A method, device and equipment for locating fault of ODN in PON and storage medium

Abstract

The application provides a method and device for locating faults of an ODN in a PON, equipment and a storage medium, relates to the field of big data processing, and can accurately establish a topology of the ODN and locate faults of the ODN based on the topology, thereby improving the accuracy of fault location. The PON includes an OLT, an ODN, and a plurality of user-side devices. The method includes: obtaining characteristic data; determining a first correlation and a second correlation based on the characteristic data; determining the topology of the ODN based on the first correlation and the second correlation; determining a fault cause based on a fault optical path index characteristic model, the transmission and reception optical power of each PON port of the OLT within a first preset time period, and the transmission and reception optical power of each user-side device in the plurality of user-side devices within the first preset time period; and locating faults of the ODN based on the fault cause and the topology of the ODN. The application can be used in the process of determining the fault point of the ODN and can solve the problem of inaccurate fault location of the ODN.

Description

Technical Field

[0001] This application relates to the field of big data processing, and in particular to a method, apparatus, device and storage medium for fault location in an optical distribution network (ODN) of a passive optical network (PON). Background Technology

[0002] With the development of network technology, internet services are booming, and new services are emerging one after another. Consequently, people's demand for network speed is also increasing. Currently, Fiber to the Home (FTTH) technology can extend fiber optic cables further into users' homes to meet their growing network needs, and is therefore gradually becoming more widely used. FTTH technology has two implementation methods: Active Optical Network (AON) and PON. Due to its low cost and high performance, PON is widely used in FTTH technology.

[0003] PON comprises an Optical Line Terminal (OLT), user-side equipment, and an Optical Distribution Network (ODN). The ODN is an optical transmission network composed of passive components such as optical fibers and splitters, used to connect the OLT and user-side equipment. In PON, the optical paths of the ODN are relatively dispersed, and the environment is complex, making it prone to optical path failures. Therefore, locating optical path failures in the ODN is crucial for determining the fault point in the PON.

[0004] Existing technologies employ intelligent ODN (Optical Distribution Network) technology for locating optical path faults in the ODN. Specifically, this involves adding electronic tags to passive components in the ODN, such as optical fibers and splitters. Technicians manually scan these tags and then fill in the fault information online to locate the fault point in the PON (Personal Optical Network). However, the electronic tags used in intelligent ODN technology have a limited lifespan and are prone to loss. Furthermore, their effectiveness relies excessively on the accuracy of manual operation, making the results unpredictable. Therefore, intelligent ODN technology cannot effectively solve the problem of locating optical path faults in the ODN. Summary of the Invention

[0005] This application provides a method, apparatus, device, and storage medium for fault location in an ODN in a PON. The method can accurately establish the topology of the ODN and locate optical path faults in the ODN based on it, thereby determining the fault point of the ODN and improving the accuracy of optical path fault location.

[0006] In a first aspect, this application provides a fault location method for an optical distribution network (ODN) in a passive optical network (PON). The PON includes an optical line terminal (OLT), an ODN, and multiple user-side devices. The method includes: acquiring feature data; wherein the feature data is used to indicate the optical power information of each PON port in all PON ports of the OLT and the optical power information of each user-side device in the multiple user-side devices within a first preset time period; determining a first correlation and a second correlation based on the feature data; the first correlation is used to indicate the correlation between each PON port of the OLT and each user-side device in the multiple user-side devices, and the second correlation is used to indicate the correlation among the multiple user-side devices; determining the topology of the ODN based on the first correlation and the second correlation; determining the cause of the fault based on the fault optical path index feature model, the received and transmitted optical power of each PON port of the OLT within the first preset time period, and the received and transmitted optical power of each user-side device in the multiple user-side devices within the first preset time period; and locating the fault in the ODN based on the cause of the fault and the topology of the ODN.

[0007] In another possible implementation, the characteristic data includes: the transmit and transmit power of each user-side device among multiple user-side devices, the time when the transmit and transmit power of each user-side device occurs, the transmit and transmit power of each PON port among all PON ports of the OLT, and the time when the transmit and transmit power of each PON port occurs.

[0008] In another possible implementation, the first correlation and the second correlation include the correlation of multiple sets of optical power. Each set of optical power includes a first optical power and a second optical power. The first optical power and the second optical power are any two different transmit and transmit power values ​​among the transmit and transmit power values ​​of each user-side equipment and each PON port. Determining the first correlation and the second correlation based on feature data includes: determining the correlation of multiple sets of optical power based on feature data. Among them, determining the correlation of a set of optical power includes: determining the average time difference between the transactions of the first optical power and the second optical power and the number of times the transactions of the first optical power and the second optical power occur simultaneously, based on the time difference between the transactions of the first optical power and the second optical power; determining the correlation of the first optical power and the second optical power based on the average time difference between the transactions of the first optical power and the second optical power, the number of times the transactions of the first optical power and the second optical power occur simultaneously, and the total number of transactions of all transmit and transmit power values.

[0009] In another possible implementation, the average time difference between the first optical power and the second optical power transactions is determined using the following formula:

[0010]

[0011] Where Δt is the average time difference between the transactions of the first optical power and the second optical power, n-1 is the number of time periods into which the first preset time is averaged, and t mi t represents the time during which the first optical power experiences a transaction within the m-th time period divided by the first preset time period. mj The time during which the second optical power occurs within the m-th time period divided by the first preset time period, where n is an integer greater than or equal to 2 and m is an integer greater than or equal to 1 and less than or equal to n-1.

[0012] In another possible implementation, the correlation between the first optical power and the second optical power is determined using the following formula;

[0013]

[0014] Among them, Cor d (i, j) represents the correlation between the first optical power and the second optical power, T ij |t| represents the number of times the first optical power and the second optical power simultaneously occur, ||T|| represents the total number of times all received and received optical powers occur, and |t| represents the total number of times transactions occur. n -t n-1 | represents the length of the time period into which the first preset time is evenly divided, N ij The first optical power and the second optical power are represented by the correlation coefficient, and Δt is the average time difference between the first optical power and the second optical power transactions.

[0015] In another possible implementation, before determining the correlation between the first optical power and the second optical power, the above method further includes: obtaining the original data of the PON, which includes the service connection relationship and physical connection relationship between the network node corresponding to the first optical power and the network node corresponding to the second optical power; and determining the correlation coefficient between the first optical power and the second optical power based on the service connection relationship and the physical connection relationship.

[0016] In one possible implementation, the ODN includes multiple optical splitters, and the feature data further includes the optical distance of each user-side device. The method further includes: determining the probability that any two or more user-side devices among the multiple user-side devices can be connected to the same optical splitter based on the optical distance of each user-side device; determining the topology of the ODN based on a first correlation and a second correlation, including: determining the topology of the ODN based on the first correlation, the second correlation, and the probability that any two or more user-side devices among the multiple user-side devices can be connected to the same optical splitter.

[0017] In another possible implementation, before determining the cause of the fault based on the fault optical path index characteristic model, the above method further includes: obtaining the transmit and transmit power of each PON port of the OLT in a first preset time period and the transmit and transmit power of each user-side device in the multiple user-side devices in the first preset time period; and training a model using artificial intelligence (AI) algorithms based on the transmit and transmit power of each PON port of the OLT in the second preset time period and the transmit and transmit power of each user-side device in the multiple user-side devices in the second preset time period to determine the fault optical path index characteristic model.

[0018] The fault location method for ODN in PON provided in this application obtains optical power information from the PON port and the user-side equipment, analyzes the optical power information to determine the correlation between the PON port and the user-side equipment, as well as among the user-side equipment. Based on the correlation, the ODN topology can be accurately reconstructed. The cause of the fault is determined through an optical path fault indicator feature model, and then, combined with the fault cause and the ODN topology, fault location in the ODN is achieved. This method uses a large amount of objective data to reconstruct the ODN topology, resulting in a more objective and accurate topology, unaffected by human factors. Thus, by determining the fault location in the ODN based on the ODN topology, the fault point in the PON can be accurately located, improving the accuracy of ODN optical path fault location. Furthermore, this allows technicians to maintain the network promptly at the fault point, providing users with a better broadband internet experience.

[0019] Furthermore, this scheme considers the correlation between optical power at both temporal and spatial levels by incorporating the time and frequency of optical power transactions, thereby obtaining the correlation between corresponding devices. It also introduces a correlation coefficient, including the original data from PON establishment in the correlation assessment, making the obtained device correlations more comprehensive and accurate, thus further improving the accuracy of the established ODN topology. Moreover, feature data can be acquired in real time; if the PON network changes, this scheme can update the ODN topology promptly, making ODN fault location more accurate. In terms of cost, compared to traditional intelligent ODN technology, this scheme does not require expensive purchases of specific equipment; it only needs to collect and analyze feature data from the existing network management system. This makes it highly usable for operators and can address the increased network maintenance costs caused by rapid network upgrades. Simultaneously, this scheme has a long service life; as the amount of feature data generated in the PON increases over time, the accuracy of the artificial intelligence algorithm continuously improves through learning, leading to increasingly better application results.

