Optical fiber network fault automatic positioning system and method

Abstract

The invention relates to the technical field of network fault detection, in particular to an optical fiber network fault automatic positioning system and method, and the method comprises the steps: collecting optical signal transmission parameters of an optical fiber link; analyzing the transmission parameters of the received optical signal, and identifying an abnormal reflection signal; determining a fault type and a corresponding fault position by adopting a support vector machine according to the characteristics of the abnormal reflection signal; and matching a corresponding maintenance suggestion based on the fault type and the fault position. According to the automatic positioning method for the optical fiber network fault, efficient and accurate identification and positioning of the optical fiber network fault are realized by integrating a data acquisition technology, intelligent data analysis and a machine learning algorithm, and the efficiency and quality of network maintenance are greatly improved.

Description

Technical Field

[0001] The present invention relates to the technical field of network fault detection, and in particular to an optical fiber network fault automatic positioning system and method. Background Art

[0002] Fiber optic network failure refers to an abnormal or interrupted state that occurs in a communication system that uses optical fiber as a transmission medium, which directly affects the high-speed transmission of data. Optical fiber is often used to build long-distance, high-speed data transmission networks due to its wide bandwidth, low loss, and strong anti-interference ability. However, when the optical fiber line suffers physical damage, loose connectors, pollution, equipment failure, or encounters external factors, it may cause network connection interruption, increased signal attenuation, and increased data transmission error rate. Fault diagnosis usually involves using professional tools to test optical power, locate the fault point, and take corresponding repair measures, such as replacing damaged optical fibers, tightening connectors, cleaning interfaces, or replacing faulty equipment to restore the normal operation of the network.

[0003] Existing optical fiber network fault location mainly adopts manual detection, which makes the work efficiency low. Summary of the invention

[0004] The purpose of the present invention is to provide a system and method for automatically locating optical fiber network faults, aiming to achieve efficient and accurate identification and locating of optical fiber network faults, thereby greatly improving the efficiency and quality of network maintenance.

[0005] To achieve the above object, the present invention provides a method for automatically locating faults in an optical fiber network, comprising collecting optical signal transmission parameters of an optical fiber link;

[0006] Analyze the transmission parameters of the received optical signal and identify abnormal reflection signals;

[0007] According to the characteristics of abnormal reflection signals, support vector machine is used to determine the fault type and the corresponding fault location;

[0008] Match the corresponding maintenance suggestions based on the fault type and fault location.

[0009] After matching the corresponding maintenance suggestions based on the fault type and the fault location, the method further includes registering and storing the fault handling results and providing scheduled maintenance reminders.

[0010] The transmission parameters include optical signal strength, optical fiber link length, loss, and reflectivity.

[0011] The specific steps of analyzing the received optical signal transmission parameters and identifying abnormal reflection signals include:

[0012] Collect optical signal data samples under fault-free conditions and establish a baseline model of normal transmission parameters;

[0013] Set the threshold for anomaly identification based on the baseline model;

[0014] Compare the current signal characteristics with the threshold of the baseline model to identify signals that deviate from the norm.

[0015] The specific steps of collecting optical signal data samples under fault-free conditions and establishing a benchmark model of normal transmission parameters include:

[0016] Remove outliers, missing values ​​and noise from optical signal data samples;

[0017] Standardize the data from different measurement points;

[0018] collecting and processing the time domain characteristics, frequency domain characteristics and statistical characteristics of the optical signal data samples;

[0019] The baseline models of time domain features, frequency domain features and statistical features are established through linear regression.

[0020] The specific steps of using a support vector machine to determine the fault type and the corresponding fault location according to the characteristics of the abnormal reflection signal include:

[0021] Selecting fault diagnosis features according to the reflection signal, wherein the fault diagnosis features include the height, width, shape, and position of the reflection peak;

[0022] Label the fault type and location for the training data set based on the actual fault conditions;

[0023] Use the labeled training data set to train the SVM model;

[0024] An independent validation dataset was used to evaluate the performance of the SVM model;

[0025] For new reflection signal data, the trained SVM model is used to predict the fault type and fault location.

[0026] The specific steps of matching the corresponding maintenance suggestions based on the fault type and the fault location include:

[0027] Establish a database containing various fault types, fault locations and corresponding maintenance suggestions;

[0028] Extract key words from fault information and fault location;

[0029] Search and match the keywords with the fault type and fault location in the database to obtain maintenance suggestions.

