FTTR service delivery quality control method and device, electronic equipment and medium

Abstract

The invention provides an FTTR service delivery quality control method and apparatus, an electronic device and a medium. The method comprises the steps of receiving a main optical modem installation scene image uploaded by a smart engineer; the main optical modem installation scene image is input into the trained AI intelligent quality inspection model, if the trained AI intelligent quality inspection model identifies that the main optical modem installation scene image has no copying features and no weak current box features and contains a main optical modem, it is judged that the main optical modem installation scene is compliant, and otherwise, it is judged that the main optical modem installation scene is not compliant; in response to the fact that the main optical modem installation scene is not compliant, a first interception instruction is automatically triggered, and a first rectification guide is pushed; by comparing the actual deployment number of the slave optical modems in the installation verification system with the order number of the slave optical modems in the front-end acceptance system and judging whether a supply source is compliant or not, the compliance of access of the slave optical modems is verified. According to the embodiment of the invention, the standardized and intelligent management and control of the FTTR service delivery quality can be realized.

Description

Technical Field

[0001] This disclosure relates to the field of whole-house optical broadband communication technology, and in particular to a method, apparatus, electronic device and medium for FTTR service delivery quality control. Background Technology

[0002] FTTR (Fiber to the Room) whole-house fiber broadband service extends fiber optic cables to various areas of a home or office, achieving high-speed network coverage through a master-slave optical modem collaborative networking system. Its delivery quality directly affects user experience and service efficiency.

[0003] However, existing FTTR business delivery quality control methods have at least the following shortcomings:

[0004] (1) The lack of a closed-loop process and reliance on manual labor result in weak constraints.

[0005] Existing methods rely heavily on the experience and self-discipline of installation and maintenance personnel in various stages of FTTR delivery (such as installation, equipment connection, and acceptance), lacking a rigid closed loop of "intelligent identification - technical interception - mandatory rectification - effect verification" implemented by a technical system. This leads to delays in the discovery and rectification of problems such as non-standard installation and non-compliant equipment, failing to ensure a mandatory technical closed loop from problem discovery to resolution, and resulting in insufficient quality control constraints.

[0006] (2) Installation compliance is easily overlooked, and source control is insufficient.

[0007] During the installation of the main optical modem, the system relies on manual judgment to determine whether the installation location of the main optical modem (such as whether it is placed in the weak current box) is compliant. This results in significant subjective judgment differences and easy omissions, making it impossible to eliminate potential network quality risks caused by improper installation from the source.

[0008] (3) The lack of automated verification for device access poses operational and maintenance risks.

[0009] In the process of accessing the optical modem, there is a lack of system-level automated real-time verification and mandatory interception mechanisms to ensure that the number of deployed devices matches the customer's order and that the source of supply is compliant. This can easily lead to over-deployment and the entry of non-compliant devices into the network, increasing network operation and maintenance risks. Summary of the Invention

[0010] This disclosure provides a method, apparatus, electronic device, and medium for FTTR service delivery quality control, which addresses the problems of existing FTTR service delivery quality control methods, such as missing process loops, easy omissions in installation compliance, and lack of automated verification for equipment access, resulting in operational and maintenance risks.

[0011] Firstly, this disclosure provides a method for quality control of FTTR service delivery, the method comprising:

[0012] During the installation and deployment of the FTTR service, images of the main optical modem installation scene uploaded by the smart home engineer are received;

[0013] The main optical modem installation scene image is input into the trained AI intelligent quality inspection model. If the trained AI intelligent quality inspection model identifies that the main optical modem installation scene image has no re-photographing characteristics, no weak current box characteristics, and contains the main optical modem, then the main optical modem installation scene is deemed compliant; otherwise, the main optical modem installation scene is deemed non-compliant.

[0014] In response to the non-compliant installation scenario of the main optical modem, the first interception command is automatically triggered, and the first rectification guide is pushed out;

[0015] By comparing the actual number of optical modems deployed in the installation verification system with the number of optical modems ordered in the front-end acceptance system, and determining whether the supply source is compliant, the compliance of the access from the optical modems is verified.

[0016] In response to non-compliant access from the optical modem, a second interception command is automatically triggered, and a second rectification guide is pushed out;

[0017] In response to the compliance of both the main optical modem installation scenario and the secondary optical modem access, the system receives the acceptance request sent by the smart home engineer, calls the converged network management platform interface according to the acceptance request, obtains the core data of the corresponding main and secondary optical modems, and performs automated completion acceptance based on the core data.

[0018] If the automated completion acceptance result is qualified, a visual report including acceptance data will be generated; otherwise, the corresponding rectification process will be triggered.

[0019] Furthermore, before inputting the main optical modem installation scene image into the trained AI intelligent quality inspection model, the method further includes:

[0020] Obtain a training dataset containing image samples of both compliant and non-compliant installation scenarios of the main optical modem;

[0021] The training dataset is augmented using data augmentation techniques to obtain an augmented training dataset, wherein the data augmentation techniques include at least one of rotating, cropping, and adjusting the brightness of the image;

[0022] The AI ​​intelligent quality inspection model based on a convolutional neural network (CNN) is trained using the expanded training dataset to obtain the trained AI intelligent quality inspection model.

[0023] Furthermore, the compliance of the access from the optical modem is verified by comparing the actual number of optical modems deployed in the installation verification system with the number of optical modems ordered in the front-end acceptance system, and determining whether the supply source is compliant. This specifically includes:

[0024] The actual number of optical modems deployed in the installation verification system is compared with the number of optical modems ordered in the front-end acceptance system. If the difference is 0, it is determined that the optical modems are deployed normally. If the difference is greater than or equal to 1, it is determined that the optical modems are deployed in excess.

[0025] Determine whether the device identifiers of each deployed slave optical modem exist in the preset list of compliant optical modem suppliers. If the device identifiers of all deployed slave optical modems exist in the list of compliant optical modem suppliers, the supply source is determined to be compliant; otherwise, the supply source is determined to be non-compliant.

[0026] Based on the normal deployment and compliant supply source of the optical modem, the access from the optical modem is deemed compliant.

[0027] In response to excessive deployment of optical modems or non-compliant supply sources, the connection from the optical modem is determined to be non-compliant.

[0028] Furthermore, the method also includes:

[0029] An interface adaptation module is deployed between the local optical modem terminal network management system and the cloud platform terminal network management system. A millisecond-level data synchronization channel is built using the Flink framework. The data synchronization channel is then used to unify and aggregate the master and slave optical modem data in the local optical modem terminal network management system and the cloud platform terminal network management system to the converged network management platform.

[0030] By utilizing the northbound interface of the converged network management platform and the data middle platform, the data of the master and slave optical modems in the converged network management platform are synchronized to the data middle platform.

[0031] Furthermore, the core data includes the online status and received light value of the master and slave optical modems, as well as the uplink connection method of the slave optical modem. The automated final acceptance testing based on the core data specifically includes:

[0032] Based on the online status and received light value of the master and slave optical modems, as well as the uplink connection mode of the slave optical modem, the following judgments are made: whether the master and slave optical modems are online, whether the received light value of the master and slave optical modems is within the preset threshold range, and whether the uplink mode of the slave optical modem is wireless.

[0033] In response to the fact that both the master and slave optical modems are online, the received light values ​​of the master and slave optical modems are within the preset threshold range, and the uplink mode of the slave optical modem is not wireless, the result of the automated completion acceptance is determined to be qualified.

[0034] Furthermore, the method also includes:

[0035] Collect multiple key technical indicators of home networks corresponding to FTTR services from the data platform at regular or irregular intervals;

[0036] Based on the collected key technical indicators of the home network, the quality analysis of the home network was carried out using the trained FTTR quality model, and the quality analysis results of the home network in different cities / provinces were obtained to indicate the causes of poor quality.

[0037] In response to receiving access requests from users with different permission levels, the system displays the home network quality analysis results and / or a list of poor-quality devices within the user's corresponding permission level, based on the access requests.

[0038] Furthermore, before performing home network quality analysis using a trained FTTR quality model based on the collected key technical indicators of the home network, the method further includes:

[0039] Several key technical indicators for home networks were selected from the technical indicators of home networks, and the key technical indicators for home networks include at least the optical receiving value of the main optical modem;

[0040] The Analytic Hierarchy Process (AHP) is used to assign weights to each of the multiple key technical indicators of home networks, and an FTTR mass model is constructed based on the multiple key technical indicators of home networks and their corresponding weights.

[0041] Based on historical data of normal FTTR operation, a threshold model built into the FTTR quality model is trained using a linear regression algorithm. This threshold model sets compliance ranges for key technical indicators of each home network, resulting in a well-trained FTTR quality model.

[0042] Secondly, this disclosure provides an FTTR service delivery quality control device, the device comprising:

[0043] Install an image receiving module to receive images of the main optical modem installation scene uploaded by smart home engineers during the FTTR service installation and deployment process;

[0044] The intelligent quality inspection module is connected to the installation image receiving module and is used to input the main optical modem installation scene image into the trained artificial intelligence (AI) intelligent quality inspection model. If the trained AI intelligent quality inspection model identifies that the main optical modem installation scene image has no re-photographing features, no weak current box features, and contains the main optical modem, then the main optical modem installation scene is deemed compliant; otherwise, the main optical modem installation scene is deemed non-compliant.

[0045] The first interception and rectification module is connected to the intelligent quality inspection module and is used to automatically trigger the first interception command and push the first rectification guide in response to non-compliance of the main optical modem installation scenario;

[0046] The access compliance judgment module is connected to the first interception and rectification module. It is used to compare the actual number of optical modems deployed in the installation verification system with the number of optical modems ordered in the front-end acceptance system, and to determine whether the supply source is compliant, so as to verify the compliance of the access from the optical modem.

[0047] The second interception and rectification module is connected to the access compliance judgment module. It is used to automatically trigger a second interception command and push a second rectification guide in response to non-compliant access from the optical modem.

