PON network-based onu gateway network configuration method, device, equipment and medium

Abstract

The application discloses an ONU gateway network configuration method and device based on a PON network, equipment and a medium. The method can be applied to network configuration in FTTR (fiber to the room), FTTR-B (enterprise-level fiber to the room) and broadband integrated terminal products. The method comprises the following steps: obtaining user input corresponding to a current network; performing intent recognition on the user input to determine a target intent and a target entity; creating a target knowledge graph based on the target intent and the target entity; generating a configuration strategy corresponding to the current network based on the target knowledge graph; executing the configuration strategy based on the current network; performing closed-loop verification on the current network to determine the execution effect of the configuration strategy; and optimizing and executing the configuration strategy based on the execution effect. The method can realize network configuration without the need for users to have professional network knowledge, helps users to realize zero learning cost, intelligent and automated network configuration, is efficient and convenient, and reduces network configuration cost.

Description

Technical Field

[0001] This invention relates to the field of network configuration technology, and in particular to a method, apparatus, device and medium for configuring an ONU gateway based on a PON network. Background Technology

[0002] With the widespread adoption of fiber optic access networks, ONUs (Optical Network Units) and home routers based on passive optical networks (PONs) have become standard equipment in countless households. These devices offer a wealth of features, such as QoS (Quality of Service) control, port forwarding, VLAN segmentation, parental controls, guest networks, and IPv6 configuration. However, configuring these features often requires specialized network knowledge. Ordinary users (such as home users and small and medium-sized enterprise network users) typically face difficulties in configuration due to a lack of expertise, leading to reduced network configuration efficiency and higher costs. Therefore, how to perform simple, efficient, and low-cost network configuration is a pressing technical problem that needs to be solved. Summary of the Invention

[0003] This invention provides a method, apparatus, device, and medium for configuring an ONU gateway network based on a PON network, in order to solve the technical problem of how to perform simple, efficient, and low-cost network configuration.

[0004] A method for configuring an ONU gateway network based on a PON network, comprising: Get the user input corresponding to the current network; The user input is subjected to intent recognition to determine the target intent and target entity; Based on the target intent and the target entity, a target knowledge graph is created; Based on the target knowledge graph, a configuration strategy corresponding to the current network is generated; The configuration policy is executed based on the current network. Perform closed-loop verification on the current network to determine the execution effect of the configuration strategy; Based on the execution results, optimize and execute the configuration strategy.

[0005] An ONU gateway network configuration device based on a PON network includes: The user input acquisition module is used to acquire user input corresponding to the current network. The intent recognition module is used to recognize the intent of the user input and determine the target intent and target entity; The knowledge graph creation module creates a target knowledge graph based on the target intent and the target entity; The configuration policy generation module generates the configuration policy corresponding to the current network based on the target knowledge graph. The configuration policy execution module executes the configuration policy based on the current network. The execution effect verification module is used to perform closed-loop verification on the current network and determine the execution effect corresponding to the configuration strategy. The configuration strategy optimization module optimizes and executes the configuration strategy based on the execution effect.

[0006] A computer device includes a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein the processor executes the computer program to implement the above-described ONU gateway network configuration method based on a PON network.

[0007] A computer-readable storage medium storing a computer program that, when executed by a processor, implements the above-described ONU gateway network configuration method based on a PON network.

[0008] The aforementioned ONU gateway network configuration method, apparatus, device, and medium based on a PON network determine the user's target intent and target entities for network configuration based on user input within the current network. Based on the target intent and target entities, a target knowledge graph is determined, and a configuration policy for network configuration is generated from the target knowledge graph. This enables the configuration of the network functions required by the user. This method does not require users to possess network expertise; it processes user input using natural language, resulting in a simple and efficient network configuration policy. Network configuration can be achieved without manual intervention, making it not only efficient and convenient but also avoiding resource waste and effectively reducing network configuration costs. During the execution of the configuration policy based on the current network, closed-loop verification is performed to determine the execution effect of the configuration policy. Based on the execution effect, the configuration policy is optimized and executed, enabling automatic optimization of the configuration policy without user debugging. This not only provides accurate network configuration but is also convenient and fast. This method can be widely applied to network configuration in terminal devices (e.g., terminal products corresponding to Fiber To The Room (FTTR) network coverage mode, enterprise-level Fiber To The Room-Business (FTTR-B) network coverage mode, and broadband converged terminal products) in home and small and medium-sized enterprise network configuration scenarios. It does not require users to have professional network knowledge. Through natural language processing and artificial intelligence technology, it automatically understands the intent of users' network configuration requirements described in everyday language, generates network configurations, verifies effects, and continuously optimizes configuration strategies, helping users achieve a truly zero-learning-cost, intelligent, and automated network management experience. Attached Figure Description

[0009] To more clearly illustrate the technical solutions of the embodiments of the present invention, the drawings used in the description of the embodiments of the present invention will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present invention. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0010] Figure 1 This is a flowchart of an ONU gateway network configuration method based on a PON network according to an embodiment of the present invention; Figure 2 This is another flowchart of an ONU gateway network configuration method based on a PON network in one embodiment of the present invention; Figure 3 This is another flowchart of an ONU gateway network configuration method based on a PON network in one embodiment of the present invention; Figure 4This is another flowchart of an ONU gateway network configuration method based on a PON network in one embodiment of the present invention; Figure 5 This is another flowchart of an ONU gateway network configuration method based on a PON network in one embodiment of the present invention; Figure 6 This is another flowchart of an ONU gateway network configuration method based on a PON network in one embodiment of the present invention; Figure 7 This is another flowchart of an ONU gateway network configuration method based on a PON network in one embodiment of the present invention; Figure 8 This is a schematic diagram of an ONU gateway network configuration device based on a PON network in one embodiment of the present invention; Figure 9 This is a schematic diagram of a computer device according to an embodiment of the present invention. Detailed Implementation

[0011] The technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some, not all, of the embodiments of the present invention. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.

[0012] This invention provides a PON-based ONU gateway network configuration method. This method involves: acquiring user input corresponding to the current network; identifying the user input intent to determine the target intent and target entity; creating a target knowledge graph based on the target intent and target entity; generating a configuration policy corresponding to the current network based on the target knowledge graph; executing the configuration policy based on the current network; performing closed-loop verification on the current network to determine the execution effect of the configuration policy; and optimizing and executing the configuration policy based on the execution effect. This PON-based ONU gateway network configuration method enables efficient and low-cost network configuration, helping users of devices connected to the network to perform simple, efficient, and low-cost network configuration.

