An RFID-based abnormal detection method and device for dumb resources

Abstract

The present application provides a method and device for abnormal detection of dumb resources based on RFID, which relates to the technical field of abnormal detection data processing. The method includes: scanning the RFID tag group of dumb resources to obtain an optical cable perception data stream, performing filtering and alignment, and establishing multiple node optical cable perception matrices, where the dumb resources include target optical cables; performing security mode deviation detection on the target optical cable based on optical cable multi-modal factors; predicting the spread risk of the target optical cable; performing communication loss detection on the first abnormal optical cable node and the second abnormal optical cable node to obtain the first node communication loss feature and the second node communication loss feature; performing abnormal correlation tracing on the first abnormal optical cable node and the second abnormal optical cable node to generate a dumb resource abnormal detection report. Through the present application, the technical objective of intelligent optical cable monitoring based on multi-modal factor fusion analysis can be achieved, and the technical effect of improving the accuracy of abnormal detection can be achieved.

Description

Technical Field

[0001] This application relates to the technical field of abnormal detection data processing, and particularly to a method and device for abnormal detection of dumb resources based on RFID. Background Art

[0002] As an important infrastructure of modern communication networks, the optical cable in dumb resources has its security and stability directly affecting the quality of information transmission. However, under complex environmental conditions, the optical cable is vulnerable to various factors, such as mechanical damage, environmental changes, electromagnetic interference, etc., resulting in signal attenuation, transmission distortion, or even communication interruption.

[0003] Currently, some existing optical cable monitoring solutions use sensors such as temperature, humidity, and strain to obtain the environmental and physical state information of the optical cable and combine big data analysis for fault prediction. However, these methods still have many deficiencies in practical applications. First, the data of a single sensor cannot comprehensively reflect the health status of the optical cable. For example, it is difficult to accurately judge the mechanical stress or vibration state of the optical cable only relying on temperature or humidity data. Second, some monitoring systems do not fully consider the mutual influence between different environmental factors during data processing, resulting in low reliability of the detection results.

[0004] In summary, there are technical problems in the prior art that due to the limitations of single-sensor data and insufficient analysis of data correlation, the evaluation of the health status of the optical cable is not comprehensive, the accuracy of abnormal detection is low, further affecting the reliability of optical cable fault prediction and the efficiency of operation and maintenance management. Summary of the Invention

[0005] The purpose of this application is to provide a method and device for abnormal detection of dumb resources based on RFID to solve the technical problems in the prior art that due to the limitations of single-sensor data and insufficient analysis of data correlation, the evaluation of the health status of the optical cable is not comprehensive, the accuracy of abnormal detection is low, further affecting the reliability of optical cable fault prediction and the efficiency of operation and maintenance management.

[0006] In view of the above problems, this application provides a method and device for abnormal detection of dumb resources based on RFID.

[0007] In a first aspect, the present application provides a method for detecting anomalies in dumb resources based on RFID, which is implemented through a device for detecting anomalies in dumb resources based on RFID, and includes: scanning the RFID tag group of the dumb resources by an RFID reader to obtain an optical cable perception data stream, and filtering and aligning the optical cable perception data stream according to optical cable multimodal factors to establish multiple node optical cable perception matrices, where the dumb resources include target optical cables; based on the optical cable multimodal factors, performing a security mode deviation detection on the target optical cable according to the multiple node optical cable perception matrices to determine a first abnormal optical cable node; predicting the spread risk of the target optical cable according to the first abnormal optical cable node to determine a second abnormal optical cable node; performing a communication loss detection on the first abnormal optical cable node and the second abnormal optical cable node to obtain a first node communication loss feature and a second node communication loss feature; and performing an abnormal correlation traceability on the first abnormal optical cable node and the second abnormal optical cable node according to the first node communication loss feature and the second node communication loss feature to generate a dumb resource anomaly detection report.

[0008] In a second aspect, the present application further provides a device for detecting anomalies in dumb resources based on RFID, which is used to execute a method for detecting anomalies in dumb resources based on RFID as described in the first aspect, and includes: a dumb resource scanning module, which is used to scan the RFID tag group of the dumb resources by an RFID reader to obtain an optical cable perception data stream, and filter and align the optical cable perception data stream according to optical cable multimodal factors to establish multiple node optical cable perception matrices, where the dumb resources include target optical cables; a deviation detection module, which is used to perform a security mode deviation detection on the target optical cable according to the multiple node optical cable perception matrices based on the optical cable multimodal factors to determine a first abnormal optical cable node; a spread risk prediction module, which is used to predict the spread risk of the target optical cable according to the first abnormal optical cable node to determine a second abnormal optical cable node; a communication loss detection module, which is used to perform a communication loss detection on the first abnormal optical cable node and the second abnormal optical cable node to obtain a first node communication loss feature and a second node communication loss feature; and an abnormal correlation traceability module, which is used to perform an abnormal correlation traceability on the first abnormal optical cable node and the second abnormal optical cable node according to the first node communication loss feature and the second node communication loss feature to generate a dumb resource anomaly detection report.

[0009] The technical solution provided in the present application has at least the following technical effects or advantages: by achieving the technical goal of intelligent optical cable monitoring based on multimodal factor fusion analysis, the technical effects of comprehensive evaluation of the health status of optical cables, improvement of the accuracy of anomaly detection, enhancement of the reliability of fault prediction, and intelligent upgrade of operation and maintenance management are achieved.

[0010] The above description is only an overview of the technical solution of the present application. In order to better understand the technical means of the present application, it can be implemented according to the content of the specification. In order to make the above and other purposes, features and advantages of the present application more obvious and understandable, the specific embodiments of the present application are specifically given below. It should be understood that the content described in this part is not intended to identify the key or important features of the embodiments of the present application, nor is it used to limit the scope of the present application. Other features of the present application will become easily understood through the following description. BRIEF DESCRIPTION OF THE DRAWINGS

[0011] In order to more clearly illustrate the technical solutions in the present application or the prior art, the following will briefly introduce the drawings required for the description of the embodiments or the prior art. Obviously, the drawings in the following description are only exemplary, and for those of ordinary skill in the art, other drawings can be obtained according to the provided drawings without creative efforts.

[0012] Figure 1 It is a schematic flowchart of a method for detecting abnormal RFID-based dumb resources in the present application;

[0013] Figure 2 It is a schematic structural diagram of a device for detecting abnormal RFID-based dumb resources in the present application.