[0048] The completion acceptance module, connected to the second interception and rectification module, is used to respond to the fact that both the main optical modem installation scenario and the secondary optical modem access are compliant, receive the acceptance request sent by the smart home engineer, call the converged network management platform interface according to the acceptance request, obtain the core data of the corresponding main and secondary optical modems, and perform automated completion acceptance based on the core data;

[0049] The visualization report generation module is connected to the completion acceptance module. If the result of the automated completion acceptance is qualified, it generates a visualization report including the acceptance data; otherwise, it triggers the corresponding rectification process.

[0050] Thirdly, this disclosure provides an electronic device comprising: at least one processor; and a memory communicatively connected to the at least one processor; wherein the memory stores one or more computer programs executable by the at least one processor, the one or more computer programs being executed by the at least one processor to enable the at least one processor to perform the FTTR service delivery quality control method described in the first aspect above.

[0051] Fourthly, this disclosure provides a computer-readable storage medium having a computer program stored thereon, wherein the computer program, when executed by a processor, implements the FTTR service delivery quality control method described in the first aspect.

[0052] The FTTR service delivery quality control methods, devices, electronic equipment, and media disclosed herein construct a rigid closed loop across the entire FTTR service delivery process by embedding AI-powered intelligent quality inspection, equipment access compliance verification, and automated completion acceptance technologies. This loop consists of "intelligent identification - technical interception - mandatory rectification - effect verification." A trained AI-powered intelligent quality inspection model automatically identifies the compliance of the main optical modem installation scenario (no duplicate photos, no weak current box, and including the main optical modem), technically preventing missed compliance checks. A real-time comparison of the actual number of optical modems deployed with the ordered number, verifying the compliance of the supply source, and triggering automatic interception commands and rectification guidelines avoids non-compliant network access and maintenance risks. Simultaneously, supported by core data from the integrated network management platform, automated completion acceptance and visualization reports are generated, replacing the traditional manual control model. This completely compensates for the deficiencies of existing methods in terms of closed-loop process and insufficient quality control constraints, ultimately achieving standardized and intelligent control of FTTR service delivery quality and significantly improving delivery efficiency and quality stability. This addresses the issues of existing FTTR service delivery quality control methods, such as missing process loops, easy omissions in installation compliance checks, and lack of automated verification for equipment access, which pose operational risks. Attached Figure Description

[0053] The accompanying drawings are provided to further illustrate the present disclosure and form part of the specification. They are used together with the embodiments of the present disclosure to explain the disclosure and do not constitute a limitation thereof. In the drawings:

[0054] Figure 1 A flowchart of a method for quality control of FTTR service delivery provided in this embodiment of the disclosure;

[0055] Figure 2 A flowchart of yet another FTTR service delivery quality control method provided in this embodiment of the disclosure;

[0056] Figure 3 This is a schematic diagram of the converged architecture of the network management platform provided in the embodiments of this disclosure;

[0057] Figure 4 A schematic diagram of the architecture of the FTTR mass state model provided in the embodiments of this disclosure;

[0058] Figure 5 A schematic diagram of the AI-powered intelligent quality inspection process provided in this embodiment of the disclosure;

[0059] Figure 6 A schematic diagram of the access verification process from the optical modem provided in an embodiment of this disclosure;

[0060] Figure 7 A schematic diagram of the FTTR automated final acceptance process provided in this embodiment of the disclosure;

[0061] Figure 8 A block diagram of an FTTR service delivery quality control device provided in this disclosure embodiment;

[0062] Figure 9 This is a block diagram of an electronic device provided in an embodiment of the present disclosure. Detailed Implementation

[0063] To enable those skilled in the art to better understand the technical solutions of this disclosure, exemplary embodiments of this disclosure are described below with reference to the accompanying drawings, including various details of the embodiments of this disclosure to aid understanding. These should be considered merely exemplary. Therefore, those skilled in the art should recognize that various changes and modifications can be made to the embodiments described herein without departing from the scope and spirit of this disclosure. Similarly, for clarity and conciseness, descriptions of well-known functions and structures are omitted in the following description.

[0064] Where there is no conflict, the various embodiments of this disclosure and the features thereof in the embodiments may be combined with each other.

[0065] As used herein, the term “and / or” includes any and all combinations of one or more related enumerated entries.

[0066] The terminology used herein is for the purpose of describing particular embodiments only and is not intended to limit this disclosure. As used herein, the singular forms “a” and “the” are intended to include the plural forms as well, unless the context clearly indicates otherwise. It will also be understood that when the terms “comprising” and / or “made of” are used in this specification, the presence of the stated feature, integral, step, operation, element, and / or component is specified, but the presence or addition of one or more other features, integrals, steps, operations, elements, components, and / or groups thereof is not excluded. Words such as “connected” or “linked” are not limited to physical or mechanical connections but can include electrical connections, whether direct or indirect.

[0067] Unless otherwise specified, all terms used herein (including technical and scientific terms) have the same meaning as commonly understood by one of ordinary skill in the art. It will also be understood that terms such as those defined in commonly used dictionaries should be interpreted as having a meaning consistent with their meaning in the context of the relevant art and this disclosure, and will not be interpreted as having an idealized or overly formal meaning, unless expressly so defined herein.

[0068] Figure 1 A flowchart illustrating a method for quality control of FTTR service delivery provided in this embodiment of the disclosure. (Refer to...) Figure 1 The method includes:

[0069] Step S101: During the FTTR service installation and deployment process, receive the main optical modem installation scene image uploaded by the smart home engineer.

[0070] Specifically, after installing the main optical modem, the smart home engineer takes pictures of the installation scene using a mobile phone / PDA (Personal Digital Assistant) and uploads them.

[0071] Step S102: Input the main optical modem installation scene image into the trained AI (Artificial Intelligence) intelligent quality inspection model. If the trained AI intelligent quality inspection model identifies that the main optical modem installation scene image has no re-photographing features, no weak current box features, and contains the main optical modem, then the main optical modem installation scene is deemed compliant; otherwise, the main optical modem installation scene is deemed non-compliant.

[0072] Specifically, after receiving the uploaded image of the main optical modem installation scene, the image can be preprocessed, for example, compressed to 1024×768 pixels, to optimize data transmission and processing efficiency and reduce system resource consumption. The compressed image is then input into a trained AI intelligent quality inspection model. This model completes feature extraction and judgment within 10 seconds. If the image is found to have no features of being copied or reproduced, no features of a low-voltage box, and contains the main optical modem, the installation scene is deemed compliant; otherwise, it is deemed non-compliant.

[0073] It should be noted that "re-photographed" specifically refers to images taken in real-time but not on-site (such as re-photographed archived photos, images taken by others that are not in the current task scene, etc.). By using the AI ​​intelligent quality inspection model to identify and exclude such re-photographed images, it can be ensured that the uploaded installation scene images are highly matched with the actual scene, equipment, and time of the current FTTR installation task, avoiding misjudgments of installation compliance due to false images. Re-photographed images cannot be used as a valid basis for judging the compliance of the main optical modem installation because they cannot truly reflect the current on-site installation situation.

[0074] It should be noted that installing the main optical modem inside the weak current box will cause the WiFi signal to attenuate by more than 30%, directly affecting the network experience. In order to eliminate the risk of signal shielding and attenuation from the source, an AI intelligent quality inspection model is used to identify whether it is installed inside the weak current box. If there are no characteristics of the weak current box (such as a metal / plastic enclosed structure, no ventilation holes in the box, etc.), it is identified as an installation outside the weak current box.

[0075] Optionally, the steps of the AI ​​intelligent quality inspection model may include: 1) determining whether the main optical modem installation scene image is a copied image; if it is, it is non-compliant; if it is not, proceed to step 2; 2) identifying whether the main optical modem exists in the main optical modem installation scene image; if there is no main optical modem, it is non-compliant; if there is a main optical modem, proceed to step 3; 3) identifying whether the main optical modem installation scene image contains features of a low-voltage box. If it exists, it is non-compliant; if it does not exist, it is compliant.

[0076] In some embodiments, before inputting the main optical modem installation scene image into the trained AI intelligent quality inspection model, the method further includes:

[0077] Obtain a training dataset containing image samples of both compliant and non-compliant installation scenarios of the main optical modem;

[0078] The training dataset is augmented using data augmentation techniques to obtain an augmented training dataset, wherein the data augmentation techniques include at least one of rotating, cropping, and adjusting the brightness of the image;

[0079] The AI ​​intelligent quality inspection model based on CNN (Convolutional Neural Network) is trained using the expanded training dataset to obtain the trained AI intelligent quality inspection model.

[0080] Specifically, images of compliant main optical modem installation scenarios (e.g., well-ventilated open areas without weak current box features, not copied images, but including the main optical modem) and non-compliant installation scenarios (e.g., weak current box features such as metal / plastic enclosed structures, copied image features) are acquired to form a training dataset. Then, data augmentation techniques such as rotation, cropping, and brightness adjustment are used to expand the training dataset, resulting in an expanded training dataset. The expanded training dataset is divided into a training set, a validation set, and a test set according to a preset ratio. The training set is used to train an AI intelligent quality inspection model based on a convolutional neural network (CNN). This AI intelligent quality inspection model includes convolutional layers, pooling layers, and fully connected layers. During training, the convolutional layers use the ReLU activation function to extract image features (e.g., the enclosed edges of the weak current box, lack of ventilation holes), the pooling layers use max pooling to reduce parameter dimensionality, and the fully connected layers use the Softmax function to output "compliant / non-compliant" judgment results. The model parameters are optimized using the validation set until the model's recognition accuracy is ≥98%, resulting in a well-trained AI intelligent quality inspection model.

[0081] It should be noted that a "data feedback channel" can be deployed in the system to automatically include the more than 100 newly added actual installation images each day into the training set. The AI ​​intelligent quality inspection model is optimized weekly using incremental training to improve the recognition accuracy of "low-voltage box-like scenarios" (such as enclosed cabinets and metal cabinets) and avoid misjudgments.

