terminals needs to include Received Signal Strength Indication (RSSI) or Reference Received Power (RSRP); Signal-to-Noise Ratio (SNR), Bit Error Rate (BER); precise geographic location (such as GPS, odometer, UWB positioning coordinates); timestamp, terminal ID, serving cell information, and other metadata. Dedicated terminals include trackside monitoring modules and fixed probe terminals.

[0052] Within the current detection period, real-time acquisition of signal strength reported by multiple terminals includes all terminals, and only signal strength is needed. Information passively reported by terminals during normal communication includes RSRP / RSSI (measured by the terminal and reported to the base station); serving cell ID, neighbor cell list; coarse location information (such as base station cell coverage area, estimated coordinates based on triangulation, or GPS location authorized by the user); and auxiliary information such as reporting time and terminal type. For example, a user's mobile phone, mobile computer, or mobile tablet can all be used as terminals.

[0053] A normal signal strength reference range is established by using signal strength data and corresponding location information reported by a dedicated terminal. The signal strength normally transmitted by a regular terminal during communication is then processed and compared with this reference range. Simultaneously, location probability is used as confidence level and weight to calculate a weighted average deviation. Based on a comparison of this weighted average deviation with a preset threshold, it is determined whether there is an anomaly in leaky cable 1. This allows for supplementary fault detection, identification, and early warning for dedicated terminals such as trackside monitoring modules and fixed probe terminals, which cannot be covered by dedicated terminals, without the need for personnel inspection. This helps to further ensure the normal working condition of leaky cable 1.

[0054] Specifically, this manual first provides an intelligent anomaly detection method for leaky cables 1. Please refer to the appendix. Figure 2 This includes the following steps:

[0055] Step S1) Obtain signal strength data and corresponding location information reported by multiple dedicated terminals along the target leaky cable 1 during its historical normal operation.

[0056] It is necessary to collect wireless signal measurement data continuously reported by multiple dedicated detection terminals deployed within the coverage area of ​​the target leaky cable 1, such as trackside monitoring modules, fixed probe terminals, and handheld terminals used for manual inspection, during historical periods when the system is known to be fault-free and operating stably. This data mainly includes key indicators characterizing signal strength, such as Received Signal Strength Indication (RSSI) and Reference Received Power (RSRP). Each signal strength record is accompanied by its corresponding high-precision geographical location information.

[0057] The method of acquiring geographic location information depends on the deployment environment and terminal type. For example, in subway tunnel scenarios, dedicated terminals are typically installed at fixed hanging points according to design drawings during the construction phase, and their mileage relative to the starting point of the line (e.g., K23+450.6) is precisely calibrated using a track engineering surveying vehicle. This mileage value is directly written into the terminal configuration as a location identifier. In mine roadways, terminals may achieve centimeter-level positioning through a pre-embedded UWB anchor network, or technicians may manually input latitude, longitude, and elevation using a laser rangefinder in conjunction with the roadway coordinate system during installation. In GNSS-available areas such as entrances and exits of surface railways or highway tunnels, some terminals can use RTK-GPS to obtain centimeter-level accurate geographic coordinates. Each record is formatted as (RSRP=-82dBm, Location=K18+723.5).

[0058] Step S2) Based on the historical signal strength data and location information, construct a probability distribution model of the terminal signal strength at each geographical location point, and determine the normal signal strength reference interval corresponding to each location point.

[0059] Constructing a probability distribution model of the terminal signal strength at each geographical location can characterize the statistical regularity of signal strength at various locations along the leaky cable 1 under healthy conditions. Because wireless signals are affected by multipath effects, environmental temperature and humidity, and minor equipment fluctuations, signal strength at the same location will fluctuate even under normal operating conditions. Therefore, using probabilistic modeling instead of a fixed threshold can more scientifically describe this natural variability.

