Broadband link fault detection method, system, electronic device and readable storage medium

Abstract

The embodiment of the present invention provides a broadband link fault detection method, system, electronic device and readable storage medium. The method includes: collecting broadband authentication and billing data; parsing the broadband authentication and billing data to obtain data of OLT devices in the broadband link; inputting data associated with the same OLT device into a state prediction model to obtain the estimated state of the same OLT device, the estimated state is: no fault or there is a risk of failure, the OLT device state prediction model uses the data of sample OLT devices and the corresponding actual state as training samples, the actual state is no fault or there is a risk of failure. It aims to improve the security of broadband services.

Description

Technical Field

[0001] The present invention relates to the technical field of fault detection, and in particular to a broadband link fault detection method, system, electronic device and readable storage medium. Background Art

[0002] In recent years, the Internet industry has developed rapidly. The fixed-line home broadband business is based on the concept of user-first, relying on the improvement of technology and services to continuously meet the increasing high-experience needs of customers. With the continuous expansion of the network market and the rapid growth of the number of user groups, the demand for network service quality is getting higher and higher, which brings new operation and maintenance challenges to network construction. The existing network operation and maintenance methods need to be investigated according to the problems that have occurred, which often have problems such as long troubleshooting cycle, high positioning difficulty, and weak overall control ability. Summary of the invention

[0003] In view of this, embodiments of the present invention provide a broadband link failure detection method, system, electronic device, and readable storage medium, aiming to improve the security of broadband services.

[0004] A first aspect of an embodiment of the present invention provides a broadband link failure detection method, the method comprising:

[0005] Collect broadband authentication and billing data;

[0006] Parsing the broadband authentication and billing data to obtain data of the OLT device in the broadband link;

[0007] Data associated with the same OLT device is input into a state prediction model to obtain an estimated state of the same OLT device, wherein the estimated state is: no fault or a risk of failure. The OLT device state prediction model uses data of sample OLT devices and corresponding actual states as training samples, wherein the actual state is: no fault or a risk of failure.

[0008] Optionally, the method further comprises:

[0009] The data of the OLT device whose estimated state is that there is a risk of failure is input into the situation prediction model to obtain the situation prediction result of the OLT device, and the situation prediction result characterizes the state change of the OLT device in a future period of time. The situation prediction model uses the data of the sample OLT device in a first historical time period and the state change in a second historical time period as training samples.

[0010] Optionally, the method further comprises:

[0011] When the status prediction result of the OLT device is a steep drop state, a prompt for troubleshooting the hardware device or network is output.

[0012] Optionally, the method further comprises:

[0013] When the situation prediction result of the OLT device is a growth state, monitor whether the service traffic reaches the expansion threshold;

[0014] When the service traffic reaches the expansion threshold, a hardware device expansion prompt is output.

[0015] Optionally, it also includes:

[0016] The online duration of the target user terminal mounted on an OLT device is input into the online duration processing model of the OLT device to determine whether the user account of the target user terminal is abnormal. The online duration processing model uses the historical online information of each user terminal mounted on the OLT device and the normal labels or abnormal labels carried by the user accounts of each user terminal as training samples.

[0017] Optionally, the method further comprises:

[0018] According to the preset time interval, count the online time and number of user terminals mounted on the same OLT device;

[0019] The terminal mounting information of the same OLT device is displayed, and the terminal mounting information includes: online time and terminal quantity.

[0020] The embodiments of the present invention include the following advantages:

[0021] A broadband link fault detection method provided by an embodiment of the present invention collects broadband authentication and billing data; analyzes the collected broadband authentication and billing data to obtain data related to an OLT device in a broadband link; inputs data associated with the same OLT device into a state prediction model to obtain an estimated state of the same OLT device, wherein the estimated state is: no fault or a risk of failure; the OLT device state prediction model uses data of a sample OLT device and a corresponding actual state as training samples, wherein the actual state is no fault or a risk of failure. The state prediction model can predict in advance whether an OLT device has a risk of failure based on the broadband authentication and billing data generated by a user, thereby improving the security of broadband services.