[0020] Secondly, this application provides a fault location device for fault location of a PON, wherein the PON includes an OLT, an ODN including multiple optical splitters, and multiple user-side devices. The device includes: an acquisition module, a determination module, and a location module; the acquisition module is used to acquire characteristic data of each PON port in all PON ports of the OLT within a first preset time period; wherein the characteristic data is used to indicate the optical power information of each PON port of the OLT and the optical power information of each user-side device among the multiple user-side devices within the first preset time period; the determination module is used to determine a first correlation and a second correlation based on the characteristic data; the first correlation is used to indicate the correlation between each PON port of the OLT and each user-side device among the multiple user-side devices, and the second correlation is used to indicate the correlation among the multiple user-side devices; the determination module is further used to determine the topology of the ODN based on the first correlation and the second correlation; the determination module is further used to determine the cause of the fault based on the fault optical path index characteristic model, the transmit and transmit power of each PON port of the OLT within the first preset time period, and the transmit and transmit power of each user-side device among the multiple user-side devices within the first preset time period; the location module is used to locate the fault in the ODN based on the cause of the fault and the topology of the ODN.

[0021] In one possible implementation, the characteristic data includes: the transmit and transmit power of each user-side device among multiple user-side devices, the time of transaction occurrence of the transmit and transmit power of each user-side device, the transmit and transmit power of each PON port among all PON ports of the OLT, and the time of transaction occurrence of the transmit and transmit power of each PON port.

[0022] In another possible implementation, the first correlation and the second correlation include the correlation of multiple sets of optical power. Each set of optical power includes a first optical power and a second optical power. The first optical power and the second optical power are any two different transmit and transmit power values ​​among the transmit and transmit power values ​​of each user-side equipment and each PON port. The determining module is specifically used to determine the correlation of multiple sets of optical power based on feature data. Specifically, determining the correlation of a set of optical power includes: the determining module is specifically used to determine the average time difference between the transactions of the first optical power and the second optical power, and the number of times the transactions of the first optical power and the second optical power occur simultaneously, based on the time of the transaction of the first optical power and the time of the transaction of the second optical power; the determining module is specifically used to determine the correlation of the first optical power and the second optical power based on the average time difference between the transactions of the first optical power and the second optical power, the number of times the transactions of the first optical power and the second optical power occur simultaneously, and the total number of transactions of all transmit and transmit power values.

[0023] In another possible implementation, the determining module is specifically used to determine the average time difference between the occurrence of transactions of the first optical power and the second optical power using the following formula:

[0024]

[0025] Where Δt is the average time difference between the transactions of the first optical power and the second optical power, n-1 is the number of time periods into which the first preset time is averaged, and t mi Let t be the time during which a transaction occurs within the m-th time period divided by the first preset time period, where the first optical power is at that time. mj The time when the second optical power occurs within the m-th time period divided by the first preset time period, where n is an integer greater than or equal to 2 and m is an integer greater than or equal to 1 and less than or equal to n-1.

[0026] In another possible implementation, the determining module is specifically used to determine the correlation between the first optical power and the second optical power using the following formula;

[0027]

[0028] Among them, Cor d (i, j) represents the correlation between the first optical power and the second optical power, T ij |t| represents the number of times the first optical power and the second optical power simultaneously occur, ||T|| represents the total number of times all received and received optical powers occur, and |t| represents the total number of times transactions occur. n -t n-1 | represents the length of the time period into which the first preset time is evenly divided, N ijThe first optical power and the second optical power are represented by the correlation coefficient, and Δt is the average time difference between the first optical power and the second optical power transactions.

[0029] In another possible implementation, the acquisition module is also used to acquire the raw data of the PON, which includes the service connection relationship and physical connection relationship between the network node corresponding to the first optical power and the network node corresponding to the second optical power; the network node is a PON port or a user-side device; the determination module is also used to determine the correlation coefficient between the first optical power and the second optical power based on the service connection relationship and the physical connection relationship.

[0030] In another possible implementation, the ODN includes multiple optical splitters, and the feature data also includes: the optical distance of each user-side device; the determination module is also used to determine the probability that any two or more user-side devices among the multiple user-side devices can be connected to the same optical splitter based on the optical distance of each user-side device; the determination module is specifically used to determine the topology of the ODN based on the first correlation, the second correlation, and the probability that any two or more user-side devices among the multiple user-side devices can be connected to the same optical splitter.

[0031] In another possible implementation, the acquisition module is further configured to acquire the transmit and transmit power of each PON port in all PON ports of the OLT during a second preset time period and the transmit and transmit power of each user-side device among multiple user-side devices during the second preset time period; the determination module is further configured to determine the fault optical path index feature model by training a model through an AI algorithm based on the transmit and transmit power of each PON port in all PON ports of the OLT during the second preset time period and the transmit and transmit power of each user-side device among multiple user-side devices during the second preset time period.

[0032] Thirdly, this application provides an electronic device comprising: a processor and a memory; the memory storing processor-executable instructions; when the processor is configured to execute the instructions, causing the electronic device to implement the method of the first aspect described above.

[0033] Fourthly, this application provides a computer-readable storage medium comprising: computer software instructions; which, when executed in an electronic device, cause the electronic device to implement the method described in the first aspect.

[0034] The beneficial effects of the second to fourth aspects mentioned above can be referred to the corresponding description of the first aspect, and will not be repeated here. Attached Figure Description

[0035] To more clearly illustrate the technical solutions in the embodiments of this application or the prior art, the drawings used in the description of the embodiments or the prior art will be briefly introduced below. Obviously, the drawings described below are only some embodiments of this application. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0036] Figure 1 A schematic diagram of a PON structure is provided for this application;

[0037] Figure 2 A schematic diagram illustrating the application environment of a fault location method for ODN in a PON provided in this application;

[0038] Figure 3 A schematic flowchart of a fault location method for ODN in PON provided in this application;

[0039] Figure 4 A schematic diagram of another fault location method for ODN in PON provided in this application;

[0040] Figure 5 A schematic diagram of an ODN topology provided in this application;

[0041] Figure 6 A schematic diagram of another ODN topology provided in this application;

[0042] Figure 7 A line graph illustrating the time-series characteristics of the optical power of a network node provided in this application;

[0043] Figure 8 A line graph illustrating the time-series characteristics of the optical power of another network node provided in this application;

[0044] Figure 9 A line graph for multi-optical path quality monitoring provided in this application;

[0045] Figure 10 A schematic diagram of another fault location method for ODN in PON provided in this application;

[0046] Figure 11 A schematic diagram of the composition of a fault location device provided in this application;

[0047] Figure 12 This is a schematic diagram of the composition of an electronic device provided in this application. Detailed Implementation

[0048] The technical solutions of the embodiments of this application will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of this application, and not all embodiments. Based on the embodiments of this application, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of this application.

[0049] It should be noted that in the embodiments of this application, the words "exemplarily" or "for example" are used to indicate examples, illustrations, or explanations. Any embodiment or design scheme described as "exemplarily" or "for example" in the embodiments of this application should not be construed as being more preferred or advantageous than other embodiments or design schemes. Specifically, the use of the words "exemplarily" or "for example" is intended to present the relevant concepts in a specific manner.

[0050] To facilitate a clear description of the technical solutions of the embodiments of this application, the terms "first" and "second" are used in the embodiments of this application to distinguish the same or similar items with essentially the same function and effect. Those skilled in the art can understand that the terms "first" and "second" are not intended to limit the quantity or execution order.

[0051] As described in the background section, ODN is a crucial component of PON, providing an optical transmission channel between the OLT and user-side equipment. Currently, PON networking in China is divided into two architectures: primary splitting network and secondary splitting network. Technically, it is divided into Ethernet Passive Optical Network (EPON) and Gigabit-Capable Passive Optical Networks (GPON). PON networks built using EPON typically have a splitting ratio of 1:32, while PON networks built using GPON typically have a splitting ratio of 1:64.

[0052] Taking a two-stage optical splitting network as an example, the basic structure of PON can be as follows: Figure 1As shown, the network includes an OLT (Optical Line Terminal), Optical Distribution Frame (ODF), optical distribution box, primary optical splitter, secondary optical splitter, and user-side equipment (such as Optical Network Units (ONUs) and Optical Network Terminals (ONTs)) connected by optical fibers. The network consisting of optical fibers and splitters between the OLT and the user-side equipment constitutes the ODN (Optical Distribution Network). The ODN comprises three parts: backbone, branches, and drop-off. Specifically, the OLT to the primary optical splitter forms the ODN backbone, the primary optical splitter to the secondary optical splitter forms the ODN branch, and the secondary optical splitter to the optical network terminal forms the ODN drop-off.

[0053] In the optical path sections near the user, namely the branch and drop-in parts of the ODN, the optical paths are relatively dispersed and the environment is complex, making them prone to optical path quality issues. This can lead to frequent broadband disconnections, substandard network speed tests, or webpage lag at the user end. Furthermore, the OLT and user-side equipment in the PON can also malfunction. If problematic optical paths cannot be located in a timely manner, the quality of the PON will only deteriorate further, and the quality of the services it carries will decline accordingly, significantly impacting the user's service experience.

[0054] In an ODN (Optical Distribution Network), components such as ODFs (Optical Distribution Frames), optical distribution boxes, splitters, and optical fibers are all passive. Because passive components do not require power, it's impossible to obtain relevant network information in real-time using technical means, making it difficult to promptly locate problematic optical paths. Currently, intelligent ODN technology is generally used to locate ODN faults. Intelligent ODN technology uses electronic tags attached to passive components. These tags are scanned and identified by corresponding electronic devices to record and update network information for relevant components in the ODN, enabling the location of problematic optical paths. For example, during network construction or maintenance, technicians use handheld electronic devices to scan the tags and fill in or update the network information of the corresponding passive components online. This tag-based approach allows for the reconstruction of the ODN topology and fault location.