[0013] In one embodiment, such as Figure 1 As shown, a method for configuring an ONU gateway network based on a PON network is provided, and this method is applied to... Figure 9 Taking a computer device as an example, the explanation includes the following steps: S101: Obtain user input corresponding to the current network; S102: Perform intent recognition on user input to determine the target intent and target entity; S103: Create a target knowledge graph based on target intent and target entities; S104: Generate the configuration strategy corresponding to the current network based on the target knowledge graph; S105: Execute the configuration policy based on the current network; S106: Perform closed-loop verification on the current network to determine the execution effect of the configuration policy; S107: Optimize and execute configuration strategies based on execution results.

[0014] Here, "current network" refers to the network required for network function configuration. In this example, the current network includes a passive optical network (PON). "User input" refers to the input from the user who needs to configure the network.

[0015] As an example, in step S101, the computer device acquires the user input corresponding to the network required for network function configuration, so as to perform network configuration based on the user input. In this example, the user input can be multi-modal, including but not limited to text, voice, and graphical interfaces. For example, the computer device acquires the user's voice dialogue or voice input through an intelligent voice dialogue system (such as a smart speaker), and uses the user's voice dialogue or voice input as the user input. For example, the computer device acquires the chat text between the user and the user through an APP (Application) or a network interface on a PC, and identifies the text as the user input. Another example is that the computer device acquires the graphical result selected by the user on a graphical interface set for the user (e.g., in the scenario of internet access time control, the user-defined visual graphical interface is a bar interface corresponding to the 0 to 24-hour time period; the user drags the bar interface to select the time period to control internet access, and obtains the selected graphical result, which is identified as the user input; or, a function selection button graphic is set for the user; the user selects the function to be configured by checking the selection button, and obtains the selected graphical result, which is identified as the user input). In this example, it can recognize and interact with multimodal user input, and is applicable to input scenarios with different user habits, demonstrating strong scenario applicability.

[0016] The target intent is used to characterize the network configuration required by the user. The target entity refers to the entity required for network configuration, which includes, but is not limited to, devices, time, values, actions, and applications / services, used to characterize the specific content of the network configuration.

[0017] As an example, in step S102, the computer device uses a preset intent recognition algorithm to identify the user input and determine the target intent and target entity corresponding to the user input. In this example, the preset intent recognition algorithm includes, but is not limited to, pre-trained large language models capable of intent recognition of multimodal data, such as the BERT model or the Chat-gpt model. For example, a computer device receives user input in a voice modal. The specific content of the user input includes the following voice dialogue: - User: "Xiao Zhi (wake word), prevent my son's phone from accessing the internet after 10 PM." - System: "Okay, I understand. You want to set up parental controls to restrict specific devices from accessing the internet after 10 PM. Do you want to restrict it every day, or only on weekdays?" - User: "Every day." - System: "Understood. What time will the restriction be lifted the next day?" - User: "7 AM." - System: "Okay, I will configure the following for you: Device: Your son's iPhone (192.168.1.105), Restriction Time: 10 PM to 7 AM the next day, Restriction Content: Internet access prohibited. Confirm execution?" - User: "Confirm." - System: "Configuration has taken effect. Verifying the effect. Verification complete. Currently working normally. It will automatically take effect at 10 PM." The computer device uses a pre-trained BERT model or Chat-gpt model to perform intent recognition on the above user input, and determines that the target intent is "parental control". The target entities include: the device entity is the son's iPhone (192.168.1.105), the time entity is 22:00 to 07:00 the next day, the numerical entity is limiting the network speed to below 10Mbps, the action entity is static network access, and the application entity / service entity is games or other apps that require network support.

[0018] In this example, the computer device employs a pre-defined intent recognition algorithm to identify various target intents corresponding to user input, including "parental control," "speed optimization," "port forwarding," "network access," "device management," "network diagnostics," "VLAN configuration," "bandwidth limiting," "scheduled restart," "firmware upgrade," and "backup and restore." The computer device can also identify various entities corresponding to user input, including device entities, time entities, numerical entities, action entities, and application / service entities, facilitating efficient network configuration.

[0019] The target knowledge graph refers to a knowledge graph that integrates target intents and target entities, used to generate network configuration strategies. The configuration strategy refers to the policy for configuring the network, used to guide network configuration.

[0020] As an example, in step S103, the computer device performs a knowledge graph creation operation based on the target intent and the target entity to obtain the target knowledge graph corresponding to the target intent and the target entity. For example, in step S102, when the target intent is "parental control," and the target entities include: the device entity is the son's iPhone (192.168.1.105), the time entity is 22:00-07:00 the next day, the numerical entity is limiting network speed to below 10Mbps, the action entity is static network access, and the application / service entity is a game software or other app that requires network support, the computer device creates a target knowledge graph, including: Parental control type: network function; Alias: "Children's Internet Management" or "Youth Protection"; Implementation method: time limit, content filtering, network speed (bandwidth) limit and application restriction; Related device: terminal device son's iPhone (192.168.1.105); Configuration parameters: time period, MAC address, ACL (Access Control List) rules; Time limit type: user expression "cannot access the Internet at night", "restricted usage time is 22:00-07:00 the next day"; Technical implementation: access control list; Maximum network bandwidth: 10Mbps; Configuration command "iptables -A FORWARD -m time ...".

[0021] As an example, in step S104, the computer device identifies the target knowledge graph corresponding to the user input, determines the configuration process that requires network configuration, and generates a configuration policy corresponding to the current network based on the configuration process. Here, the configuration process refers to the process corresponding to different types of network configurations. For example, for a target knowledge graph requiring parental control, the computer device needs to determine the configuration order of time limits, content filtering, network speed (bandwidth) limits, and application limits; that is, it needs to determine the configuration process and generate a configuration policy based on the configuration process. As another example, for a target knowledge graph requiring network optimization, the computer device needs to determine the configuration process for network optimization and generate a corresponding configuration policy based on the configuration process.

[0022] In this example, steps S101 to S104 can be completed in just tens of seconds, which can significantly improve network configuration efficiency compared to the existing technology that requires professional personnel for configuration.

[0023] As an example, in step S105, the computer device sends the generated configuration policy to the device corresponding to the current network, generates the configuration command corresponding to the configuration policy, and executes the configuration command to implement the configuration policy. For example, for the configuration policy corresponding to "parental control," the computer device sends the configuration policy to the home router corresponding to the current network used by the son's mobile phone, generates the configuration command corresponding to the configuration policy, and executes the configuration command to implement the configuration policy. In this example, the configuration policy required by the user can be generated without the user performing specific network configuration. This method generates network configuration policies based on user input using natural language, without requiring manual intervention in the specific network configuration process. This allows ordinary users without professional network configuration knowledge to obtain configuration policies simply and efficiently, which is not only efficient and convenient but also avoids resource waste and effectively reduces network configuration costs.