[0022] A second aspect of an embodiment of the present invention provides a broadband link fault detection system, the system comprising:

[0023] Data collection module, used to collect broadband authentication and billing data;

[0024] A data analysis module, used to analyze the broadband authentication and billing data to obtain data of the OLT device in the broadband link;

[0025] The state prediction module is used to input data associated with the same OLT device into a state prediction model to obtain an estimated state of the same OLT device, wherein the estimated state is: no fault or a risk of failure. The OLT device state prediction model uses data of sample OLT devices and corresponding actual states as training samples, and the actual state is no fault or a risk of failure.

[0026] A third aspect of an embodiment of the present invention provides an electronic device, comprising a memory, a processor, and a program stored in the memory and executable on the processor, wherein when the processor executes the program, the steps in a broadband link failure detection method described in the first aspect of the present invention are implemented.

[0027] A fourth aspect of an embodiment of the present invention provides a readable storage medium, on which a program is stored. When the program is executed by a processor, the steps in the broadband link failure detection method described in the first aspect of the present invention are implemented. BRIEF DESCRIPTION OF THE DRAWINGS

[0028] In order to more clearly illustrate the technical solutions of the embodiments of the present invention, the accompanying drawings required for use in the description of the embodiments of the present invention will be briefly introduced below. Obviously, the accompanying drawings in the following description are only some embodiments of the present invention. For ordinary technicians in this field, other accompanying drawings can be obtained based on these accompanying drawings without paying creative labor.

[0029] Figure 1 is a flow chart of a broadband link failure detection method shown in one embodiment of the present invention;

[0030] Figure 2 It is a schematic diagram of an authentication and accounting message transmission path in a broadband link failure detection method shown in an embodiment of the present invention;

[0031] Figure 3 It is a schematic diagram of a broadband link fault detection system shown in an embodiment of the present invention. DETAILED DESCRIPTION

[0032] The following will be combined with the drawings in the embodiments of the present invention to clearly and completely describe the technical solutions in the embodiments of the present invention. Obviously, the described embodiments are part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0033] Before describing the present invention, the background of the present invention is described. With the rapid development of the Internet industry, fixed-line home broadband services are user-first and rely on the improvement of technology and services to continuously meet customers' increasing demand for high experience. With the continuous expansion of the network market and the rapid growth of the number of user groups, the demand for network service quality is getting higher and higher, bringing new operation and maintenance challenges to network construction. The existing network operation and maintenance methods need to be investigated based on the problems that have occurred, and often have problems such as long troubleshooting cycles, high positioning difficulty, and weak global control capabilities. Specific problems include: 1. The dual uplink technology of OLT equipment has disaster tolerance for link functions. When an equipment failure occurs, the service will not reflect the fault situation. At this time, the link may have a single point problem, and once the number of users surges, it may cause capacity failure of the equipment; 2. Fault troubleshooting is mainly based on phenomenon and record tracing, most of which are post-location, and the traceability content is widely distributed, which increases the difficulty of positioning and causes the fault resolution cycle to become longer; 3. It is difficult to collect and analyze evidence for poor quality terminal users and PCDN users who occupy system resources and may cause harm; 4. The investment in the construction of home broadband equipment needs to be based on the growth and changes in the number of user groups. Therefore, it is necessary to have a clear and intuitive perception of the actual status of equipment operation in the current broadband authentication link.

[0034] In order to solve the above problems, ensure the security of broadband services and improve user experience, the present invention proposes a broadband link fault detection method, which mainly analyzes the link data in the authentication and billing message to analyze the real state of the OLT equipment in the broadband service and perceive the operating status of the OLT equipment, so as to realize fast and accurate location of the link fault; and through real-time online situation analysis, actively and timely discover the OLT capacity demand and security risks, and assist the operation and maintenance personnel in reasonable resource allocation, security risk investigation and other tasks.