[0082] Step S103: In response to the non-compliant installation scenario of the main optical modem, the first interception command is automatically triggered and the first rectification guide is pushed.

[0083] Specifically, if the main optical modem is installed in a non-compliant environment (such as being installed inside a low-voltage box), the system will automatically trigger the first interception command to restrict the progress of the completion process and push the first rectification guidelines (such as suggesting placement in a well-ventilated and open area or including a "scheme diagram of the suggested placement area"). After the smart home engineer completes the rectification according to the rectification guidelines, he will retake the image of the main optical modem installation scene and upload it. The AI ​​intelligent quality inspection model will perform a second verification on the newly uploaded image until it is determined that the main optical modem installation scene is compliant.

[0084] Step S104: By comparing the actual number of optical modems deployed in the installation verification system with the number of optical modems ordered in the front-end acceptance system, and determining whether the supply source is compliant, the compliance of the access from the optical modems is verified.

[0085] Specifically, during the installation process, smart home engineers need to scan the serial number (SN) on the bottom of the optical modem and submit it. At this time, the installation verification system will automatically calculate and record the actual number of optical modems deployed and their SNs. To solve compatibility issues caused by non-compliant optical modems entering the network, the system compares the actual number of optical modems deployed in the installation verification system with the number ordered in the front-end acceptance system to determine whether the supply source is compliant, thereby verifying the compliance of the optical modem access and eliminating the potential risks of non-standard equipment entering the network from the source.

[0086] It should be noted that the front-end acceptance system and the installation verification system use the WebSocket protocol to achieve bidirectional real-time communication.

[0087] In some embodiments, the process of verifying the compliance of the access from the optical modem by comparing the actual number of optical modems deployed in the installation verification system with the number of optical modems ordered in the front-end acceptance system, and determining whether the supply source is compliant, specifically includes:

[0088] The actual number of optical modems deployed in the installation verification system is compared with the number of optical modems ordered in the front-end acceptance system. If the difference is 0, it is determined that the optical modems are deployed normally. If the difference is greater than or equal to 1, it is determined that the optical modems are deployed in excess.

[0089] Determine whether the device identifiers of each deployed slave optical modem exist in the preset list of compliant optical modem suppliers. If the device identifiers of all deployed slave optical modems exist in the list of compliant optical modem suppliers, the supply source is determined to be compliant; otherwise, the supply source is determined to be non-compliant.

[0090] Based on the normal deployment and compliant supply source of the optical modem, the access from the optical modem is deemed compliant.

[0091] In response to excessive deployment of optical modems or non-compliant supply sources, the connection from the optical modem is determined to be non-compliant.

[0092] Specifically, the actual number of optical modems deployed in the installation verification system is compared with the number of optical modems ordered in the front-end acceptance system: if the difference is 0, it is determined that the optical modems are deployed normally; if the difference is greater than or equal to 1, it is determined that the optical modems are deployed in excess.

[0093] Specifically, the compliant optical modem supply list (i.e., the terminal database) stores the compliant optical modem model, SN code, and supplier information. The data is updated at a frequency of ≤1 hour. By comparing the device identifier (such as SN code) of each slave optical modem with the compliance information (such as SN code) in the compliant optical modem supply list, if the device identifier of any slave optical modem does not match the compliance information in the compliant optical modem supply list, the slave optical modem access is determined to be non-compliant. Otherwise, if the device identifiers of all slave optical modems can match the compliance information in the compliant optical modem supply list, the slave optical modem access is determined to be compliant.

[0094] Step S105: In response to non-compliant access from the optical modem, a second interception command is automatically triggered, and a second rectification guide is pushed.

[0095] Specifically, if the access from the optical modem is non-compliant (such as excessive deployment), the system will automatically trigger a second interception command to restrict the progress of the completion process and push a second rectification guide (such as "Excessive deployment: 1 optical modem needs to be deleted"). After the smart home engineer completes the rectification, the system will re-verify, and the completion process can continue only after compliance is achieved.

[0096] Specifically, the system can automatically collect and store the account information corresponding to the "over-deployed" devices, providing the market side with accurate business suggestions for "adding optical modems", and achieving synergy between "control efficiency" and "business growth".

[0097] Step S106: In response to the compliance of both the main optical modem installation scenario and the secondary optical modem access, receive the acceptance request sent by the smart home engineer, call the converged network management platform interface according to the acceptance request, obtain the core data of the corresponding main and secondary optical modems, and perform automated completion acceptance based on the core data.

[0098] Specifically, after the main optical modem is installed in compliance with regulations and the secondary optical modem is connected in compliance with regulations, the smart home engineer initiates an "acceptance request". The system responds to the acceptance request and calls the core data of the main and secondary optical modems of the converged network management platform within 2 seconds through the RESTful API (Application Programming Interface) interface to complete the verification of key indicators, so as to achieve automated completion acceptance.

[0099] In some embodiments, the core data includes the online status and received optical values ​​of the master and slave optical modems, as well as the uplink connection method of the slave optical modem. The automated final acceptance testing based on the core data specifically includes:

[0100] Based on the online status and received light value of the master and slave optical modems, as well as the uplink connection mode of the slave optical modem, the following judgments are made: whether the master and slave optical modems are online, whether the received light value of the master and slave optical modems is within the preset threshold range, and whether the uplink mode of the slave optical modem is wireless.

[0101] In response to the fact that both the master and slave optical modems are online, the received light values ​​of the master and slave optical modems are within the preset threshold range, and the uplink mode of the slave optical modem is not wireless, the result of the automated completion acceptance is determined to be qualified.

[0102] Specifically, the automation completion acceptance requires verification of the following three key indicators:

[0103] (a) Optical modem online status: Determine whether the master and slave optical modems are in an "online" state (based on real-time heartbeat data from the network management system);

[0104] (b) Compliance of received light value: Compare the received light value of the optical modem with the preset threshold range, such as -8dBm to -28dBm).

[0105] (c) Compliance of connection method: Confirm whether the uplink method from the optical modem is wireless; if it is wireless, it is not compliant.

[0106] In some embodiments, the method further includes:

[0107] An interface adaptation module is deployed between the local optical modem terminal network management system and the cloud platform terminal network management system. A millisecond-level data synchronization channel is built using the Flink framework. The data synchronization channel is then used to unify and aggregate the master and slave optical modem data in the local optical modem terminal network management system and the cloud platform terminal network management system to the converged network management platform.

[0108] By utilizing the northbound interface of the converged network management platform and the data middle platform, the data of the master and slave optical modems in the converged network management platform are synchronized to the data middle platform.

[0109] Specifically, to address the data fragmentation issue between traditional dual-network management platforms—namely, the local optical modem terminal network management system and the cloud platform terminal network management system—and to achieve unified aggregation, real-time sharing, and efficient management of all optical modem data, a RESTful API interface protocol is adopted for cross-system communication. An interface adaptation module is deployed between the local optical modem terminal network management system and the cloud platform terminal network management system, enabling data compatibility between the two systems through the RESTful API. A data synchronization channel is built using the Flink framework, with a millisecond-level (e.g., 100ms-level) synchronization frequency. This unifies over 20 types of data, including optical modem operating status (online / offline), access type (10GPON / LAN), and received optical value (dBm), to the converged network management platform, eliminating data silos. Simultaneously, a northbound interface between the converged network management platform and the data middleware is developed, using RESTful APIs to achieve high-concurrency data transmission, ensuring that all optical modem information is automatically synchronized to the data middleware.

[0110] Step S107: If the automated completion acceptance result is qualified, a visualization report including acceptance data will be generated; otherwise, the corresponding rectification process will be triggered.

[0111] Specifically, if the automated completion acceptance result is satisfactory, a visual report including acceptance data (such as device serial numbers, indicator data, etc.) will be automatically generated based on the UniApp UI framework. After user confirmation, compliance completion is completed. Simultaneously, all acceptance data can be transmitted back to the data platform in real time via HTTP (Hypertext Transfer Protocol) POST requests, supporting operations and maintenance personnel in analyzing "regional acceptance pass rate" and "high-frequency abnormal indicators," and optimizing network planning (e.g., if the received optical value is generally low in a certain area, it is recommended to investigate line attenuation). If the automated completion acceptance result is unsatisfactory, the corresponding rectification process will be triggered.

[0112] In some embodiments, the method further includes:

[0113] Collect multiple key technical indicators of home networks corresponding to FTTR services from the data platform at regular or irregular intervals;

[0114] Based on the collected key technical indicators of the home network, the quality analysis of the home network was carried out using the trained FTTR quality model, and the quality analysis results of the home network in different cities / provinces were obtained to indicate the causes of poor quality.

[0115] In response to receiving access requests from users with different permission levels, the system displays the home network quality analysis results and / or a list of poor-quality devices within the user's corresponding permission level, based on the access requests.

[0116] Specifically, the data platform executes scheduled data collection tasks according to a preset cycle (e.g., hourly / daily), while also supporting unscheduled data collection triggered by business needs. By inputting multiple key technical indicators of home networks collected into a trained FTTR quality model, the model outputs quantitative statistical results such as the current values, anomaly percentages, and month-on-month changes of key technical indicators of home networks at different city / province levels. These statistical results directly constitute a quantitative representation of the causes of quality problems. For example, taking "abnormal uplink optical received value of the main optical modem (above -8dBm or below -27dBm)" as an example, the model can output the number of FTTR-managed users in each city / province, the number of users with abnormal uplink optical received values ​​(including users above -8dBm and below -27dBm), the anomaly percentage (including the percentage of users above -8dBm and below -27dBm), and the month-on-month change (e.g., the month-on-month change in increment). Based on this, it can directly locate areas with high incidence of abnormal optical received values ​​and clearly identify core causes of quality problems such as "non-compliant installation location leading to optical power attenuation" and "equipment hardware failure".