[0055] However, current intelligent ODN technology suffers from the following problems: 1. Electronic tags have a limited lifespan and are prone to loss. 2. The equipment used in intelligent ODN technology is expensive, and the rapid pace of ODN construction and iteration leads to a faster rate of equipment replacement, significantly increasing the cost of intelligent ODN technology. 3. The effectiveness of fault location in intelligent ODN technology depends heavily on the responsibility and accuracy of the technical personnel; excessive reliance on manual labor makes fault location unpredictable. Therefore, existing intelligent ODN technology cannot effectively solve the fault location problem in ODN.

[0056] Against this background, this application provides a method for fault location in an ODN (Optical Distribution Network) within a PON (Passive Optical Network). This method can accurately and effectively reconstruct the ODN topology (or network topology diagram), and based on the reconstructed topology, optical path faults in the ODN can be located to pinpoint the fault location, thus improving the accuracy of optical path fault location. This facilitates operators in maintaining the normal operation of the passive optical network and provides users with a better broadband internet experience.

[0057] The fault location method provided in this application can be applied to, for example... Figure 2 The application environment shown. For example... Figure 2 As shown, the application environment may include: fault location device 201, network management system 202, and passive optical network (i.e., PON) 203.

[0058] The fault location device 201 can be a server cluster consisting of multiple servers, a single server, a computer, or a processor or processing chip in a server or computer. This application does not limit the specific device form of the fault location device 201. Figure 2 The fault location device 201 is illustrated as a single server. The aforementioned network management system 202 can be a server cluster composed of multiple servers, or a single server (such as...). Figure 2 (As shown), or a computer, or a server, or a processor or processing chip in a computer, etc. This application does not limit the specific device form of the network management system 202.

[0059] In some embodiments of this application, the network management system 202 can monitor the network status of the passive optical network 203 to store relevant data of the OLT and user-side equipment in the passive optical network 203. When fault location of the ODN is required, the fault location device 201 can obtain relevant data of the PON where the ODN is located from the network management system 202. After obtaining the relevant data, the fault location device 201 can establish the topology of the ODN based on this data using artificial intelligence algorithms, and perform fault location of the ODN based on the established topology and fault optical path indicator feature model to determine the fault point of the ODN.

[0060] In addition, the above Figure 2The application environment illustrated is based on the example where the network management system 202 and the fault location device 201 are different devices. In other embodiments, the network management system 202, which stores relevant data of the passive optical network 203, and the fault location device 201, which determines the ODN fault point, can also be the same device, or two different modules of the same device. This embodiment does not impose specific limitations here. When the network management system 202 and the fault location device 201 are two different modules of the same device, the fault location method in this embodiment is similar to the example above, and will not be described in detail here.

[0061] Figure 3 This is a flowchart illustrating a fault location method for an ODN in a PON, provided as an embodiment of this application. The PON may include an OLT, an ODN, and multiple user-side devices. Figure 3 As shown, the fault location method for ODN in PON provided in this application can be implemented by the above-mentioned fault location device, and may specifically include the following steps:

[0062] S301. Obtain feature data.

[0063] Among them, the feature data is used to indicate the optical power information of each PON port of the OLT and the optical power information of each user-side device among multiple user-side devices within a first preset time period.

[0064] In a PON (Poly-Oriented Network), all components of the ODN (Optical Distribution Network) are passive devices. Passive devices cannot actively report their network status, making it difficult for network operators to obtain the network status of each node in the ODN and hindering effective ODN monitoring. However, the network status of the OLT (Optical Line Terminal) and user-side equipment at both ends of the ODN can be obtained and is generally stored in the PON network management system. The OLT has multiple PON ports, each connecting to multiple user-side devices. Under normal circumstances, the optical power of each device is relatively stable. If the optical power of one device in the network changes, the optical power of other devices connected to that device will be affected. Therefore, based on the above principle, information about the optical power of the devices at both ends can be obtained, such as the optical power of each device over a period of time and the time of transactions. By analyzing the optical power information, the connection relationships between devices can be reconstructed, the ODN topology can be restored, and the fault point of the ODN can be located.

[0065] When fault location is required in the ODN, the fault location device can acquire information on the optical power of each PON port of the OLT and the optical power of each user-side device among multiple user-side devices within a certain period of time, such as a first preset time period.

[0066] S302. Based on the feature data, determine the first correlation and the second correlation.

[0067] The first correlation is used to indicate the correlation between each PON port of the OLT and each user-side device among multiple user-side devices, and the second correlation is used to indicate the correlation among multiple user-side devices.

[0068] After acquiring the feature data, the fault location device can determine the first correlation between each PON port and each user-side device, and the second correlation between each user-side device, based on the feature data. The first correlation characterizes the connection relationship between the PON port and the user-side device, such as the probability of a connection. The second correlation characterizes the connection relationship between the user-side devices, such as the probability of a connection.

[0069] S303. Determine the topology of the ODN based on the first correlation and the second correlation.

[0070] After determining the first correlation between each PON port and each user-side device, and the second correlation between each user-side device, the fault location device can determine the topology of the ODN based on the first and second correlations.

[0071] S304. Based on the fault optical path index characteristic model, the transmit and transmit power of each PON port of the OLT in the first preset time period and the transmit and transmit power of each user-side device in the multiple user-side devices in the first preset time period, determine the cause of the fault.

[0072] S305. Based on the cause of the fault and the topology of the ODN, locate the fault in the ODN.

[0073] After determining the ODN topology, the fault location device can determine the cause of the fault based on a pre-trained fault optical path indicator feature model, the transmit and transmit power of each PON port of the OLT within a first preset time period, and the transmit and transmit power of each user-side device among multiple user-side devices within the first preset time period. Furthermore, the fault location device can locate the fault in the ODN based on the fault cause and the ODN topology to pinpoint the location of the fault point.

[0074] The technical solution provided by the above embodiments brings at least the following beneficial effects: By acquiring optical power information from the PON port and the user-side equipment, and analyzing the optical power information, the correlation between the PON port and the user-side equipment, as well as among the user-side equipment, can be determined. Based on the correlation, the ODN topology can be accurately reconstructed. The cause of the fault is determined through an optical path fault indicator feature model, and then, combined with the fault cause and the ODN topology, fault location in the ODN can be achieved. This method reconstructs the ODN topology using a large amount of objective data, resulting in a more objective and accurate topology, unaffected by human factors. Thus, by determining the ODN fault location based on the ODN topology, the fault point of the PON can be accurately located, improving the accuracy of ODN optical path fault location. Furthermore, it enables technicians to maintain the network promptly at the fault point, providing users with a good broadband internet experience.

[0075] The following detailed description, in conjunction with specific embodiments, illustrates a fault location method for an ODN in a PON provided by this application. This method can be applied to fault location devices. Figure 4 As shown, the method may include S401-S406.

[0076] PON can include OLT, ODN, and multiple user-side devices.

[0077] S401. Obtain feature data.

[0078] Among them, the feature data is used to indicate the optical power information of each PON port of the OLT and the optical power information of each user-side device among multiple user-side devices within a first preset time period.

[0079] Specifically, the characteristic data may include: the transmit and transmit power of each user-side device among multiple user-side devices, the time of transaction occurrence of the transmit and transmit power of each user-side device, the transmit and transmit power of each PON port among all PON ports of the OLT, and the time of transaction occurrence of the transmit and transmit power of each PON port. In this embodiment, the transmit and transmit power may include transmit power and transmit power.

[0080] The fault location device can obtain slice data from all PON ports of the OLT, which is the characteristic data within a time period.

[0081] In some embodiments, both the OLT and user-side equipment in a PON can periodically and proactively report their own receive and transmit power to the PON network management system. The network management system can store the receive and transmit power of each device received at various times. When fault location is required in the ODN, the fault location device can access the PON network management system to obtain slice data under all PON ports of the OLT, that is, to obtain characteristic data within a time period, such as the receive and transmit power of all PON ports of the OLT, and the receive and transmit power of multiple user-side equipment connected to the PON ports. The receive and transmit power of the PON ports and user-side equipment can also be referred to as time-series characteristics, used to represent the changes in optical power within a given time period.

[0082] In addition, the network management system also stores the times of the aforementioned optical power events (i.e., the times when optical power changes). Therefore, the fault location device can also obtain the times of the PON port's receive and transmit optical power events and the times of the user-side equipment's receive and transmit optical power events within the aforementioned time period from the network management system. The aforementioned time period can be predefined. If the aforementioned time period is the first preset time period, the fault location device can obtain the aforementioned characteristic data from the PON's network management system.

[0083] For example, in a PON, the OLT has two PON ports, namely PON port A and PON port B. This PON includes 10 user-side devices, namely user-side device 1, user-side device 2, ..., user-side device 10. The fault location device can access the PON network management system to obtain the received and transmitted optical power and transaction times of PON ports A and B over the past week, as well as the received and transmitted optical power and transaction times of user-side devices 1 through 10 over the past week.

[0084] In some embodiments, after acquiring feature data, the fault location device can use artificial intelligence (AI) feature recognition (e.g., unsupervised algorithms) to process the acquired data and exclude abnormal data (i.e., the denoising process in data processing technology). For details on the denoising process of specific unsupervised algorithms, please refer to relevant technical documents; they will not be elaborated upon here.

[0085] After acquiring the feature data, the fault location device can use a multi-level clustering algorithm to determine the topology of the ODN, such as by executing S402-S404 below.

[0086] S402. Based on the feature data, determine the probability that any two or more user-side devices among multiple user-side devices are connected to the same optical splitter.

[0087] As described in the foregoing embodiments, an ODN may include multiple optical splitters. User-side devices (PSDs) can connect to an OLT via these ODN splitters, enabling users to access broadband internet through their PDNs. During connection, different PSDs may connect to the OLT via the same splitter or different splitters. PSDs connected to the same splitter share similar characteristics; for example, if the optical distance of PSDs connected to the same splitter meets certain conditions, the optical distance of the PSDs can also be included in the aforementioned characteristic data. Therefore, the fault location device can determine the likelihood that multiple PSDs within a PON are connected to the same splitter based on the characteristic data.