[0035] In the present invention, Figure 1 FIG. 1 is a flow chart of a broadband link failure detection method according to an embodiment of the present invention. Figure 1 The present invention provides a broadband link fault detection method, comprising:

[0036] Step S11: Collect broadband authentication and billing data;

[0037] Step S12: parsing the broadband authentication and billing data to obtain data of the OLT device in the broadband link;

[0038] Step S13: inputting data associated with the same OLT device into a state prediction model to obtain an estimated state of the same OLT device, wherein the estimated state is: no fault or a risk of fault. The OLT device state prediction model uses data of sample OLT devices and corresponding actual states as training samples, wherein the actual state is no fault or a risk of fault.

[0039] In the embodiment of the present invention, OLT (optical line terminal) is an optical line terminal, which refers to a terminal device used to connect to an optical fiber trunk line. Step S11 collects broadband authentication and billing data of broadband home users. Figure 2 As shown in the figure, after the OLT device turns on the PPPoE+ function, it will insert a RAIO (Relay Agent Information Option) message into the authentication request message of the PPPoE user, carrying relevant location information. The broadband home user starts dialing from the ONU, and its link data will be finally aggregated to the AAA platform through various layers of equipment to achieve accurate binding of broadband home users and link attributes. Due to the stability of the home broadband installation address and link equipment, its link information will basically not change, so the link information in each authentication and billing message of the user is the same. The broadband authentication system maintains the user's online information data based on the user's authentication status, and updates the online information in real time through the intermediate billing package. This online information includes data such as the user's Internet access time and traffic usage, which can reflect the use of the network by broadband home users. When the OLT device of the authentication link is abnormal, its online data will fluctuate, resulting in data deviation year-on-year or month-on-month.

[0040] After collecting the broadband authentication and billing data, step S12 parses and processes each broadband authentication and billing data collected to obtain user information data, bound device information data, authentication information data, and traffic information data in each broadband authentication and billing data. Then, data integration is performed based on the OLT device dimension, and data belonging to the same OLT device is integrated to obtain the data of the OLT device. That is, the OLT device through which each broadband authentication and billing data will pass is determined according to the message transmission path of the broadband authentication and billing data, and the broadband authentication and billing data is integrated into the corresponding OLT device according to the OLT device through which each broadband authentication and billing data will pass.

[0041] For example, the collected broadband authentication and billing data includes data 1, data 2, data 3, data 4, data 5, and data 6, and the OLT device includes device A and device B. The collected broadband authentication and billing data is parsed and processed to obtain the user information data, bound device information data, authentication information data, and traffic information data in each broadband authentication and billing data. The association relationship between each broadband authentication and billing data and the OLT device is determined according to the link through which the broadband authentication and billing data is transmitted, such as determining that data 1, data 2, and data 3 are authenticated by OLT device A, and data 4, data 5, and data 6 are authenticated by OLT device B. When data integration is performed with the OLT device dimension as the main body, the user information data, binding device information data, authentication information data and flow information data corresponding to data 1, data 2 and data 3 belonging to OLT device A obtained through analysis and processing are integrated into the data of OLT device A, that is, the user information data, binding device information data, authentication information data and flow information data corresponding to data 1, data 2 and data 3 are data associated with OLT device A; the user information data, binding device information data, authentication information data and flow information data corresponding to data 4, data 5 and data 6 belonging to OLT device B obtained through analysis and processing are integrated into the data of OLT device B, that is, the user information data, binding device information data, authentication information data and flow information data corresponding to data 4, data 5 and data 6 are data associated with OLT device B.

[0042] Step S13 inputs the associated data belonging to the same OLT device into the pre-trained state prediction model of the present invention to obtain the predicted state of the same OLT device. The predicted state of the OLT device includes two states: no fault or the risk of fault.

[0043] In an embodiment of the present invention, a state prediction model for predicting the state of an OLT device uses data of a sample OLT device and actual states corresponding to the data of the sample OLT device as training samples, wherein the actual states include two states: no fault or a risk of fault.