[0030] The specific steps of extracting keywords from the fault information and the fault location include:

[0031] Remove punctuation, special characters and numbers from text;

[0032] Split the text into words or phrases;

[0033] Apply TF-IDF to identify keywords.

[0034] The specific steps of searching and matching the keywords with the fault types and fault locations in the database to obtain maintenance suggestions include:

[0035] Format the extracted fault information into a data format consistent with that in the database;

[0036] Use keywords as query parameters to perform fuzzy matching in the database and obtain maintenance suggestions.

[0037] In a second aspect, the present invention also provides an automatic optical fiber network fault location system, comprising: a data collection module, an abnormal signal recognition module, a fault recognition module and a suggestion module;

[0038] The data collection module is used to collect optical signal transmission parameters of the optical fiber link;

[0039] The abnormal signal identification module is used to analyze the received optical signal transmission parameters and identify abnormal reflection signals;

[0040] The fault identification module is used to determine the fault type and the corresponding fault location using a support vector machine according to the characteristics of the abnormal reflection signal;

[0041] The suggestion module is used to match corresponding maintenance suggestions based on the fault type and the fault location.

[0042] The present invention discloses an automatic optical fiber network fault location system and method, which uses professional equipment such as a highly sensitive spectrum analyzer and an optical time domain reflectometer (OTDR) to comprehensively scan each link in the optical fiber network. The collected data includes, but is not limited to, key transmission parameters such as optical power, optical attenuation coefficient, dispersion characteristics, signal delay time, and reflection loss. This process ensures the integrity and accuracy of the information and provides a solid foundation for subsequent analysis. The collected data is then input into a specially designed data processing system. The system uses advanced algorithms to deeply analyze the optical signal transmission parameters, and uses pattern recognition technology to filter out abnormal signal patterns from massive data, especially those abnormal reflection signals indicating the presence of faults. These anomalies are usually manifested as a sudden drop in signal strength, irregular reflection waveforms, or unexpected changes in transmission parameters. After identifying the abnormal reflection signal, the method further uses the support vector machine (SVM) algorithm in machine learning to classify and analyze the characteristics of the abnormal signal. SVM can accurately determine the type of fault (such as fiber breakage, excessive bending, connector contamination, etc.) and the exact location of the fault based on the trained model and the specific characteristics of the abnormal signal, such as waveform profile, frequency distribution, etc. This process is highly automated, reducing the subjectivity and errors of human judgment. After determining the detailed information of the fault, the system will automatically match the corresponding maintenance strategy and suggestions. These suggestions take into account the fault type, the particularity of the location, and the successful experience of historical maintenance cases, and may include specific operation guidelines such as replacing specific sections of optical fiber, cleaning connectors, and adjusting the optical fiber layout to reduce bending loss. The generation of maintenance suggestions is intended to provide field engineers with direct and feasible operation plans to speed up the fault repair process and reduce network downtime.

[0043] In summary, this fiber optic network fault automatic location method realizes efficient and accurate identification and location of fiber optic network faults by integrating advanced data acquisition technology, intelligent data analysis and machine learning algorithms, greatly improving the efficiency and quality of network maintenance. BRIEF DESCRIPTION OF THE DRAWINGS

[0044] In order to more clearly illustrate the embodiments of the present invention or the technical solutions in the prior art, the drawings required for use in the embodiments or the description of the prior art will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying creative work.

[0045] Figure 1 It is a flow chart of a method for automatically locating optical fiber network faults according to the first embodiment of the present invention.

[0046] Figure 2This is a flow chart of analyzing the transmission parameters of a received optical signal and identifying an abnormal reflection signal according to the first embodiment of the present invention.

[0047] Figure 3 It is a flow chart of collecting optical signal data samples under a fault-free condition and establishing a reference model of normal transmission parameters according to the first embodiment of the present invention.

[0048] Figure 4 This is a flow chart of the first embodiment of the present invention for determining a fault type and a corresponding fault location using a support vector machine according to the characteristics of an abnormal reflection signal.

[0049] Figure 5 This is a flowchart of matching corresponding maintenance suggestions based on fault type and fault location according to the first embodiment of the present invention.

[0050] Figure 6 This is a flowchart of extracting keywords from fault information and fault locations according to the first embodiment of the present invention.