[0088] In some embodiments, the acquired feature data may further include the optical distance between each user-side device. The optical distance can be used to characterize the distance between the user-side device and the OLT. The fault location device can determine the probability that two or more user-side devices are connected to the same splitter based on the optical distance between each user-side device. For example, the closer the optical distances of different user-side devices are, the greater the probability that they are connected to the same splitter.

[0089] For example, taking the example from the above embodiment, user-side devices 1-5 are each approximately 10 kilometers away from the OLT, meaning their optical distances are all approximately 10 kilometers. Similarly, user-side devices 6-10 are each approximately 20 kilometers away from the OLT, meaning their optical distances are all approximately 20 kilometers. Therefore, the fault location device can determine that the optical distances of user-side devices 1-5 are relatively close, indicating a high probability that they are connected to the same optical splitter. Likewise, it is also highly likely that user-side devices 6-10 are connected to the same optical splitter. Conversely, the probability that user-side devices 1-5 and user-side devices 6-10 are connected to the same optical splitter is relatively low.

[0090] S403. Based on the feature data, determine the first correlation and the second correlation.

[0091] The first correlation is used to indicate the correlation between each PON port of the OLT and each user-side device among multiple user-side devices, and the second correlation is used to indicate the correlation among multiple user-side devices.

[0092] Correlation refers to the likelihood of connections between network nodes in a PON network, including PON ports and user-side equipment. If connections exist between network nodes, a change in the optical power of one network node will affect other network nodes connected to it. The transmit and receive power and the time of the transmit and receive power transactions in the characteristic data reflect the time when the transmit and receive power changes, which can reflect the likelihood of connections between network nodes. Therefore, the fault location device can determine the correlation between network nodes based on the characteristic data, such as determining a first correlation to indicate the correlation between PON ports and user-side equipment, and a second correlation to indicate the correlation between user-side equipment.

[0093] The first and second correlations can include the correlations of multiple sets of optical power. For example, each set of optical power includes a first optical power and a second optical power, where the first and second optical power are any two different transmit / receive power values ​​among the transmit / receive power values ​​of each user-side device and each PON port. In other words, the correlation between each PON port and each user-side device, as well as the correlation between each user-side device, can be obtained by determining the correlation between every two transmit / receive power values ​​(referred to as a set of optical power values) in the feature data.

[0094] The methods for determining the correlation of different groups of optical power are similar. Specifically, the fault location device uses S403a-S403d as follows to determine the correlation of a group of optical power. For each group of optical power, S403a-S403d is executed to determine its correlation, thereby determining the aforementioned first correlation and the aforementioned second correlation.

[0095] S403a. Based on the time when the first optical power and the second optical power occur, determine the average time difference between the occurrence of transactions by the first optical power and the second optical power, as well as the number of times the first optical power and the second optical power occur simultaneously.

[0096] As described in the foregoing embodiments, it can be concluded that the closer the times when two optical powers occur, the greater their correlation. As an example, before determining the correlation of optical powers, a first preset time period can be divided into multiple average time periods. The average time difference between the occurrences of transactions between the two optical powers within the first preset time period can be determined based on the time difference between the occurrences of transactions within each time period, thereby determining the correlation between the two optical powers. This can improve the accuracy of the correlation determination result.

[0097] That is, in some embodiments, the fault location device can divide the first preset time period into n-1 time periods. Then, the fault location device can determine the time difference between the occurrence of transactions of the first optical power and the second optical power in each of the n-1 time periods based on the time of occurrence of transactions of the first optical power and the second optical power. Afterwards, the fault location device can determine the average time difference between the occurrence of transactions of the first optical power and the second optical power within the first preset time period based on the time difference between the occurrence of transactions of the first optical power and the second optical power in each time period.

[0098] For example, the fault location device can determine the average time difference between the occurrence of transactions at the first optical power and the second optical power using the following formula:

[0099]

[0100] Where Δt is the average time difference between the transactions of the first optical power and the second optical power, n-1 is the number of time periods into which the first preset time is averaged, and t mi Let t be the time during which a transaction occurs within the m-th time period divided by the first preset time period, where the first optical power is at that time. mj The time when the second optical power occurs within the m-th time period divided by the first preset time period, where n is an integer greater than or equal to 2 and m is an integer greater than or equal to 1 and less than or equal to n-1.

[0101] It should be noted that the first preset time can be divided according to actual needs. That is, the size of n can be set according to actual needs. The finer the division (i.e., the larger n is), the more accurate the determined average time difference Δt will be, and thus the better the fault location effect of the scheme will be.

[0102] Furthermore, the fault location device can also count the number of times the first optical power and the second optical power simultaneously occur within a first preset time period based on the time when the first optical power and the second optical power occur. For example, if both the first optical power and the second optical power occur within a certain time period divided within the first preset time period, they can be considered to have occurred simultaneously. For instance, if both the first optical power and the second optical power occur within the first time period divided within the first preset time period (regardless of whether the specific times of the occurrences are the same), the fault location device can count the number of times the first optical power and the second optical power simultaneously occur as 1. Similarly, the fault location device can perform the above statistics for each of the remaining n-1 time periods to count the number of times the first optical power and the second optical power simultaneously occur within the first preset time period.

[0103] For example, consider determining the optical power correlation between PON port A and user-side device 1 in the above example. The first optical power can be the optical power of PON port A (e.g., received power or emitted power), and the second optical power can be the optical power of user-side device 1 (e.g., received power or emitted power).

[0104] The fault location device can determine the average time difference between optical power transactions at PON port A and user-side equipment 1, as well as the number of times transactions occur simultaneously between PON port A and user-side equipment 1 within a preset time period. For example, the preset time period is one week, and n-1 is 168. The fault location device divides one week into 168 time periods, with one hour as the granularity. Furthermore, the fault location device determines that in 100 of these 168 time periods, the time difference between optical power transactions at PON port A and those at user-side equipment 1 is 5 minutes. In the remaining 68 time periods, no transactions occur between the optical power at PON port A and those at user-side equipment 1, and the determined time difference between their optical power transactions is 0. Therefore, the average time difference between transactions for the first and second optical power can be obtained using the following formula:

[0105]

[0106] That is, the fault location device can determine that the average time difference Δt between optical power transactions at PON port A and user-side equipment 1 is approximately 2.98 min (rounded to two decimal places). In addition, the fault location device can also determine the number of times optical power transactions occur simultaneously at PON port A and user-side equipment 1, such as 100 times.

[0107] S403b: Obtain the raw data of PON, including the service connection relationship and physical connection relationship of the network node corresponding to the first optical power and the network node corresponding to the second optical power.

[0108] S403c. Determine the correlation coefficient between the first optical power and the second optical power based on the service connection relationship and the physical connection relationship.

[0109] Physical connectivity refers to the direct connection between network nodes during PON establishment. Service connectivity refers to the business transactions between network nodes, which may be services registered with the operator by the network node during PON establishment (such as broadband services). The service and physical connectivity relationships of each network node help to reconstruct the ODN topology. Therefore, this solution can introduce a correlation coefficient based on these relationships as one of the important criteria for assessing relevance.

[0110] In some examples, when establishing a PON, the operator's network management system stores the raw PON data, which includes the physical connection relationships and service connection relationships between various network nodes in the PON (such as PON ports and user-side devices, or between user devices).

[0111] The fault location device can access the network management system to obtain the service connection relationship and physical connection relationship between the network node corresponding to the first optical power and the network node corresponding to the second optical power. Then, based on the service connection relationship and physical connection relationship between the network node corresponding to the first optical power and the network node corresponding to the second optical power, the fault location device can determine the correlation coefficient between the first optical power and the second optical power.

[0112] For example, the correlation coefficient between the first optical power and the second optical power ranges from {0, 1, 2}. Specifically, if there is both a physical connection and a service connection between the first optical power and the second optical power, the correlation coefficient is 2. If there is only one of a service connection or a physical connection between the first optical power and the second optical power, the correlation coefficient is 1. If there is no connection between the first optical power and the second optical power, the correlation coefficient is 0.

[0113] For example, continuing with the example in the above embodiment, the first optical power is the optical power of PON port A, and the second optical power is the optical power of user-side equipment 1. The fault location device obtains from the network management system that there is both a physical connection and a service connection between PON port A and user-side equipment 1, so the correlation coefficient between the optical power of PON port A and the optical power of user-side equipment 1 is 2.

[0114] S403d. Determine the correlation between the first optical power and the second optical power based on the average time difference between transactions occurring at the first optical power and the second optical power, the number of transactions occurring simultaneously at the first optical power and the second optical power, the total number of transactions occurring at all received and received optical powers, and the correlation coefficient.

[0115] In some embodiments, the fault location device can calculate the total number of optical power transactions based on the time of each transaction. For example, for an optical power included in the feature data, if the feature data includes a record of the time of an optical power transaction, the fault location device counts the number of such transactions as 1, and then calculates the total number of such transactions based on the number of records of such optical power transactions in the feature data. Similarly, the above calculation is performed on other optical powers included in the feature data to obtain the total number of transactions for each optical power. Then, the total number of transactions for each optical power is summed to determine the total number of transactions for all optical power transactions.