[0014] Description of reference numerals: dumb resource scanning module 11, deviation detection module 12, affected risk prediction module 13, communication loss detection module 14, abnormal correlation tracing module 15. DETAILED DESCRIPTION OF THE EMBODIMENTS

[0015] By providing a method and device for detecting abnormal RFID-based dumb resources, the present application solves the technical problems in the prior art that due to the limitations of single-sensor data and insufficient analysis of data correlation, the evaluation of the health status of optical cables is not comprehensive, the accuracy of abnormal detection is low, and further affects the reliability of optical cable fault prediction and the efficiency of operation and maintenance management. The technical goal of intelligent optical cable monitoring based on multi-modal factor fusion analysis is achieved, and the technical effects of comprehensive evaluation of the health status of optical cables, improvement of abnormal detection accuracy, enhancement of the reliability of fault prediction, and intelligent upgrade of operation and maintenance management are achieved.

[0016] Next, the technical solutions in the present application will be clearly and completely described with reference to the accompanying drawings. Obviously, the described embodiments are only a part of the embodiments of the present application, rather than all the embodiments of the present application. It should be understood that the present application is not limited by the example embodiments described herein. All other embodiments obtained by those of ordinary skill in the art based on the embodiments of the present application without creative efforts belong to the scope of protection of the present application. Additionally, it should be noted that for the sake of description, only the parts related to the present application are shown in the accompanying drawings rather than all of them.

[0017] Embodiment 1. Please refer to the attached Figure 1 , the present application provides a method for detecting anomalies in dumb resources based on RFID, which is applied to a device for detecting anomalies in dumb resources based on RFID, and specifically includes the following steps:

[0018] S1: According to the RFID reader scanning the RFID tag group of the dumb resources, obtain the optical cable perception data stream, and filter and align the optical cable perception data stream according to the optical cable multimodal factors to establish multiple node optical cable perception matrices, where the dumb resources include the target optical cable.

[0019] Further, S1 includes: S11: Perform feature recognition on the optical cable perception data stream according to the optical cable multimodal factors to obtain multiple node optical cable feature data; S12: Perform Kalman filtering on the multiple node optical cable feature data to obtain multiple node optical cable filtered data; S13: Perform time synchronization alignment on the multiple node optical cable filtered data to obtain multiple node optical cable aligned data; S14: Perform matrix processing on the multiple node optical cable aligned data to obtain the multiple node optical cable perception matrices.

[0020] Further, the optical cable multimodal factors include optical cable vibration, optical cable curvature, optical cable stress, optical cable strain, optical cable temperature, optical cable humidity, and optical cable environment.

[0021] Specifically, an RFID reader is a device used for wireless communication with RFID tags. It can emit electromagnetic waves of a specific frequency and interact with RFID tags for data. When the RFID reader approaches the RFID tag, by emitting an electromagnetic signal, it activates the passive RFID tag or wakes up the active RFID tag to make it return the stored data.

[0022] Among them, the dumb resources include the target optical cable. The target optical cable is the resource to be detected for anomalies. The target optical cable includes multiple optical cable nodes, and each optical cable node corresponds to an RFID tag.

[0023] Radio signals are sent to the tags by the RFID reader and the returned information is received. The RFID reader can identify multiple RFID tags in batches and transmit them to the back-end management system for processing. Since the scanning is usually non-contact, the efficiency of data collection can be improved and the need for manual intervention can be reduced. Dumb resources refer to physical facilities such as optical cables or fiber optic jumpers, which do not have the ability to actively send or process information by themselves, but can be intelligently managed through external means (such as RFID tags). The characteristics of dumb resources are static existence, and usually rely on additional technical means for monitoring. For example, an optical cable itself cannot directly feedback its status information, but if an RFID tag is attached to the optical cable, the remote acquisition of information such as its location, type, and environmental status can be achieved. An RFID tag group refers to a collection of multiple RFID tags, which may be distributed at different positions or key nodes of the optical cable to ensure the integrity and accuracy of data. The optical cable sensing data stream refers to the optical cable status information obtained through RFID tags and is transmitted and processed in the form of a data stream. The sensed data usually includes the identification information, location information, type of the optical cable, and environmental parameters (such as temperature, humidity, etc.) obtained through integrated sensors.

[0024] The optical cable multi-modal factors refer to multiple different types of factors that affect the optical cable sensing data stream. Among the optical cable multi-modal factors, optical cable vibration refers to the vibration generated when the optical cable is affected by external forces or environmental factors, which may affect the transmission performance of the optical cable. Vibration will cause signal attenuation, distortion, and even damage to the optical cable, so its impact needs to be monitored and analyzed. The optical cable bend reflects the degree of bending of the optical cable during installation or use. The bending of the optical cable will affect the quality of signal transmission. Especially in optical fiber communication, excessive bending may cause signal loss or breakage. The optical cable stress is the internal resistance generated when the optical cable is under stress. Excessive stress will cause the optical cable to deform or even break, affecting the stability of the communication system. Stress analysis helps to predict possible failures of the optical cable during use. The optical cable strain refers to the deformation of the optical cable material under stress. Strain analysis can help evaluate the damaged degree and service life of the optical cable. Long-term strain accumulation may lead to the decline of the optical cable performance. The optical cable temperature refers to the temperature change of the environment where the optical cable is located. Too high or too low temperature will have an adverse impact on the physical performance of the optical cable. Temperature change will cause the expansion or contraction of the optical cable material, thus affecting the use effect of the optical cable. The optical cable humidity refers to the humidity in the environment where the optical cable is located. Too high humidity may cause corrosion or other damages inside the optical cable, affecting its transmission performance. The optical cable environment refers to the external environmental conditions where the optical cable is located, including but not limited to factors such as temperature, humidity, and electromagnetic interference. Environmental conditions have an important impact on the long-term use of the optical cable. Harsh environments may accelerate the aging and damage of the optical cable.

[0025] Due to possible problems such as noise and inconsistent timing during the data acquisition process, it is necessary to filter and align the optical cable sensing data stream to improve the reliability and availability of the optical cable sensing data stream. Feature recognition refers to extracting important feature data that can represent the state of the optical cable from the optical cable sensing data stream. By analyzing the multimodal factors of the optical cable, the optical cable feature data of multiple nodes can be identified, such as the temperature change trend at different positions, humidity level, electromagnetic interference intensity in the optical cable environment, etc. Redundant information in the original data is removed, and only the features valuable for judging the health state of the optical cable are retained to obtain the optical cable feature data of multiple nodes.

[0026] Kalman filtering is an algorithm for dynamic data optimization that can predict and correct the optimal state when the data is disturbed by noise. Performing Kalman filtering on the optical cable feature data of multiple nodes can reduce data fluctuations, improve data smoothness, and obtain the optical cable filtered data of multiple nodes. For example, if the optical cable temperature data of a certain node changes violently in a short period of time while the temperatures of other nodes remain stable, Kalman filtering can determine that the mutant data may be noise and correct it to make the data smoother and more reasonable.