[0024] Closed-loop verification refers to a verification method that compares the execution results of the configuration strategy with the expected results to determine the effectiveness of the configuration strategy. Execution results refer to the effect of the configuration strategy during execution.

[0025] As an example, in step S106, during the execution of the configuration policy, the computer device collects network performance indicators corresponding to the target intent, determines the current network performance indicators, and compares and analyzes the current network performance indicators with the expected network performance indicators corresponding to the target intent to determine the execution effect of the configuration policy. In this example, network performance indicators include network latency and packet loss rate. During the execution of the configuration policy, the computer device collects network latency and packet loss rate, determines the network latency and packet loss rate during the execution of the configuration policy, compares the network latency during the execution of the configuration policy with the expected network latency, and determines whether the network latency during the execution of the configuration policy is not greater than the expected network latency. It also compares the packet loss rate during the execution of the configuration policy with the expected packet loss rate, and determines whether the packet loss rate during the execution of the configuration policy is greater than the expected packet loss rate. Based on whether the network latency and packet loss rate during the execution of the configuration policy are greater than the expected network latency and the expected packet loss rate, the execution effect of the configuration policy is determined. In this example, the computer device determines that the execution effect of the configuration policy has achieved the expected result when it determines that the network latency during the execution of the configuration policy is no greater than the expected network latency and the packet loss rate during the execution of the configuration policy is no greater than the expected packet loss rate. Conversely, if it determines that the network latency during the execution of the configuration policy is greater than the expected network latency, or the packet loss rate during the execution of the configuration policy is greater than the expected packet loss rate, it determines that the execution effect of the configuration policy has not achieved the expected result. This example determines the execution effect during the execution of the configuration policy in order to optimize the configuration policy based on the execution effect and achieve accurate network configuration.

[0026] As an example, in step S107, the computer device analyzes the execution effect of the configuration strategy. If it determines that the execution effect of the configuration strategy meets the expected requirements, it determines that the configuration strategy generated in step S104 can achieve the expected execution effect during execution and does not require optimization. In this case, the computer device continues to execute the configuration strategy generated in step S104. If the computer device determines that the execution effect of the configuration strategy does not meet the expected requirements, it determines that the configuration strategy generated in step S104 cannot achieve the expected execution effect during execution and needs optimization. In this case, the computer device performs performance optimization on the configuration strategy generated in step S104 to obtain an optimized configuration strategy, and executes the optimized configuration strategy to facilitate accurate network configuration.

[0027] In this embodiment, based on user input within the current network, the user's target intent and target entities for network configuration are determined. Based on these intents and entities, a target knowledge graph is identified, and a configuration strategy for network configuration is generated from this graph. This enables the configuration of the network functions required by the user. This method does not require users to possess network expertise; it processes user input using natural language, resulting in a simple and efficient network configuration strategy. Network configuration can be achieved without manual intervention, making it not only efficient and convenient but also avoiding resource waste and effectively reducing network configuration costs. During the execution of the configuration strategy based on the current network, closed-loop verification is performed to determine the execution effect of the configuration strategy. Based on the execution effect, the configuration strategy is optimized and executed, enabling automatic optimization without user debugging. This not only ensures accurate network configuration but is also convenient and fast. This method can be widely applied to network configuration in terminal devices (e.g., terminal products corresponding to Fiber To The Room (FTTR) network coverage mode, enterprise-level Fiber To The Room-Business (FTTR-B) network coverage mode, and broadband converged terminal products) in home and small and medium-sized enterprise network configuration scenarios. It does not require users to have professional network knowledge. Through natural language processing and artificial intelligence technology, it automatically understands the intent of users' network configuration requirements described in everyday language, generates network configurations, verifies effects, and continuously optimizes configuration strategies, helping users achieve a truly zero-learning-cost, intelligent, and automated network management experience.

[0028] In one embodiment, such as Figure 2 As shown, step S102, which involves performing intent recognition on the user input to determine the target intent and target entity, includes: S201: Perform text conversion on the user input to obtain the initial converted text; S202: Use a preset intent classification model to identify the intent of the initial converted text and determine the target intent corresponding to the initial converted text; S203: Based on the target intent and the initial transformed text, perform entity recognition to determine the initial entity corresponding to the initial transformed text; S204: Fill the slots based on the initial entity to form the target filling slots, and determine whether the target filling slots are complete; S205: If the target filling slot is incomplete, generate a user follow-up dialogue, perform entity recognition based on the user follow-up dialogue, determine the initial entity, and repeat the process of filling the slot based on the initial entity. S206: If the target fill slot is complete, then the initial entity in the target slot is determined as the target entity.

[0029] The initial converted text refers to the text after the user input has been converted.

[0030] As an example, in step S201, the computer device performs modal recognition on the user input. When it determines that the user input is in text mode, it determines the text mode user input as the initial converted text. When it determines that the user input is in graphic or audio mode, it performs modal conversion on the graphic or audio mode user input to obtain the initial converted text corresponding to the user input. For example, the computer device obtains the chat text between the user and an app (application) or PC-based web interface and determines the text mode chat text as the initial converted text. For example, the computer device obtains the interactive voice between the user and a smart speaker recorded and performs modal conversion on the audio mode interactive voice to obtain the initial converted text. As another example, the computer device obtains the graphic content selected by the user in a graphical interface (e.g., a bar graph corresponding to the length of the online time period, used to represent the length of the online time period), performs modal conversion on the graphic content of the graphic mode, and obtains the initial converted text corresponding to the user input.

[0031] The target intent refers to the user's intention to perform network configuration.

[0032] As an example, in step S202, the computer device uses a pre-defined intent classification model to identify the intent of the initial converted text, thereby obtaining the target intent corresponding to the initial converted text. In this example, during the process of identifying the intent of the initial converted text using the pre-defined intent classification model, the computer device increases the accuracy of intent identification by semantically enhancing the initial converted text. For example, the computer device performs semantic enhancement processing on the initial converted text to obtain semantically enhanced text, and then uses the pre-defined intent classification model to identify the intent of the semantically enhanced text, thereby obtaining the target intent corresponding to the initial converted text.

[0033] In this example, the computer device enhances semantics by standardizing colloquial expressions in the initial converted text, parsing ambiguous time periods, and disambiguating device references. For instance, for the initial converted text "Children shouldn't play on their phones at night," the computer device standardizes the colloquial expression by using the initial standardized expression "Restrict device network access during nighttime hours." For the initial converted text "late at night," the computer device uses "22:00-06:00" to parse the ambiguous time period. For the initial converted text "my phone," if it detects that the user has two phones, it continues to ask the user which phone it is to eliminate device ambiguity.