[0117] Specifically, the system is based on the RBAC (role-based access control) model and pre-sets a multi-level user permission system (such as supporting two levels: province-city, or three levels: province-city-county). When it receives access requests from users with different permission levels, it first verifies the user's identity and permission level, and then filters out the home network quality analysis results within the user's permission range according to the permission level rules (such as provincial users can view data for the entire province and its subordinate cities, while county users can only view data for their own county) and / or a list of poor-quality devices (including the SN code of the poor-quality device, the reason for the poor quality, abnormal indicators, rectification suggestions, etc.).

[0118] In some embodiments, before performing home network quality analysis using a trained FTTR quality model based on collected key technical indicators of the home network, the method further includes:

[0119] Several key technical indicators for home networks were selected from the technical indicators of home networks, and the key technical indicators for home networks include at least the optical receiving value of the main optical modem;

[0120] The Analytic Hierarchy Process (AHP) is used to assign weights to each of the multiple key technical indicators of home networks, and an FTTR quality model is constructed based on the multiple key technical indicators of home networks and their corresponding weights.

[0121] Based on historical data of normal FTTR operation, a threshold model built into the FTTR quality model is trained using a linear regression algorithm. This threshold model sets compliance ranges for key technical indicators of each home network, resulting in a well-trained FTTR quality model.

[0122] Specifically, in order to solve the problems of fragmentation and lag in traditional manual analysis, and in order to achieve more precise and dynamic quality control, multi-dimensional indicator modeling is used to realize automatic analysis and anomaly detection of home network quality.

[0123] Specifically, 28 key technical indicators for home networks were first selected from the existing home network technical specifications. These indicators cover basic information, device identification, network topology information, device operating status, performance indicators, connection methods, and network information. For example, these 28 key home network technologies could include: city / prefecture, manufacturer, FTTR main gateway SN code, FTTR main gateway MAC (Media Access Control) address, broadband account, OLT (optical line terminal) name, OLT_IP, and PON (Passive Optical Networking). Network (Passive Optical Network) board type, PON port, FTTR master gateway device model, FTTR software version information, work order status, current online status of the master gateway, first online time of the master gateway, uplink connection method of the master gateway, number of FTTR slave optical modems, first online method of slave optical modems, most recent online method of slave optical modems, uplink optical power exceeding the limit of the master optical modem, uplink optical power of the master optical modem (i.e., the received optical value of the master optical modem), uplink optical power exceeding the limit of slave optical modems, uplink optical power of slave optical modems, wireless signal strength of slave optical modems, home network type (single gateway, single gateway + managed AP, etc.), most recent online time of the master gateway, most recent offline time of the master gateway, first online method of the master gateway, and the distance between the master and slave optical modems is too close. Based on the 28 key technical indicators of home networks selected, the Analytic Hierarchy Process (AHP) was used to determine the weight of each indicator (such as 30% for the uplink optical power of the master optical modem and 20% for the wireless signal strength of the slave optical modem), and an FTTR quality model was constructed. Based on historical normal operation data, a threshold model is trained using a linear regression algorithm to set compliance ranges for each indicator (e.g., optical modem received light value -8dBm to -28dBm, wireless signal strength from the optical modem ≥ -70dBm). When real-time data exceeds the range threshold, the model automatically records the "reason for poor quality" and stores the data. Among these, the threshold model, as the core component of the FTTR quality model, is responsible for dynamically determining reasonable compliance ranges for each home network's key technical indicators.

[0124] It should be noted that the FTTR service delivery quality control method disclosed herein not only solves the core pain points of traditional FTTR delivery, but also achieves synergy between "quality control" and "business growth", providing a standardized implementation paradigm for the high-quality development of whole-house optical broadband services for telecommunications operators.

[0125] In one specific embodiment, a method for quality control of FTTR service delivery based on data processing, artificial intelligence, and network collaboration technologies is provided. This method is applicable to the automated and precise quality control of FTTR services by telecommunications operators throughout the entire process from equipment access to final acceptance. It can realize full aggregation of optical modem data, dynamic analysis of home network quality, intelligent identification of installation scenarios, compliance judgment of access equipment, and automatic verification of acceptance processes. This method can at least solve the problems of insufficient automation, low accuracy, and delayed response in broadband delivery quality control under the traditional management model. This method constructs five core technical dimensions: "converged network management platform - FTTR quality model - AI intelligent quality inspection model - optical modem access compliance verification - automated final acceptance," and implements FTTR service delivery quality control in five key steps. Each step is sequentially linked and works in a closed-loop coordination: Step 1 (converged network management platform) provides real-time, full-volume optical modem data as a foundation for all subsequent steps; Step 2 (FTTR quality model) uses the data from Step 1 to achieve dynamic network quality analysis; Steps 3 (AI intelligent quality inspection) and 4 (optical modem access compliance verification) operate in parallel during the installation phase, respectively controlling "installation scenario compliance" and "device access compliance"; Step 5 (automated final acceptance) completes the final quality verification based on the control results of the first four steps, forming a closed loop of "data-analysis-control-acceptance." The overall process is as follows: Figure 2 As shown, steps one through five form a comprehensive intelligent management system for the delivery quality of whole-house fiber broadband services. The specific steps are as follows:

[0126] Step 1: Build an integrated network management platform to solidify the data foundation.

[0127] (1) Purpose

[0128] By developing cross-system data interfaces and using real-time synchronization technology, the problem of data fragmentation between the local optical modem terminal network management system (hereinafter referred to as local network management) and the cloud platform terminal network management system (hereinafter referred to as cloud platform) in the traditional dual network management platform is solved, realizing the unified aggregation, real-time sharing and efficient management of all optical modem data, and providing a data foundation for subsequent quality analysis.

[0129] (2) Conditions

[0130] ① Hardware support (cloud server cluster, CPU ≥ 32 cores, memory ≥ 128GB, to ensure data processing efficiency).

[0131] ② Software framework (using RESTful API interface protocol to achieve cross-system communication, and using MQTT real-time data stream to complete data synchronization);

[0132] ③ Data security technology (TLS encryption is used at the transport layer).

[0133] (3) Specific operations

[0134] ① Data fusion between the two platforms;

[0135] An interface adaptation module is deployed between the cloud platform and the local network management system to achieve data compatibility between the two systems through RESTful API. A data synchronization channel is built using the Flink framework, with a synchronization frequency of 100ms. More than 20 types of data, such as the optical modem's operating status (online / offline), access type (10GPON / LAN), and received optical value (dBm), are uniformly aggregated to the converged network management platform to eliminate data barriers.

[0136] ② Northbound data connection;

[0137] Develop a northbound interface that integrates the network management platform and the data middle platform, and use RESTful API to achieve high-concurrency data transmission, ensuring that all information of the optical modem (including historical operating data and real-time quality data) is automatically synchronized to the data middle platform to support subsequent quality analysis.

[0138] ③ Hierarchical authorization of permissions;

[0139] Based on the RBAC (Role-Based Access Control) model, an access control module is configured within the converged network management platform: "Data Query - Quality Analysis" permissions are assigned to provincial / municipal backend personnel (supporting multi-dimensional data filtering, such as region, time, and device model), and "Terminal Status Query - Quality Remediation" viewing permissions are configured for frontline smart home engineers.

[0140] Among them, the network management platform converged architecture is as follows: Figure 3 As shown, it includes the optical modem device layer, the converged platform layer, and the data and permission layer. The optical modem device layer uses 10GPON uplink and LAN uplink optical modems as data sources and transmits its own data to the converged platform layer. This layer collects data through two types of optical modem terminal network management systems: local and cloud platforms. The converged network management platform then completes the data integration. Finally, the integrated full data is synchronized to the data middle platform of the data and permission layer, and permissions are assigned according to roles—the provincial / municipal backend can query global data, while smart home engineers can only manage the terminal status.

[0141] It should be noted that the converged network management platform can view the quality status of a single or partial optical modem, while the data platform can view the quality status of all optical modems and perform quality status analysis.

[0142] (4) Function

[0143] It enables "one-time data collection and full-terminal sharing" of optical modem data, with data synchronization latency ≤100ms, improving the efficiency of quality analysis by 60%, and increasing the response speed of first-line maintenance by 50%, providing a precise and real-time data foundation for subsequent technical management and control.

[0144] Step 2: Apply the FTTR mass state model to achieve dynamic quality improvement.

[0145] (1) Purpose

[0146] By modeling multi-dimensional indicators, the system can automatically analyze and identify anomalies in the quality of home networks, solving the problems of fragmentation and lag in traditional manual analysis, and achieving more precise and dynamic quality control.

[0147] (2) Conditions

[0148] ①Indicator database (containing over 800,000 data points on home network operation, covering different house types, equipment models, and network environments);

[0149] ② Algorithm framework (using shell scripts to collect FTTR data daily; analyzing and generating indicators such as incremental month-on-month percentage);

[0150] ③ Visualization tools (real-time reports built with Vue.js, and Excel dashboards exported using Apache POI).

[0151] (3) Specific operations

[0152] ① Modeling of all-factor indicators;

[0153] Twenty-eight key technical indicators for home networks were selected (uplink mode of the main optical modem, wireless signal strength of the secondary optical modem, optical signal received value of the main and secondary optical modems, number of user-end devices connected, network access status of user-owned routers, etc.). The Analytic Hierarchy Process (AHP) was used to determine the weight of each indicator (e.g., optical signal received value accounts for 30%, uplink access type of the secondary optical modem accounts for 25%), etc., and an FTTR quality model was constructed.

[0154] ② Determining a reasonable threshold;

[0155] Based on historical normal operation data, a threshold model is trained using a linear regression algorithm to set compliance ranges for each indicator (such as optical modem received light value -8dBm to -28dBm, and wireless signal strength from optical modem ≥ -70dBm). When real-time data exceeds the range threshold, the model automatically records the "reason for poor quality" and stores the data.

[0156] ③ Secondary data penetration;

[0157] Deploy a data penetration module to support access via web browser for users at both the provincial and municipal levels; enable fast data query based on the ClickHouse database (query response time ≤ 500ms); frontline personnel can directly export a "list of poor-quality equipment" (including equipment SN code, abnormal indicators, rectification suggestions, etc.) to initiate targeted rectification.