[0027] Since there may be acquisition time delays in the optical cable filtered data of different nodes, direct comparison may lead to errors. Therefore, it is necessary to perform time alignment on the optical cable filtered data. Time synchronization alignment refers to adjusting the time scale of the optical cable filtered data of multiple nodes to ensure that the optical cable filtered data of all nodes is compared and analyzed on the same time basis, obtaining the optical cable aligned data of multiple nodes.

[0028] Matrix processing refers to organizing the optical cable aligned data of multiple nodes into a matrix form to obtain the optical cable sensing matrix of multiple nodes for subsequent calculation and analysis. Each row or column of the optical cable sensing matrix can represent the feature data of different nodes, such as temperature, humidity, electromagnetic interference, etc., making data processing more intuitive. For example, if there are ten optical cable nodes and each node collects at least three types of data, namely temperature, humidity, and electromagnetic interference intensity, the final optical cable sensing matrix is convenient for subsequent fault detection or trend prediction through machine learning or statistical analysis methods. The numerical records of the optical cable sensing matrix of 10 nodes in the most recent processed data record are shown in Table 1.

[0029]

[0030] Table 1: Numerical records of the optical cable sensing matrix of 10 nodes in the most recent processed data record

[0031]

[0032] S2: Based on the multimodal factors of the optical cable, perform security mode deviation detection on the target optical cable according to the multiple node optical cable sensing matrices, and determine the first abnormal optical cable node.

[0033] Further, S2 includes: S21: Perform security mode mining on multiple optical cable nodes of the target optical cable according to the multimodal factors of the optical cable, and establish multiple node security mode matrices; S22: Perform anomaly detection on the multiple node optical cable sensing matrices according to the multiple node security mode matrices to obtain multiple node optical cable anomaly detection results; S23: Perform security mode deviation risk assessment according to the multiple node optical cable anomaly detection results to obtain multiple node optical cable risk coefficients; S24: If any one of the multiple node optical cable risk coefficients is greater than or equal to the optical cable risk threshold, obtain the marked node optical cable risk coefficient, and record the optical cable node corresponding to the marked node optical cable risk coefficient as the first abnormal optical cable node.

[0034] Further, S21 includes: S211: Retrieve the normal monitoring records of each optical cable node according to the multimodal factors of the optical cable to obtain the normal characteristic record sets of each node optical cable; S212: Perform confidence evaluation on each parameter in the normal characteristic record sets of each node optical cable to obtain the confidence evaluation sets of each node parameter; S213: Based on the confidence evaluation sets of each node parameter, perform confidence parameter selection on the normal characteristic record sets of each node optical cable according to the predetermined confidence constraint to obtain the security characteristic sets of each node optical cable; S214: Perform central tendency analysis according to the security characteristic sets of each node optical cable to obtain the multiple node security mode matrices.

[0035] Specifically, the multiple optical cable nodes of the target optical cable refer to the detection points at multiple key positions along its laying path, which can be optical cable joints, branch points, or other important positions that need to be monitored. A sensing device, such as an RFID tag, a sensor, etc., can be installed at each optical cable node to monitor the status of the optical cable in real time. All optical cable nodes together constitute a distributed monitoring system, enabling the status of the optical cable to be compared and analyzed at different spatial positions.

[0036] Security mode mining refers to using data analysis and pattern recognition technologies to extract the patterns or characteristics related to safe operation from the multimodal factor data of the optical cable, identify the data patterns of the optical cable in the normal operation state, distinguish the possible abnormal situations, and establish multiple node security mode matrices.

[0037] Extract the normal operation records of each node with optical cable multimodal factors as indicators from historical monitoring data for subsequent feature analysis. Since the state of the optical cable may fluctuate with time and environment changes, it is necessary to establish a database of monitoring normal records to ensure that a stable and reliable data set can be used for subsequent analysis. For example, within the past month, the temperature, humidity, and interference signal strength data of each optical cable node every day are stored, forming the basis of the normal feature record set of each node's optical cable.

[0038] Confidence evaluation refers to analyzing the credibility of each parameter in the normal feature record set of each node's optical cable to evaluate the stability and reliability of different parameters and obtain the confidence evaluation set of each node's parameters. The confidence evaluation set of each node's parameters refers to the set that calculates and stores the confidence information for all monitoring parameters of each optical cable node, which is used to screen out feature parameters with higher credibility to improve the reliability of subsequent analysis. Confidence is usually calculated based on statistical methods. For example, the dispersion degree of data is evaluated through the mean and standard deviation, or the Bayesian inference method is used to calculate the credibility of parameters. For example, if the humidity data of a certain node varies within a small range over a period of time, its confidence is high; on the contrary, if the electromagnetic interference signal of a certain node fluctuates greatly, its confidence may be low.

[0039] For example, perform confidence evaluation on 3 parameters (temperature, humidity, electromagnetic interference) in the normal feature record set of 10-node optical cables. Confidence can be calculated by the mean ± confidence interval. The confidence interval is based on the mean and standard deviation of the data and determines the confidence range through the normal distribution confidence level. Calculate the confidence interval at a 95% confidence level (i.e., Z = 1.96), and the formula is: . Among them, is the confidence interval, is the mean, is the standard deviation, n is the number of parameters, and Z = 1.96 corresponds to a 95% confidence level. The data record of the most recent confidence evaluation is shown in Table 2.

[0040] Table 2: Data record of the most recent confidence evaluation

[0041]

[0042] Analysis of the temperature parameter shows that the confidence interval is small, which may indicate that the temperature parameter is stable and the temperature fluctuation range is acceptable. Analysis of the humidity parameter shows that the confidence interval of most humidity is about ±0.7%, which may indicate that the data is generally stable. Analysis of the electromagnetic interference parameter shows that the confidence intervals are generally small, which may indicate that the interference changes little and the data is credible. The specific situation is analyzed by those skilled in the art according to the actual situation.

[0043] The final confidence score calculation can define the confidence score using the width of the confidence interval: Confidence = , if the maximum allowable temperature fluctuation range is set to 2.0 °C, and the temperature confidence interval for a certain node is ±0.31 °C, then the temperature confidence of this node: , similarly, if the maximum allowable humidity fluctuation range is 5%, and the humidity confidence interval is ±0.74%, then the humidity confidence: .

[0044] Confidence parameter selection refers to screening out a set of parameters with relatively high credibility based on the confidence evaluation set of each node's parameters according to the preset confidence constraints, in order to construct the optical cable safety feature set of each node, which can represent the long-term stable state of the optical cable. The confidence constraint can be set according to engineering requirements. For example, only parameters with a confidence above 0.7 are selected into the safety feature set.