[0051] Figure 7 This is a flow chart of the first embodiment of the present invention for searching and matching keywords with fault types and fault locations in a database to obtain maintenance suggestions. DETAILED DESCRIPTION

[0052] Embodiments of the present invention are described in detail below, examples of which are shown in the accompanying drawings, wherein the same or similar reference numerals throughout represent the same or similar elements or elements having the same or similar functions. The embodiments described below with reference to the accompanying drawings are exemplary and are intended to be used to explain the present invention, and should not be construed as limiting the present invention.

[0053] First embodiment

[0054] See also Figure 1 to Figure 7 The present invention provides a method for automatically locating optical fiber network faults, comprising:

[0055] S101 collects optical signal transmission parameters of the optical fiber link;

[0056] The transmission parameters include optical signal strength, length, loss, and reflectivity of the optical fiber link.

[0057] Optical signal strength is an indicator of the energy of optical signals transmitted in optical fiber links, usually in decibel milliwatts (dBm). Optical signal strength is directly related to the quality and distance of data transmission. A signal that is too weak may increase the data bit error rate, while a signal that is too strong may cause nonlinear effects and affect the integrity of the signal. Therefore, accurately measuring the signal strength on each link segment helps to identify the location of signal attenuation and take timely reinforcement measures.

[0058] Understanding the actual length of the optical fiber is crucial for calculating the signal attenuation budget, designing compensation solutions, and planning network expansion. The length of the optical fiber is proportional to the signal attenuation, and longer links require more sophisticated power budget management or the deployment of relay equipment to ensure that the signal can reach the receiving end without loss.

[0059] Fiber loss refers to the reduction of optical signal energy during transmission, which is mainly composed of absorption loss, scattering loss and connection loss. By measuring the difference in optical power at both ends of the link and combining it with the link length, the total loss value can be calculated. Understanding the specific loss of each link helps to optimize network design, select the appropriate fiber type, and arrange regular cleaning and maintenance work to reduce unnecessary signal loss.

[0060] It reflects the degree of reflection of the optical signal in the fiber interface or link. Excessive reflection will cause signal distortion and affect the communication quality. In particular, improper connector installation or contamination of the fiber end face may cause enhanced reflection. Therefore, measuring and controlling reflectivity is a key link to ensure signal purity and system stability. By using professional instruments such as optical time domain reflectometer (OTDR), the reflection characteristics in the fiber link can be effectively evaluated and the reflection source can be located.

[0061] S102 analyzes the transmission parameters of the received optical signal to identify abnormal reflection signals;

[0062] This step aims to separate abnormal reflection signals that may be caused by line defects, connection problems or external interference from the complex received optical signals through precise analysis. Identifying these anomalies is crucial to preventing system failures and reducing data packet loss. The specific steps include:

[0063] S201 collects optical signal data samples under a fault-free condition and establishes a baseline model of normal transmission parameters;

[0064] The specific steps include:

[0065] S301 removes outliers, missing values ​​and noise in the optical signal data sample;

[0066] Statistical methods such as box plot analysis are used to identify and remove outliers in the data, while filling in missing values ​​to ensure the integrity and accuracy of the data set. Noise filtering is performed through filtering algorithms such as low-pass filters to eliminate high-frequency random interference.

[0067] S302 performs standardization processing on data at different measurement points;

[0068] In order to eliminate the deviations introduced by equipment differences or environmental changes at different measurement points, all data were standardized to ensure that all features were comparable on the same scale.

[0069] S303 collects time domain features, frequency domain features and statistical features of the processed optical signal data samples;

[0070] Time domain features (such as the rising edge and falling edge characteristics of the signal waveform), frequency domain features (spectral distribution obtained by Fourier transform) and statistical features (mean, variance, etc.) are extracted from the processed data. These features can fully reflect the transmission quality of the optical signal.

[0071] S304 establishes a baseline model of time domain features, frequency domain features and statistical features through linear regression.

[0072] Using linear regression analysis, the above characteristics are used as input variables to establish a prediction model associated with normal optical signal transmission performance. This model will become a reference standard for subsequent detection of abnormal signals.

[0073] S202 sets a threshold for abnormality recognition based on the benchmark model;

[0074] After completing the comprehensive collection and analysis of the optical signal transmission parameters of the fiber link under fault-free conditions, the next step is to build a baseline model that represents the normal operating state. This model combines the statistical distribution and typical characteristics of key parameters such as optical signal strength, link length, loss, and reflectivity. Based on this baseline model, technicians or algorithms will set a series of thresholds that are used to define the boundaries between normal signal characteristics and abnormal signal characteristics that may indicate faults or performance degradation.