[0158] The architecture of the FTTR mass model is as follows: Figure 4As shown, it includes a core indicator layer, a permission system layer, and a data display layer. The core indicator layer covers 28 key indicators, such as the uplink mode of the main optical modem, the received light value, and the connection status of the slave optical modem. These indicators correspond to the three-level permission division of the permission system layer (provincial level is responsible for full data monitoring, municipal level is responsible for regional data management, and county level focuses on the rectification of problematic equipment). Finally, the data display layer presents the compliance rate, anomaly warnings, and optimization progress of each city and prefecture branch through the data dashboard, realizing full-process quality management from indicator collection and hierarchical control to data visualization.

[0159] (4) Function

[0160] It achieves 100% automation in home network quality analysis, anomaly identification accuracy of ≥95%, and reduces the remediation cycle for poor-quality equipment from 72 hours to 24 hours, supporting refined technical management and control of home broadband networks.

[0161] Step 3: Implement AI-powered intelligent quality inspection to solve installation standardization problems.

[0162] (1) Purpose

[0163] By using convolutional neural network (CNN) image recognition technology, the system automatically determines whether the installation scenario of the main optical modem is compliant (whether it is installed in the weak current box), solving the problems of missed detection and low efficiency in traditional manual quality inspection, and eliminating the risk of signal shielding and attenuation from the source.

[0164] (2) Conditions

[0165] ① Training dataset (over 5,000 installation scene images, including low-voltage boxes (metal / plastic enclosed structures) and compliant scenes (ventilated open areas), expanded to over 20,000 images using data augmentation techniques (rotation, cropping, brightness adjustment);

[0166] ② AI model framework (building CNN models based on TensorFlow 2.x);

[0167] ③ Terminal support (The mobile phone / PDA used by smart home engineers must have a camera with a resolution of 13 megapixels or higher, and support real-time image upload (using 5G / 4G network, upload speed ≥10Mbps)).

[0168] (3) Specific operations

[0169] ① AI model training;

[0170] The expanded dataset was divided into training, validation, and test sets in a 7:2:1 ratio and input into the CNN model for training. The convolutional layers used the ReLU activation function to extract image features (such as the closed edges of the weak current box and the absence of ventilation holes), the pooling layers used max pooling to reduce the dimensionality of parameters, and the fully connected layers used the Softmax function to output the "compliant / non-compliant" judgment result. The model parameters were optimized through iterative training to achieve a recognition accuracy of ≥98%.

[0171] ② Real-time scene recognition and process interception;

[0172] After the smart home engineer installs the main optical modem, he takes a picture of the scene with his mobile phone / PDA and uploads it. The system compresses the image to 1024×768 pixels and then inputs it into the AI ​​module. The model completes feature extraction and judgment within 10 seconds: if it is identified as "low voltage box installation" (non-compliant), the system automatically triggers an interception command and pushes rectification guidelines (including "suggested placement area diagram"); ​​after rectification, the image is taken again and the model verifies it a second time until it is compliant.

[0173] ③ Dynamic model iteration;

[0174] A "data feedback channel" is deployed in the module to automatically include more than 100 newly added actual installation images into the training set every day. The model is optimized weekly using incremental training to improve the recognition accuracy of "low-voltage box-like scenarios" (such as enclosed cabinets and metal cabinets) and avoid misjudgments.

[0175] The AI-powered intelligent quality inspection process is as follows: Figure 5 As shown, after the smart home engineer installs the main optical modem, they take a photo and upload it. The system then performs an AI-powered quality inspection (determining within 10 seconds whether it is installed inside a low-voltage box, etc.: characteristics of the low-voltage box → metal / plastic enclosed structure, etc.). If the quality inspection fails (e.g., installed inside a low-voltage box), the system intercepts the inspection and sends a rectification guide (suggesting placement in a well-ventilated and open area). The smart home engineer then rectifys the issue or redoes the installation and takes another photo for re-inspection. If the quality inspection passes, the process is completed and proceeds to the final stage, where the user confirms the installation.

[0176] (4) Function

[0177] The compliance rate of main optical modem installation has increased from 85% to 100%, WiFi signal attenuation has been reduced by 90%, and manual quality inspection costs have been reduced by 70%, thus eliminating network quality risks caused by non-compliance with installation scenarios from a technical perspective.

[0178] Step 4: Compliance verification of optical modem access, driving operational compliance upgrades.

[0179] (1) Purpose

[0180] By using real-time data comparison technology between the front-end acceptance system and the installation verification system, the system automatically determines whether the access from the optical modem is compliant (terminal source, number of deployments), resolves compatibility issues caused by non-compliant optical modems entering the network, and achieves full lifecycle technical management of access devices.

[0181] (2) Conditions

[0182] ① Data interface (the front-end acceptance system and the installation verification system use the WebSocket protocol to achieve bidirectional real-time communication), obtain the number of terminals based on the front-end order receiving interface;

[0183] ② Terminal database (stores compliant optical modem model, SN code, and supplier information; data update frequency ≤ 1 hour);

[0184] ③ Interception trigger module (supports linkage with completion process, response time ≤50ms).

[0185] (3) Specific operations

[0186] ①Preset compliance verification rules;

[0187] The system has a preset quantity verification rule: the "quantity ordered from optical modem" in the front-end acceptance system is obtained in real time via the WebSocket protocol and compared with the "deployment quantity" actually entered. If the difference is ≥1, it is judged as "over-deployment".

[0188] ② Automatic interception and rectification guidelines;

[0189] When a verification anomaly is triggered, the system automatically sends an interception command to the completion process module to prevent the process from closing. At the same time, it generates a "rectification guide" (such as "Over-deployment: 1 optical modem needs to be deleted") and pushes it to the smart home engineer terminal. After the rectification is completed, the system re-verifies, and the completion process can only continue after compliance is achieved.

[0190] ③ Data value mining;

[0191] The system automatically collects and stores the account information corresponding to the "over-deployed" devices, providing the market side with accurate business suggestions for "adding optical modems", and achieving synergy between "control efficiency" and "business growth".

[0192] The verification process from the optical modem is as follows: Figure 6As shown, the order acceptance process first clarifies the quantity ordered from the optical modem, and then proceeds to the installation verification process: the system automatically compares the actual number of units deployed from the optical modem, the ordered quantity, and the source of the terminal supply. If everything is normal, it proceeds to the compliance completion and user confirmation stage; if non-compliant terminals or excessive installations are found, the system will intercept and push rectification guidelines (confirming whether the terminal is online, the source of the terminal, etc.), and the smart home engineer will rectify or redo the work (such as replacing compliant terminals or adjusting the number of units installed); and further value can be explored (additional installation marketing can be carried out for users with excessive deployments).

[0193] (4) Function

[0194] The optical modem has a 100% compliance rate for network access, while providing data support for business growth and achieving synergy between "technical control and business optimization".

[0195] Step 5: FTTR final acceptance and establishment of stringent quality standards.

[0196] (1) Purpose

[0197] By using network management data auto-retrieval and report generation technology, the entire process of FTTR delivery and acceptance is automated, solving the problems of "flexible standards and lagging data" in traditional manual acceptance and ensuring rigid control over delivery quality.

[0198] (2) Conditions

[0199] ① Network management data call interface (using RESTful API interface to achieve high-speed data interaction with the converged network management platform).

[0200] ② Report generation (based on UniApp's UI framework, reports are automatically generated);

[0201] (3) Specific operations

[0202] ① Automated verification process;

[0203] After the smart home engineer completes the installation, they initiate an "acceptance request" in the tool. The system then uses a RESTful API interface to retrieve core data from the converged network management platform within 2 seconds, completing the verification of 3 key indicators:

[0204] (a) Optical modem online status: Determine whether the master and slave optical modems are in an "online" state (based on real-time heartbeat data from the network management system);

[0205] (b) Compliance of received light value: Compare the received light value of the optical modem with the preset threshold (-8dBm to -28dBm);

[0206] (c) Compliance of connection method: Confirm whether the uplink method from the optical modem is wireless; if it is wireless, it does not meet the standard, and the system will automatically prompt "Acceptance failed" and guide rectification.

[0207] ②Automatic generation of visual reports

[0208] After acceptance, a report containing "device information (model, SN code) and indicator data (light reception value, online status, etc.)" is automatically generated based on the UniApp UI framework. After confirmation, the acceptance loop is completed.

[0209] ③ Acceptance data feedback and application

[0210] All acceptance data (including equipment serial numbers, indicator data, etc.) are transmitted back to the data platform in real time via HTTP POST requests, supporting the operation and maintenance side to analyze "regional acceptance pass rate" and "high-frequency abnormal indicators" and optimize network planning (e.g., if the optical reception value in a certain area is generally low, it is recommended to investigate line attenuation).

[0211] The FTTR automated completion acceptance process is as follows: Figure 7 As shown, after the installation is completed, the system will start a one-click test (30 seconds to check the online status of the optical modem, the light reception value, and the connection method). If the test fails, the system will intercept the test and push a rectification guide, which will be rectified or redone by the smart home engineer. If the test passes, a visual report will be generated, and the user will confirm the completion of the compliant project.

[0212] (4) Function

[0213] FTTR achieves 100% automation in final acceptance, reducing acceptance time from 30 minutes to 5 minutes, and ensuring 100% accuracy of acceptance data. This reduces potential after-sales quality issues at the source and provides precise data support for network optimization.

[0214] In another specific embodiment, based on the large-scale operation needs of a provincial telecommunications operator's whole-house optical broadband service, and addressing the five major pain points in traditional FTTR delivery—"data fragmentation due to dual network management, delayed manual quality analysis, missed compliance checks during installation, non-compliant optical modem access, and low acceptance efficiency"—a comprehensive management solution was implemented, comprising a "converged network management platform - FTTR quality model - AI intelligent quality inspection - optical modem access verification - automated final acceptance." Specific implementation steps include:

[0215] Step 1: Build a converged network management platform to achieve unified management and control of optical modem data;

[0216] To address the pain points of the traditional "distributed management across two platforms," ​​we implemented platform integration and data connectivity to improve management efficiency.