[0045] Central tendency analysis refers to statistically analyzing the optical cable safety feature set of each node to find the distribution trend of the overall data, including calculating statistical indicators such as mean, median, and variance to judge the degree of data concentration. Organize the safety feature data of all optical cable nodes in the form of a matrix to obtain multiple node safety mode matrices for visual analysis and intelligent decision-making. For example, if the average temperature of all optical cable nodes is 26 °C (Celsius) and the variance is small, it indicates that the overall temperature state is relatively stable; but if the variance is large, it may mean that there are abnormal conditions in some nodes and further monitoring is required.

[0046] Multiple node safety mode matrices are matrices established based on the historical normal data of the optical cable, which contain the safety modes of each optical cable node, including the normal ranges of parameters such as temperature, humidity, and electromagnetic interference. The multiple node optical cable sensing matrix is an optical cable data matrix obtained through real-time monitoring, which reflects the current state of each optical cable node. Anomaly detection refers to comparing real-time data with safety mode data to judge whether the current data deviates from the normal range. If the temperature of a certain node exceeds the normal range, or the electromagnetic interference signal shows abnormal fluctuations, it is considered that there may be an anomaly at this node. Through comparative analysis, abnormal conditions of multiple optical cable nodes can be identified, and finally, multiple node optical cable anomaly detection results can be obtained.

[0047] When the multiple node optical cable anomaly detection results indicate that the data of some optical cable nodes deviate from the normal range, it is necessary to conduct a safety mode deviation risk assessment, that is, calculate the severity of the abnormal deviation to obtain the multiple node optical cable risk coefficients, which are used to quantify the abnormal degree of each optical cable node. The calculation of the risk coefficient can be based on statistical methods, such as standard deviation calculation or machine learning-based methods, to quantify the degree of deviation.

[0048] Preset an optical cable risk threshold, which is a standard for distinguishing normal nodes and abnormal nodes. For example, if the optical cable risk threshold is set to 0.7, then any optical cable node with a risk coefficient greater than or equal to 0.7 will be considered an abnormal node. The risk coefficient of the abnormal node will be marked and regarded as the key object of concern. If multiple optical cable nodes exceed the optical cable risk threshold, they are marked as the first abnormal optical cable nodes and require further diagnosis and repair to prevent more serious failures in the optical cable system.

[0049] S3: Predict the spread risk of the target optical cable based on the first abnormal optical cable node, and determine the second abnormal optical cable node.

[0050] Furthermore, S3 includes: S31: Record the node optical cable anomaly detection result corresponding to the first abnormal optical cable node as the first optical cable anomaly detection result; S32: Perform abnormal propagation prediction on the first optical cable anomaly detection result according to the ARIMA-LSTM combined model to obtain multiple node spread risk coefficients; S33: Determine whether the multiple node spread risk coefficients are greater than or equal to the node spread risk threshold; S34: If any one of the multiple node spread risk coefficients is greater than or equal to the node spread risk threshold, obtain the marked node spread risk coefficient, and record the optical cable node corresponding to the marked node spread risk coefficient as the second abnormal optical cable node.

[0051] Furthermore, S32 includes: S321: Retrieve the optical cable abnormal propagation event set corresponding to the first abnormal optical cable node; S322: Perform short-term abnormal propagation feature learning on the optical cable abnormal propagation event set according to the ARIMA model in the ARIMA-LSTM combined model to obtain the first optical cable abnormal propagation prediction model; S323: Perform long-term abnormal propagation feature learning on the optical cable abnormal propagation event set according to the LSTM model in the ARIMA-LSTM combined model to obtain the second optical cable abnormal propagation prediction model; S324: Perform fusion learning on the first optical cable abnormal propagation prediction model and the second optical cable abnormal propagation prediction model according to the optical cable abnormal propagation event set to obtain the third optical cable abnormal propagation prediction model; S325: Input the first optical cable anomaly detection result into the third optical cable abnormal propagation prediction model to obtain the multiple node spread risk coefficients.

[0052] Specifically, when an abnormal node is detected, it is necessary to record the abnormal situation of the abnormal node for further analysis and processing. The anomaly detection result of the first abnormal optical cable node, that is, the abnormal situation identified during the anomaly detection process, such as too high temperature, abnormal humidity, excessive electromagnetic interference, etc., needs to be stored and recorded to form the first optical cable anomaly detection result. The first optical cable anomaly detection result is a data set that contains detailed abnormal information of the first abnormal optical cable node.

[0053] In an optical cable network, an anomaly in a certain node may affect other surrounding nodes and even trigger failures on a larger scale. Therefore, it is necessary to predict the propagation of anomalies in order to take preventive measures in advance. The ARIMA-LSTM combined model combines time series analysis and deep learning methods, which can handle short-term trends and long-term dependencies simultaneously to improve the accuracy of anomaly propagation prediction.

[0054] The optical cable anomaly propagation event set refers to the optical cable anomalies corresponding to the first anomalous optical cable node that occurred in history and their propagation conditions, including the starting position, propagation path, propagation time, and affected scope of the anomaly. For example, if an anomaly starts from node 3 and affects nodes 5 and 6 within five minutes, it will be stored in the optical cable anomaly propagation event set.

[0055] The ARIMA (AutoRegressive Integrated Moving Average) model is a statistical method for time series prediction, suitable for short-term prediction, and can identify trends, periodic changes, and short-term fluctuations in time series. In the prediction of optical cable anomaly propagation, the ARIMA model is used to learn the propagation characteristics of optical cable anomalies in a short period, such as whether a certain anomaly will quickly affect surrounding nodes within a few minutes or hours. By learning from the anomaly propagation event set, the ARIMA model can generate the first optical cable anomaly propagation prediction model, which can predict the short-term diffusion trend of anomalies.

[0056] The LSTM (Long Short-Term Memory) is a deep learning model for processing long-term time series data, which is good at capturing long-term dependencies. In the prediction of optical cable anomaly propagation, the LSTM model is used to analyze the propagation characteristics of anomalies in a longer time range, such as whether a certain anomaly will affect more distant optical cable nodes after several hours, days, or longer. By learning from the anomaly propagation event set, the LSTM model can generate the second optical cable anomaly propagation prediction model, which can predict the long-term anomaly diffusion trend.