[0044] In an embodiment of the present invention, the present invention mainly constructs corresponding 6 features based on the data of the OLT device to train the state prediction model. The 6 features include feature 1: taking the OLT device as the recording unit, recording the feature of the change trend of the number of broadband online at the minute level; feature 2: taking the OLT device as the recording unit, recording the feature of the change trend of the number of broadband authentication successes at the minute level; feature 3: taking the OLT device as the recording unit, recording the feature of the change trend of the number of broadband authentication failures at the minute level; feature 4: taking the OLT device as the recording unit, recording the feature of the change trend of the number of broadband offline at the minute level; feature 5: taking the OLT device as the recording unit, recording the feature of the change trend of the number of broadband abnormal offline at the minute level; feature 6: taking the OLT device as the recording unit, recording the feature of the change trend of the number of broadband online at the minute level. By acquiring a large amount of historical data of OLT devices, and processing the large amount of historical data of OLT devices to obtain the feature quantities of the above 6 main features corresponding to the data of the large amount of historical OLT devices, the initial model is trained according to the feature quantities of the obtained 6 main features and the actual fault results to obtain the state prediction model of the present invention. The working state of the OLT device is predicted by using the state prediction model established by the present invention to determine whether the OLT device has a risk of failure.

[0045] In an embodiment of the present invention, since the changing trends of a large amount of data are needed when predicting faults of OLT equipment, and in order to improve the prediction efficiency of fault prediction, it is particularly important to quickly count the changing trends. Therefore, in order to quickly realize the statistics of changing trends of massive data, the present invention simultaneously creates asynchronous warehousing and a column-based database, creates online data, authentication data, and inventory data in batches every minute, and imports the data into the column-based database in batches. The data imported into the column-based database is quickly statistically analyzed in the column-based database to obtain minute-level feature change trends, and then the feature change data corresponding to the feature change trends are extracted, and the feature change data is input into the state prediction model created by the present invention to predict the working state of the OLT equipment.

[0046] In an embodiment of the present invention, when the number of online users drops sharply and is cleared to zero after a period of time, it means that the OLT device in the link fails, resulting in the inability to transmit the authentication information backward in the link; there is an OLT device in the authentication information, and the device information check shows that it is offline, indicating that the OLT is in an out-of-control state; and when the number of online users drops sharply and is not cleared to zero after a period of time, it means that some optical ports of the OLT device in the link are abnormal, resulting in a sharp drop in the number of online users.

[0047] For example, continuing with the above example, by inputting the user information data, binding device information data, authentication information data and traffic information data corresponding to the above data 1, data 2 and data 3 into the state prediction model, the estimated state of OLT device A is obtained as no fault; by inputting the user information data, binding device information data, authentication information data and traffic information data corresponding to the above data 4, data 5 and data 6 into the state prediction model, the estimated state of OLT device B is obtained as the risk of fault.

[0048] A broadband link fault detection method provided by an embodiment of the present invention collects broadband authentication and billing data; analyzes the collected broadband authentication and billing data to obtain data related to an OLT device in a broadband link; inputs data associated with the same OLT device into a state prediction model to obtain an estimated state of the same OLT device, wherein the estimated state is: no fault or a risk of failure; the OLT device state prediction model uses data of a sample OLT device and a corresponding actual state as training samples, wherein the actual state is no fault or a risk of failure. The state prediction model can predict in advance whether an OLT device has a risk of failure based on the broadband authentication and billing data generated by a user, thereby improving the security of broadband services.

[0049] In the present invention, the method also includes: inputting data of an OLT device whose estimated state is that there is a risk of failure into a situation prediction model to obtain a situation prediction result of the OLT device, wherein the situation prediction result characterizes the state change of the OLT device in a future period of time, and the situation prediction model uses the data of the sample OLT device in a first historical time period and the state change in a second historical time period as training samples.