[0060] Please see the appendix Figure 3 The method for constructing a probability distribution model of terminal signal strength at each geographic location point based on the historical signal strength data and location information includes:

[0061] Step S21) The historical signal strength data is aggregated into a grid according to geographical location to obtain several discrete observation points. To facilitate modeling and calculation, the continuous geographic space along the leaky cable 1 is first discretized. For example, in a subway tunnel with a total length of 10 kilometers, the track direction can be divided into grids at 50-meter intervals, generating a total of 200 equidistant observation points (such as K0+000, K0+050, ..., K9+950). Each grid represents a spatial unit, and its location is usually identified by the center point coordinates or the corresponding kilometer marker. This gridding not only simplifies spatial indexing but also accommodates data reported by terminals with different densities, making it particularly suitable for scenarios where dedicated terminals are not uniformly deployed.

[0062] Step S22) For each observation point, collect all historical signal strength values ​​reported by terminals within the grid where the observation point is located, forming a signal strength sample set for that observation point. Taking observation point K5+300 as an example (corresponding to grid range K5+275 to K5+325), select all RSRP records falling within this range from the historical database. Assume that during a 3-month normal operation period, this grid received a total of 4120 valid reported data from 3 fixed trackside terminals and onboard probes of passing trains. The timestamps of these data are evenly distributed, covering different time periods, weather conditions, and operational load conditions. The resulting sample set can reflect the overall signal strength of this location under normal conditions.

[0063] Step S23) Perform statistical analysis on the signal strength sample set and fit to obtain the probability density function.

[0064] For the 4120 RSRP samples at K5+300, data cleaning was first performed to remove obvious outliers, such as <-120dBm, and then the distribution characteristics were analyzed. Histogram observation revealed a unimodal, approximately symmetrical shape. Further maximum likelihood estimation was used to fit a Gaussian distribution, yielding a mean μ = -81.6dBm and a standard deviation σ = 2.1dBm. Alternatively, a better model can be automatically selected. If the AIC / BIC criterion indicates that a Gaussian mixture model provides a better fit (e.g., there is a bimodal distribution due to differences in early and late signals), then the Gaussian mixture model is preferred. The probability density function f at this point is output. (K5+300) (x) describes the random distribution of signal strength at that location.

[0065] Step S24) Obtain the probability distribution model based on the grid and probability density function.

[0066] Each grid cell is bound to its corresponding probability density function, forming a "location-distribution" mapping relationship, i.e., a complete probability distribution model. For example, each of the 200 grid cells in the entire line has an independent fi(x), (i=1,2,…,200), constituting a signal strength probability map covering the entire line. This probability distribution model not only stores the expected signal level but also implicitly includes local fluctuation characteristics, providing a basis for the subsequent generation of dynamic reference intervals.

[0067] The method for determining the normal signal strength reference range for each location point includes: determining the signal strength value range covering a preset confidence level based on the probability density function; obtaining the normal signal strength reference range based on the signal strength value range. Further, the normal signal strength reference range is determined based on the probability density function of each point. For example, for K5+300, if the confidence level is set to 95%, the reference range is [μ-1.96σ,μ+1.96σ]=[﹣85.7"dBm",﹣77.5"dBm"]. When the cable is operating normally, approximately 95% of the observed values ​​should fall within this range. If real-time observations consistently exceed this limit, it is considered a potential anomaly.

[0068] Step S3) Generate the location distribution probability of the terminal, and select several locations with location distribution probabilities higher than the preset reference probability value as reference locations 2.

[0069] Ordinary users' terminals upload approximate locations, which can be used to generate the terminal's location distribution probability. When approximate location data uploaded by ordinary users' terminals is unavailable, the location distribution probability of terminals can be received manually based on the actual geographical and architectural environment.

[0070] When sufficient location information reported by historical terminals is available, these historical terminals include both dedicated terminals and terminals used by ordinary users. Please refer to the appendix. Figure 4 The method of selecting several locations whose positional probability is higher than a preset reference probability value as reference location 2 includes:

[0071] Step S31) Based on the location information reported by historical terminals, count the number of times the terminal is located within the grid of each observation point, and use this count as the number of visits to the observation point. Count the total frequency of terminal visits for each defined geographic grid. An visit refers to a terminal successfully reporting signal strength data within that grid. For example, on a 12-kilometer-long subway line 5 in a certain city, 240 grids are divided at 50-meter intervals. Statistics show that grid K7+200 (corresponding to kilometer 7.2) received 18,650 terminal reports in three months of historical data, while grid K9+850, located near a branch line junction, received only 210 reports. This difference reflects the spatial imbalance in operational density, train stop frequency, or user activity levels.