[0057] Since the ARIMA model is good at short-term prediction and the LSTM model is good at long-term prediction, the prediction results of the two models can be fused to improve the overall prediction accuracy. Fusion learning can combine the advantages of multiple models, making the final prediction result more reliable. In the prediction of optical cable anomaly propagation, fusion learning will perform weighted combination or non-linear fusion on the results of the first optical cable anomaly propagation prediction model and the second optical cable anomaly propagation prediction model based on the anomaly propagation event set, so as to obtain the third optical cable anomaly propagation prediction model, which can take into account both short-term and long-term anomaly propagation characteristics and improve the comprehensiveness and accuracy of the prediction.

[0058] Finally, the first result of the optical cable anomaly detection is input into the third model of the optical cable anomaly propagation prediction to predict how the anomaly propagates to other nodes and calculate the influence risk coefficients of multiple nodes. The influence risk coefficient is a value that measures the possibility and severity of a certain node being affected by the anomaly. For example, if the influence risk coefficient of a certain optical cable node is 0.8, it indicates that the node has a high possibility of being affected by the anomaly and may cause system failures, while the influence risk coefficient of another node is 0.2, indicating a lower degree of being affected.

[0059] During the process of optical cable anomaly propagation prediction, the influence risk coefficient of each optical cable node represents the possibility and severity of the node being affected by the anomaly. To distinguish normal nodes from high-risk nodes, a node influence risk threshold is preset in advance to define the safe range of optical cable nodes. For example, assume the threshold is set to 0.6, then all nodes with influence risk coefficients greater than or equal to 0.6 may be in a dangerous state and need special attention. Check the influence risk coefficients of all nodes one by one and determine whether there are nodes whose influence risk coefficients exceed or equal this threshold.

[0060] When it is detected that the influence risk coefficients of one or more optical cable nodes exceed the node influence risk threshold, these nodes will be regarded as the nodes most severely affected by the anomaly, and the influence risk coefficients of these nodes are obtained as the marked node influence risk coefficients, which are used to quantify the severity of the anomaly influence. At this time, these high-risk nodes are marked as the second abnormal optical cable nodes.

[0061] S4: Conduct communication loss detection on the first abnormal optical cable nodes and the second abnormal optical cable nodes to obtain the first node communication loss characteristics and the second node communication loss characteristics.

[0062] Furthermore, S4 includes: S41: Conduct transmission tests on the first abnormal optical cable nodes according to multiple predetermined test signals to obtain multiple signal transmission monitoring data; S42: Input the multiple signal transmission monitoring data into the optical cable communication performance evaluation model to obtain multiple communication performance evaluation results, where the optical cable communication performance evaluation model includes multi-dimensional indicators of optical cable communication performance, and the multi-dimensional indicators of optical cable communication performance include transmission attenuation degree, transmission distortion degree, and transmission delay degree; S43: Calculate the central value of each dimension for the multiple communication performance evaluation results based on the multi-dimensional indicators of optical cable communication performance to obtain the first node communication performance evaluation result; S44: Obtain the expected communication performance evaluation result corresponding to the first abnormal optical cable node, and calculate the loss of the first node communication performance evaluation result according to the expected communication performance evaluation result to obtain the first node communication loss characteristics.

[0063] Specifically, communication loss detection is a process of evaluating the transmission performance of optical cable nodes to understand the actual communication situation. The communication performance of abnormal nodes may be affected by faults or external interferences, resulting in a decline in signal transmission quality. By conducting communication loss detection, it can be determined whether there are obvious performance degradations in these nodes, and then the overall health status of the optical cable network can be judged.

[0064] To detect the communication loss of the first abnormal optical cable node, multiple predetermined test signals are used for transmission tests, including signals with different frequencies, different intensities, and different waveforms, which are used to simulate various situations in actual communication. By conducting tests on the first abnormal optical cable node, multiple signal transmission monitoring data can be obtained, which record characteristics such as attenuation and distortion during the signal transmission process and reflect the transmission quality of the optical cable node.

[0065] Next, the optical cable communication performance evaluation model considers multiple dimensions of indicators, including transmission attenuation degree, transmission distortion degree, and transmission delay degree. The transmission attenuation degree reflects the degree of signal loss during transmission; the transmission distortion degree measures the change or distortion of the signal waveform; the transmission delay degree describes the time required for the signal to travel from the sending end to the receiving end. Inputting multiple signal transmission monitoring data into the optical cable communication performance evaluation model, through the analysis of the monitoring data, multiple communication performance evaluation results are obtained.

[0066] By processing multiple communication performance evaluation results, the central values of each dimension can be obtained. For example, the central value of the transmission attenuation degree may be the average attenuation degree of all test signals, and the transmission distortion degree and delay degree are also calculated respectively according to the test results to obtain the central value of the transmission attenuation degree, the central value of the transmission distortion degree, and the central value of the transmission delay degree. The central values of each dimension can provide a general evaluation result, helping to understand the overall communication performance of the first node, obtaining the communication performance evaluation result of the first node, and providing a basis for subsequent loss calculation.

[0067] The expected communication performance evaluation result is an expected result obtained based on the theoretical value or historical data of the first abnormal optical cable node in the normal working state, including the expected transmission attenuation degree, the expected transmission distortion degree, and the expected transmission delay degree. For example, under normal circumstances, the transmission attenuation degree of the optical cable node may be within a certain range, and the expected communication performance evaluation result represents the normal value of this range. By comparing the difference between the actual communication performance evaluation result of the first node and the expected value, the communication loss can be calculated, and then the loss degree of the communication performance can be identified, and the basis for subsequent fault troubleshooting and repair can be provided according to the loss characteristics.

[0068] According to the method of calculating the communication loss characteristics of the first node, the communication loss detection of the second abnormal optical cable node is carried out, including transmission test, input of the optical cable communication performance evaluation model, calculation of the central value of each dimension, and loss calculation, to obtain the communication loss characteristics of the second node.

[0069] S5: According to the communication loss characteristics of the first node and the communication loss characteristics of the second node, perform abnormal correlation tracing on the first abnormal optical cable node and the second abnormal optical cable node, and generate a dummy resource abnormal detection report.

[0070] Further, S5 includes: S51: Perform correlation analysis on the communication loss characteristics of the first node according to the first result of optical cable abnormal detection to obtain the first correlation factor of optical cable abnormality; S52: Perform correlation analysis on the communication loss characteristics of the second node according to the first result of optical cable abnormal detection to obtain the second correlation factor of optical cable abnormality; S53: Denote the node optical cable abnormal detection result pair corresponding to the second abnormal optical cable node as the second result of optical cable abnormal detection; S54: Perform correlation analysis on the communication loss characteristics of the second node according to the second result of optical cable abnormal detection to obtain the third correlation factor of optical cable abnormality; S55: Organize the first result of optical cable abnormal detection, the second result of optical cable abnormal detection, the communication loss characteristics of the first node, the communication loss characteristics of the second node, the first correlation factor of optical cable abnormality, the second correlation factor of optical cable abnormality, and the third correlation factor of optical cable abnormality to obtain the dummy resource abnormal detection report.

