Method and device for identifying FTTR edge device deployment abnormity

Abstract

The invention provides a method and a device for identifying abnormal deployment of FTTR edge equipment. The method comprises the following steps: acquiring signal intensity of the edge equipment and behavior characteristic data of a terminal; and determining whether deployment of the edge device is abnormal or not according to the signal strength of the edge device and the behavior characteristic data of the terminal. According to the method provided by the embodiment of the invention, the multi-dimensional joint judgment of the FTTR edge device deployment abnormity is carried out through the signal strength of the edge device and the behavior characteristic data of the terminal, so that the problem that the prior art only depends on RSSI (Received Signal Strength Indicator) and is easily influenced by environment shielding and device power difference is solved, false alarms and missing alarms are effectively reduced, and the judgment accuracy is greatly improved.

Description

Technical Field

[0001] This invention relates to the field of communication technology, and in particular to a method and apparatus for identifying abnormal deployments of FTTR edge devices. Background Technology

[0002] With the widespread deployment of Fiber to the Room (FTTR) technology in home settings, a master-slave optical network unit (ONU) collaborative networking approach is typically adopted to achieve high-speed broadband coverage throughout the house.

[0003] In related technologies, during actual deployments, improper installation, special room layouts, or user misoperation often result in master and slave ONU devices being placed too close together. This leads to wasted coverage resources, increased channel interference, and degraded network performance, negatively impacting user experience. Therefore, how to efficiently and accurately identify the problem of FTTR edge devices being placed too close together is a technical problem that urgently needs to be solved by those skilled in the art. Summary of the Invention

[0004] This invention provides a method and apparatus for identifying abnormal deployment of FTTR edge devices. It performs multi-dimensional joint judgment of abnormal deployment of FTTR edge devices by using the signal strength of the edge devices and the behavioral feature data of the terminals. This avoids the problem of existing technologies relying solely on RSSI, which is susceptible to environmental obstruction and differences in device power. It also solves the defect of focusing only on device status and ignoring the actual user usage scenario, effectively reducing false alarms and false negatives, and significantly improving the accuracy of judgment.

[0005] This invention provides a method for identifying abnormal deployment of FTTR edge devices, comprising the following steps.

[0006] Acquire signal strength data from edge devices and behavioral characteristic data of terminals; Based on the signal strength of the edge device and the behavioral characteristics data of the terminal, it is determined whether the deployment of the edge device is abnormal.

[0007] According to a method for identifying abnormal FTTR edge device deployment provided by the present invention, the behavioral characteristic data of the terminal includes data on the terminal accessing edge devices and data on the terminal switching between edge devices; the step of determining whether the deployment of the edge device is abnormal based on the signal strength of the edge device and the behavioral characteristic data of the terminal includes: Based on the data accessed by the terminal to the edge device, the overlap of terminals served by different edge devices is determined; The number of times the terminal switches between edge devices is determined based on the data of the terminal switching between edge devices; The deployment of the edge devices is determined to be abnormal based on the signal strength of the edge devices, the overlap of the terminals served by the different edge devices, and the number of times the terminals are switched.

[0008] According to a method for identifying abnormal FTTR edge device deployment provided by the present invention, determining whether the edge device deployment is abnormal based on the signal strength of the edge device, the overlap of terminals served by different edge devices, and the number of handovers of the terminals includes: If the signal strength of the edge device is greater than a first threshold, the overlap of terminals served by different edge devices is greater than a second threshold, and the number of handovers of the terminals is greater than a third threshold, then the edge device deployment is determined to be abnormal.

[0009] According to a method for identifying FTTR edge device deployment anomalies provided by the present invention, when determining that the edge device deployment is abnormal, the method further includes: Determine the historical number of times the edge device has been tuned; The handling method for the edge device is determined based on the number of historical optimizations.

[0010] According to a method for identifying FTTR edge device deployment anomalies provided by the present invention, the step of determining the handling method of the edge device based on the historical tuning count includes: If the number of historical optimization attempts is less than the fourth threshold, adjust the power of the edge device and update the number of historical optimization attempts. If the number of historical optimizations is greater than or equal to the fourth threshold, cloud devices are triggered to collaboratively handle the deployment anomalies of the edge devices.