[0050] In an embodiment of the present invention, when it is determined through a state prediction model that an OLT device is at risk of failure, a further prediction is made of the subsequent development of the situation of the OLT device, so that the technician can make corresponding countermeasures based on the predicted subsequent development of the situation of the OLT device. Specifically, the data of the OLT device whose estimated state is at risk of failure is continuously input into the situation prediction model to obtain the situation prediction result of the OLT device, and the situation prediction result characterizes the state change of the OLT device in the future period of time, such as whether the future online state of the OLT is a stable state, a steep drop state, a growth state, etc. By analyzing the data of the existing OLT device, the trend prediction result of the online state of the future OLT device is obtained, and the development trend of the OLT device is measured in three states: a stable state, a steep drop state, and a growth state, so as to determine the subsequent response strategy for the OLT device.

[0051] The situation prediction model uses the data of the sample OLT device in the first historical time period and the state change in the second historical time period as training samples. The first historical time period is earlier than the second historical time period, and the first historical time period and the second historical time period are two independent time periods, and there is no intersection between the two.

[0052] For example, continuing with the above example, when the estimated state of OLT device B is obtained as a risk of failure, the data of LOT device B is input into the situation prediction model to obtain the situation prediction result of OLT device B, such as whether the online status of the OLT device in the future period is stable, sharply declining, or growing.

[0053] In the present invention, the method further comprises: when the situation prediction result of the OLT device is a steep drop state, outputting a prompt for troubleshooting the hardware device or the network.

[0054] In an embodiment of the present invention, when the dynamic prediction result of the OLT device is a steep drop state, it indicates that there is a fault in the OLT device, resulting in a sharp drop in the number of online users. At this time, a prompt for troubleshooting the hardware device and / or the network is output so that the technical staff can respond in time to quickly troubleshoot and resolve the abnormality to avoid further expansion of the abnormal situation.

[0055] In the present invention, the method further includes: when the situation prediction result of the OLT device is a growth state, monitoring whether the service flow reaches the expansion threshold; when the service flow reaches the expansion threshold, outputting a hardware device expansion prompt.

[0056] In an embodiment of the present invention, when the status prediction result of the OLT device is a growth state, it is further monitored whether the service traffic has reached the expansion threshold. When the service traffic has reached the expansion threshold, it means that the current OLT device has reached the upper limit of its working capacity. At this time, a prompt message of hardware device expansion is output to meet subsequent user needs.

[0057] In the present invention, the online duration of a target user terminal mounted on an OLT device is input into an online duration processing model of the OLT device to determine whether a user account of the target user terminal is abnormal. The online duration processing model uses historical online information of each user terminal mounted on the OLT device and normal labels or abnormal labels carried by user accounts of each user terminal as training samples.

[0058] In an embodiment of the present invention, since the existence of abnormal accounts will cause frequent authentication and / or traffic anomalies, in order to quickly locate abnormal accounts, the present invention inputs the online duration of the target user terminal mounted on an OLT device into the online duration processing model of the OLT device, and processes it through the online duration processing model of the OLT device. The online duration processing model of the OLT device will output the result of whether the user account of the target user terminal is abnormal, thereby timely locating the location of the abnormal account. The online duration processing model is obtained after qualified training using the historical online information of each user terminal mounted on the OLT device and the normal labels or abnormal labels carried by the user accounts of each user terminal as training samples. Through the online duration processing model of the present invention, the user status within the data statistical period is analyzed, so as to identify accounts that may have risks such as frequent authentication and traffic anomalies, thereby achieving rapid positioning of fault problems.

[0059] In the present invention, the method further includes: counting the online duration and the number of terminals of user terminals mounted on the same OLT device at preset time intervals; and displaying the terminal mounting information of the same OLT device, wherein the terminal mounting information includes: the online duration and the number of terminals.