[0071] Specifically, during the abnormal correlation tracing process, first analyze the communication loss characteristics of the first abnormal optical cable node. Denote the node optical cable abnormal detection result corresponding to the first abnormal optical cable node as the first result of optical cable abnormal detection. According to the first result of optical cable abnormal detection, evaluate the correlation between the communication loss (such as signal attenuation, distortion degree, or delay degree, etc.) of the first abnormal optical cable node and the abnormal state. Through correlation analysis, generate the first correlation factor of optical cable abnormality, which can indicate the correlation degree between the communication loss of the first abnormal optical cable node and its abnormal state. For example, if the signal attenuation degree of the node is relatively high and the abnormal detection result indicates a fault, then this correlation factor will reflect the relationship between the attenuation degree and the fault.

[0072] Similarly, the communication loss characteristics of the second abnormal optical cable node also need to be correlated and analyzed. Based on the first result of optical cable abnormal detection, evaluate how the communication loss characteristics of the second abnormal optical cable node affect its abnormal detection state, generate the second correlation factor of optical cable abnormality, and illustrate the connection between the communication loss characteristics (such as delay degree, distortion degree) of the second abnormal optical cable node and its abnormal detection result, to help confirm the fault mode of the second node.

[0073] Record the abnormal detection results of the second abnormal optical cable node and mark them as the second result of optical cable abnormal detection, providing basic data for subsequent analysis of communication loss characteristics.

[0074] After recording the abnormal detection results of the second node, conduct a correlation analysis on the communication loss characteristics of the second abnormal optical cable node to obtain the third correlation factor of optical cable abnormality. By relating the second result of optical cable abnormal detection of the second abnormal optical cable node to its communication loss characteristics (such as signal attenuation, distortion degree, or delay degree, etc.), the potential correlation with node abnormality can be analyzed.

[0075] Finally, after completing the collection and analysis of all data, organize information such as the optical cable abnormal detection results, communication loss characteristics, and correlation factors, and generate a final abnormal detection report for dumb resources, including the correlation analysis between the communication loss characteristics and abnormal detection results of the first abnormal optical cable node and the second abnormal optical cable node, which can understand the status and fault causes of the abnormal optical cable node, provide decision-making support for network maintenance personnel, and enable faster identification of problem nodes and necessary repairs.

[0076] In summary, the method for abnormal detection of dumb resources based on RFID provided by this application has the following technical effects: achieving the technical goal of intelligent optical cable monitoring based on multi-modal factor fusion analysis, and achieving the technical effects of comprehensive assessment of optical cable health status, improvement of abnormal detection accuracy, enhancement of reliability of fault prediction, and intelligent upgrade of operation and maintenance management.

[0077] Embodiment 2. Based on the same inventive concept as the method for abnormal detection of dumb resources based on RFID in the foregoing embodiment, this application also provides an apparatus for abnormal detection of dumb resources based on RFID. Please refer to the appendix Figure 2, including: a dumb resource scanning module 11, which is used to scan the RFID tag group of the dumb resource according to the RFID reader to obtain the optical cable perception data stream, filter and align the optical cable perception data stream according to the optical cable multi-modal factor, and establish multiple node optical cable perception matrices, where the dumb resource includes the target optical cable; a deviation detection module 12, which is used to perform a security mode deviation detection on the target optical cable based on the optical cable multi-modal factor according to the multiple node optical cable perception matrices to determine the first abnormal optical cable node; a spread risk prediction module 13, which is used to predict the spread risk of the target optical cable according to the first abnormal optical cable node to determine the second abnormal optical cable node; a communication loss detection module 14, which is used to detect the communication loss of the first abnormal optical cable node and the second abnormal optical cable node to obtain the first node communication loss feature and the second node communication loss feature; an abnormal association tracing module 15, which is used to trace the abnormality of the first abnormal optical cable node and the second abnormal optical cable node according to the first node communication loss feature and the second node communication loss feature to generate a dumb resource abnormality detection report.

[0078] Furthermore, the dumb resource abnormality detection device based on RFID is also used to: perform a security mode mining on multiple optical cable nodes of the target optical cable according to the optical cable multi-modal factor to establish multiple node security mode matrices; perform an abnormality detection on the multiple node optical cable perception matrices according to the multiple node security mode matrices to obtain multiple node optical cable abnormality detection results; perform a security mode deviation risk evaluation according to the multiple node optical cable abnormality detection results to obtain multiple node optical cable risk coefficients; if any one of the multiple node optical cable risk coefficients is greater than or equal to the optical cable risk threshold, obtain the marked node optical cable risk coefficient, and record the optical cable node corresponding to the marked node optical cable risk coefficient as the first abnormal optical cable node.

[0079] Furthermore, the dumb resource abnormality detection device based on RFID is also used to: retrieve the normal record of monitoring for each optical cable node according to the optical cable multi-modal factor to obtain a normal feature record set of each node optical cable; perform a confidence evaluation on each parameter in the normal feature record set of each node optical cable to obtain a confidence evaluation set of each node parameter; based on the confidence evaluation set of each node parameter, select confidence parameters for the normal feature record set of each node optical cable according to a predetermined confidence constraint to obtain a security feature set of each node optical cable; perform a central tendency analysis according to the security feature set of each node optical cable to obtain the multiple node security mode matrices.

[0080] Further, the RFID-based dumb resource anomaly detection device is further configured to: record the node optical cable anomaly detection result corresponding to the first anomalous optical cable node as the first optical cable anomaly detection result; perform anomaly propagation prediction on the first optical cable anomaly detection result according to the ARIMA-LSTM combined model to obtain multiple node impact risk coefficients; determine whether the multiple node impact risk coefficients are greater than or equal to the node impact risk threshold; if any one of the multiple node impact risk coefficients is greater than or equal to the node impact risk threshold, obtain the marked node impact risk coefficient, and record the optical cable node corresponding to the marked node impact risk coefficient as the second anomalous optical cable node.

[0081] Further, the RFID-based dumb resource anomaly detection device is further configured to: retrieve the optical cable anomaly propagation event set corresponding to the first anomalous optical cable node; perform short-term anomaly propagation feature learning on the optical cable anomaly propagation event set according to the ARIMA model in the ARIMA-LSTM combined model to obtain the first optical cable anomaly propagation prediction model; perform long-term anomaly propagation feature learning on the optical cable anomaly propagation event set according to the LSTM model in the ARIMA-LSTM combined model to obtain the second optical cable anomaly propagation prediction model; perform fusion learning on the first optical cable anomaly propagation prediction model and the second optical cable anomaly propagation prediction model according to the optical cable anomaly propagation event set to obtain the third optical cable anomaly propagation prediction model; input the first optical cable anomaly detection result into the third optical cable anomaly propagation prediction model to obtain the multiple node impact risk coefficients.