[0011] According to the present invention, a method for identifying FTTR edge device deployment anomalies, wherein triggering cloud devices to collaboratively process the edge device deployment anomalies includes: Send a first message to the cloud device; the first message is used to instruct the cloud device to determine whether the edge device is deployed abnormally based on the abnormal placement analysis model; the abnormal placement analysis model is trained based on sample data; the sample data includes the signal strength of the edge device, the behavioral characteristic data of the terminal, and the data of edge device deployment abnormal / normal confirmed by the operation and maintenance personnel on site.

[0012] The present invention also provides a device for identifying abnormal deployment of FTTR edge devices, comprising the following modules: The acquisition module is used to acquire signal strength data of edge devices and behavioral characteristic data of terminals; The determination module is used to determine whether the deployment of the edge device is abnormal based on the signal strength of the edge device and the behavioral characteristic data of the terminal.

[0013] The present invention also provides an edge device, including a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein the processor executes the program to implement the identification method for abnormal FTTR edge device deployment as described above.

[0014] The present invention also provides a non-transitory computer-readable storage medium having a computer program stored thereon, which, when executed by a processor, implements the identification method for abnormal FTTR edge device deployment as described above.

[0015] The present invention also provides a computer program product, including a computer program that, when executed by a processor, implements the identification method for abnormal FTTR edge device deployment as described above.

[0016] The method and apparatus for identifying FTTR edge device deployment anomalies provided by this invention perform multi-dimensional joint judgment of FTTR edge device deployment anomalies by using the signal strength of the edge device and the behavioral feature data of the terminal. This avoids the problem that existing technologies rely solely on RSSI and are easily affected by environmental obstruction and differences in device power. It also solves the defect of only focusing on device status and ignoring the actual user usage scenario, effectively reducing false alarms and false negatives, and significantly improving the accuracy of judgment. Attached Figure Description

[0017] To more clearly illustrate the technical solutions in this invention or the prior art, the drawings used in the description of the embodiments or the prior art will be briefly introduced below. Obviously, the drawings described below are some embodiments of this invention. For those skilled in the art, other drawings can be obtained from these drawings without creative effort.

[0018] Figure 1 This is one of the flowcharts illustrating the method for identifying abnormal deployment of FTTR edge devices provided by the present invention.

[0019] Figure 2 This is the second flowchart of the method for identifying abnormal deployment of FTTR edge devices provided by the present invention.

[0020] Figure 3 This is the third flowchart of the method for identifying abnormal deployment of FTTR edge devices provided by the present invention.

[0021] Figure 4 This is the fourth flowchart of the method for identifying abnormal deployment of FTTR edge devices provided by the present invention.

[0022] Figure 5This is a schematic diagram of the structure of the FTTR edge device deployment anomaly identification device provided by the present invention.

[0023] Figure 6 This is a schematic diagram of the structure of the electronic device provided by the present invention. Detailed Implementation

[0024] To make the objectives, technical solutions, and advantages of this invention clearer, the technical solutions of this invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some, not all, of the embodiments of this invention. All other embodiments obtained by those skilled in the art based on the embodiments of this invention without creative effort are within the scope of protection of this invention.

[0025] The following is combined Figures 1-6 The present invention describes a method and apparatus for identifying abnormal deployments of FTTR edge devices.

[0026] To facilitate a clearer understanding of the technical solutions of the various embodiments of this application, some technical content related to the various embodiments of this application will be introduced first.

[0027] With the widespread deployment of FTTR technology in home scenarios, a master-slave ONU collaborative networking approach is typically used to achieve whole-house high-speed broadband coverage. However, in actual deployments, improper installation, special room layouts, or user misoperation often result in the master and slave ONU devices being placed too close together, leading to wasted coverage resources, increased channel interference, and degraded network performance, thus affecting user experience. Currently, the judgment of edge devices being placed too close mainly relies on manual experience or a single threshold for master-slave signal strength, lacking a dynamic and automated judgment and handling mechanism. Its main shortcomings are as follows: 1. The lack of a multi-dimensional feature comprehensive judgment mechanism and reliance solely on signal strength RSSI can easily lead to false alarms or missed alarms.