[0034] The initial entity refers to the entity obtained by directly identifying the target intent and the initial converted text, including but not limited to device entities, time entities, numerical entities, action entities, and application entities / service entities.

[0035] As an example, in step S203, the computer device uses a preset entity recognition model (e.g., NER model) to perform entity recognition on the target intent and the initial conversion text, and determines the initial entity corresponding to the initial conversion text, such as recognizing device entity, time entity, numerical entity, action entity, and application entity / service entity.

[0036] Slot filling refers to a method used to determine whether entity recognition of the initial converted text is complete. Target-filled slots refer to the slots after entity filling.

[0037] As an example, in step S204, the computer device fills the identified initial entity into the preset slot to obtain the target filling slot, and determines whether the target filling slot is complete in order to determine whether the slot identification is complete.

[0038] In this context, the user's follow-up questions are used to obtain more initial entities.

[0039] As an example, in step S205, when the computer device determines that the target filling slot is incomplete, i.e., there are still empty slots, it determines that the initial entity recognition is not yet complete. The computer device continues to generate a user follow-up dialogue, performs text conversion on the user follow-up dialogue to obtain the converted text, performs entity recognition on the converted text, continues to determine the initial entity, and repeats step S204, i.e., fills the slots based on the initial entities to form the target filling slots, and determines whether the target filling slots are complete, until the target filling slots are complete. For example, when the computer device determines that the target filling slots are incomplete, i.e., there are still empty slots, it determines the entity type corresponding to the empty slots. If it determines that it is a time entity, it generates a user follow-up dialogue for inquiring about time, performs text conversion on the user follow-up dialogue to obtain the converted text, performs entity recognition on the converted text, and obtains the initial entity related to time.

[0040] As an example, in step S206, when the computer device determines that the target filling slot is complete, that is, there is no empty slot, it determines that the initial entity recognition is complete, and determines all the initial entities that have been recognized as target entities, thereby realizing entity recognition of the initial converted text.

[0041] In this embodiment, user input is used to identify intent, and the target entity is accurately determined by filling slots.

[0042] In one embodiment, such as Figure 3 As shown, step S202, which involves using a preset intent classification model to identify the intent of the initial converted text and determine the initial intent and initial entity corresponding to the initial converted text, includes: S301: Use a preset intent classification model to identify the intent of the initial converted text and generate the initial intent; S302: Perform a confidence analysis on the initial intent and determine the results of the confidence analysis; S303: If the confidence analysis result does not meet the preset confidence conditions, then generate intent confirmation data, obtain user input based on the intent confirmation data, perform text conversion on the user input to obtain initial converted text, and repeat the process of using the preset intent classification model to identify intent in the initial converted text. S304: If the confidence analysis results meet the preset confidence conditions, then the initial intent is determined as the target intent corresponding to the initial converted text.

[0043] The initial intent refers to the user's desired network configuration intent, which is initially determined after the user performs intent recognition on the initial converted text using a preset intent classification model.

[0044] As an example, in step S301, the computer device directly uses a preset intent classification model to perform intent recognition on the initial converted text, and initially identifies the initial intent. For example, after the computer device recognizes "I want my child's phone to not be able to play games at night" in the initial converted text, it determines that the initial intent is "parental control".

[0045] The confidence analysis results are used to characterize the credibility of the initial intent.

[0046] As an example, in step S302, the computer device uses a preset confidence algorithm (e.g., a pre-trained neural network model) to perform confidence analysis on the initial intention, determine the reliability of the initial intention, and obtain the confidence analysis result corresponding to the initial intention, which is used to determine whether the initial intention is accurate and credible.

[0047] Among these, pre-set reliability conditions refer to the pre-defined conditions under which the initial intent is believed. Intent confirmation data refers to data used to help users confirm their intent.

[0048] As an example, in step S303, when the computer device determines that the confidence analysis result does not meet the preset confidence condition, it generates intent confirmation data and sends the intent confirmation data to the user terminal to obtain user input. The user input is then converted into text to obtain initial converted text. Step S301 is then repeated, that is, the intent recognition of the initial converted text is repeatedly performed using the preset intent classification model, and the initial intent is updated until the initial intent meets the preset confidence condition. In this example, the preset confidence condition is that the confidence value in the confidence analysis result is greater than 0.8.

[0049] As an example, in step S304, if the computer device determines that the confidence analysis result meets the preset confidence condition, then it determines the initial intent as the target intent corresponding to the initial converted text. In this example, the preset confidence condition is that the confidence value in the confidence analysis result is greater than 0.8. Understandably, the higher the confidence value in the confidence analysis result, the higher the credibility of the initial intent. Determining the initial intent that meets the preset confidence condition in the confidence analysis result as the target intent corresponding to the initial converted text achieves a more accurate and reliable determination of the target intent.

[0050] In this embodiment, by performing confidence analysis on the initially identified initial intent, the target intent corresponding to the initial converted text is accurately determined.

[0051] In one embodiment, such as Figure 4 As shown, step S104, which is to generate the configuration strategy corresponding to the current network based on the target knowledge graph, includes: S401: Identify the target intent type based on the target knowledge graph; S402: If the target intent type is optimization type, then based on the target knowledge graph, determine the optimization process and generate the first configuration strategy corresponding to the optimization process; S403: If the target intent type is a diagnosis type, then based on the target knowledge graph, determine the diagnosis process and generate the second configuration strategy corresponding to the diagnosis process; S404: If the target intent type is configuration type, then based on the target knowledge graph, determine the configuration process and generate the third configuration strategy corresponding to the configuration process.

[0052] The target intent type refers to the type of network configuration required by the user, including but not limited to optimization, diagnostic, and configuration types. Optimization types are used for network optimization, diagnostic types are used for diagnosing network problems, and configuration types are used for network rule matching.

[0053] As an example, in step S401, the computer device performs intent type identification on the target knowledge graph to determine the target intent type corresponding to the target knowledge graph. For example, the target knowledge graph includes: Parental control type: network function; Alias: "Children's Internet Management" or "Youth Protection"; Implementation method: time limit, content filtering, network speed (bandwidth) limit, and application restriction; Related device: son's iPhone (192.168.1.105); Configuration parameters: time period, MAC address, ACL (Access Control List) rules; Time limit type: user expression "cannot access the Internet at night", "restricted usage time is 22:00-07:00 the next day"; Technical implementation: Access Control List; Maximum network bandwidth: 10Mbps; Configuration command "iptables -A FORWARD -m time ...". The computer device performs intent type identification on the target knowledge graph and determines that the target intent type corresponding to the target knowledge graph is a configuration type.