[0072] Step S32) Normalize the number of visits to each observation point to obtain the probability of the terminal appearing at each observation point, which is used as the location distribution probability. To eliminate the influence of the total number of visits and obtain comparable probability values, it is necessary to normalize the number of visits to all grids. Specifically, let the number of visits to the i-th grid be ni, and the total number of visits be... Let M be the total number of grid cells, then the probability of the location distribution of that grid cell is Pi = ni / N. For example, if the total number of reports across the entire line is N = 2,150,000, then the probability of K7+200 is P = 18,650 / 2,150,000 ≈ 0.00867 (i.e., 0.867%), while the probability of K9+850 is only 0.0098%. The resulting location distribution probability characterizes the spatial activity level of the terminal.

[0073] Step S33) Based on the preset reference probability value, select the observation points whose position distribution probability is greater than or equal to the reference probability value and use them as reference positions 2.

[0074] A preset reference probability value is set (e.g., Pref=0.003, i.e., 0.3%). This threshold can be dynamically determined based on historical data distribution (e.g., taking the 80th percentile of all Pi), or it can be fixed based on experience. All grids that satisfy Pi≥Pref are marked as reference location 2. In the above case, K7+200 (0.867%>0.3%) is selected, while K9+850 (0.0098%<0.3%) is excluded. Ultimately, a total of 162 reference locations 2 are selected, mainly concentrated in terminal dense areas such as station sections and sections with uniform train speed.

[0075] Step S4) During the current detection period, the signal strength reported by multiple terminals is collected in real time to form the current signal strength observation dataset.

[0076] Unlike the historical modeling phase, the data collection during the detection period emphasizes timeliness, continuity, and broad coverage, and is used for dynamic comparison with the established normal behavior model.

[0077] Within a set detection cycle (e.g., every 5 minutes, every 10 minutes, or real-time streaming), wireless signal measurement information is continuously received from various terminals deployed within the target area, including dedicated detection terminals (e.g., trackside fixed probes) and ordinary user terminals (e.g., train-mounted equipment, staff handheld terminals, or passenger mobile phones). Each reported record must contain at least a signal strength value (e.g., RSRP, RSSI). Where conditions permit, corresponding location information (e.g., GPS coordinates, track mileage markers, or base station cell identifiers) can also be uploaded to generate the terminal location distribution probability. After timestamp alignment, format standardization, and preliminary validity verification, the data is aggregated into a current signal strength observation dataset.

[0078] Step S5) Associate the actual signal strength in the current signal strength observation dataset with reference position 2 and the position distribution probability of reference position 2 to form associated data. The associated data includes the actual signal strength, reference position 2 and the corresponding position distribution probability.

[0079] Since ordinary users' terminals do not upload location information (when not authorized to obtain location information) or can only upload coarse location information (ordinary users cannot obtain precise location information in environments such as tunnels, but can only obtain coarse location information based on information such as cell information), this specification uses a scheme to compensate for the lack of available accurate location information for ordinary users' terminals. Specifically, it guesses that the current user terminal is located at each reference location 2 with a corresponding probability, i.e., location distribution probability. This probabilistic approach is used to determine and predict whether there is an anomaly in the leaky cable 1. Using a dataset of current signal strength observations formed by data reported from a large number of ordinary users' terminals, probabilistic methods are employed to achieve location compensation.

[0080] When the signal strength information reported by the terminal includes location information, it usually indicates that the terminal is a train-mounted terminal or other terminal belonging to the staff. This part of the data is relatively small, and apart from being used to generate the location distribution probability of the terminal, this part of the data can be ignored in the current detection cycle.

[0081] For example, if the actual signal strength reported by the user terminal is -87.3dBm, then the probability of the user being located at the reference position 2K3+100 is 1.25%, which is much higher than that at K8+750 in the middle of the interval (the probability of the position distribution is 0.21%). Therefore, the actual signal strength reported this time is compared with the normal signal strength reference interval at the reference position 2K3+100 with a weight of 1.25% to obtain the deviation.