[0060] In an embodiment of the present invention, in order to intuitively reflect the situation of the number of online devices of each device, so that technical personnel can make correct response strategies. The present invention counts the online duration and terminal quantity of user terminals mounted on the same OLT device according to preset time intervals, and displays the online duration and terminal quantity of user terminals under the unified OLT device obtained by statistics, and clearly displays the situation of the online quantity of each OLT device in a visual manner, and intuitively reflects the situation of the online quantity of each OLT device, so as to facilitate real-time prediction of the development of broadband services. The terminal mounting information includes: the online duration of each device and the terminal quantity of terminal devices under the same OLT device. It should be understood that the preset time interval can be set according to actual needs, such as setting 5 minutes or 10 minutes or 15 minutes as the preset time interval, which is not specifically limited here.

[0061] In an embodiment of the present invention, an implementation method of a broadband link fault detection method provided by the present invention is: collecting broadband authentication and billing data, parsing and processing the collected broadband authentication and billing data to obtain user information data, binding device information data, authentication information data and flow information data, and then determining the OLT devices through which each broadband authentication and billing data passes according to the transmission link of each broadband authentication and billing data. Then, taking the OLT device dimension as the main body, the parsing and processing results corresponding to all broadband authentication and billing data passing through the same OLT device are integrated into the same OLT device to form the data of the same OLT device for subsequent state prediction.

[0062] After completing data integration to obtain the data of the OLT device, the data of the OLT device obtained is input into the state prediction model pre-created by the present invention to obtain the estimated state of the OLT device. When the estimated state of the OLT device is fault-free, the operation is terminated, and when the estimated state of the OLT device is at risk of failure, the situation prediction of the OLT device is continued. Specifically, the data of the OLT device is continuously input into the situation prediction model pre-created by the present invention to obtain the situation prediction result of the OLT device, and the situation prediction result characterizes the state change of the OLT device in the future period of time, such as the number of online users belonging to the OLT device will drop sharply or will increase or will not fluctuate greatly in the future period of time. In the case where the situation prediction result of the OLT device is a steep drop state, prompt information for troubleshooting the hardware device or network is output accordingly. In the case where the situation prediction result of the OLT device is a growth state, whether the service flow reaches the amplification threshold is further monitored, and when the service flow reaches the amplification threshold, the hardware device amplification prompt information is output. At the same time, the online duration of the target user terminal mounted on the same OLT device is input into the online duration processing model of the OLT device to determine whether the user account of the target user terminal is abnormal. At the same time, in order to intuitively reflect the status of the number of online devices and predict the business development in real time, the present invention also counts the online duration and terminal quantity of the user terminals mounted on the same OLT device according to a preset time interval, and visualizes the statistical results.

[0063] A broadband link fault detection method provided by an embodiment of the present invention collects broadband authentication and billing data; analyzes the collected broadband authentication and billing data to obtain data related to the OLT device in the broadband link; inputs the data associated with the same OLT device into a state prediction model to obtain the estimated state of the same OLT device. Through the state prediction model, it is possible to predict in advance whether the OLT device is at risk of failure based on the broadband authentication and billing data generated by the user, thereby improving the security of broadband services. The broadband link fault detection method provided by an embodiment of the present invention takes broadband link binding as the technical basis, collects and analyzes home broadband authentication and billing message data, uses the user's online information and link binding attributes, and combines the information of the OLT device to extract, identify, clean, build a model and perform streaming calculation on the service data, thereby realizing dynamic situation perception of the broadband access platform with the OLT device dimension as the main body. And through the online duration processing model, the real-time online data of the user is analyzed to realize the abnormal perception of the broadband network, provide a basis for user group fault warning and capacity expansion suggestions, and further improve customer satisfaction during broadband use.

[0064] A second aspect of an embodiment of the present invention provides a broadband link fault detection system, such as Figure 3 As shown, the system 300 includes:

[0065] Data collection module 301, used to collect broadband authentication and billing data;

[0066] The data analysis module 302 is used to analyze the broadband authentication and billing data to obtain the data of the OLT device in the broadband link;

[0067] The state prediction module 303 is used to input data associated with the same OLT device into a state prediction model to obtain an estimated state of the same OLT device, wherein the estimated state is: no fault or a risk of failure. The OLT device state prediction model uses data of sample OLT devices and corresponding actual states as training samples, and the actual state is no fault or a risk of failure.