[0054] The optimization process refers to the process used to configure user intents for optimization types. The first configuration strategy is the configuration strategy corresponding to the optimization process.

[0055] As an example, in step S402, when the computer device determines that the target intent type is an optimization type, it identifies the target knowledge graph, determines the target indicator data and optimization steps that need to be optimized, obtains the optimization process, and collects the current indicator data that has not been optimized in the current network. According to the optimization steps, it configures the current indicator data that needs to be optimized and the target indicator data that has not been optimized, generating a first configuration strategy. For example, for a game scenario that requires network speed optimization, when the target intent type is determined to be an optimization type, the target knowledge graph is identified, and the optimization steps are determined to be: increasing the network usage priority of the game and ensuring that the minimum bandwidth is greater than a preset threshold. The computer device obtains the priority level a that the game scenario needs to be increased to and the current priority level b of the game, the current minimum network bandwidth c, and the minimum bandwidth d that needs to be increased. According to the optimization steps, the first configuration strategy is set as: increasing the game priority from level b to level a and increasing the network bandwidth from c to d.

[0056] The diagnostic process refers to the process used to configure user intents for diagnostic types. The second configuration strategy is the configuration strategy corresponding to the diagnostic process.

[0057] As an example, in step S403, when the computer device determines that the target intent type is a diagnosis type, it determines the diagnosis process based on the target knowledge graph and generates a second configuration strategy based on the diagnosis process. For example, for a scenario requiring diagnosis of slow network speed, the computer device determines the diagnosis process as network speed testing, network speed bottleneck identification, and device traffic analysis based on the target knowledge graph, and further determines the second configuration strategy as follows: perform network speed testing, network speed bottleneck identification, and device traffic analysis in sequence, and finally output the reason for the slow network speed.

[0058] The configuration process refers to the process used to configure user intents of the optimization type. The third configuration strategy is the configuration strategy corresponding to the configuration process.

[0059] As an example, in step S404, when the computer device determines that the target intent type is a configuration type, it determines the configuration process based on the target knowledge graph and generates a third configuration policy corresponding to the configuration process. For example, in a scenario requiring parental control, network usage rules need to be controlled. Based on the target knowledge graph, the configuration process is determined to be limiting the network speed of the device corresponding to the MAC address within a target time period. Therefore, the configuration process is determined as follows: determine the MAC address corresponding to the device and the network speed limiting time period, and limit the network speed within a threshold range.

[0060] In this embodiment, the target intent types corresponding to different target knowledge graphs are identified, different target intent types are configured according to different processes, and configuration strategies corresponding to different target intent types are reasonably generated to achieve accurate strategy configuration for different target intent types.

[0061] In one embodiment, such as Figure 5 As shown, step S105, which is to execute the configuration policy based on the current network, includes: S501: Obtain the device capabilities of the device corresponding to the configuration policy; S502: Based on device capabilities, determine whether the device being sent supports the configuration policy; S503: If the device being sent supports configuration policies, then perform conflict detection on the current network and determine the conflict detection results; S504: If the conflict detection result indicates that a conflict exists, a target policy is generated based on the preset conflict adjustment strategy, the target policy is sent to the sending device, and the target policy is executed. S505: If the conflict detection result is that there is no conflict, the configuration policy will be sent to the sending device through the current network and the configuration policy will be executed.

[0062] In this context, the distributing device refers to the device that executes the configuration policy, such as a router. Device capability refers to the ability of the distributing device to execute the configuration policy.

[0063] As an example, in step S501, the computer device obtains the device capabilities corresponding to the issuing device that needs to execute the configuration policy. In this example, the computer device determines the issuing device that needs to execute the configuration policy and obtains the device capabilities, such as the configuration functions supported by the issuing device.

[0064] As an example, in step S502, the computer device determines whether the configuration policy is within the scope of configuration functions supported by the issuing device. If it determines that the configuration policy is within the scope of configuration functions supported by the issuing device, it determines that the issuing device supports the configuration policy. If it determines that the configuration policy is not within the scope of configuration functions supported by the issuing device, it determines that the issuing device does not support the configuration policy. For example, when the issuing device is a router, if the router supports network configuration with diagnostic functions, it determines that the router can support the configuration of the second configuration policy corresponding to the diagnostic process. If the issuing device is a router, but the router does not support network configuration with diagnostic functions, it determines that the router cannot support the configuration of the second configuration policy corresponding to the diagnostic process.

[0065] Conflict detection refers to detecting whether there are execution conflicts between different devices in the current network. The conflict detection result indicates whether the configuration policy can be issued immediately.

[0066] As an example, in step S503, when the computer device determines that the issuing device corresponding to the configuration policy supports the configuration policy, it further detects whether there is a conflict in the current network for policy execution, that is, it performs conflict detection on the current network and obtains a conflict detection result. The conflict detection result includes two cases: no network conflict and network conflict exists. In this example, the computer device detects whether there are multiple working devices working simultaneously in the current network. If there are multiple working devices, the conflict detection result is determined to be a network conflict; if there are no multiple working devices, the conflict detection result is determined to be no network conflict.

[0067] Among them, the preset conflict adjustment strategy refers to the pre-defined strategy used to adjust network conflicts. The target strategy refers to the configured strategy. As an example, in step S504, when the computer device determines that a conflict exists in the conflict detection result, it obtains a preset conflict adjustment strategy, generates a target strategy based on the preset conflict adjustment strategy, sends the target strategy to the sending device, and executes the target strategy. In this example, when the computer device determines that a conflict exists, it can resolve the conflict and generate a strategy using the following three methods: First method: Obtain the configuration strategy sending time selected by the user, determine the configuration strategy as the target strategy, send the target strategy to the sending device at the user-selected sending time, and execute the target strategy. Second method: Merge the configuration strategy to be sent with the strategy currently being sent by the network to obtain the merged target strategy, send the target strategy to the corresponding sending device, and execute the target strategy. This allows for the simultaneous sending of multiple strategies, improving network configuration efficiency. Third method: Determine the priority of the conflicting configuration strategies based on preset priorities, generate multiple target strategies corresponding to the configuration strategies according to the priority order, and send and execute the target strategies according to the priority order.

[0068] As an example, in step S505, when the computer device determines that there is no conflict in the conflict detection result, it determines that the configuration policy to be executed this time can be directly issued and executed, the configuration policy is issued to the issuing device through the current network, and the configuration policy is executed.