[0116] After the fault location device determines the average time difference between the transactions of the first optical power and the second optical power, the number of transactions of the first optical power and the second optical power occurring simultaneously, the total number of transactions of all received and received optical powers, and the correlation coefficient, the correlation between the first optical power and the second optical power can be determined based on the average time difference between the transactions of the first optical power and the second optical power, the number of transactions of the first optical power and the second optical power occurring simultaneously, the total number of transactions of all received and received optical powers, and the correlation coefficient.

[0117] For example, the fault location device can determine the correlation between the first optical power and the second optical power using the following formula:

[0118]

[0119] Among them, Cor d (i, j) represents the correlation between the first optical power and the second optical power, T ij |t| represents the number of transactions occurring simultaneously at the first and second optical powers, ||T|| represents the total number of transactions occurring at all optical powers, and |t| represents the total number of transactions occurring at all optical powers. n -t n-1 | represents the length of the time period into which the first preset time is evenly divided, N ij The first optical power represents the correlation coefficient between the second optical power, and Δt is the average time difference between the first optical power and the second optical power.

[0120] For example, continuing with the example in the above embodiment, the fault location device determines the number T of transactions that occur simultaneously on PON port A and user-side device 1. ij The value is 100, the average time difference Δt between transactions is 2.98 min, and the total number of transactions involving all transmit and receive power, as counted by the fault location device, is 1000. n -t n-1 | represents the length of the time period to which the above-preset time period is divided, which is 60 minutes, and the correlation coefficient N. ij The correlation coefficient Cor between the optical power of PON port A and user-side device 1 is 2. d (i,j)=100 / 1000*60 / 2.93*2=4.10.

[0121] Similarly, the fault location device can also determine the correlation between the optical power of PON port A and other user-side devices, such as user-side devices 2 to 10, the correlation between the optical power of PON port B and each user-side device, such as user-side device 1 to 10, and the correlation between the optical power of every two user-side devices in user-side devices 1 to 10, in order to obtain the aforementioned first correlation and second correlation.

[0122] S404. Determine the topology of the ODN based on the first correlation, the second correlation, and the probability that any two or more user-side devices among multiple user-side devices are connected to the same optical splitter.

[0123] As described in the preceding embodiments, the greater the correlation between optical powers, the greater the likelihood of a connection between the network nodes containing those optical powers. For example, if the correlation between optical powers is greater than a preset threshold, it indicates a strong correlation between the corresponding network nodes, meaning that a connection can be considered to exist between them. If the correlation between optical powers is less than the preset threshold, it indicates a weak correlation between the corresponding network nodes, meaning that no connection can be considered to exist between them.

[0124] In PON, multiple user-side devices connected to the same optical splitter share similar characteristics (e.g., consistent optical power variation trends), indicating strong correlation between each pair of these devices. Therefore, multiple user-side devices connected to the same optical splitter can be considered a group (or community). Community detection algorithms are classic clustering algorithms that determine the group relationships (or graphs) between network nodes based on their correlations (e.g., edge weights in the algorithm). This process moves from edges to a graph, with dynamic weight accumulation, gradually reconstructing the ODN topology.

[0125] Therefore, in some embodiments, after determining the first correlation and the second correlation, the fault location device can determine the topology of the ODN based on the first correlation and the second correlation using a community detection algorithm. The specific usage of the community detection algorithm can be found in relevant technical documents and will not be elaborated here. Additionally, as described in the foregoing embodiments, the fault location device can also determine the probability that any two or more user-side devices among multiple user-side devices are connected to the same optical splitter based on the optical distance of the user-side devices. This probability can be used to verify the accuracy of the ODN topology determined using the community detection algorithm.

[0126] For example, referring to the examples in the above embodiments, the fault location device, by executing S403a-S403d, determines that the correlation between the optical power of PON port A and the optical power of user-side devices 1-5 is greater than a preset threshold. Based on this result, the fault location device can determine that there is a connection relationship between PON port A and user-side devices 1-5. Similarly, the fault location device determines that the correlation between the optical power of PON port B and user-side devices 6-10 is greater than a preset threshold. Based on this result, the fault location device can determine that there is a connection relationship between PON port B and user-side devices 6-10.

[0127] Furthermore, the fault location device also determined that the correlation of optical power between user-side device 1 and user-side devices 2 and 3 is greater than a preset threshold, while the correlation of optical power between user-side device 1 and user-side devices 4-10 is less than a preset threshold. Similarly, the fault location device determined that the correlation of optical power between user-side device 6 and user-side devices 7 and 8 is greater than a preset threshold, while the correlation of optical power between user-side device 6 and user-side devices 1-5, and between user-side device 9 and user-side device 10, is less than a preset threshold. Likewise, for each user-side device, the correlation of optical power between that user-side device and other user-side devices is determined.

[0128] Based on the above analysis, the fault location device can determine that there is a connection between user-side equipment 1 and user-side equipment 3, and that there is no connection between user-side equipment 1 and user-side equipment 3 and other user-side equipment. Therefore, the fault location device can determine that user-side equipment 1, user-side equipment 2, and user-side equipment 3 are connected to the same optical splitter. Similarly, if there is a connection between user-side equipment 4 and user-side equipment 5, and there is no connection between user-side equipment 4 and user-side equipment 5 and other user-side equipment, then user-side equipment 4 and user-side equipment 5 are connected to the same optical splitter.

[0129] Similarly, if user-side devices 6 and 8 are connected but not connected to other user-side devices, and user-side devices 9 and 10 are connected but not connected to other user-side devices, then the fault location device can determine that user-side devices 6 and 8 are connected to the same optical splitter, and user-side devices 9 and 10 are connected to the same optical splitter.

[0130] The above analysis shows that there are at least two optical splitters of the same level under PON port A, and at least two optical splitters of the same level under PON port B. Since there are multiple optical splitters of the same level under one PON port, it can be determined that the user-side equipment is connected to the OLT in a two-level network structure.

[0131] Based on the above analysis, the fault location device can determine the topology of the ODN as follows: Figure 5As shown, user-side devices 1-3 are connected to secondary optical splitter 1, and user-side devices 4 and 5 are connected to secondary optical splitter 2. Secondary optical splitters 1 and 2 are connected to primary optical splitter 1, which is connected to PON port A. Additionally, user-side devices 6-8 are connected to secondary optical splitter 3, and user-side devices 9 and 10 are connected to secondary optical splitter 4. Secondary optical splitters 3 and 4 are connected to primary optical splitter 2, which is connected to PON port B. PON ports A and B are located on the OLT.

[0132] Optionally, in conjunction with the examples in the above embodiments, the probability that any two or more user-side devices among multiple user-side devices are connected to the same optical splitter can verify the above conclusion, making the results more accurate. According to the example in S402, user-side devices 1-5 are more likely to be connected to the same optical splitter. Figure 5 As can be seen from the topology of the ODN, user-side device 1 to user-side device 5 are indirectly connected to the same first-level optical splitter 1. Therefore, the conclusion of the above correlation analysis is relatively accurate.

[0133] The above example uses a reconstructed ODN topology as a secondary network. In other examples, the reconstructed topology, based on determined correlations, may also be a primary network. For instance, a PON OLT includes two ports, PON port C and PON port D. The PON includes five user-side devices connected to the OLT, user-side devices 11 through 15. Based on the determined correlations, the fault location device can determine that PON port C is connected to user-side devices 11 and 12, and PON port D is connected to user-side devices 13 through 15.

[0134] Furthermore, based on the correlation, it is determined that there is a connection between user-side equipment 11 and user-side equipment 12, and that user-side equipment 11 and user-side equipment 12 are not connected to other user-side equipment. Therefore, the fault location device can determine that user-side equipment 11 and user-side equipment 12 are connected to the same optical splitter. Based on the correlation, it is determined that there is a connection between user-side equipment 13 and user-side equipment 15, and that user-side equipment 13 and user-side equipment 15 are not connected to other user-side equipment. Therefore, the fault location device can determine that user-side equipment 13 and user-side equipment 15 are connected to the same optical splitter. From the above analysis, it can be seen that there is only one optical splitter under each PON port. Therefore, it can be determined that the user-side equipment is connected to the OLT in a first-level network structure.

[0135] Based on the above analysis, the topology of the ODN established by the fault location device can be as follows: Figure 6As shown, user-side devices 11 and 12 are connected to PON port C via optical splitter 5, and user-side devices 13-15 are connected to PON port D via optical splitter 6. PON port C and PON port D are configured on the OLT.

[0136] The above example illustrates how the ODN topology is reconstructed based on the first correlation, the second correlation, and the probability that any two or more user-side devices among multiple user-side devices will connect to the same optical splitter. This method of reconstructing the ODN topology takes into account the fact that user-side devices are relatively concentrated under the same optical splitter in real-world scenarios. Therefore, by dividing the devices into multiple groups based on the probability of connecting to the same splitter, the connection relationships between user-side devices determined by the correlation can be verified and checked, improving the accuracy of reconstructing the group relationships between user-side devices and further enhancing the accuracy of the reconstructed ODN topology. In some other embodiments, the ODN topology may also be reconstructed based solely on the first and second correlations; this application does not impose specific limitations on this approach.

[0137] After reconstructing the topology of the ODN, the fault location device can locate the fault in the ODN based on the fault optical path index feature model and the reconstructed topology of the ODN, specifically including the following S405-S406.

[0138] S405. Based on the fault optical path index characteristic model, the transmit and transmit power of each PON port in all PON ports of the OLT included in the PON during the first preset time period, and the transmit and transmit power of each user-side device in the multiple user-side devices included in the PON during the first preset time period, determine the cause of the fault.