[0217] ① Dual-platform integration and breaking down data barriers

[0218] The original "local optical modem terminal network management (LAN uplink optical modem)" and "cloud platform terminal network management (10GPON uplink optical modem)" are unified into the integrated network management platform to solve the problem of "two sets of network management independently managed" and realize centralized management and control of optical modem devices and unified data aggregation.

[0219] ② Northbound interface connection and full data synchronization

[0220] Develop a northbound interface that integrates the network management platform and the data middleware to achieve automatic synchronization of all information such as the optical modem's operating status, access type, and quality data. This solves the problem that the original cloud platform could not connect to the data middleware and provides accurate data support for digital operation decisions.

[0221] ③RBAC hierarchical permission empowerment and operation optimization

[0222] Global data query permissions are granted to multiple middleware administrators at the provincial / municipal level, supporting quality analysis and indicator monitoring; terminal status query and quality improvement operation permissions are configured for multiple front-line smart home engineers, enabling them to directly manage master and slave optical modems through a visual interface without switching systems, thus improving maintenance response speed.

[0223] Step 2: Construct an FTTR quality model to achieve dynamic quality improvement of the home network;

[0224] Construct a comprehensive quality status model to achieve precise control and continuous optimization of home network quality.

[0225] ① Construction of a comprehensive operational quality model

[0226] It integrates 28 key indicators, including the uplink mode of the main optical modem (fiber optic / network cable), the wireless connection status of the slave optical modem, the optical reception value of the main and slave optical modems (abnormal threshold: higher than -8dB or lower than -27dB), and the network access compatibility of the user's own router, covering all scenarios of home network and solving the fragmentation problem of traditional quality management.

[0227] ② Configuration of a three-tier access control system

[0228] It penetrates the three-tiered access control system of provincial, municipal, and county-level maintenance teams, supporting real-time querying of full-volume quality data via webpage or mobile device. Frontline installation and maintenance personnel can directly export a list of equipment with poor quality, quickly locate problems such as abnormal light reception and unstable connections, and carry out precise rectification.

[0229] ③ Establishment of a dynamic closed-loop optimization mechanism

[0230] The data dashboard displays the compliance status of each city and prefecture branch in real time, and triggers early warnings for abnormal indicators such as optical fiber reception values ​​exceeding thresholds, initiating a closed loop of "discovery-optimization-iteration". Cross-professional optimization is achieved through network and customer service collaboration, such as batch adjusting fiber optic routes or improving the quality of primary and secondary lines in cells with abnormal optical fiber reception, making the quality standards more aligned with the actual user experience.

[0231] Step 3: Apply AI intelligent quality inspection technology to solve the problem of main optical modem installation specifications;

[0232] Introducing AI image recognition technology to solve the signal attenuation problem caused by the installation of the main optical modem's low-voltage box:

[0233] ① AI-powered intelligent scene recognition

[0234] Smart home engineers capture and upload panoramic images of the main optical modem installation scene via mobile or PDA terminals. The system instantly calls upon a pre-trained AI intelligent quality inspection model, relying on a feature library formed by training on massive samples, to accurately identify core characteristics of the low-voltage box (reflective metal enclosures, plastic boxes without ventilation holes, etc.), completing the compliance determination of the installation location within 10 seconds. By replacing manual experience-based judgment with technological identification, the risk of missed or incorrect judgments is completely avoided, achieving standardization and accuracy in scene verification.

[0235] ② Installation process is locked and controlled

[0236] If the AI ​​identifies the installation as "installed inside a low-voltage box" (a scenario that can easily lead to WiFi signal attenuation exceeding 30%, directly impacting network experience), the system immediately triggers a rigid interception mechanism: automatically blocking the completion process, locking closed-loop permissions, and simultaneously pushing visual rectification guidelines (such as explicitly recommending placement in a well-ventilated and open area of ​​the living room). Smart home engineers must adjust the installation location according to the guidelines and re-take photos for re-inspection until the AI ​​determines the location is completely compliant before final delivery can be completed, ensuring installation compliance through a mandatory process.

[0237] ③ Dynamic iterative optimization of the quality inspection model

[0238] A model training library is built based on a full set of installation and maintenance image data. The AI ​​model is continuously trained iteratively by adding new scenario samples (such as enclosed cabinets, metal cabinets, embedded TV cabinets, and other low-voltage box scenarios) to continuously improve the accuracy of edge scene recognition. Through a cyclical mechanism of "data accumulation - model optimization - standard upgrade", the management and control standards are dynamically adapted to the evolution of actual installation scenarios, ensuring that the AI ​​quality inspection capabilities continuously meet business needs.

[0239] Step 4: Establish a compliance management system for optical modem access to drive standardized operations;

[0240] Establish an optical modem access control system to achieve the dual goals of ensuring compliance and unlocking business value:

[0241] ① Rigorous verification of compliance of newly installed optical modems

[0242] The compliance verification of optical modems is embedded in the completion process. The system compares the data received from the front end (the number of optical modems ordered) with the actual deployment information entered by the engineers in real time. For issues such as "non-compliant supply terminals (poor compatibility that easily causes failures)" and "deployment quantity exceeding the order," the system immediately triggers completion interception and pushes rectification guidelines to eliminate the potential risks of non-standard equipment entering the network from the source.

[0243] ② In-depth mining of the value of management and control data

[0244] By monitoring "over-deployment behavior," we can accurately identify users' WiFi coverage expansion needs (such as large-house users requiring additional optical modems), and conduct targeted installation marketing services to achieve synergistic progress in "improved management efficiency" and "business growth." After non-standard equipment was taken offline, network compatibility was significantly improved, and user service satisfaction improved accordingly.

[0245] Step 5: Implement FTTR automated final acceptance and establish strict quality standards;

[0246] Innovate the acceptance model to achieve rigid control over the entire process and two-way data empowerment:

[0247] ① Rigid quality control throughout the entire process

[0248] The "one-click quality inspection" function is deeply embedded into the entire order construction process, forming a "quality inspection upon installation" linkage mechanism. After the smart home engineer completes the installation on-site, they retrieve real-time data from the network management system and intelligently verify key indicators such as the optical modem's online status, the compliance of the received light value, and the compliance of the connection method (fiber / network cable / wireless type matching) within 30 seconds. For items that do not meet the standards, a "process lockout" is implemented—directly blocking the completion process and simultaneously pushing targeted rectification guidance. The smart home engineer must complete the problem repair and re-initiate the inspection until all indicators are fully qualified before acceptance can be passed. This rigid control ensures that the completion acceptance pass rate remains stable at 100%, eliminating potential after-sales quality problems from the source.

[0249] ②Empowering the Value of Acceptance Data in Two Directions

[0250] Upon completion of acceptance testing, the system instantly generates a visual delivery report, presenting users with a comprehensive overview of core information such as terminal model, real-time operating status, and network speed, ensuring both service transparency and users' right to know. Simultaneously, all acceptance data (including device model, connection method, and optical transmission records) is synchronized to the data platform in real time, forming a home network quality database. This provides precise decision support for operations and maintenance, facilitating the rapid identification of "high-incidence areas of optical transmission anomalies," enabling batch optimization of optical fiber routing layouts and reducing recurrence rates. It also provides dynamic data references for the marketing side, allowing analysis of terminal activity based on device online status, adjustment of marketing and control strategies, and improvement of business development quality. Through the two-way flow of data—"user-end display + operations-end application"—acceptance data is transformed from a "process byproduct" into a core asset driving service optimization and business growth.

[0251] It should be noted that the FTTR service delivery quality control method provided in this disclosure has the following characteristics:

[0252] a) Construct a rigid closed loop throughout the entire process of "intelligent identification - technical interception - mandatory rectification - effect verification": By embedding a verification mechanism of "real-time intelligent identification of anomalies, rigid technical interception of violations, and mandatory verification of rectification effects" into the entire construction process, an unbreakable rigid closed loop is constructed. This solves the problems of existing technologies only being limited to monitoring and early warning, lacking rigid constraints, which leads to delayed problem rectification, ineffective implementation, and "information silos". This ensures a closed loop throughout the entire chain from the discovery of quality anomalies to their resolution.

[0253] b) Establish a complete lifecycle link from "data foundation - quality analysis - installation control - access verification - operation optimization": By building a solid data foundation across the entire link, connecting the control of all aspects of construction, and linking quality inspection and operation optimization, we can achieve full lifecycle coverage control from "planning to execution to monitoring to optimization" and solve the problems of "single coverage and fragmented data" in existing solutions.

[0254] c) Construct a comprehensive management and control system based on "data-driven + intelligent technology": By collecting all optical modem data, introducing multi-dimensional intelligent recognition technology, and constructing an FTTR quality model, a virtuous cycle of "intelligent technology supporting the implementation of standards and standards guiding intelligent judgment" is formed, achieving a dual improvement in management and control efficiency and accuracy, and solving problems such as "insufficient intelligence and data utilization".

[0255] The FTTR service delivery quality control method provided in this disclosure constructs a rigid closed loop of "intelligent identification - technical interception - mandatory rectification - effect verification" by embedding AI intelligent quality inspection, equipment access compliance verification, and automated completion acceptance into the entire FTTR service delivery process. This involves: automatically identifying the compliance of the main optical modem installation scenario (no duplicate photos, no weak current box, and including the main optical modem) using a trained AI intelligent quality inspection model, thus technically preventing missed compliance checks; comparing the actual number of optical modems deployed with the ordered number in real time through an installation verification system to verify the compliance of the supply source, triggering automatic interception commands and rectification guidelines to avoid non-compliant network access and maintenance risks; and achieving automated completion acceptance and generating visual reports based on core data from the integrated network management platform, replacing the traditional manual control mode. This completely compensates for the deficiencies of existing methods in terms of missing process loops and insufficient quality control constraints, ultimately achieving standardized and intelligent control of FTTR service delivery quality, significantly improving delivery efficiency and quality stability. This addresses the issues of existing FTTR service delivery quality control methods, such as missing process loops, easy omissions in installation compliance checks, and lack of automated verification for equipment access, which pose operational risks.