[0069] In this embodiment, before issuing the configuration policy, the device capabilities of the issuing device and whether there are any conflicts with the current network are detected. This ensures the feasibility of policy issuance and effectively improves the efficiency of policy issuance and execution.

[0070] In one embodiment, such as Figure 6 As shown, step S106, which involves performing closed-loop verification on the current network to determine the execution effect of the configuration policy, includes: S601: Get the distribution status of the configuration policy; S602: If the distribution status is a distribution failure status, then execute the rollback strategy; S603: If the distribution status is successful, then within the preset detection period, performance indicators are obtained at preset time intervals. Based on the performance indicators, the configuration strategy is analyzed to determine the execution effect of the configuration strategy.

[0071] The distribution status is used to characterize the distribution status of the configuration policy.

[0072] As an example, in step S601, after a period of time following the issuance of the configuration policy, the computer device obtains the issuance status of the configuration policy and determines whether the issuance status of the configuration policy is a successful issuance status, so as to monitor whether the configuration policy has been successfully issued in real time based on the issuance status.

[0073] As an example, in step S602, when the computer device determines that the distribution status is a distribution failure status, it executes a rollback strategy to ensure the atomicity of the configuration strategy, so as to prevent network failures caused by incorrect configuration and improve the security performance of network configuration.

[0074] The preset detection period refers to a pre-defined time interval after the configuration policy begins execution. The preset time interval refers to the time interval for conducting detection. Performance metrics are indicators used to characterize the effectiveness of network configuration.

[0075] As an example, in step S603, when the computer device determines that the distribution status is successful, it acquires performance indicators at preset time intervals during a preset detection period after successful distribution. Based on the performance indicators, it analyzes the configuration strategy to determine the execution effect of the configuration strategy. Understandably, the required performance indicators differ depending on the target intent. For example, when the target intent is game optimization, the performance indicators are network latency, network jitter, and packet loss rate; when the target intent is video optimization, the performance indicators are buffer time and bitrate stability; when the target intent is parental control, the performance indicators are whether MAC address restrictions on internet access are effective, whether the network restriction time period is accurate, whether network bandwidth is limited, and the connectivity of port forwarding. In this example, when the computer device determines that the target intent is game optimization, it determines that the configuration strategy execution effect meets the expected requirements when the network latency is less than a preset latency value, the jitter duration is less than a preset duration, and the packet loss rate is less than a preset packet loss rate threshold. When the computer device determines that the target intent is video optimization, and among the performance metrics, the buffer time is less than the preset buffer time threshold and the bitrate stability is high (small changes in sampling bitrate), the configuration policy execution effect is determined to meet the expected requirements. When the computer device determines that the target intent is parental control, and among the performance metrics, MAC address internet access restrictions are effective, network restriction time periods are accurate, network bandwidth restrictions are within the preset bandwidth range, and port forwarding connectivity is low, the configuration policy execution effect is determined to meet the expected requirements.

[0076] In this embodiment, when the distribution status is in the distribution failure state, a rollback strategy is executed to ensure the atomicity of the configuration strategy, thereby preventing network failures caused by incorrect configuration and improving the security performance of network configuration. When the distribution status is in the distribution success state, the configuration strategy is analyzed based on performance indicators to determine the execution effect of the configuration strategy, so as to facilitate real-time monitoring of network configuration and ensure the configuration effect of network configuration.

[0077] In one embodiment, such as Figure 7 As shown, step S107, namely, optimizing and executing the configuration strategy based on the execution effect, includes: S701: If the execution effect does not meet the expectations, then determine the optimization indicators based on the expected effect, optimize the configuration strategy based on the optimization indicators, and determine and execute the optimized configuration strategy. S702: If the execution effect achieves the expected result, continue to execute the configuration strategy until the configuration strategy is completed and the execution effect is displayed.

[0078] The expected effect refers to the execution effect of the pre-configured equipment strategy.

[0079] As an example, in step S701, when the computer device determines that the execution effect has not met expectations, it determines the optimization indicators that need to be optimized based on the expected effect, optimizes the configuration strategy according to the optimization indicators, and determines and executes the optimized configuration strategy to ensure the network configuration effect. For example, if the computer device's target intention is game optimization, and it determines that the network latency is not less than a preset latency value among the performance indicators, and it determines that the execution effect of network latency has not met expectations, it determines that network latency control is the target that needs to be optimized, optimizes the network latency, obtains and executes the optimized configuration strategy to achieve the purpose of game optimization.

[0080] As an example, in step S702, when the computer device determines that the execution effect has achieved the expected effect, it continues to execute the configuration policy until the configuration policy is completed, and displays the execution effect on the user interface so that the user can intuitively understand the effect of the network configuration.

[0081] In this embodiment, different methods are used to execute the configuration strategy according to whether the execution effect achieves the expected effect, so as to ensure that the network configuration has a good configuration effect. This method can automatically optimize the execution strategy without requiring manual debugging by the user. It is not only convenient and fast, but also reduces debugging costs and has a strong adaptability to network environment.

[0082] It should be understood that the sequence number of each step in the above embodiments does not imply the order of execution. The execution order of each process should be determined by its function and internal logic, and should not constitute any limitation on the implementation process of the embodiments of the present invention.

[0083] In one embodiment, an ONU gateway network configuration device based on a PON network is provided, which corresponds one-to-one with the ONU gateway network configuration method based on a PON network in the above embodiments. For example... Figure 8 As shown, the ONU gateway network configuration device based on a PON network includes a user input acquisition module 801, an intent recognition module 802, a knowledge graph creation module 803, a configuration policy generation module 804, a configuration policy execution module 805, an execution effect verification module 806, and a configuration policy optimization module 807. Detailed descriptions of each functional module are as follows: User input acquisition module 801 is used to acquire user input corresponding to the current network. The intent recognition module 802 is used to recognize the user input intent and determine the target intent and target entity. The knowledge graph creation module 803 creates a target knowledge graph based on the target intent and target entities; The configuration policy generation module 804 generates the configuration policy corresponding to the current network based on the target knowledge graph. Configuration policy execution module 805 executes configuration policies based on the current network; The execution effect verification module 806 is used to perform closed-loop verification of the current network and determine the execution effect corresponding to the configuration strategy. Configuration strategy optimization module 807 optimizes and executes configuration strategies based on execution results.