[0139] Among them, the fault optical path index feature model has the function of performing similarity analysis, mutation analysis and overall trend analysis on the transmit and receive transmit power data based on the transmit and receive transmit power of each network node in PON, thereby determining the cause of the fault.

[0140] S406. Based on the cause of the fault and the topology of the ODN, locate the fault in the ODN.

[0141] In some embodiments, the fault location device can obtain the transmit and receive power of each network node in the PON within a first preset time period from feature data, that is, the transmit and receive power of each PON port among all PON ports of the OLT included in the PON within the first preset time period, and the transmit and receive power of each user-side device among the multiple user-side devices included in the PON within the first preset time period. Then, the fault location device can input the transmit and receive power of the aforementioned network nodes into a pre-trained fault optical path indicator feature model to determine the cause of the fault. Furthermore, the fault location device can perform fault location on the ODN based on the cause of the fault and the topology of the ODN.

[0142] For example, Figure 7 This application provides a line graph illustrating the time-series characteristics of the optical power of a network node. Taking a network node including the user-side device 1 and PON port A as exemplified in the above embodiments as an example, Figure 7 Specifically, this refers to the time series characteristics of the received optical power of the user-side device 1 (taking ONT as an example) and the received optical power of the corresponding PON port A over a period of time.

[0143] from Figure 7 It can be seen that on September 5th, the received optical power of user-side device 1 and the corresponding received optical power of PON port A decreased simultaneously, indicating consistent time-series characteristics. Figure 7 By inputting the time-series features into the faulty optical path index feature model, the cause of the fault can be determined to be damage to the drop fiber. Combined with the above... Figure 5 The topology diagram of the ODN shows that the fault point is the damaged drop fiber between the secondary optical splitter 1 and the user-side equipment 1.

[0144] This application also provides a line graph of the time-series characteristics of the optical power of another network node, such as... Figure 8 As shown, taking the user-side device 6 in the above embodiment as an example, Figure 8 This represents the time-series characteristics of the received optical power of the user-side device 6 (taking ONT as an example) and the received and emitted optical power of the corresponding PON port B over a period of time.

[0145] from Figure 8 It can be seen that on September 5th, the received optical power of user-side device 6 suddenly decreased, but the received optical power of the corresponding PON port B remained basically stable, indicating inconsistent time series characteristics. Figure 8 By inputting Zhang Zhong's time-series features into the faulty optical path index feature model, the cause of the fault can be determined to be a user-side equipment failure. Combined with the above... Figure 5 The topology diagram of the ODN shows that the fault point is user-side device 6.

[0146] The faulty optical path index characteristic model can be determined based on the optical power information in the network management system. The following steps describe the process of determining the faulty optical path index characteristic model.

[0147] 1. Obtain the transmit and transmit power of each PON port in all PON ports of the OLT during the second preset time period, and the transmit and transmit power of each user-side device in multiple user-side devices during the second preset time period.

[0148] In some embodiments, the fault location device can access the network management system to obtain the transmit and receive power of the PON port and the transmit and receive power of the user-side equipment, referred to as training data. The training data is essentially the same as the feature data described above, except that the time period for acquisition differs; for example, the first preset time is the past week, and the second preset time is the past month. The training data is used to train the fault optical path indicator feature model.

[0149] 2. Based on the transmit and transmit power of each PON port in all PON ports of the OLT during the second preset time period and the transmit and transmit power of each user-side device in multiple user-side devices during the second preset time period, a fault optical path index feature model is determined by training the model through AI algorithm.

[0150] In some embodiments, after acquiring training data, the fault location device can determine the theoretical received optical power of the user-side equipment and the theoretical received optical power of the PON port based on the training data. An AI algorithm is used to train a model on the training data, the theoretical received optical power of the user-side equipment, and the theoretical received optical power of the PON port to obtain one or more fault optical path index feature models. The input to the trained fault optical path index model is the time-series characteristics of the PON port and the user-side equipment (the received and transmitted optical power of the PON port and the user-side equipment). Depending on the fault optical path index feature model, the output fault cause will also be different. These fault causes may include fiber bending, user-side equipment failure, optical module degradation, and damage to the drop fiber. Alternatively, a single fault optical path index feature model can be trained, and different fault causes can be output depending on the input data. This application does not impose specific limitations in this regard.

[0151] For example, Figure 9 This application provides a multi-path quality monitoring line graph, from... Figure 9The system can obtain the changes in optical power over a period of time, i.e., the time-series characteristics of each optical power. Since the time-series characteristics of optical power are relatively stable under normal operating conditions, the changes in optical power over certain time periods can be used to analyze the causes of faults such as unstable optical paths, equipment malfunctions, and changes in the splitting ratio due to engineering construction. Therefore, the time-series characteristics can be used as input and the causes of faults as output to train various faulty optical path indicator characteristic models, thereby achieving fault localization in the ODN.

[0152] This solution provides a method for determining the theoretical received optical power of the user-side equipment and the theoretical received optical power of the PON port required for training the faulty optical path characteristic model. For example, Table 1 below shows a method for calibrating the optical attenuation of various optical devices provided by this solution:

[0153] Table 1

[0154]

[0155]

[0156] Table 1 provides data on the average attenuation per unit length (1 km) of optical fiber, the average attenuation of the optical splitter at different splitting ratios, and the output power of the optical splitter. Based on the average attenuation of each optical device in Table 1, and using the formulas: Theoretical received optical power of user-side equipment = PON port emitted power - total line optical attenuation, and Theoretical received optical power of PON port = User-side equipment emitted power - total line optical attenuation, the theoretical optical power of the PON port and user-side equipment can be determined. The theoretical optical power of the PON port and user-side equipment helps determine whether there are weak light problems in the optical path, thus improving the effectiveness of training the determined fault optical path characteristic model.

[0157] Figure 10This application provides a flowchart of another fault location method for ODN in PON. It involves acquiring a large amount of optical path data (i.e., the feature data and second feature data in the above embodiments) from the network management system. Noise removal is performed using AI feature recognition (i.e., the unsupervised algorithm in the above embodiments eliminates abnormal data), followed by transformation and feature analysis to establish the topology (i.e., the ODN topology is restored based on correlation and community detection algorithms in the above embodiments). Furthermore, AI algorithms are used to establish various fault optical path indicator feature models, performing similarity analysis, mutation analysis, and overall trend analysis on the optical path data to locate various fault causes, such as trunk, branch, drop fiber, optical module, or excessive splitting ratio faults. Finally, ODN monitoring results are output (i.e., in the above embodiments, the fault cause is determined based on the fault optical path indicator feature model, and then the ODN fault location is determined by combining the ODN topology). Based on the results, resource correction work orders are dispatched so that relevant technical personnel can perform network maintenance based on the fault cause and fault location.

[0158] The technical solution provided by the above embodiments brings at least the following beneficial effects: By acquiring optical power information from the PON port and the user-side equipment, and analyzing the optical power information, the correlation between the PON port and the user-side equipment, as well as among the user-side equipment, can be determined. Based on the correlation, the ODN topology can be accurately reconstructed. The cause of the fault is determined through an optical path fault indicator feature model, and then, combined with the fault cause and the ODN topology, fault location in the ODN can be achieved. This method reconstructs the ODN topology using a large amount of objective data, resulting in a more objective and accurate topology, unaffected by human factors. Thus, by determining the ODN fault location based on the ODN topology, the fault point of the PON can be accurately located, improving the accuracy of ODN optical path fault location. Furthermore, it enables technicians to maintain the network promptly at the fault point, providing users with a good broadband internet experience.

[0159] Furthermore, this scheme considers the correlation between optical power at both temporal and spatial levels by incorporating the time and frequency of optical power transactions, thereby obtaining the correlation between corresponding devices. It also introduces a correlation coefficient, including the original data from PON establishment in the correlation assessment, making the obtained device correlations more comprehensive and accurate, thus further improving the accuracy of the established ODN topology. Moreover, feature data can be acquired in real time; if the PON network changes, this scheme can update the ODN topology promptly, making ODN fault location more accurate. In terms of cost, compared to traditional intelligent ODN technology, this scheme does not require expensive purchases of specific equipment; it only needs to collect and analyze feature data from the existing network management system. This makes it highly usable for operators and can address the increased network maintenance costs caused by rapid network upgrades. Simultaneously, this scheme has a long service life; as the amount of feature data generated in the PON increases over time, the accuracy of the artificial intelligence algorithm continuously improves through learning, leading to increasingly better application results.

[0160] Based on the results of fault location in the ODN, this application also provides a fault handling method, as follows:

[0161] 1. If the cause of the fault is weak light in the trunk fiber, the fault point may occur at both ends of the trunk and along the trunk optical path. Therefore, this application provides the following solution.

[0162] The first step is to check if the primary splitter is malfunctioning.

[0163] Use an optical power meter to check the output optical power of the first-stage splitter: If the attenuation [the optical power at the splitter's input minus the measured optical power at the splitter's output] is greater than the theoretical value of the splitter [1.5 dB], the loss reference values ​​are: 1:8 splitter 10.5 dB, 1:16 splitter 13.5 dB, 1:32 splitter 16.5 dB.

[0164] Recommended solutions: Replace the splitter, unplug and replug the pigtails, clean the pigtail ends, and correct any bends in the pigtails. After completing these steps, log in to the network management system to check if the ONU / ONT optical power meets the standards. Determine if the problem is resolved or proceed to the next step.

[0165] The second step is to check for any abnormalities in the main optical path.

[0166] Use an optical power meter to check the received optical power at the entrance of the first-level splitter: if the attenuation [the emitted power at the PON port minus the measured optical power at the entrance of the first-level splitter] is greater than the theoretical value [3dB], then it is confirmed that there is an abnormality in the backbone optical path.