[0082] Step S6) Compare the actual signal strength with the normal signal strength reference range of the reference position 2, calculate the deviation, and calculate the confidence level according to the corresponding position distribution probability.

[0083] The method of comparing the actual signal strength with the normal signal strength reference range at reference position 2, calculating the deviation, and calculating the confidence level based on the corresponding positional probability distribution includes:

[0084] When the actual signal strength does not exceed the normal signal strength reference range, the deviation is 0;

[0085] When the actual signal strength exceeds the normal signal strength reference range, calculate the absolute value of the difference between the actual signal strength and the normal signal strength reference range, and calculate the ratio of the absolute value to the actual signal strength. The ratio is used as the deviation.

[0086] The position distribution probability of the reference position 2 is used as the confidence level.

[0087] By combining the statistical reliability of spatial location, deviation and confidence levels are provided for subsequent weighted decision-making. A potential anomaly is only considered when the signal strength deviates significantly from the historical reasonable range of that location; and the more comprehensive the historical observations of that location, i.e., the higher the probability of the location distribution, the higher the weight of its judgment result.

[0088] For example, consider reference position 2 K5+300, whose normal signal strength reference range is [-85.7dBm, -77.5dBm], with a position distribution probability of 0.78%; and reference position 2 K2+100, whose normal signal strength reference range is [-83.2, -76.0]dBm, with a position distribution probability of 1.52%. In the current detection period, two actual signal strengths are received. Actual signal strength A: RSRP = -82.1dBm. Since the normal signal strength reference range for reference position 2 K5+300 is -82.1 ∈ [-85.7, -77.5], it does not exceed the range, the deviation is 0, and the confidence level is 0.0078. Since the normal signal strength reference range for reference position 2 K2+100 is -82.1∈[-83.2,-76.0], and it exceeds the range, the deviation is 1.1 / |-83.2|≈0.0132, and the confidence level is 0.0152.

[0089] Step S7) Using the confidence level as the weight, calculate the weighted average deviation of the current signal strength observation dataset.

[0090] Please see the appendix Figure 5 The method for calculating the weighted average deviation of the current signal strength observation dataset using the confidence level as the weight includes:

[0091] Step S71) Calculate the weighted sum of the current signal strength observation dataset using the confidence level as the weight.

[0092] For each record, the deviation Di is multiplied by its corresponding confidence level Ci (i.e., the probability of location distribution), and then summed to obtain the weighted sum. , where N is the number of valid observations successfully associated with reference position 2 within the current detection period.

[0093] Step S72) Calculate the sum of all confidence scores in the current signal strength observation dataset as the total confidence score. The total confidence score is calculated by summing all confidence scores. The total confidence level reflects the overall credible information content of valid observations within the current period, avoiding distortion of weighted results due to fluctuations in the number of terminals.

[0094] Step S73) Obtain the weighted average deviation based on the quotient of the weighted sum and the total confidence level.

[0095] Weighted average deviation = weighted sum / total confidence level.

[0096] Taking the 8:15-8:20 detection cycle as an example, two related data points were selected as shown in Table 1.

[0097] Based on the two related data points shown in Table 1, the weighted sum is calculated as: 0.0545*0.0152 + 0.1193*0.0078 + 0.1592*0.0152 + 0.0520*0.0078 = 0.00458438. The total confidence level is calculated as: 0.0152 + 0.0078 + 0.0152 + 0.0078 = 0.046. The weighted average deviation is calculated as: weighted sum / total confidence level = 0.00458438 / 0.046 = 0.09966. #1, being located in the high-activity area of ​​the station (with a confidence level of 0.0152), has a greater contribution weight.

[0098] Table 1 Related Data Table

[0099]

[0100] Step S8) When the weighted average deviation exceeds the preset deviation threshold, it is determined that there is an abnormality in the leaky cable 1 and an abnormality alarm is issued.