[0082] Further, the RFID-based dumb resource anomaly detection device is further configured to: perform transmission tests on the first anomalous optical cable node according to multiple predetermined test signals to obtain multiple signal transmission monitoring data; input the multiple signal transmission monitoring data into an optical cable communication performance evaluation model to obtain multiple communication performance evaluation results, where the optical cable communication performance evaluation model includes multi-dimensional indexes for optical cable communication performance evaluation, and the multi-dimensional indexes for optical cable communication performance evaluation include transmission attenuation, transmission distortion, and transmission delay; calculate the centralized value of each dimension for the multiple communication performance evaluation results based on the multi-dimensional indexes for optical cable communication performance evaluation to obtain the first node communication performance evaluation result; obtain the expected communication performance evaluation result corresponding to the first anomalous optical cable node, and calculate the loss of the first node communication performance evaluation result according to the expected communication performance evaluation result to obtain the first node communication loss feature.

[0083] Further, the RFID-based dumb resource anomaly detection device is further configured to: perform correlation analysis on the first node communication loss feature according to the first result of the optical cable anomaly detection to obtain a first optical cable anomaly correlation factor; perform correlation analysis on the second node communication loss feature according to the first result of the optical cable anomaly detection to obtain a second optical cable anomaly correlation factor; record the node optical cable anomaly detection result pair corresponding to the second abnormal optical cable node as the second result of the optical cable anomaly detection; perform correlation analysis on the second node communication loss feature according to the second result of the optical cable anomaly detection to obtain a third optical cable anomaly correlation factor; collate the first result of the optical cable anomaly detection, the second result of the optical cable anomaly detection, the first node communication loss feature, the second node communication loss feature, the first optical cable anomaly correlation factor, the second optical cable anomaly correlation factor, and the third optical cable anomaly correlation factor to obtain the dumb resource anomaly detection report.

[0084] Further, the RFID-based dumb resource anomaly detection device is further configured to: perform feature recognition on the optical cable perception data stream according to the optical cable multimodal factor to obtain a plurality of node optical cable feature data; perform Kalman filtering on the plurality of node optical cable feature data to obtain a plurality of node optical cable filtered data; perform time synchronization alignment on the plurality of node optical cable filtered data to obtain a plurality of node optical cable aligned data; perform matrix processing on the plurality of node optical cable aligned data to obtain the plurality of node optical cable perception matrices.

[0085] Further, the RFID-based dumb resource anomaly detection device is further configured to: the optical cable multimodal factor includes optical cable vibration, optical cable curvature, optical cable stress, optical cable strain, optical cable temperature, optical cable humidity, and optical cable environment.

[0086] The various embodiments in this specification are described in a progressive manner, and the key point of each embodiment is the difference from other embodiments. The RFID-based dumb resource anomaly detection method and specific examples in the foregoing Embodiment 1 are equally applicable to the RFID-based dumb resource anomaly detection device in this embodiment. Through the foregoing detailed description of the RFID-based dumb resource anomaly detection method, those skilled in the art can clearly know the RFID-based dumb resource anomaly detection device in this embodiment. Therefore, for the sake of brevity of the specification, it will not be described in detail here.

[0087] The foregoing description of the disclosed embodiments enables those skilled in the art to practice or use the present application. Various modifications to these embodiments will be readily apparent to those skilled in the art, and the general principles defined herein may be implemented in other embodiments without departing from the spirit or scope of the present application. Thus, the present application is not intended to be limited to the embodiments shown herein, but is to be accorded the widest scope consistent with the principles and novel features disclosed herein.

[0088] Obviously, those skilled in the art can make various changes and modifications to the present application without departing from the spirit and scope of the present application. Thus, if these modifications and variations of the present application fall within the scope of the present application and its equivalent technologies, the present application is also intended to include these changes and modifications.

Claims

1. An abnormal detection method for dumb resources based on RFID, characterized in that Including: Scanning the RFID tag group of the dumb resources by an RFID reader to obtain an optical cable perception data stream, filtering and time synchronization alignment of the optical cable perception data stream according to the optical cable multimodal factors, and establishing an optical cable perception matrix for multiple nodes, where the dumb resources include the target optical cable; Based on the optical cable multimodal factors, performing a security mode deviation detection on the target optical cable according to the optical cable perception matrix for multiple nodes to determine the first abnormal optical cable node; Performing a spread risk prediction on the target optical cable according to the first abnormal optical cable node to determine the second abnormal optical cable node; Performing a communication loss detection on the first abnormal optical cable node and the second abnormal optical cable node to obtain a first node communication loss feature and a second node communication loss feature; According to the first node communication loss feature and the second node communication loss feature, performing an abnormal association traceability on the first abnormal optical cable node and the second abnormal optical cable node to generate a dumb resource abnormal detection report.

2. The method for detecting abnormal dumb resources based on RFID according to claim 1, wherein Based on the optical cable multimodal factors, performing a security mode deviation detection on the target optical cable according to the optical cable perception matrix for multiple nodes to determine the first abnormal optical cable node, including: Performing a security mode mining on multiple optical cable nodes of the target optical cable according to the optical cable multimodal factors to establish a security mode matrix for multiple nodes; Performing an abnormal detection on the optical cable perception matrix for multiple nodes according to the security mode matrix for multiple nodes to obtain an optical cable abnormal detection result for multiple nodes; Performing a security mode deviation risk evaluation according to the optical cable abnormal detection result for multiple nodes to obtain a risk coefficient for multiple nodes of the optical cable; If any one of the risk coefficients of multiple nodes of the optical cable is greater than or equal to the optical cable risk threshold, obtaining the risk coefficient of the marked node optical cable, and recording the optical cable node corresponding to the risk coefficient of the marked node optical cable as the first abnormal optical cable node.

3. The method for detecting abnormal dumb resources based on RFID according to claim 2, wherein Performing a security mode mining on multiple optical cable nodes of the target optical cable according to the optical cable multimodal factors to establish a security mode matrix for multiple nodes, including: Retrieving the normal monitoring records of each optical cable node according to the optical cable multimodal factors to obtain a normal feature record set for each node of the optical cable; Performing a confidence evaluation on each parameter in the normal feature record set for each node of the optical cable to obtain a confidence evaluation set for each node parameter; Based on the confidence evaluation set for each node parameter, performing a confidence parameter selection on the normal feature record set for each node of the optical cable according to a predetermined confidence constraint to obtain a security feature set for each node of the optical cable; Performing a central tendency analysis according to the security feature set for each node of the optical cable to obtain the security mode matrix for multiple nodes.