[0256] It is understood that the various method embodiments mentioned above in this disclosure can be combined with each other to form combined embodiments without violating the principle and logic. Due to space limitations, this disclosure will not elaborate further. Those skilled in the art will understand that in the above methods of specific implementation, the specific execution order of each step should be determined by its function and possible internal logic.

[0257] Figure 8 A block diagram of an FTTR service delivery quality control device provided in an embodiment of this disclosure.

[0258] Reference Figure 8 This disclosure provides an FTTR service delivery quality control device for executing the above-described FTTR service delivery quality control method. The device includes:

[0259] Install image receiving module 11 to receive images of the main optical modem installation scene uploaded by smart home engineers during the FTTR service installation and deployment process;

[0260] The intelligent quality inspection module 12 is connected to the installation image receiving module 11 and is used to input the main optical modem installation scene image into the trained artificial intelligence AI intelligent quality inspection model. If the trained AI intelligent quality inspection model identifies that the main optical modem installation scene image has no re-photographing features, no weak current box features, and contains the main optical modem, then the main optical modem installation scene is determined to be compliant; otherwise, the main optical modem installation scene is determined to be non-compliant.

[0261] The first interception and rectification module 13 is connected to the intelligent quality inspection module 12 and is used to automatically trigger the first interception command and push the first rectification guide in response to the non-compliance of the main optical modem installation scenario.

[0262] The access compliance judgment module 14 is connected to the first interception and rectification module 13. It is used to compare the actual number of optical modems deployed in the installation verification system with the number of optical modems ordered in the front-end acceptance system, and to determine whether the supply source is compliant, so as to verify the compliance of the access from the optical modem.

[0263] The second interception and rectification module 15 is connected to the access compliance judgment module 14 and is used to automatically trigger a second interception command and push a second rectification guide in response to non-compliant access from the optical modem.

[0264] The completion acceptance module 16 is connected to the second interception and rectification module 15. It is used to respond to the fact that the installation scenario of the main optical modem and the access of the secondary optical modem are both compliant, receive the acceptance request sent by the smart home engineer, call the interface of the converged network management platform according to the acceptance request, obtain the core data of the corresponding main and secondary optical modems, and perform automated completion acceptance based on the core data.

[0265] The visualization report generation module 17 is connected to the completion acceptance module 16. If the result of the automated completion acceptance is qualified, it generates a visualization report including the acceptance data; otherwise, it triggers the corresponding rectification process.

[0266] Optionally, the device further includes:

[0267] The training dataset acquisition module is used to acquire a training dataset containing image samples of compliant and non-compliant installation scenarios of the main optical modem.

[0268] A data augmentation module is used to augment the training dataset using data augmentation techniques to obtain an augmented training dataset, wherein the data augmentation techniques include at least one of rotating, cropping, and adjusting the brightness of the image;

[0269] The quality inspection model training module is used to train an AI intelligent quality inspection model based on a convolutional neural network (CNN) using the expanded training dataset, and obtain the trained AI intelligent quality inspection model.

[0270] Optionally, the access compliance judgment module 14 includes:

[0271] The quantity comparison unit is used to compare the actual number of optical modems deployed in the installation verification system with the number of optical modems ordered in the front-end acceptance system. If the difference is 0, it is determined that the optical modems are deployed normally. If the difference is greater than or equal to 1, it is determined that the optical modems are deployed in excess.

[0272] The source compliance judgment unit is used to determine whether the device identifier of each deployed slave optical modem exists in the preset compliant optical modem supply list. If the device identifier of all deployed slave optical modems exists in the compliant optical modem supply list, the supply source is determined to be compliant; otherwise, the supply source is determined to be non-compliant.

[0273] The first determination unit is used to determine the compliance of access from the optical modem in response to the normal deployment and compliance of the supply source of the optical modem.

[0274] The second determination unit is used to determine that the access from the optical modem is non-compliant in response to excessive deployment from the optical modem or non-compliant supply sources.

[0275] Optionally, the device further includes:

[0276] The optical modem data aggregation module is used to deploy interface adaptation modules on the local optical modem terminal network management system and the cloud platform terminal network management system. It uses the Flink framework to build a millisecond-level data synchronization channel and uses the data synchronization channel to uniformly aggregate the master and slave optical modem data in the local optical modem terminal network management system and the cloud platform terminal network management system to the converged network management platform.

[0277] The data platform synchronization module is used to synchronize the data of the master and slave optical modems in the converged network management platform to the data platform using the northbound interface of the converged network management platform and the data platform.

[0278] Optionally, the core data includes the online status and received light value of the master and slave optical modems, as well as the uplink connection method of the slave optical modem. The completion acceptance module 16 includes:

[0279] The acceptance judgment unit is used to make the following judgments based on the online status and received light value of the master and slave optical modems and the uplink connection mode of the slave optical modem: to determine whether the master and slave optical modems are online, to determine whether the received light value of the master and slave optical modems is within the preset threshold range, and to determine whether the uplink mode of the slave optical modem is wireless.

[0280] The qualification judgment unit is used to determine the result of the automated completion acceptance as qualified in response to the fact that both the master and slave optical modems are online, the received light values ​​of the master and slave optical modems are within the preset threshold range, and the uplink mode of the slave optical modem is not wireless.

[0281] Optionally, the device further includes:

[0282] The key indicator collection module is used to collect multiple key technical indicators of home networks corresponding to FTTR services from the data platform at regular or irregular intervals.

[0283] The network quality analysis module is used to perform quality analysis of home networks based on the collected key technical indicators of home networks and using a trained FTTR quality model to obtain quality analysis results of home networks in different cities / provinces to indicate the causes of quality problems.

[0284] The access control module is used to respond to access requests from users with different access control levels, and to display the home network quality analysis results and / or a list of poor-quality devices within the user's corresponding access control level range according to the access request.

[0285] Optionally, the device further includes:

[0286] The key indicator screening module is used to screen out multiple key technical indicators of home network from home network technical indicators. The multiple key technical indicators of home network include at least the optical receiving value of the main optical modem.

[0287] The quality model construction module is used to assign weights to each of the multiple key technical indicators of home networks using the analytic hierarchy process, and to construct an FTTR quality model based on the multiple key technical indicators of home networks and their corresponding weights.

[0288] The threshold model training module is used to train the built-in threshold model of the FTTR quality model based on historical data of normal FTTR operation, using a linear regression algorithm. This allows the threshold model to set compliance ranges for key technical indicators of each home network, resulting in the trained FTTR quality model.

[0289] Figure 9 This is a block diagram of an electronic device provided in an embodiment of the present disclosure.

[0290] Reference Figure 9 This disclosure provides an electronic device, which includes: at least one processor 701; at least one memory 702; and one or more I / O interfaces 703 connected between the processor 701 and the memory 702; wherein the memory 702 stores one or more computer programs that can be executed by the at least one processor 701, and the one or more computer programs are executed by the at least one processor 701 to enable the at least one processor 701 to execute the above-described FTTR service delivery quality control method.

[0291] This disclosure also provides a computer-readable storage medium storing a computer program thereon, wherein the computer program, when executed by a processor, implements the aforementioned FTTR service delivery quality control method. The computer-readable storage medium may be volatile or non-volatile.

[0292] In summary, the FTTR service delivery quality control method, device, electronic equipment, and medium provided in this disclosure construct a rigid closed loop of "intelligent identification - technical interception - mandatory rectification - effect verification" by embedding AI intelligent quality inspection, equipment access compliance verification, and automated completion acceptance into the entire FTTR service delivery process. This achieves this by: automatically identifying the compliance of the main optical modem installation scenario (no duplicate photos, no weak current box, and including the main optical modem) using a trained AI intelligent quality inspection model, thus technically preventing missed compliance checks; comparing the actual number of optical modems deployed with the ordered number in real time through the installation verification system, verifying the compliance of the supply source, and triggering automatic interception commands and rectification guidelines to avoid non-compliant network access and maintenance risks; and using core data from the integrated network management platform to achieve automated completion acceptance and generate visual reports, replacing the traditional manual control mode. This completely compensates for the deficiencies of existing methods in terms of missing process loops and insufficient quality control constraints, ultimately achieving standardized and intelligent control of FTTR service delivery quality, significantly improving delivery efficiency and quality stability. This addresses the issues of existing FTTR service delivery quality control methods, such as missing process loops, easy omissions in installation compliance checks, and lack of automated verification for equipment access, which pose operational risks.

[0293] Those skilled in the art will understand that all or some of the steps, systems, and apparatuses disclosed above, and their functional modules / units, can be implemented as software, firmware, hardware, or suitable combinations thereof. In hardware implementations, the division between functional modules / units mentioned above does not necessarily correspond to the division of physical components; for example, a physical component may have multiple functions, or a function or step may be performed collaboratively by several physical components. Some or all physical components may be implemented as software executed by a processor, such as a central processing unit, digital signal processor, or microprocessor, or as hardware, or as an integrated circuit, such as an application-specific integrated circuit (ASIC). Such software can be distributed on a computer-readable storage medium, which may include computer storage media (or non-transitory media) and communication media (or transient media).

[0294] As is known to those skilled in the art, the term computer storage medium includes volatile and non-volatile, removable and non-removable media implemented in any method or technology for storing information, such as computer-readable program instructions, data structures, program modules, or other data. Computer storage media includes, but is not limited to, random access memory (RAM), read-only memory (ROM), erasable programmable read-only memory (EPROM), static random access memory (SRAM), flash memory or other memory technologies, portable compact disc read-only memory (CD-ROM), digital versatile disc (DVD) or other optical disc storage, magnetic cartridges, magnetic tape, disk storage or other magnetic storage devices, or any other medium that can be used to store desired information and is accessible to a computer. Furthermore, it is known to those skilled in the art that communication media typically contain computer-readable program instructions, data structures, program modules, or other data in modulated data signals such as carrier waves or other transmission mechanisms, and may include any information delivery medium.