[0028] Currently, some intelligent operation and maintenance methods in FTTR networks attempt to determine the deployment relationship based on the RSSI (Radio Signal Strength Index) of the WiFi signal between master and slave ONUs; that is, the higher the RSSI value, the closer the perceived physical distance. However, RSSI only reflects the instantaneous signal strength and is greatly affected by environmental obstructions, wall structure, antenna direction, bandwidth usage, etc., and cannot accurately reflect the actual deployment status of the devices. For example, in an environment without wall obstructions and with the door open, the RSSI of two devices in separate rooms may still be higher than -35dBm; some master and slave devices placed in the same corner of a room may have a low RSSI due to channel interference or power limitations, and may be mistakenly judged as normal; some high-power device models inherently have a high RSSI, leading to an increased false alarm rate. Therefore, using only RSSI thresholds as a judgment standard is prone to situations in complex home environments where "deployment is normal but misjudged as too close" or "deployment is too close but not detected," lacking versatility.

[0029] 2. Unable to dynamically detect deployment redundancy issues based on terminal / user behavior.

[0030] The ultimate goal of FTTR is to provide distributed coverage services for user terminals. Therefore, terminal connection behavior data is an important indicator of the rationality of the network structure. However, existing systems often neglect the analysis of access terminal data, focusing only on the device's own status, resulting in the inability to identify master-slave relationships with severely overlapping service areas. For example, if the same terminal frequently switches between master and slave (e.g., more than 3 times within 5 minutes), it indicates that the coverage areas of the two devices overlap, resulting in deployment redundancy. If the MAC address lists of terminals served by the master and slave devices are highly consistent (overlap rate ≥80%), it indicates that the terminals cannot distinguish between master and slave devices, leading to signal overlap. However, because existing solutions do not establish a data behavior linkage analysis mechanism between the ONU and the terminal, they cannot achieve quantitative modeling and dynamic judgment of terminal connection behavior, thus failing to perceive problems in actual network use and leading to a decline in user experience.

[0031] 3. Lack of a collaborative mechanism between real-time edge detection and cloud-based intelligent analysis.

[0032] Currently, the FTTR network structure suffers from insufficient timeliness in problem perception. Most intelligent judgment functions rely on cloud platform data aggregation and periodic analysis, lacking real-time performance and on-site adaptability. Simultaneously, the computing power of edge ONU devices is severely underestimated, failing to realize their value in pre-judgment. For example, most current judgment models are deployed in the cloud, requiring unified analysis after user behavior data is uploaded, resulting in high judgment latency and an inability to provide real-time feedback on deployment issues; edge devices lack local judgment logic, and even in situations of strong interference or overlapping deployments, they cannot autonomously perceive the problem and can only wait for cloud-side policies to be issued; edge devices cannot actively participate in the feedback process, requiring significant cloud resources to complete actions that could be performed by edge devices, necessitating substantial cloud computing power. Consequently, the overall judgment system cannot achieve rapid, accurate, and closed-loop problem identification and handling.

[0033] Figure 1 This is one of the flowcharts illustrating the method for identifying abnormal deployments of FTTR edge devices provided by the present invention. The method includes the following: Step 101: Obtain signal strength data of edge devices and behavioral characteristic data of terminals.

[0034] Specifically, traditional methods rely solely on the static RSSI strength values ​​between master and slave devices to determine the deployment distance of FTTR edge devices. This method is susceptible to environmental interference, device differences, building structure, and other factors, resulting in a high misjudgment rate.

[0035] To address the aforementioned issues, this embodiment first acquires the signal strength of the edge device and the behavioral characteristic data of the terminal. Optionally, the signal strength of the edge device includes the mutual scan received signal strength (RSSI) between the master ONU and the slave ONU, and the user's behavioral characteristic data includes access data and handover data of the terminal connected to the edge device, and may also include data such as the terminal's rate and throughput.

[0036] Step 102: Determine whether the deployment of the edge devices is abnormal based on the signal strength of the edge devices and the behavioral characteristics of the terminals.

[0037] Specifically, after acquiring the signal strength data of the edge device and the behavioral characteristic data of the terminal, this application determines whether the deployment of the edge device is abnormal based on the signal strength data of the edge device and the behavioral characteristic data of the terminal. In other words, this application integrates multi-dimensional features such as the signal strength data of the edge device and the behavioral characteristic data of the terminal to determine whether the FTTR edge devices are placed too close together, thereby significantly improving the accuracy of identifying abnormal FTTR edge device deployments. This solves the problem of false alarms or missed alarms easily generated by relying solely on RSSI, reducing network resource waste, user experience degradation, and after-sales complaints caused by improper device deployment, helping operators effectively control network maintenance costs and improve service reputation and user satisfaction. Furthermore, it should be noted that data acquisition and processing in this application embodiment can be completed locally on the edge device without relying on cloud data transmission and aggregation, which can significantly reduce data latency and improve the real-time performance of identifying abnormal FTTR edge device deployments.