[0068] Optionally, the system 300 further includes:

[0069] The situation prediction module is used to input the data of the OLT device whose estimated state is that there is a risk of failure into the situation prediction model to obtain the situation prediction result of the OLT device, and the situation prediction result characterizes the state change of the OLT device in the future period of time. The situation prediction model uses the data of the sample OLT device in the first historical time period and the state change in the second historical time period as training samples.

[0070] Optionally, the system 300 further includes:

[0071] The first information output module is used to output a prompt for troubleshooting hardware devices or networks when the situation prediction result of the OLT device is a steep drop state.

[0072] Optionally, the system 300 further includes:

[0073] A monitoring module, used to monitor whether the service traffic reaches the expansion threshold when the situation prediction result of the OLT device is a growth state;

[0074] The second information output module is used to output a hardware device expansion prompt when the business traffic reaches the expansion threshold.

[0075] Optionally, the system 300 further includes:

[0076] The abnormality monitoring module is used to input the online time of the target user terminal mounted on an OLT device into the online time processing model of the OLT device to determine whether the user account of the target user terminal is abnormal. The online time processing model uses the historical online information of each user terminal mounted on the OLT device and the normal label or abnormal label carried by the user account of each user terminal as training samples.

[0077] Optionally, the system 300 further includes:

[0078] The data statistics module is used to count the online duration and number of user terminals mounted on the same OLT device according to preset time intervals;

[0079] The third information output module is used to display the terminal mounting information of the same OLT device, and the terminal mounting information includes: online time and terminal quantity.

[0080] A third aspect of an embodiment of the present invention provides an electronic device, comprising a memory, a processor, and a program stored in the memory and executable on the processor, wherein when the processor executes the program, the steps in a broadband link failure detection method described in the first aspect of the present invention are implemented.

[0081] A fourth aspect of an embodiment of the present invention provides a readable storage medium, on which a program is stored. When the program is executed by a processor, the steps in the broadband link failure detection method described in the first aspect of the present invention are implemented.

[0082] The various embodiments in this specification are described in a progressive manner, and each embodiment focuses on the differences from other embodiments. The same or similar parts between the various embodiments can be referenced to each other.

[0083] Finally, it should be noted that, in this article, relational terms such as first and second, etc. are only used to distinguish one entity or operation from another entity or operation, and do not necessarily require or imply any such actual relationship or order between these entities or operations. Moreover, the terms "include", "comprise" or any other variants thereof are intended to cover non-exclusive inclusion, so that a process, method, article or terminal device including a series of elements includes not only those elements, but also other elements not explicitly listed, or also includes elements inherent to such process, method, article or terminal device. In the absence of further restrictions, the elements defined by the sentence "comprise a ..." do not exclude the existence of other identical elements in the process, method, article or terminal device including the elements.

[0084] The above is a detailed introduction to a broadband link fault detection method, system, electronic device and readable storage medium provided by the present invention. Specific examples are used in this article to illustrate the principles and implementation methods of the present invention. The description of the above embodiments is only used to help understand the method of the present invention and its core idea. At the same time, for those skilled in the art, according to the idea of ​​the present invention, there will be changes in the specific implementation method and application scope. In summary, the content of this specification should not be understood as a limitation on the present invention.