[0101] Dynamic early warning based on statistical regularities considers the signal strength of a large number of terminals and the probability distribution of reference location 2 to achieve early and progressive fault identification and warning. It can be achieved by analyzing the weighted average deviation distribution of a large number of normal cycles and selecting its 99th percentile as a threshold (e.g., 0.35) to ensure that the false alarm rate is less than 1% under normal fluctuations; it can also be dynamically adjusted based on fault simulation experiments or expert experience to balance the risk of missed alarms and maintenance costs. When the weighted average deviation is 0.09966, no warning is required.

[0102] On the other hand, this specification provides an intelligent anomaly detection system for leaky cables 1. Please refer to the appendix. Figure 6 ,include:

[0103] The acquisition module 100 acquires signal strength data and corresponding location information reported by multiple dedicated terminals along the target leaky cable 1 during its historical normal operation.

[0104] The construction module 200 constructs a probability distribution model of the terminal signal strength at each geographical location point based on the historical signal strength data and location information, and determines the normal signal strength reference interval corresponding to each location point.

[0105] Select module 300 to generate the terminal's location distribution probability, and select several locations whose location distribution probability is higher than the preset reference probability value as reference locations 2;

[0106] The acquisition module 400 collects the signal strength reported by multiple terminals in real time during the current detection period to form the current signal strength observation dataset;

[0107] The association module 500 associates the actual signal strength in the current signal strength observation dataset with reference position 2 and the position distribution probability of reference position 2 to form association data. The association data includes the actual signal strength, reference position 2 and the corresponding position distribution probability.

[0108] The comparison module 600 compares the actual signal strength with the normal signal strength reference range of the reference position 2, calculates the deviation, and calculates the confidence level based on the corresponding position distribution probability.

[0109] The calculation module 700 uses the confidence level as the weight to calculate the weighted average deviation of the current signal strength observation dataset;

[0110] The judgment module 800 determines that the leaky cable 1 is abnormal when the weighted average deviation exceeds the preset deviation threshold and issues an abnormality alarm.

[0111] Please see Figure 7 The diagram shown is a structural schematic of an electronic device provided in an embodiment of this specification.

[0112] like Figure 7 As shown, the electronic device 1100 may include: at least one processor 1101, at least one network interface 1104, a user interface 1103, a memory 1105, and at least one communication bus 1102. The communication bus 1102 can be used to connect and communicate with the various components mentioned above. The user interface 1103 may include buttons, and optionally may include standard wired or wireless interfaces. The network interface 1104 may include, but is not limited to, a Bluetooth module, an NFC module, or a Wi-Fi module. The processor 1101 may include one or more processing cores. The processor 1101 connects to various parts within the electronic device 1100 using various interfaces and lines, and performs various functions of the routing device and processes data by running or executing instructions, programs, code sets, or instruction sets stored in the memory 1105, and by calling data stored in the memory 1105. Optionally, the processor 1101 may be implemented using at least one hardware form of DSP, FPGA, or PLA. The processor 1101 may integrate one or more combinations of CPU, GPU, and modem. The CPU primarily handles the operating system, user interface, and applications; the GPU is responsible for rendering and drawing the content that the display screen needs to show; and the modem is used for wireless communication.

[0113] It is understandable that the aforementioned modem may not be integrated into the processor 1101, but may be implemented using a separate chip.

[0114] The memory 1105 may include RAM or ROM. Optionally, the memory 1105 may include a non-transitory computer-readable medium. The memory 1105 may be used to store instructions, programs, code, code sets, or instruction sets. The memory 1105 may include a program storage area and a data storage area, wherein the program storage area may store instructions for implementing an operating system, instructions for at least one function (such as touch function, sound playback function, image playback function, etc.), instructions for implementing the above-described method embodiments, etc.; the data storage area may store data involved in the above-described method embodiments, etc. Optionally, the memory 1105 may also be at least one storage device located remotely from the aforementioned processor 1101. As a computer storage medium, the memory 1105 may include an operating system, a network communication module, a user interface module, and application programs. The processor 1101 may be used to call the application programs stored in the memory 1105 and execute the methods in the above-described embodiments.