[0075] The process of setting thresholds involves the following aspects:

[0076] Statistical analysis: Conduct in-depth statistics on the data in the benchmark model, such as calculating the mean value, standard deviation, etc., to provide a quantitative basis for setting thresholds.

[0077] Balance between sensitivity and specificity: The setting of the threshold needs to find a balance between the false negative rate and the false positive rate. Too high or too low a threshold may lead to increased maintenance costs or untimely fault detection.

[0078] Dynamic adjustment: Taking into account factors such as environmental changes and equipment aging, threshold settings should have a certain dynamic adjustment mechanism to adapt to long-term changes in network status.

[0079] S203 compares the current signal feature with the threshold of the reference model to identify signals that deviate from the normal state.

[0080] After the threshold is clear, the real-time monitoring system is used to continuously collect optical signal data in the fiber link. For each data point, the system compares it with the threshold set by the benchmark model to identify abnormal signals.

[0081] S103 determines the fault type and the corresponding fault location using a support vector machine according to the characteristics of the abnormal reflection signal;

[0082] The specific steps include:

[0083] S401 selects a fault diagnosis feature according to the reflection signal, wherein the fault diagnosis feature includes the height, width, shape, and position of the reflection peak;

[0084] At this stage, the first task is to extract the most diagnostically valuable features from the abnormal reflection signal. These features can effectively distinguish different types of faults and locate the fault point. Key features usually include:

[0085] Height of reflection peak: reflects the strength of signal reflection and is related to the severity of the fault.

[0086] Width: A broad peak may mean a larger fault area or more severe signal attenuation.

[0087] Shape: Specific shape patterns may be associated with specific types of failures.

[0088] Position: The distance of the reflection peak relative to the test point directly indicates the approximate location of the fault.

[0089] S402: marking the fault type and fault location for the training data set according to the actual fault situation;

[0090] In order to train the SVM model, a high-quality training data set needs to be prepared. This involves collecting a large number of reflection signal samples of known fault types and accurately labeling the fault type (such as broken fiber, loose connector, fiber contamination, etc.) and the specific location of the fault for each sample based on actual inspection and maintenance records. This labeled information is the basis for model learning.

[0091] S403 uses the labeled training data set to train the SVM model;

[0092] After selecting appropriate features and completing data annotation, use these data to train the SVM model. In this process, it is necessary to determine the appropriate kernel function (such as linear kernel, RBF kernel) and adjust the model parameters (such as penalty factor C, kernel function parameter γ) to achieve the best classification effect. By optimizing the hyperparameters, ensure that the model can accurately learn the mapping relationship between different fault features and fault types and locations.

[0093] S404 evaluates the performance of the SVM model using an independent validation dataset;

[0094] In order to verify the generalization ability of the model, it is necessary to use a validation data set that is independent of the training data for testing. By calculating performance indicators such as accuracy, recall, and F1 score, the performance of the model on unknown data can be evaluated. If the performance does not meet the requirements, it may be necessary to go back to the previous steps and adjust the feature selection, model parameters, or data set until the model performance reaches the expected level.

[0095] S405 predicts the fault type and fault location for the new reflection signal data using the trained SVM model.

[0096] After the model training is completed, it is deployed to the fault detection system. When the system receives new reflected signal data, it first extracts the same feature set, and then inputs these features into the trained SVM model. The model can then output the predicted fault type and approximate fault location. In this way, the maintenance team can quickly obtain fault information and take timely action, effectively shortening the troubleshooting and repair time and ensuring the stable operation of the network.

[0097] S104 matches corresponding maintenance suggestions based on the fault type and the fault location.

[0098] The specific steps include:

[0099] S501 establishes a database including various fault types, fault locations and corresponding maintenance suggestions;

[0100] Create a structured database that records all known fault types (such as fiber breakage, loose connectors, contamination, etc.), their typical locations where they may occur, and recommended repair steps or solutions for each situation. This database is the basis for subsequent automatic matching of repair recommendations.

[0101] S502 extracts keywords from the fault information and the fault location;

[0102] This process involves text processing of fault descriptions and location information to accurately extract key information for easy matching with records in the database. The specific steps include:

[0103] S601 deletes punctuation marks, special characters and numbers in the text;

[0104] First, remove all non-alphabetic characters, such as commas, periods, punctuation marks such as brackets, numbers, and special characters from the fault description and location information to focus on the plain text content.