Claims

1. A broadband link failure detection method, characterized in that: The method comprises: Collect broadband authentication and billing data; Parsing the broadband authentication and billing data to obtain data of the OLT device in the broadband link; Inputting data associated with the same OLT device into a state prediction model to obtain an estimated state of the same OLT device, wherein the estimated state is: no fault or a risk of fault; the OLT device state prediction model uses data of sample OLT devices and corresponding actual states as training samples, wherein the actual state is no fault or a risk of fault; Among them, the state prediction model is trained by constructing corresponding 6 features based on the data of the OLT device, and the 6 features include feature 1: taking the OLT device as the recording unit, recording the characteristics of the change trend of the number of broadband online devices at the minute level; feature 2: taking the OLT device as the recording unit, recording the characteristics of the change trend of the number of broadband authentication successes at the minute level; feature 3: taking the OLT device as the recording unit, recording the characteristics of the change trend of the number of broadband authentication failures at the minute level; feature 4: taking the OLT device as the recording unit, recording the characteristics of the change trend of the number of broadband offline devices at the minute level; feature 5: taking the OLT device as the recording unit, recording the characteristics of the change trend of the number of broadband abnormal offline devices at the minute level; feature 6: taking the OLT device as the recording unit, recording the characteristics of the change trend of the number of broadband online devices at the minute level.

2. The method according to claim 1, characterized in that The method further comprises: The data of the OLT device whose estimated state is that there is a risk of failure is input into the situation prediction model to obtain the situation prediction result of the OLT device, and the situation prediction result characterizes the state change of the OLT device in a future period of time. The situation prediction model uses the data of the sample OLT device in a first historical time period and the state change in a second historical time period as training samples.

3. The method according to claim 2, characterized in that The method further comprises: When the status prediction result of the OLT device is a steep drop state, a prompt for troubleshooting the hardware device or network is output.

4. The method according to claim 2, characterized in that: The method further comprises: When the situation prediction result of the OLT device is a growth state, monitor whether the service traffic reaches the expansion threshold; When the service traffic reaches the expansion threshold, a hardware device expansion prompt is output.

5. The method according to claim 1, characterized in that Also includes: The online duration of the target user terminal mounted on an OLT device is input into the online duration processing model of the OLT device to determine whether the user account of the target user terminal is abnormal. The online duration processing model uses the historical online information of each user terminal mounted on the OLT device and the normal labels or abnormal labels carried by the user accounts of each user terminal as training samples.

6. The method according to claim 1, characterized in that The method further comprises: According to the preset time interval, count the online time and number of user terminals mounted on the same OLT device; The terminal mounting information of the same OLT device is displayed, and the terminal mounting information includes: online time and terminal quantity.

7. A broadband link fault detection system, characterized in that: The system comprises: Data collection module, used to collect broadband authentication and billing data; A data analysis module, used to analyze the broadband authentication and billing data to obtain data of the OLT device in the broadband link; The state prediction module is used to input data associated with the same OLT device into a state prediction model to obtain the estimated state of the same OLT device, wherein the estimated state is: no fault or the risk of fault. The OLT device state prediction model uses the data of the sample OLT device and the corresponding actual state as training samples, wherein the actual state is no fault or the risk of fault. The state prediction model is trained based on the data of the OLT device to construct corresponding 6 features, wherein the 6 features include feature 1: taking the OLT device as the recording unit, recording the feature of the trend of the number of broadband online changes at the minute level; feature 2: taking the OLT device as the recording unit, recording the feature of the trend of the number of broadband successful authentication changes at the minute level; feature 3: taking the OLT device as the recording unit, recording the feature of the trend of the number of broadband authentication failure changes at the minute level; feature 4: taking the OLT device as the recording unit, recording the feature of the trend of the number of broadband offline changes at the minute level; feature 5: taking the OLT device as the recording unit, recording the feature of the trend of the number of broadband abnormal offline changes at the minute level; feature 6: taking the OLT device as the recording unit, recording the feature of the trend of the number of broadband online changes at the minute level.

8. An electronic device comprising a memory, a processor, and a program stored in the memory and executable on the processor, characterized in that: When the processor executes the program, the steps in the broadband link failure detection method as described in any one of claims 1-6 are implemented.

9. A readable storage medium having a program stored thereon, characterized in that: When the program is executed by a processor, the steps in a broadband link failure detection method as described in any one of claims 1 to 6 are implemented.

Images