[0084] In one embodiment, the intent recognition module 802 includes: The text conversion submodule is used to convert user input into text and obtain the initial converted text; The target intent determination submodule is used to identify the intent of the initial converted text using a preset intent classification model, and to determine the target intent corresponding to the initial converted text. The initial entity determination submodule is used to perform entity recognition based on the target intent and the initial transformed text, and to determine the initial entity corresponding to the initial transformed text. The slot filling submodule is used to fill slots based on the initial entity, form the target filling slot, and determine whether the target filling slot is complete. The first judgment submodule is used to generate a user follow-up dialogue if the target filling slot is incomplete, perform entity recognition based on the user follow-up dialogue, determine the initial entity, and repeatedly perform slot filling based on the initial entity. The second judgment submodule is used to determine the initial entity in the target slot as the target entity if the target filling slot is complete.

[0085] In one embodiment, the target intent determination submodule includes: The initial intent generation unit is used to identify the intent of the initial transformed text using a preset intent classification model and generate the initial intent. The confidence analysis unit is used to perform confidence analysis on the initial intent and determine the confidence analysis results; The first confidence judgment unit is used to generate intent confirmation data if the confidence analysis result does not meet the preset confidence conditions, and to obtain user input based on the intent confirmation data, perform text conversion on the user input to obtain initial converted text, and repeatedly perform intent recognition on the initial converted text using a preset intent classification model. The second confidence judgment unit is used to determine the initial intent as the target intent corresponding to the initial converted text if the confidence analysis result meets the preset confidence conditions.

[0086] In one embodiment, the configuration policy generation module 804 includes: The intent type identification submodule identifies the target intent type based on the target knowledge graph; The first configuration strategy determination submodule is used to determine the optimization process based on the target knowledge graph and generate the first configuration strategy corresponding to the optimization process if the target intent type is optimization type. The second configuration strategy determination submodule is used to determine the diagnostic process based on the target knowledge graph and generate the second configuration strategy corresponding to the diagnostic process if the target intent type is a diagnostic type. The third configuration strategy determination submodule is used to determine the configuration process based on the target knowledge graph and generate the third configuration strategy corresponding to the configuration process if the target intent type is a configuration type.

[0087] In one embodiment, the configuration policy execution module 805 includes: The device capability acquisition submodule is used to acquire the device capabilities of the devices corresponding to the configuration policy. The device capability judgment submodule determines whether the device supports the configuration policy based on its capabilities. The conflict detection submodule is used to perform conflict detection on the current network and determine the conflict detection result if the issuing device supports the configuration policy. The first strategy execution submodule is used to generate a target strategy based on the preset conflict adjustment strategy if the conflict detection result is that a conflict exists, and then send the target strategy to the sending device and execute the target strategy. The second strategy execution submodule is used to send the configuration policy to the sending device through the current network and execute the configuration policy if the conflict detection result is that there is no conflict.

[0088] In one embodiment, the performance verification module 806 includes: The distribution status acquisition submodule is used to obtain the distribution status of the configuration policy; The rollback strategy execution submodule is used to execute the rollback strategy if the distribution status is a distribution failure status. The execution effect determination submodule is used to obtain performance indicators at preset time intervals within a preset detection period if the distribution status is successful. Based on the performance indicators, the configuration strategy is analyzed to determine the execution effect corresponding to the configuration strategy.

[0089] In one embodiment, the configuration policy optimization module 807 includes: The first optimization module is used to determine optimization indicators based on the expected results if the execution effect does not meet the expectations, optimize the configuration strategy based on the optimization indicators, and determine and execute the optimized configuration strategy. The second optimization module is used to continue executing the configuration strategy if the execution effect meets the expectations, until the configuration strategy is completed and the execution effect is displayed.

[0090] Specific limitations regarding the ONU gateway network configuration device based on a PON network can be found in the limitations of the ONU gateway network configuration method based on a PON network described above, and will not be repeated here. Each module in the aforementioned ONU gateway network configuration device based on a PON network can be implemented entirely or partially through software, hardware, or a combination thereof. These modules can be embedded in or independent of the processor in the computer device in hardware form, or stored in the memory of the computer device in software form, so that the processor can call and execute the operations corresponding to each module.

[0091] In one embodiment, a computer device is provided, which may be a server, and its internal structure diagram may be as follows: Figure 9 As shown, the computer device includes a processor, memory, network interface, and database connected via a system bus. The processor provides computing and control capabilities. The memory includes non-volatile storage media and internal memory. The non-volatile storage media stores the operating system, computer programs, and database. The internal memory provides an environment for the operation of the operating system and computer programs stored in the non-volatile storage media. The database stores data used or generated during the execution of a PON-based ONU gateway network configuration method. The network interface communicates with external terminals via a network connection. When the computer program is executed by the processor, it implements a PON-based ONU gateway network configuration method.

[0092] In one embodiment, a computer device is provided, including a memory, a processor, and a computer program stored in the memory and executable on the processor. When the processor executes the computer program, it implements the ONU gateway network configuration method based on a PON network as described in the above embodiments, for example... Figure 1 As shown in S101-S107, or Figures 2 to 7 As shown, to avoid repetition, it will not be described again here. Alternatively, when the processor executes the computer program, it implements the functions of each module / unit in this embodiment of the ONU gateway network configuration device based on a PON network, for example... Figure 8 The functions of the user input acquisition module 801, intent recognition module 802, knowledge graph creation module 803, configuration strategy generation module 804, configuration strategy execution module 805, execution effect verification module 806, and configuration strategy optimization module 807 shown are not described in detail here to avoid repetition.

[0093] In one embodiment, a computer-readable storage medium is provided, on which a computer program is stored. When executed by a processor, the computer program implements the ONU gateway network configuration method based on a PON network as described in the above embodiment, for example... Figure 1As shown in S101-S107, or Figures 2 to 7 As shown, to avoid repetition, it will not be described again here. Alternatively, when the computer program is executed by the processor, it implements the functions of each module / unit in this embodiment of the ONU gateway network configuration device based on a PON network, for example... Figure 8 The functions of the user input acquisition module 801, intent recognition module 802, knowledge graph creation module 803, configuration strategy generation module 804, configuration strategy execution module 805, execution effect verification module 806, and configuration strategy optimization module 807 shown are not described again here to avoid repetition. The computer-readable storage medium may be non-volatile or volatile.

[0094] Those skilled in the art will understand that all or part of the processes in the methods of the above embodiments can be implemented by a computer program instructing related hardware. This computer program can be stored in a non-volatile computer-readable storage medium. When executed, the computer program can include the processes of the embodiments of the above methods. Any references to memory, storage, databases, or other media used in the embodiments provided in this application can include non-volatile and / or volatile memory. Non-volatile memory may include read-only memory (ROM), programmable ROM (PROM), electrically programmable ROM (EPROM), electrically erasable programmable ROM (EEPROM), or flash memory. Volatile memory may include random access memory (RAM) or external cache memory. By way of illustration and not limitation, RAM is available in a variety of forms, such as static RAM (SRAM), dynamic RAM (DRAM), synchronous DRAM (SDRAM), dual data rate SDRAM (DDRSDRAM), enhanced SDRAM (ESDRAM), synchronous link DRAM (SLDRAM), RAMbus direct RAM (RDRAM), direct memory bus dynamic RAM (DRDRAM), and memory bus dynamic RAM (RDRAM), etc.