[0105] S602 divides the text into words or phrases;

[0106] Segmenting the cleaned text into single words or meaningful phrase units helps in identifying and extracting keywords in subsequent steps.

[0107] S603 applies TF-IDF to identify keywords.

[0108] The TF-IDF (Term Frequency-Inverse Document Frequency) algorithm is used to identify the keywords that best represent the text topic from the word segmentation results. TF-IDF measures the importance of each word in the text, and high-frequency words that are not common in other documents are given higher weights, thereby selecting the most representative words as keywords for matching the database.

[0109] S503 searches and matches the keyword with the fault type and fault location in the database to obtain maintenance suggestions.

[0110] The specific steps include:

[0111] S701 formats the extracted fault information into a data format consistent with that in the database;

[0112] This step is to organize and format the key information extracted from the original fault description (such as fault type, location, specific code, etc.) to ensure that it is consistent with the record format in the fault database. This usually involves data cleaning, normalization and standardization, such as converting text descriptions into unified codes or labels, and ensuring that the format of information such as time and location is unified so that the database can accurately identify and process this information.

[0113] S702 uses the keyword as a query parameter to perform fuzzy matching in the database to obtain maintenance suggestions.

[0114] In this step, the previously formatted fault information keywords are used to search in the fault database through a fuzzy matching algorithm. Fuzzy matching allows a certain degree of flexibility, and relevant fault cases and solutions can be found even if the description is not completely accurate. Based on the matching results, the system will provide a series of possible maintenance suggestions, including fault cause analysis, maintenance steps, required tools or spare parts, etc., to help maintenance personnel quickly determine the best maintenance strategy.

[0115] S105 registers and stores the fault handling result and provides a scheduled maintenance reminder.

[0116] The detailed information of the entire handling process, including fault description, measures taken, maintenance results, consumables usage, maintenance personnel information, etc., is entered into the system for registration and long-term storage. This not only serves as a historical record for future reference, but also helps with failure mode analysis, maintenance cost control and optimization of maintenance strategies.

[0117] Second embodiment

[0118] On the basis of the first embodiment, the present invention further provides an automatic optical fiber network fault location system, comprising: a data collection module, an abnormal signal recognition module, a fault recognition module and a suggestion module; the data collection module is used to collect optical signal transmission parameters of the optical fiber link; the abnormal signal recognition module is used to analyze the received optical signal transmission parameters and identify abnormal reflection signals; the fault recognition module is used to determine the fault type and the corresponding fault location using a support vector machine according to the characteristics of the abnormal reflection signal; the suggestion module is used to match the corresponding maintenance suggestions based on the fault type and the fault location.

[0119] In this embodiment, the data collection module is the first process of the system. The module uses advanced optical time domain reflectometer (OTDR), spectrum analyzer and other equipment to continuously monitor the optical signal transmission parameters in the optical fiber link, including but not limited to optical signal intensity, transmission rate, loss rate, reflection coefficient, etc. These data are the basis for subsequent analysis and processing, and are transmitted to the central processing unit of the system in real time through a high-speed data interface. The abnormal signal recognition module receives the original data from the data collection module and uses a complex signal processing algorithm to perform in-depth analysis on the optical signal transmission parameters. It identifies abnormal reflection patterns in the signal by comparing it with the preset normal signal feature model, such as irregular reflection peaks, abnormal signal attenuation areas, etc., which are often direct manifestations of faults in the optical fiber link. Based on the data provided by the abnormal signal recognition module, the fault recognition module introduces machine learning technology, especially the support vector machine (SVM) algorithm, to perform detailed classification and pattern recognition on the features of abnormal reflection signals. By learning a large number of labeled fault samples, SVM can accurately distinguish different types of faults (such as breakage, loose joints, contamination, etc.), and calculate the exact location of the fault in combination with signal features. This process not only improves the accuracy of fault identification, but also greatly speeds up the speed of fault location. Once the fault type and location are determined, the suggestion module is immediately started. It matches the most suitable maintenance suggestions and operation guidelines for the currently identified fault type and location based on the system's embedded fault knowledge base and historical maintenance records. These suggestions include recommended maintenance methods, lists of required tools and spare parts, safe operating procedures, etc., to provide immediate and targeted guidance to the on-site maintenance team, helping to quickly troubleshoot and restore network operations. In summary, this fiber optic network fault automatic location system realizes the automation of the entire chain from data collection to fault location to maintenance suggestion generation through highly integrated data processing processes and intelligent analysis technologies, greatly improving the efficiency and accuracy of fiber optic network maintenance.