[0115] This specification also provides a computer-readable storage medium storing instructions that, when executed on a computer or processor, cause the computer or processor to perform multiple steps as described in the above embodiments. If the constituent modules of the above-described electronic device are implemented as software functional units and sold or used as independent products, they can be stored in the computer-readable storage medium.

[0116] This specification also provides a computer program product, including a computer program that, when executed by a processor, implements the multiple steps described in the above embodiments.

[0117] Where there is no conflict, the technical features in this embodiment and implementation scheme can be combined arbitrarily.

[0118] In the above embodiments, implementation can be achieved, in whole or in part, through software, hardware, firmware, or any combination thereof. When implemented using software, it can be implemented, in whole or in part, as a computer program product. The computer program product includes multiple computer instructions. When the computer program instructions are loaded and executed on a computer, all or part of the processes or functions described in the embodiments of this specification are generated. The computer can be a general-purpose computer, a special-purpose computer, a computer network, or other programmable device. The computer instructions can be stored in a computer-readable storage medium or transmitted through the computer-readable storage medium. The computer instructions can be transmitted from one website, computer, server, or data center to another website, computer, server, or data center via wired (e.g., coaxial cable, fiber optic, Digital Subscriber Line (DSL)) or wireless (e.g., infrared, wireless, microwave, etc.) means. The computer-readable storage medium can be any available medium accessible to a computer or a data storage device such as a server or data center integrating multiple available media. The available media may be magnetic media (e.g., floppy disks, hard disks, magnetic tapes), optical media (e.g., digital versatile discs (DVDs)), or semiconductor media (e.g., solid-state drives (SSDs)).

[0119] When implemented through hardware or firmware, the aforementioned method flow is programmed into the hardware circuit to obtain the corresponding hardware circuit structure and achieve the corresponding function. For example, a Programmable Logic Device (PLD) (such as a Field Programmable Gate Array (FPGA)) is such an integrated circuit, whose logic function is determined by the user programming the device. Designers can program a digital system onto a PLD themselves, eliminating the need for chip manufacturers to design and fabricate dedicated integrated circuit chips. Furthermore, nowadays, instead of manually fabricating integrated circuit chips, this programming is mostly implemented using "logic compiler" software, similar to the software compiler used in program development. The original code before compilation must also be written in a specific programming language, called a Hardware Description Language (HDL). There is not just one HDL, but many. Those skilled in the art should understand that by simply performing some logic programming on the method flow using one of the aforementioned hardware description languages ​​and programming it into an integrated circuit, the hardware circuit implementing the logic method flow can be easily obtained.

[0120] The embodiments described above are merely preferred embodiments of this specification and are not intended to limit the scope of this specification. Any modifications and improvements made by those skilled in the art to the technical solutions of this specification without departing from the spirit of this specification should fall within the protection scope defined by the claims of this specification.

Claims

1. An intelligent anomaly detection system for leaky cables, characterized in that, include: The acquisition module acquires signal strength data and corresponding location information reported by multiple dedicated terminals along the target leaky cable during its historical normal operation. The construction module, based on the historical signal strength data and location information, constructs a probability distribution model of the terminal signal strength at each geographical location point, and determines the normal signal strength reference interval corresponding to each location point; Select the module to generate the terminal's location distribution probability, and select several locations whose location distribution probability is higher than the preset reference probability value as reference locations; The acquisition module collects signal strength data reported by multiple terminals in real time during the current detection period, forming a current signal strength observation dataset. The association module associates the actual signal strength in the current signal strength observation dataset with a reference location and the location distribution probability of the reference location to form association data. The association data includes the actual signal strength, the reference location, and the corresponding location distribution probability. The comparison module compares the actual signal strength with the normal signal strength reference range at the reference location, calculates the deviation, and calculates the confidence level based on the corresponding location distribution probability. The calculation module uses the confidence level as the weight to calculate the weighted average deviation of the current signal strength observation dataset; The judgment module determines that there is an abnormality in the leaky cable when the weighted average deviation exceeds a preset deviation threshold and issues an abnormality alarm.