[0095] Those skilled in the art will clearly understand that, for the sake of convenience and brevity, the above-described division of functional units and modules is used as an example. In practical applications, the above functions can be assigned to different functional units and modules as needed, that is, the internal structure of the device can be divided into different functional units or modules to complete all or part of the functions described above.

[0096] The above-described embodiments are only used to illustrate the technical solutions of the present invention, and are not intended to limit it. Although the present invention has been described in detail with reference to the foregoing embodiments, those skilled in the art should understand that modifications can still be made to the technical solutions described in the foregoing embodiments, or equivalent substitutions can be made to some of the technical features. Such modifications or substitutions do not cause the essence of the corresponding technical solutions to deviate from the spirit and scope of the technical solutions of the embodiments of the present invention, and should all be included within the protection scope of the present invention.

Claims

1. A method for configuring an ONU gateway network based on a PON network, characterized in that, include: Get the user input corresponding to the current network; The user input is subjected to intent recognition to determine the target intent and target entity; Based on the target intent and the target entity, a target knowledge graph is created; Based on the target knowledge graph, a configuration strategy corresponding to the current network is generated; The configuration policy is executed based on the current network. Perform closed-loop verification on the current network to determine the execution effect of the configuration strategy; Based on the execution results, optimize and execute the configuration strategy.

2. The ONU gateway network configuration method based on a PON network according to claim 1, characterized in that, The step of performing intent recognition on the user input to determine the target intent and target entity includes: The user input is converted into text to obtain the initial converted text; A preset intent classification model is used to identify the intent of the initial converted text and determine the target intent corresponding to the initial converted text. Based on the target intent and the initial converted text, entity recognition is performed to determine the initial entity corresponding to the initial converted text; Based on the initial entity, slots are filled to form target filling slots, and it is determined whether the target filling slots are complete. If the target filling slot is incomplete, a user follow-up dialogue is generated, entity recognition is performed based on the user follow-up dialogue to determine the initial entity, and the process of filling the slot based on the initial entity is repeated. If the target filling slot is complete, the initial entity in the target slot is determined as the target entity.

3. The ONU gateway network configuration method based on a PON network according to claim 2, characterized in that, The step of using a preset intent classification model to identify the intent of the initial converted text and determine the target intent corresponding to the initial converted text includes: The initial intent is generated by using the preset intent classification model to identify the intent of the initial converted text. A confidence analysis is performed on the initial intent to determine the confidence analysis results; If the confidence analysis result does not meet the preset confidence conditions, then intent confirmation data is generated, and the user input is obtained based on the intent confirmation data. The user input is then converted into text to obtain initial converted text, and the process of using the preset intent classification model to identify intent in the initial converted text is repeated. If the confidence analysis result meets the preset confidence conditions, then the initial intent is determined as the target intent corresponding to the initial converted text.

4. The ONU gateway network configuration method based on a PON network according to claim 1, characterized in that, The step of generating the configuration strategy corresponding to the current network based on the target knowledge graph includes: Based on the target knowledge graph, intent type identification is performed to determine the target intent type; If the target intent type is an optimization type, then based on the target knowledge graph, an optimization process is determined, and a first configuration strategy corresponding to the optimization process is generated; If the target intent type is a diagnosis type, then based on the target knowledge graph, a diagnosis process is determined, and a second configuration strategy corresponding to the diagnosis process is generated; If the target intent type is a configuration type, then based on the target knowledge graph, a configuration process is determined, and a third configuration strategy corresponding to the configuration process is generated.

5. The ONU gateway network configuration method based on a PON network according to claim 1, characterized in that, The execution of the configuration policy based on the current network includes: Obtain the device capabilities of the device corresponding to the configuration policy; Based on the device capabilities, determine whether the issuing device supports the configuration strategy; If the issuing device supports the configuration policy, then perform conflict detection on the current network and determine the conflict detection result; If the conflict detection result indicates that a conflict exists, a target strategy is generated based on a preset conflict adjustment strategy, and the target strategy is sent to the sending device for execution. If the conflict detection result indicates that there is no conflict, the configuration policy is distributed to the distributing device through the current network and the configuration policy is executed.

6. The ONU gateway network configuration method based on a PON network according to claim 1, characterized in that, The step of performing closed-loop verification on the current network to determine the execution effect of the configuration policy includes: Obtain the distribution status of the configuration strategy; If the delivery status is a delivery failure status, then the rollback strategy is executed; If the distribution status is a successful distribution status, then within a preset detection period, performance indicators are obtained at preset time intervals. Based on the performance indicators, the configuration strategy is analyzed to determine the execution effect corresponding to the configuration strategy.

7. The ONU gateway network configuration method based on a PON network according to claim 1, characterized in that, The step of optimizing and executing the configuration strategy based on the execution effect includes: If the execution effect does not achieve the expected effect, then an optimization index is determined based on the expected effect, the configuration strategy is optimized based on the optimization index, and the optimized configuration strategy is determined and executed. If the execution effect achieves the expected result, the configuration strategy will continue to be executed until the configuration strategy is completed, and the execution effect will be displayed.

8. A network configuration device for an ONU gateway based on a PON network, characterized in that, include: The user input acquisition module is used to acquire user input corresponding to the current network. The intent recognition module is used to recognize the intent of the user input and determine the target intent and target entity; The knowledge graph creation module creates a target knowledge graph based on the target intent and the target entity; The configuration policy generation module generates the configuration policy corresponding to the current network based on the target knowledge graph. The configuration policy execution module executes the configuration policy based on the current network. The execution effect verification module is used to perform closed-loop verification on the current network and determine the execution effect corresponding to the configuration strategy. The configuration strategy optimization module optimizes and executes the configuration strategy based on the execution effect.

9. A computer device comprising a memory, a processor, and a computer program stored in the memory and executable on the processor, characterized in that, When the processor executes the computer program, it implements the ONU gateway network configuration method based on a PON network as described in any one of claims 1 to 7.

10. A computer-readable storage medium storing a computer program, characterized in that, When the computer program is executed by the processor, it implements the ONU gateway network configuration method based on the PON network as described in any one of claims 1 to 7.

Images