[0038] The method described above uses signal strength data from edge devices and behavioral characteristic data from terminals to perform multi-dimensional joint judgment of FTTR edge device deployment anomalies. This avoids the problem that existing technologies rely solely on RSSI, which is susceptible to environmental obstruction and differences in device power. It also solves the defect of focusing only on device status and ignoring the actual user scenario, effectively reducing false alarms and false negatives, and significantly improving the accuracy of judgment.

[0039] In some embodiments, the terminal's behavioral characteristic data includes data on the terminal's access to edge devices and data on the terminal's switching between edge devices; determining whether the deployment of the edge devices is abnormal based on the signal strength of the edge devices and the terminal's behavioral characteristic data includes: Based on the data from the terminal access edge devices, determine the degree of overlap between the terminals served by different edge devices; The number of handovers for a terminal is determined based on the data from the terminal's switching between edge devices; The deployment of edge devices is determined based on the signal strength of the edge devices, the degree of overlap of the terminals served by different edge devices, and the number of terminal handovers.

[0040] Specifically, when the master and slave ONUs are placed too close together, their wireless signal coverage areas will overlap significantly, forming an overlapping coverage area. In this overlapping coverage area, the signal strength difference between the master and slave ONUs is minimal, and terminals may randomly connect to either the master or slave device, thus increasing the overlap of terminals connected to the master and slave ONUs.

[0041] Optionally, in this embodiment, the terminal access status is analyzed based on the data from the terminal access edge device to generate a list of access terminals for the master and slave ONUs, such as terminal device MAC address + edge device MAC address + access time + device action (access / disconnect). A data association algorithm is used to compare the MAC addresses of terminals connected to the master and slave ONUs, and the terminal overlap rate (TOR) is calculated by determining the ratio of the number of duplicate MAC address records to the total number of terminal records.

[0042] For example, if the set of access terminals of a master device is M={m1,m2,...,mn}, and the set of access terminals of a certain slave device is S={s1,s2,...,sk}, then the overlap rate TOR of the two terminals is defined as: TOR=(|M∩S|) / (|M∪S|).

[0043] Optionally, placing FTTR edge devices too close together can lead to coverage redundancy. In overlapping coverage areas, the signal strength difference between the master and slave ONUs is minimal, which can increase the overlap of terminals connected to the master and slave ONUs and cause frequent switching of terminals connected to the master and slave ONUs.

[0044] Optionally, in this embodiment, when the same terminal (uniquely identified by its MAC address) disconnects from device A (e.g., the master optical modem) and connects to device B (e.g., the slave optical modem) within a short period, it is considered a handover event. Optionally, if the number of access handovers between the master device and any slave device within the period [t0,t1] is fi, then the terminal handover frequency (TRF) for the current master-slave combination is: Where N represents the number of terminals participating in the handover behavior statistics within the analysis period, and fi is the number of times a terminal switches between devices within the time window.

[0045] Optionally, after determining the mutual scan received signal strength (RSSI) between the master ONU and slave ONU, the overlap of terminals served by different edge devices, and the number of terminal handovers, the problem of edge devices being placed too close can be accurately judged based on data from multiple dimensions. This significantly improves the accuracy of anomaly identification and avoids false alarms and missed alarms caused by the existing technology's reliance solely on RSSI.

[0046] The method described above combines the mutual scanning signal strength between master and slave ONUs, the overlap of terminals served by different edge devices, and terminal switching behavior to form a multi-dimensional joint judgment system for edge device deployment anomalies. This effectively avoids the shortcomings of relying solely on RSSI, which is susceptible to environmental interference, and achieves accurate identification of master and slave ONUs being placed too close together in the FTTR system.

[0047] In some embodiments, determining whether the deployment of the edge devices is abnormal based on the signal strength of the edge devices, the overlap of terminals served by different edge devices, and the number of terminal handovers includes: If the signal strength of the edge device is greater than the first threshold, the overlap of terminals served by different edge devices is greater than the second threshold, and the number of terminal handovers is greater than the third threshold, the edge device deployment is determined to be abnormal.