2. A method for intelligent anomaly detection of leaky cables, characterized in that, Includes the following steps: Acquire signal strength data and corresponding location information reported by multiple dedicated terminals along the target leaky cable during its historical normal operation. Based on the historical signal strength data and location information, a probability distribution model of the terminal signal strength at each geographical location is constructed, and a reference interval for normal signal strength corresponding to each location is determined. Generate the location distribution probability of the terminal, and select several locations whose location distribution probability is higher than the preset reference probability value as reference locations; During the current detection period, the signal strength reported by multiple terminals is collected in real time to form the current signal strength observation dataset. The actual signal strength in the current signal strength observation dataset is associated with a reference location and the location distribution probability of the reference location to form associated data. The associated data includes the actual signal strength, the reference location and the corresponding location distribution probability. The actual signal strength is compared with the normal signal strength reference range at the reference location to calculate the deviation, and the confidence level is calculated based on the corresponding location distribution probability. Using the confidence level as the weight, the weighted average deviation of the current signal strength observation dataset is calculated; When the weighted average deviation exceeds a preset deviation threshold, the leaky cable is determined to be abnormal, and an abnormality alarm is issued.

3. The intelligent anomaly detection method for leaky cables according to claim 2, characterized in that, The method for constructing a probability distribution model of terminal signal strength at each geographic location point based on the historical signal strength data and location information includes: Historical signal strength data is aggregated in a grid according to geographical location to obtain several discrete observation points; For each observation point, collect the signal strength values ​​reported by all historical terminals within the grid where the observation point is located, forming a signal strength sample set for the observation point; Statistical analysis is performed on the signal strength sample set to obtain a probability density function. Based on the grid and probability density function, a probability distribution model is obtained.

4. The intelligent anomaly detection method for leaky cables according to claim 3, characterized in that, Methods for determining the reference range of normal signal strength for each location point include: Based on the probability density function, determine the range of signal strength values ​​that cover the preset confidence level; The normal signal strength reference range is obtained based on the signal strength value range.

5. The intelligent anomaly detection method for leaky cables according to claim 3, characterized in that, Methods for selecting several locations whose location distribution probability is higher than a preset reference probability value as reference locations include: Based on the location information reported by the historical terminals, the number of times the terminal visited each of the observation points in the grid is counted, which is taken as the number of visits to the observation points. The number of visits to each observation point is normalized to obtain the probability of the terminal appearing at each observation point, which is used as the location distribution probability. Based on a preset reference probability value, points with a location distribution probability greater than or equal to the reference probability value are selected as reference locations.

6. The intelligent anomaly detection method for leaky cables according to claim 2, characterized in that, The method of comparing the actual signal strength with the normal signal strength reference range at the reference location, calculating the deviation, and calculating the confidence level based on the corresponding location distribution probability includes: When the actual signal strength does not exceed the normal signal strength reference range, the deviation is 0; When the actual signal strength exceeds the normal signal strength reference range, calculate the absolute value of the difference between the actual signal strength and the normal signal strength reference range, and calculate the ratio of the absolute value to the actual signal strength. The ratio is used as the deviation. The probability distribution of the reference location is used as the confidence level.

7. The intelligent anomaly detection method for leaky cables according to claim 2, characterized in that, The method for calculating the weighted average deviation of the current signal strength observation dataset using the confidence level as a weight includes: Using the confidence level as the weight, calculate the weighted sum of the current signal strength observation dataset; Calculate the sum of all confidence scores in the current signal strength observation dataset, and use this as the total confidence score; The weighted average deviation is obtained by dividing the weighted sum by the total confidence level.

8. An electronic device, characterized in that, Including the processor and memory; The processor is connected to the memory; The memory is used to store executable program code; The processor runs a program corresponding to the executable program code stored in the memory to perform the method as described in any one of claims 2-7.

9. A computer-readable storage medium having a computer program stored thereon, characterized in that, When the computer program is executed by a processor, it implements the method as described in any one of claims 2-7.

10. A computer program product, comprising a computer program, characterized in that, When the computer program is executed by the processor, it implements the method as described in any one of claims 2-7.

Images