[0167] Troubleshooting suggestions: a. Test the fiber core of the same line and try replacing it. b. Verify the flange or ODF adapter for abnormalities; replace it if abnormalities are confirmed. c. Use an optical time-domain reflectometer (OTDR) to test the problem area and repair the faulty point or re-sew it. After troubleshooting, log in to the network management system to check if the ONU / ONT optical power meets the standards to determine if the rectification is complete or proceed to the next step.

[0168] The third step is to check if the optical module of the OLT's PON port is malfunctioning.

[0169] Use an optical power meter to check if the PON port's luminous power meets the standard. Address any PON ports that do not meet the standard or have weak luminous power. Simultaneously check the optical power on the ODF rack side of the equipment room; the difference between the optical power and the PON port test value should be less than 1 dB.

[0170] Recommended troubleshooting steps: Replace the optical module, unplug and replug the pigtail, clean the pigtail end face, and correct any bends in the pigtail. After troubleshooting, log in to the network management system to check if the ONU / ONT optical power meets the standards, and determine if the rectification is complete or proceed to the next step.

[0171] Fourth, if the above three steps are normal, proceed to verify the branch optical paths connected to the first-level optical splitter and test the optical power of each port after passing through the first-level optical splitter.

[0172] Recommended course of action: Record and report the test data, and contact the technical supervisor for assistance.

[0173] 2. If the cause of the fault is weak light in the branch fiber, the fault point may be in the branch optical path or the secondary splitter. Therefore, this application provides the following solution.

[0174] Step 1: Check if the secondary splitter is malfunctioning.

[0175] Use an optical power meter to check the output optical power of the secondary beam splitter: If the attenuation [the optical power at the beam splitter input minus the measured optical power at the beam splitter output] is greater than the theoretical value of the beam splitter [1.5 dB], the loss reference values ​​are: 1:8 beam splitter 10.5 dB, 1:16 beam splitter 13.5 dB, 1:32 beam splitter 16.5 dB.

[0176] Recommended troubleshooting steps: Unplug and replug the splitter connectors, clean and remove dust, and replace the splitter. After troubleshooting, log in to the network management system to check if the ONU / ONT optical power meets the standards, and determine if the rectification is complete or proceed to the next step.

[0177] The second step is to check for any abnormalities in the branch optical path.

[0178] Locate the secondary optical splitter based on the ONU / ONT information, and use an optical power meter to check the received optical power at the secondary optical splitter's inlet: if the attenuation [the optical power at the outlet of the primary optical splitter minus the optical power measured at the inlet of the secondary optical splitter] is greater than the theoretical value [3dB], then it is confirmed that there is an abnormality in this branch optical path.

[0179] Troubleshooting suggestions: a. Test the same fiber cores on the same line; try replacing the fiber cores. b. Use an OTDR to test the problem area; repair the bad points or re-sponsor the fiber. After troubleshooting, log in to the network management system to check if the ONU / ONT optical power meets the standards. Determine if the problem is resolved or proceed to the next step.

[0180] Step 3: If all the above checks are normal, proceed to the in-home low light treatment step. You can contact the technical supervisor for assistance.

[0181] 3. If the cause of the fault is weak light in the drop fiber, the fault point may be at both ends of the drop fiber or along its optical path. Therefore, this application provides the following solution.

[0182] Step 1: Check if the optical power at the port of the secondary splitter is abnormal.

[0183] Use an optical power meter to check whether the output optical power of the secondary beam splitter connected to a low-light user exceeds the theoretical value.

[0184] Recommended solution: If the problem persists, switch to another secondary optical splitter port that has normal optical power and is not in use. After troubleshooting, log in to the network management system to check if the ONU / ONT optical power meets the standards. Determine if the problem is resolved or proceed to the next step.

[0185] Step 2: Check if there is any abnormality in the incoming fiber optic cable.

[0186] Measure the optical power output of the secondary optical splitter in the stairwell and the optical power received by the user's ONU / ONT. If the difference is greater than 2dB of the theoretical value, it is confirmed that there is an abnormality in the incoming optical path.

[0187] Recommended solutions: a. Replace the drop fiber; b. Re-fusion the incoming fiber; c. Fix excessive bending in the incoming fiber; d. Re-process the cold splice. After processing, log in to the network management system to check if the ONU / ONT optical power meets the standards to determine if the rectification is complete.

[0188] Step 3: Troubleshoot ONU / ONT faults.

[0189] Check if the ONU / ONT is working properly.

[0190] Troubleshooting suggestions: a. If the flange is dusty, wipe it with a fiber optic cleaner or alcohol. b. If the optical module is malfunctioning, replace it. c. If the ONU / ONT hardware is malfunctioning, replace the ONU / ONT.

[0191] In an exemplary embodiment, this application also provides a fault location device for fault location of the ODN in a PON, wherein the PON includes an OLT, an ODN including multiple optical splitters, and multiple user-side devices. The fault location device may include one or more functional modules for implementing the fault location method of the ODN in the PON described in the above method embodiments.

[0192] For example, Figure 11 This is a schematic diagram illustrating the composition of a fault location device provided in an embodiment of this application. Figure 11 As shown, the fault location device includes: an acquisition module 1101, a determination module 1102, and a location module 1103. The acquisition module 1101, the determination module 1102, and the location module 1103 are connected to each other.

[0193] The acquisition module 1101 acquires feature data; wherein the feature data is used to indicate the optical power information of each PON port in all PON ports of the OLT and the optical power information of each user-side device in multiple user-side devices within a first preset time period.

[0194] The determination module 1102 is used to determine a first correlation and a second correlation based on the feature data; the first correlation is used to indicate the correlation between each PON port of the OLT and each user-side device among multiple user-side devices, and the second correlation is used to indicate the correlation among multiple user-side devices.

[0195] The determination module 1102 is also used to determine the topology of the ODN based on the first correlation and the second correlation.

[0196] The determination module 1102 is also used to determine the cause of the fault based on the fault optical path index characteristic model, the transmit and transmit power of each PON port of the OLT in the first preset time period, and the transmit and transmit power of each user-side device among multiple user-side devices in the first preset time period.

[0197] Location module 1103 is used to locate faults in the ODN based on the cause of the fault and the topology of the ODN.

[0198] In some embodiments, the feature data includes: the transmit and transmit power of each user-side device among a plurality of user-side devices, the time when the transmit and transmit power of each user-side device occurs, the transmit and transmit power of each PON port among all PON ports of the OLT, and the time when the transmit and transmit power of each PON port occurs.

[0199] In some embodiments, the first correlation and the second correlation include the correlation of multiple sets of optical power, each set of optical power including a first optical power and a second optical power, wherein the first optical power and the second optical power are any two different transmit and receive power among the transmit and receive power of each user-side device and the transmit and receive power of each PON port.

[0200] The determination module 1102 is specifically used to determine the correlation of multiple sets of optical power based on feature data.

[0201] The determination module 1102 determines the correlation of a set of optical powers by: determining the average time difference between the transactions of the first optical power and the second optical power, and the number of times the transactions of the first optical power and the second optical power occur simultaneously, based on the time of the transaction of the first optical power and the time of the transaction of the second optical power; and determining the correlation between the first optical power and the second optical power based on the average time difference between the transactions of the first optical power and the second optical power, the number of times the transactions of the first optical power and the second optical power occur simultaneously, and the total number of transactions of all received and transmitted optical powers.

[0202] In some embodiments, the determining module 1102 is specifically used to determine the average time difference between the occurrence of transactions of the first optical power and the second optical power using the following formula:

[0203]

[0204] Where Δt is the average time difference between the transactions of the first optical power and the second optical power, n-1 is the number of time periods into which the first preset time is averaged, and t mi Let t be the time during which a transaction occurs within the m-th time period divided by the first preset time period, where the first optical power is at that time. mj The time when the second optical power occurs within the m-th time period divided by the first preset time period, where n is an integer greater than or equal to 2 and m is an integer greater than or equal to 1 and less than or equal to n-1.

[0205] In some embodiments, the determining module 1102 is specifically used to determine the correlation between the first optical power and the second optical power using the following formula;

[0206]

[0207] Among them, Cor d (i, j) represents the correlation between the first optical power and the second optical power, T ij |t| represents the number of transactions occurring simultaneously at the first and second optical powers, ||T|| represents the total number of transactions occurring at all optical powers, and |t| represents the total number of transactions occurring at all optical powers. n -t n-1 | represents the length of the time period into which the first preset time is evenly divided, N ij The first optical power represents the correlation coefficient between the second optical power, and Δt is the average time difference between the first optical power and the second optical power.

[0208] In some embodiments, the acquisition module 1101 is further configured to acquire raw data of the PON, including the service connection relationship and physical connection relationship of the network node corresponding to the first optical power and the network node corresponding to the second optical power; the network node is a PON port or a user-side device.

[0209] The determination module 1102 is also used to determine the correlation coefficient between the first optical power and the second optical power based on the service connection relationship and the physical connection relationship.

[0210] In some embodiments, the ODN includes multiple optical splitters, and the feature data also includes the optical distance of each user-side device.

[0211] The determination module 1102 is also used to determine the possibility that any two or more user-side devices among multiple user-side devices can be connected to the same optical splitter based on the optical distance of each user-side device.

[0212] The determination module 1102 is specifically used to determine the topology of the ODN based on the first correlation, the second correlation, and the probability that any two or more user-side devices among multiple user-side devices are connected to the same optical splitter.