[0048] Specifically, in this embodiment, the edge device deployment is determined to be abnormal only when the signal strength of the edge device is greater than the first threshold, the overlap of the terminals served by different edge devices is greater than the second threshold, and the number of terminal switching is greater than the third threshold, i.e., all three conditions are met simultaneously; if any one condition is not met, the deployment is determined to be normal. This can effectively eliminate the judgment bias caused by fluctuations in a single indicator and significantly improve the accuracy of FTTR edge device deployment anomaly identification in complex home environments.

[0049] In some embodiments, when an edge device deployment anomaly is determined, the method further includes: Determine the historical number of times edge devices have been optimized; Based on the number of historical optimizations, determine the handling methods for abnormal edge device deployments.

[0050] Specifically, by analyzing the historical number of optimizations, the history of interventions for the same abnormal scenario can be clearly understood. In this embodiment, when an edge device deployment anomaly is determined, the handling method for the anomaly is adjusted specifically based on the historical number of optimizations performed on the edge device, effectively avoiding repeated, ineffective optimizations and resource waste.

[0051] For example, a smaller number of historical optimization attempts indicates that there is still room for trial and error and improvement, and optimization can continue based on the original direction. A larger number of historical optimization attempts indicates that the anomaly has not been eliminated after multiple optimization adjustments, and the original optimization method needs to be changed. Therefore, this application adjusts the handling method for edge device deployment anomalies based on the number of historical optimization attempts, which can effectively balance the real-time nature of anomaly handling, the efficiency of resource utilization, and the thoroughness of problem resolution, thus achieving efficient handling of edge device deployment anomalies.

[0052] In some embodiments, determining the handling method for edge devices based on the number of historical optimizations includes: If the number of historical optimization attempts is less than the fourth threshold, adjust the power of the edge devices and update the number of historical optimization attempts. If the number of historical optimizations is greater than or equal to the fourth threshold, cloud devices will be used to collaboratively handle edge device deployment anomalies.

[0053] Specifically, in this embodiment, when the number of historical optimization attempts is less than the fourth threshold, the embodiment performs a local, autonomous, lightweight handling operation for edge device deployment anomalies. Optionally, the master ONU generates a power optimization strategy to adjust the transmit power of the master and slave ONUs in a step-by-step manner, such as reducing it by 10% each time. This reduces the wireless signal coverage radius and alleviates coverage overlap and channel interference caused by edge devices being placed too close together. The entire adjustment process requires no cloud intervention; the power adjustment is completed locally at the edge. The master ONU directly issues power configuration commands to the slave ONU and executes them synchronously. This results in low response latency, quickly reducing signal interference and coverage redundancy between the master and slave ONUs, preventing anomalies from continuously affecting the connection stability of user terminals. This solves the pain points of high latency and slow response in traditional cloud-based centralized handling, achieving real-time and rapid mitigation of edge device deployment anomalies.

[0054] Optionally, after the edge device completes the power adjustment operation locally, the main ONU immediately increments the locally stored historical tuning count counter by 1 to ensure that each autonomous tuning behavior is accurately recorded, providing a basis for the next round of anomaly judgment and handling decisions.

[0055] Optionally, if the number of historical optimization attempts is greater than or equal to the fourth threshold, a cloud-based collaborative handling process for edge device deployment anomalies is triggered. That is, when the number of historical optimization attempts is greater than or equal to the fourth threshold, the edge device determines that local power adjustments are insufficient to resolve the edge device deployment anomalies and immediately initiates a cloud-based collaborative handling process. Optionally, the cloud, trained on massive samples, possesses stronger generalization capabilities, effectively addressing the limitations of edge device capabilities and significantly improving the rationality of home broadband network deployment and the intelligence level of anomaly handling in the FTTR system. It should be noted that the cloud can perform training and iterative optimization based on actual network operation data and user terminal behavior data. Compared to traditional rule-based judgment methods, it possesses stronger generalization and learning capabilities, making it particularly suitable for home networking environments with complex terminal behavior and dynamically changing network states. Furthermore, this application adopts a collaborative architecture of lightweight edge judgment + strong cloud analysis. Edge devices achieve rapid prediction and preliminary optimization through rule combinations (RSSI + overlap rate + handover frequency), and the cloud analysis module performs secondary optimization judgment based on this. This design effectively improves the real-time performance of judgment while maintaining recognition accuracy, effectively balancing efficiency and accuracy.