[0295] The computer-readable program instructions described herein can be downloaded from computer-readable storage media to various computing / processing devices, or downloaded via a network, such as the Internet, local area network, wide area network, and / or wireless network, to an external computer or external storage device. The network may include copper transmission cables, fiber optic transmission, wireless transmission, routers, firewalls, switches, gateway computers, and / or edge servers. A network adapter card or network interface in each computing / processing device receives the computer-readable program instructions from the network and forwards them to the computer-readable storage media in the respective computing / processing device.

[0296] Computer program instructions used to perform the operations of this disclosure may be assembly instructions, instruction set architecture (ISA) instructions, machine instructions, machine-dependent instructions, microcode, firmware instructions, status setting data, or source code or object code written in any combination of one or more programming languages, including object-oriented programming languages ​​such as Smalltalk, C++, etc., and conventional procedural programming languages ​​such as the "C" language or similar programming languages. The computer-readable program instructions may execute entirely on the user's computer, partially on the user's computer, as a standalone software package, partially on the user's computer and partially on a remote computer, or entirely on a remote computer or server. In cases involving a remote computer, the remote computer may be connected to the user's computer via any type of network—including a local area network (LAN) or a wide area network (WAN)—or may be connected to an external computer (e.g., via the Internet using an Internet service provider). In some embodiments, electronic circuitry, such as programmable logic circuitry, field-programmable gate arrays (FPGAs), or programmable logic arrays (PLAs), is personalized by utilizing the status information of the computer-readable program instructions to implement various aspects of this disclosure.

[0297] This disclosure has disclosed exemplary embodiments, and although specific terminology has been used, it is for general illustrative purposes only and should not be construed as limiting. In some instances, it will be apparent to those skilled in the art that features, characteristics, and / or elements described in conjunction with particular embodiments may be used alone, or in combination with features, characteristics, and / or elements described in conjunction with other embodiments, unless otherwise expressly indicated. Therefore, those skilled in the art will understand that various changes in form and detail may be made without departing from the scope of this disclosure as set forth by the appended claims.

Claims

1. A method for quality control of fiber-to-the-room (FTTR) service delivery, characterized in that, The method includes: During the installation and deployment of the FTTR service, images of the main optical modem installation scene uploaded by the smart home engineer are received; The main optical modem installation scene image is input into the trained AI intelligent quality inspection model. If the trained AI intelligent quality inspection model identifies that the main optical modem installation scene image has no re-photographing characteristics, no weak current box characteristics, and contains the main optical modem, then the main optical modem installation scene is deemed compliant; otherwise, the main optical modem installation scene is deemed non-compliant. In response to the non-compliant installation scenario of the main optical modem, the first interception command is automatically triggered, and the first rectification guide is pushed out; By comparing the actual number of optical modems deployed in the installation verification system with the number of optical modems ordered in the front-end acceptance system, and determining whether the supply source is compliant, the compliance of the access from the optical modems is verified. In response to non-compliant access from the optical modem, a second interception command is automatically triggered, and a second rectification guide is pushed out; In response to the compliance of both the main optical modem installation scenario and the secondary optical modem access, the system receives the acceptance request sent by the smart home engineer, calls the converged network management platform interface according to the acceptance request, obtains the core data of the corresponding main and secondary optical modems, and performs automated completion acceptance based on the core data. If the automated completion acceptance result is qualified, a visual report including acceptance data will be generated; otherwise, the corresponding rectification process will be triggered.

2. The method according to claim 1, characterized in that, Before inputting the main optical modem installation scene image into the trained AI intelligent quality inspection model, the method further includes: Obtain a training dataset containing image samples of both compliant and non-compliant installation scenarios of the main optical modem; The training dataset is augmented using data augmentation techniques to obtain an augmented training dataset, wherein the data augmentation techniques include at least one of rotating, cropping, and adjusting the brightness of the image; The AI ​​intelligent quality inspection model based on a convolutional neural network (CNN) is trained using the expanded training dataset to obtain the trained AI intelligent quality inspection model.

3. The method according to claim 1, characterized in that, The compliance of the access to the optical modem is verified by comparing the actual number of optical modems deployed in the installation verification system with the number of optical modems ordered in the front-end acceptance system, and determining whether the supply source is compliant. This specifically includes: The actual number of optical modems deployed in the installation verification system is compared with the number of optical modems ordered in the front-end acceptance system. If the difference is 0, it is determined that the optical modems are deployed normally. If the difference is greater than or equal to 1, it is determined that the optical modems are deployed in excess. Determine whether the device identifiers of each deployed slave optical modem exist in the preset list of compliant optical modem suppliers. If the device identifiers of all deployed slave optical modems exist in the list of compliant optical modem suppliers, the supply source is determined to be compliant; otherwise, the supply source is determined to be non-compliant. Based on the normal deployment and compliant supply source of the optical modem, the access from the optical modem is deemed compliant. In response to excessive deployment of optical modems or non-compliant supply sources, the connection from the optical modem is determined to be non-compliant.

4. The method according to claim 1, characterized in that, The method further includes: An interface adaptation module is deployed between the local optical modem terminal network management system and the cloud platform terminal network management system. A millisecond-level data synchronization channel is built using the Flink framework. The data synchronization channel is then used to unify and aggregate the master and slave optical modem data in the local optical modem terminal network management system and the cloud platform terminal network management system to the converged network management platform. By utilizing the northbound interface of the converged network management platform and the data middle platform, the data of the master and slave optical modems in the converged network management platform are synchronized to the data middle platform.

5. The method according to claim 1, characterized in that, The core data includes the online status and received optical values ​​of the master and slave optical modems, as well as the uplink connection method of the slave optical modem. The automated final acceptance testing based on this core data specifically includes: Based on the online status and received light value of the master and slave optical modems, as well as the uplink connection mode of the slave optical modem, the following judgments are made: whether the master and slave optical modems are online, whether the received light value of the master and slave optical modems is within the preset threshold range, and whether the uplink mode of the slave optical modem is wireless. In response to the fact that both the master and slave optical modems are online, the received light values ​​of the master and slave optical modems are within the preset threshold range, and the uplink mode of the slave optical modem is not wireless, the result of the automated completion acceptance is determined to be qualified.

6. The method according to claim 1, characterized in that, The method further includes: Collect multiple key technical indicators of home networks corresponding to FTTR services from the data platform at regular or irregular intervals; Based on the collected key technical indicators of the home network, the quality analysis of the home network was carried out using the trained FTTR quality model, and the quality analysis results of the home network in different cities / provinces were obtained to indicate the causes of poor quality. In response to receiving access requests from users with different permission levels, the system displays the home network quality analysis results and / or a list of poor-quality devices within the user's corresponding permission level, based on the access requests.

7. The method according to claim 6, characterized in that, Before performing quality analysis of the home network using a trained FTTR quality model based on the collected key technical indicators of the home network, the method further includes: Several key technical indicators for home networks were selected from the technical indicators of home networks, and the key technical indicators for home networks include at least the optical receiving value of the main optical modem; The Analytic Hierarchy Process (AHP) is used to assign weights to each of the multiple key technical indicators of home networks, and an FTTR mass model is constructed based on the multiple key technical indicators of home networks and their corresponding weights. Based on historical data of normal FTTR operation, a threshold model built into the FTTR quality model is trained using a linear regression algorithm. This threshold model sets compliance ranges for key technical indicators of each home network, resulting in a well-trained FTTR quality model.

8. A fiber-to-the-room (FTTR) service delivery quality control device, characterized in that, The device includes: Install an image receiving module to receive images of the main optical modem installation scene uploaded by smart home engineers during the FTTR service installation and deployment process; The intelligent quality inspection module is connected to the installation image receiving module and is used to input the main optical modem installation scene image into the trained artificial intelligence (AI) intelligent quality inspection model. If the trained AI intelligent quality inspection model identifies that the main optical modem installation scene image has no re-photographing features, no weak current box features, and contains the main optical modem, then the main optical modem installation scene is deemed compliant; otherwise, the main optical modem installation scene is deemed non-compliant. The first interception and rectification module is connected to the intelligent quality inspection module and is used to automatically trigger the first interception command and push the first rectification guide in response to non-compliance of the main optical modem installation scenario; The access compliance judgment module is connected to the first interception and rectification module. It is used to compare the actual number of optical modems deployed in the installation verification system with the number of optical modems ordered in the front-end acceptance system, and to determine whether the supply source is compliant, so as to verify the compliance of the access from the optical modem. The second interception and rectification module is connected to the access compliance judgment module. It is used to automatically trigger a second interception command and push a second rectification guide in response to non-compliant access from the optical modem. The completion acceptance module, connected to the second interception and rectification module, is used to respond to the fact that both the main optical modem installation scenario and the secondary optical modem access are compliant, receive the acceptance request sent by the smart home engineer, call the converged network management platform interface according to the acceptance request, obtain the core data of the corresponding main and secondary optical modems, and perform automated completion acceptance based on the core data; The visualization report generation module is connected to the completion acceptance module. If the result of the automated completion acceptance is qualified, it generates a visualization report including the acceptance data; otherwise, it triggers the corresponding rectification process.

9. An electronic device, characterized in that, The electronic device includes: At least one processor; and A memory communicatively connected to the at least one processor; wherein, The memory stores one or more computer programs that can be executed by the at least one processor, the one or more of the computer programs being executed by the at least one processor to enable the at least one processor to perform the fiber-to-the-room (FTTR) service delivery quality control method as described in any one of claims 1-7.

10. A computer-readable storage medium having a computer program stored thereon, characterized in that, When the computer program is executed by the processor, it implements the fiber-to-the-room (FTTR) service delivery quality control method as described in any one of claims 1-7.

Images