[0120] What is disclosed above is only a preferred embodiment of the present invention, and it certainly cannot be used to limit the scope of rights of the present invention. Ordinary technicians in this field can understand that all or part of the processes of the above embodiment and equivalent changes made according to the claims of the present invention still fall within the scope of the invention.

Claims

1. A method for automatically locating optical fiber network faults, characterized in that: include: Collect optical signal transmission parameters of the optical fiber link; Analyze the transmission parameters of the received optical signal and identify abnormal reflection signals; According to the characteristics of abnormal reflection signals, support vector machine is used to determine the fault type and the corresponding fault location; Match the corresponding maintenance suggestions based on the fault type and fault location.

2. The method for automatically locating a fault in an optical fiber network according to claim 1, characterized in that: After matching the corresponding maintenance suggestions based on the fault type and the fault location, the method further includes registering and storing the fault handling results and providing a scheduled maintenance reminder.

3. A method for automatically locating optical fiber network faults according to claim 2, characterized in that: The transmission parameters include optical signal strength, length, loss, and reflectivity of the optical fiber link.

4. A method for automatically locating optical fiber network faults according to claim 3, characterized in that: The specific steps of analyzing the received optical signal transmission parameters and identifying abnormal reflection signals include: Collect optical signal data samples under fault-free conditions and build a baseline model of normal transmission parameters; Set the threshold for anomaly identification based on the baseline model; Compare the current signal characteristics with the threshold of the baseline model to identify signals that deviate from the norm.

5. The method for automatically locating optical fiber network faults according to claim 4, characterized in that: The specific steps of collecting optical signal data samples under the fault-free condition and establishing a benchmark model of normal transmission parameters include: Remove outliers, missing values ​​and noise from optical signal data samples; Standardize the data from different measurement points; collecting and processing the time domain characteristics, frequency domain characteristics and statistical characteristics of the optical signal data samples; The baseline models of time domain features, frequency domain features and statistical features are established through linear regression.

6. A method for automatically locating optical fiber network faults according to claim 5, characterized in that: The specific steps of using a support vector machine to determine the fault type and the corresponding fault location according to the characteristics of the abnormal reflection signal include: Selecting fault diagnosis features according to the reflection signal, wherein the fault diagnosis features include the height, width, shape, and position of the reflection peak; Label the fault type and location for the training data set based on the actual fault conditions; Use the labeled training data set to train the SVM model; An independent validation dataset was used to evaluate the performance of the SVM model; For new reflection signal data, the trained SVM model is used to predict the fault type and fault location.

7. A method for automatically locating optical fiber network faults according to claim 6, characterized in that: The specific steps of matching the corresponding maintenance suggestions based on the fault type and the fault location include: Establish a database containing various fault types, fault locations and corresponding maintenance suggestions; Extract key words from fault information and fault location; Search and match the keywords with the fault type and fault location in the database to obtain maintenance suggestions.

8. The method for automatically locating optical fiber network faults according to claim 7, characterized in that: The specific steps of extracting keywords from the fault information and the fault location include: Remove punctuation, special characters and numbers from text; Split the text into words or phrases; Apply TF-IDF to identify keywords.

9. The method for automatically locating optical fiber network faults according to claim 8, characterized in that: The specific steps of searching and matching the keywords with the fault types and fault locations in the database to obtain maintenance suggestions include: Format the extracted fault information into a data format consistent with that in the database; Use keywords as query parameters to perform fuzzy matching in the database and obtain maintenance suggestions.

10. An automatic optical fiber network fault location system, characterized in that: include: Data collection module, abnormal signal recognition module, fault recognition module and suggestion module; The data collection module is used to collect optical signal transmission parameters of the optical fiber link; The abnormal signal identification module is used to analyze the received optical signal transmission parameters and identify abnormal reflection signals; The fault identification module is used to determine the fault type and the corresponding fault location using a support vector machine according to the characteristics of the abnormal reflection signal; The suggestion module is used to match corresponding maintenance suggestions based on the fault type and the fault location.

Images