[0056] The method described in the above embodiment adjusts the power of the edge device and updates the historical optimization count when the number of historical optimizations is less than the fourth threshold; when the number of historical optimizations is greater than or equal to the fourth threshold, it triggers cloud-device collaborative processing of edge device deployment anomalies. This achieves cloud-edge collaboration and on-demand cloud access, leveraging the real-time response advantage of the edge while relying on the cloud for intelligent analysis. This significantly improves the intelligence level of anomaly handling in the FTTR system, avoids the large amount of redundant calculations caused by continuous polling and analysis of all devices in the cloud in traditional solutions, and improves the overall system operating efficiency and processing speed.

[0057] In some embodiments, triggering cloud devices to collaboratively handle edge device deployment anomalies includes: Send the first message to the cloud device; the first message is used to instruct the cloud device to determine whether the edge device is deployed abnormally based on the abnormal placement analysis model; the abnormal placement analysis model is trained based on sample data; the sample data includes the signal strength of the edge device, the behavioral characteristic data of the terminal, and the data of edge device deployment abnormal / normal confirmed by the operation and maintenance personnel on site.

[0058] Specifically, in this embodiment, when the number of historical optimizations is greater than or equal to a fourth threshold, a first message is sent to the cloud device. The first message is used to instruct the cloud device to determine whether the edge device is deployed abnormally based on the abnormal placement analysis model. That is, this application uses a machine learning model to perform in-depth analysis of multi-dimensional historical and real-time data and outputs a secondary judgment result on whether the edge device is deployed abnormally. It should be noted that, compared with the rule judgment based on a fixed threshold on the edge side, the cloud-based abnormal placement analysis model is trained on massive labeled samples, which can capture the non-linear correlation between features, effectively avoid false alarms caused by environmental interference in the single rule judgment on the edge side, significantly improve the judgment reliability, and reduce the false alarm rate of operation and maintenance. Optionally, if the cloud-based abnormal placement analysis model determines that the deployment is abnormal, the cloud generates an operation and maintenance work order through the work order management module and pushes it to the dispatch system, clearly marking the device location, abnormal characteristics, and historical optimization status, guiding the operation and maintenance personnel to carry out on-site rectification; the rectification result is fed back to the cloud by the operation and maintenance personnel as a new sample to supplement the model training library. Optionally, if the cloud-based abnormal placement analysis model determines that the abnormality is not a deployment anomaly, the cloud records the judgment result, terminates the subsequent handling process, and feeds the result back to the edge device to assist the edge side in optimizing the threshold or judgment logic.

[0059] For example, such as Figure 2 As shown, the method for identifying abnormal deployment of FTTR edge devices in this application has the following specific process: This application adopts a cloud-edge collaborative judgment architecture, embedding a lightweight combined rule judgment module in the edge devices (master and slave ONUs) of the FTTR network. Simultaneously, it further improves judgment accuracy through a cloud-based intelligent analysis module and machine learning model, and synchronizes the judgment results to the dispatch system to complete the closed loop. Specifically, the edge side completes data acquisition and processing, the master optical modem completes the analysis and optimization of the aggregated data, and the master optical modem obtains the threshold for the "too close" rule judgment from the cloud side. The cloud side completes the aggregation of network-wide data and the threshold management of edge-side rule judgments, while deploying a machine learning model to infer abnormal signals generated by the edge side. The signals generated by machine learning are then processed through the work order management module to complete the closed-loop processing of work order generation, dispatch, and receiving manual feedback.

[0060] The method described above, based on the collaborative judgment mechanism of lightweight edge rule judgment and cloud machine learning, realizes dynamic perception and closed-loop processing of the problem of FTTR being placed too close together. It not only leverages the advantages of real-time edge response, but also significantly reduces the number of invalid on-site visits caused by edge misjudgment through secondary verification of the cloud model, significantly improves the reliability of edge device deployment anomaly identification, and reduces the operator's manpower and time costs.

[0061] For example, such as Figure 3 and Figure 4 As shown, the method for identifying abnormal deployment of FTTR edge devices in this application has the following specific process: I. Edge-side combination judgment process (1) Periodic execution task startup: The system starts the analysis task of placing too close together according to the set period.