[0213] In some embodiments, the acquisition module 1101 is further configured to acquire the transmit and transmit power of each PON port in all PON ports of the OLT during a second preset time period and the transmit and transmit power of each user-side device in a plurality of user-side devices during the second preset time period.

[0214] The determination module 1102 is further configured to determine the fault optical path index feature model by training a model through artificial intelligence (AI) algorithm based on the transmit and transmit power of each PON port in all PON ports of the OLT in a second preset time period and the transmit and transmit power of each user-side device in multiple user-side devices in a second preset time period.

[0215] In an exemplary embodiment, this application also provides an electronic device, which may be the fault location device in the above method embodiments. Figure 12 This is a schematic diagram illustrating the composition of an electronic device provided in an embodiment of this application. For example... Figure 12 As shown, the electronic device may include a processor 1201 and a memory 1202; the memory 1202 stores instructions executable by the processor 1201; when the processor 1201 is configured to execute the instructions, the electronic device implements the method described in the foregoing method embodiments.

[0216] In an exemplary embodiment, this application also provides a computer-readable storage medium storing computer program instructions thereon; when the computer program instructions are executed by a computer, the computer causes the computer to implement the method described in the foregoing embodiments. The computer-readable storage medium may be a non-transitory computer-readable storage medium, such as a ROM, random access memory (RAM), CD-ROM, magnetic tape, floppy disk, and optical data storage device.

[0217] In an exemplary embodiment, this application also provides a computer program product that, when run on a computer, causes the computer to execute the aforementioned related method steps to implement the fault location method for ODN in PON as described in the above embodiment.

[0218] The above are merely specific embodiments of this application, but the scope of protection of this application is not limited thereto. Any variations or substitutions within the technical scope disclosed in this application should be included within the scope of protection of this application. Therefore, the scope of protection of this application should be determined by the scope of the claims.

Claims

1. A fault location method for the optical distribution network (ODN) in a passive optical network (PON), characterized in that, The PON includes an optical line terminal (OLT), an ODN, and multiple user-side devices, and the method includes: Acquire feature data; wherein, the feature data is used to indicate the optical power information of each PON port in all PON ports of the OLT and the optical power information of each user-side device in the plurality of user-side devices within a first preset time period; the feature data includes: the transmit and transmit power of each user-side device in the plurality of user-side devices, the time when the transmit and transmit power of each user-side device occurs, the transmit and transmit power of each PON port in all PON ports of the OLT, and the time when the transmit and transmit power of each PON port occurs; Based on the feature data, a first correlation and a second correlation are determined; the first correlation is used to indicate the correlation between each PON port of the OLT and each user-side device among the plurality of user-side devices, and the second correlation is used to indicate the correlation among the plurality of user-side devices. The topology of the ODN is determined based on the first correlation and the second correlation. Based on the fault optical path index feature model, the transmit and transmit power of each PON port of the OLT during the first preset time period and the transmit and transmit power of each user-side device among the plurality of user-side devices during the first preset time period are used to determine the cause of the fault; the fault optical path index feature model is used to perform similarity analysis, mutation analysis and overall trend analysis on the transmit and transmit power, and output the cause of the fault. Based on the cause of the fault and the topology of the ODN, the fault location of the ODN is performed; The first correlation and the second correlation include the correlation of multiple sets of optical power. Each set of optical power includes a first optical power and a second optical power. The first optical power and the second optical power are any two different transmit and receive power values ​​among the transmit and receive power values ​​of each user-side device and each PON port. Determining the first correlation and the second correlation based on the feature data includes: determining the correlation of the multiple sets of optical power based on the feature data; Determining the correlation of a set of optical powers includes: Based on the time when the first optical power and the time when the second optical power occur, determine the average time difference between the transactions of the first optical power and the second optical power, as well as the number of times the first optical power and the second optical power occur simultaneously. The correlation between the first optical power and the second optical power is determined based on the average time difference between transactions between the first optical power and the second optical power, the number of transactions between the first optical power and the second optical power occurring simultaneously, and the total number of transactions between all optical power and receiver.

2. The method according to claim 1, characterized in that, The average time difference between transactions occurring between the first optical power and the second optical power is determined using the following formula: in, Let n be the average time difference between the transactions occurring between the first optical power and the second optical power, and n-1 be the number of time periods into which the first preset time is divided. The time during which the first optical power experiences transactions within the m-th time period divided by the first preset time period. The time during which the second optical power occurs within the m-th time period divided by the first preset time period, where n is an integer greater than or equal to 2 and m is an integer greater than or equal to 1 and less than or equal to n-1.

3. The method according to claim 1, characterized in that, The correlation between the first optical power and the second optical power is determined using the following formula; in, This indicates the correlation between the first optical power and the second optical power. This indicates the number of times that the first optical power and the second optical power occur simultaneously. This represents the total number of transactions occurring across all received and transmitted power levels. This indicates the length of the time period into which the first preset time is divided equally. This represents the correlation coefficient between the first optical power and the second optical power. The average time difference between the first optical power and the second optical power transactions.

4. The method according to claim 3, characterized in that, Before determining the correlation between the first optical power and the second optical power, the method further includes: Obtain the raw data of the PON, which includes the service connection relationship and physical connection relationship between the network node corresponding to the first optical power and the network node corresponding to the second optical power; the network node is the PON port or the user-side equipment. Based on the service connection relationship and the physical connection relationship, the correlation coefficient between the first optical power and the second optical power is determined.

5. The method according to any one of claims 1-4, characterized in that, The ODN includes multiple optical splitters, and the feature data also includes: the optical distance of each user-side device; The method further includes: Based on the optical distance of each user-side device, determine the probability that any two or more user-side devices among the plurality of user-side devices can be connected to the same optical splitter. Determining the topology of the ODN based on the first correlation and the second correlation includes: The topology of the ODN is determined based on the first correlation, the second correlation, and the probability that any two or more user-side devices among the plurality of user-side devices are connected to the same optical splitter.

6. The method according to claim 1, characterized in that, Before determining the cause of the fault based on the fault optical path index characteristic model, the method further includes: Obtain the transmit and transmit power of each PON port in all PON ports of the OLT during a second preset time period and the transmit and transmit power of each user-side device in the plurality of user-side devices during the second preset time period; Based on the transmit and transmit power of each PON port in all PON ports of the OLT during the second preset time period and the transmit and transmit power of each user-side device among the multiple user-side devices during the second preset time period, a model is trained using artificial intelligence (AI) algorithms to determine the fault optical path index characteristic model.

7. A fault location device, characterized in that, For fault location in optical distribution network (ODN) in passive optical network (PON), the PON includes optical line terminal (OLT), the ODN, and multiple user-side devices, and the device includes: an acquisition module, a determination module, and a location module. The acquisition module is used to acquire feature data; wherein, the feature data is used to indicate the optical power information of each PON port in all PON ports of the OLT and the optical power information of each user-side device in the plurality of user-side devices within a first preset time period; the feature data includes: the transmit and transmit power of each user-side device in the plurality of user-side devices, the time of transaction of the transmit and transmit power of each user-side device, the transmit and transmit power of each PON port in all PON ports of the OLT, and the time of transaction of the transmit and transmit power of each PON port; The determining module is used to determine a first correlation and a second correlation based on the feature data; the first correlation is used to indicate the correlation between each PON port of the OLT and each user-side device among the plurality of user-side devices, and the second correlation is used to indicate the correlation among the plurality of user-side devices. The determining module is further configured to determine the topology of the ODN based on the first correlation and the second correlation; The determining module is further configured to determine the cause of the fault based on the fault optical path index feature model, the transmit and transmit power of each PON port of the OLT in the first preset time period, and the transmit and transmit power of each user-side device among the plurality of user-side devices in the first preset time period; the fault optical path index feature model is used to perform similarity analysis, mutation analysis and overall trend analysis on the transmit and transmit power, and output the cause of the fault. The positioning module is used to locate the fault in the ODN based on the cause of the fault and the topology of the ODN. The first correlation and the second correlation include the correlation of multiple sets of optical power. Each set of optical power includes a first optical power and a second optical power. The first optical power and the second optical power are any two different transmit and receive power values ​​among the transmit and receive power values ​​of each user-side device and each PON port. The determining module is specifically used to determine the correlation of the multiple sets of optical power based on the feature data; The determining module determines the correlation of a set of optical powers, including: Based on the time when the first optical power and the time when the second optical power occur, determine the average time difference between the transactions of the first optical power and the second optical power, as well as the number of times the first optical power and the second optical power occur simultaneously. The correlation between the first optical power and the second optical power is determined based on the average time difference between transactions between the first optical power and the second optical power, the number of transactions between the first optical power and the second optical power occurring simultaneously, and the total number of transactions between all optical power and receiver.

8. The apparatus according to claim 7, characterized in that, The acquisition module is also used to acquire the transmit and transmit power of each PON port in all PON ports of the OLT during a second preset time period and the transmit and transmit power of each user-side device in the plurality of user-side devices during the second preset time period. The determining module is further configured to perform model training using artificial intelligence (AI) algorithms based on the transmit and transmit power of each PON port in all PON ports of the OLT during a second preset time period and the transmit and transmit power of each user-side device among the plurality of user-side devices during the second preset time period, so as to determine the fault optical path index feature model.

9. An electronic device, characterized in that, The electronic device includes: a processor and a memory; The memory stores instructions that the processor can execute; When the processor is configured to execute the instructions, the electronic device performs the method as described in any one of claims 1-6.

10. A computer-readable storage medium, characterized in that, The computer-readable storage medium includes: computer software instructions; When the computer software instructions are executed in an electronic device, the electronic device causes the electronic device to perform the method as described in any one of claims 1-6.

Images