4. The method for detecting abnormal dumb resources based on RFID according to claim 1, wherein, Performing a spread risk prediction on the target optical cable according to the first abnormal optical cable node to determine the second abnormal optical cable node, including: Recording the optical cable abnormal detection result corresponding to the first abnormal optical cable node as the first optical cable abnormal detection result; Performing an abnormal propagation prediction on the first optical cable abnormal detection result according to the ARIMA-LSTM combined model to obtain a spread risk coefficient for multiple nodes; Judging whether the spread risk coefficient for multiple nodes is greater than or equal to the node spread risk threshold; If any of the risk coefficients of the affected nodes among the multiple nodes is greater than or equal to the risk threshold of the affected node, obtain the risk coefficient of the identified node affected, and record the optical cable node corresponding to the risk coefficient of the identified node affected as the second abnormal optical cable node.

5. The method for detecting abnormal dumb resources based on RFID according to claim 4, characterized in that, Perform abnormal propagation prediction on the first optical cable anomaly detection result according to the ARIMA-LSTM combined model to obtain risk coefficients of multiple nodes, including: Retrieve the optical cable abnormal propagation event set corresponding to the first abnormal optical cable node; Perform short-term abnormal propagation feature learning on the optical cable abnormal propagation event set according to the ARIMA model in the ARIMA-LSTM combined model to obtain the first optical cable abnormal propagation prediction model; Perform long-term abnormal propagation feature learning on the optical cable abnormal propagation event set according to the LSTM model in the ARIMA-LSTM combined model to obtain the second optical cable abnormal propagation prediction model; Perform fusion learning on the first optical cable abnormal propagation prediction model and the second optical cable abnormal propagation prediction model according to the optical cable abnormal propagation event set to obtain the third optical cable abnormal propagation prediction model; Input the first optical cable anomaly detection result into the third optical cable abnormal propagation prediction model to obtain the risk coefficients of the multiple nodes.

6. The abnormal detection method for dumb resources based on RFID according to claim 1, wherein Perform communication loss detection on the first abnormal optical cable node and the second abnormal optical cable node to obtain the first node communication loss feature and the second node communication loss feature, including: Perform transmission tests on the first abnormal optical cable node according to multiple predetermined test signals to obtain multiple signal transmission monitoring data; Input the multiple signal transmission monitoring data into the optical cable communication performance evaluation model to obtain multiple communication performance evaluation results, where the optical cable communication performance evaluation model includes multi-dimensional indexes for optical cable communication performance evaluation, and the multi-dimensional indexes for optical cable communication performance evaluation include transmission attenuation degree, transmission distortion degree, and transmission delay degree; Based on the multi-dimensional indexes for optical cable communication performance evaluation, calculate the central value of each dimension for the multiple communication performance evaluation results to obtain the first node communication performance evaluation result; Obtain the expected communication performance evaluation result corresponding to the first abnormal optical cable node, and perform loss calculation on the first node communication performance evaluation result according to the expected communication performance evaluation result to obtain the first node communication loss feature.

7. The abnormal detection method for dumb resources based on RFID according to claim 1, characterized in that Perform abnormal correlation tracing on the first abnormal optical cable node and the second abnormal optical cable node according to the first node communication loss feature and the second node communication loss feature, and generate a dummy resource anomaly detection report, including: Perform correlation analysis on the first node communication loss feature according to the first optical cable anomaly detection result to obtain the first optical cable anomaly correlation factor; Perform correlation analysis on the second node communication loss feature according to the first optical cable anomaly detection result to obtain the second optical cable anomaly correlation factor; Record the node optical cable anomaly detection result pair corresponding to the second abnormal optical cable node as the second optical cable anomaly detection result; Perform correlation analysis on the second node communication loss feature according to the second optical cable anomaly detection result to obtain the third optical cable anomaly correlation factor; Collate the first result of the optical cable anomaly detection, the second result of the optical cable anomaly detection, the communication loss characteristics of the first node, the communication loss characteristics of the second node, the first correlation factor of the optical cable anomaly, the second correlation factor of the optical cable anomaly, and the third correlation factor of the optical cable anomaly to obtain the dumb resource anomaly detection report.

8. The method for detecting abnormal dumb resources based on RFID according to claim 1, characterized in that Filter and perform time synchronization alignment on the optical cable perception data stream according to the optical cable multimodal factors to establish optical cable perception matrices for multiple nodes, including: Perform feature recognition on the optical cable perception data stream according to the optical cable multimodal factors to obtain optical cable feature data for multiple nodes; Perform Kalman filtering on the optical cable feature data for multiple nodes to obtain optical cable filtered data for multiple nodes; Perform time synchronization alignment on the optical cable filtered data for multiple nodes to obtain optical cable aligned data for multiple nodes; Perform matrix processing on the optical cable aligned data for multiple nodes to obtain the optical cable perception matrices for multiple nodes.

9. The abnormal detection method for dumb resources based on RFID according to claim 1, wherein The optical cable multimodal factors include optical cable vibration, optical cable bend, optical cable stress, optical cable strain, and optical cable environment, and the optical cable environment at least includes optical cable temperature and optical cable humidity.

10. An RFID-based abnormal detection device for dumb resources, characterized in that, Steps for implementing the method for detecting anomalies of dumb resources based on RFID according to any one of claims 1 to 9, including: A dumb resource scanning module, which is used to scan the RFID tag group of the dumb resource according to the RFID reader to obtain the optical cable perception data stream, and filter and perform time synchronization alignment on the optical cable perception data stream according to the optical cable multimodal factors to establish optical cable perception matrices for multiple nodes, and the dumb resource includes the target optical cable; A deviation detection module, which is used to perform a safety mode deviation detection on the target optical cable based on the optical cable multimodal factors according to the optical cable perception matrices for multiple nodes to determine the first abnormal optical cable node; A spread risk prediction module, which is used to perform a spread risk prediction on the target optical cable according to the first abnormal optical cable node to determine the second abnormal optical cable node; A communication loss detection module, which is used to perform a communication loss detection on the first abnormal optical cable node and the second abnormal optical cable node to obtain the communication loss characteristics of the first node and the communication loss characteristics of the second node; An anomaly correlation tracing module, which is used to perform an anomaly correlation tracing on the first abnormal optical cable node and the second abnormal optical cable node according to the communication loss characteristics of the first node and the communication loss characteristics of the second node to generate a dumb resource anomaly detection report.

Images