[0062] (2) The analysis begins. The data acquisition and processing module calls the local cached data or pulls the currently collected data and enters the data judgment process.

[0063] (3) Determine whether the RSSI strength between the master and slave devices exceeds the RSSI threshold RSSIth, which is -35dBm by default. If the RSSI value is greater than RSSIth, it is considered that there is a risk of deployment being too close, and the analysis result is saved as analysis result R1. Analyze the terminal access situation within the period and generate a list of access terminals of the current device (terminal device MAC + device device MAC + access time + device action (access / disconnect)); (4) Determine whether the current device is the main optical modem. If it is the main optical modem, wait for the secondary optical modem to report the third step of the analysis data within the cycle; if it is the secondary optical modem, report the preliminary analysis data to the main optical modem.

[0064] (5) After obtaining all data information of the slave devices and their downstream terminals, including access time and MAC address, construct an analysis rule model.

[0065] Specifically, a data association algorithm is used to compare the MAC addresses of terminals connected to the master and slave devices. By calculating the ratio of the number of duplicate MAC address records to the total number of terminal records, the formula for calculating the terminal overlap rate (TOR) is derived: Let the set of access terminals of the master device be M={m1,m2,...,mn}, and the set of access terminals of a certain slave device be S={s1,s2,...,sk}. Then the overlap rate TOR of the two terminals is defined as: TOR=(|M∩S|) / (|M∪S|).

[0066] Let fi be the number of access handovers that a terminal Ti performs between the master device and any slave device within the analysis period [t0, t1]. Then, the terminal handover frequency TRF for the current master-slave combination is: Where N represents the number of terminals participating in the handover behavior statistics within the analysis period, and fi is the number of times a terminal switches between devices within the time window.

[0067] A handover event is defined as follows: when the same terminal (uniquely identified by its MAC address) disconnects from device A (such as the main optical modem) and connects to device B (such as the secondary optical modem) within a short period of time, it is considered a handover event.

[0068] Based on the previously obtained RSSI, TOR, and TRF, a judgment is made according to the thresholds configured in the cloud. If RSSI > RSSIth, TOR ≥ TORth, and TRF ≥ TRFth, a signal of placement too close is generated, and step six is ​​performed. If the conditions are not met, the judgment within this window period is terminated directly, and the optimization count is reset to 0. In other words, this application combines the edge combination rules of RSSI, terminal overlap, and handover frequency to make proactive optimization attempts, significantly improving the accuracy of identifying scenarios where placement is too close.

[0069] (6) After generating an abnormal placement signal, determine whether the signal has reached the threshold of the optimization count. If the threshold has been reached, generate an abnormal placement signal for the device directly and send it to the cloud to end. If the threshold has not been reached, increment the optimization count by 1 and generate an automatic power optimization strategy for the optical modem. Reduce the current device's transmit power by 10% and perform optimization operation. This round of analysis process ends and waits for the next cycle to start.

[0070] II. Cloud-based machine learning enhances judgment. (7) After receiving a signal that the edge side is placed too close, sort out the data within the cycle of the involved equipment, including RSSI statistics of master and slave devices, terminal access data, etc., and at the same time standardize the raw data, fill in missing fields, denoise, type conversion, etc.

[0071] (8) Based on the preprocessed data obtained, construct a feature vector suitable for the model input for the abnormal scenario of devices being placed too close together.

[0072] Table 1 Feature name Description RSSI_mean Master-slave mutual scan signal strength mean RSSI_std Signal strength standard deviation characterizes link volatility. TOR Terminal Overlap Ratio TRF Terminal Roaming Frequency (9) Call the pre-deployed XGBoost abnormal placement model in the cloud, input the feature vector, and output the judgment result of whether the placement is too close. If the placement is too close, generate a FTTR placement too close rectification work order to the dispatch system. The dispatch system arranges installation and maintenance personnel to carry out on-site handling and feeds back the rectification result and on-site confirmation result to the cloud. The cloud feeds back the sample and label to the placement too close model for reinforcement learning to continuously optimize and iterate; if it does not exist, it ends directly. That is, in this application, the edge actively analyzes the FTTR placement too close, and jointly models the FTTR network environment and user equipment behavior features such as the signal strength statistics between master and slave devices, terminal switching frequency, etc., to generate abnormal placement signals and then perform secondary analysis through cloud inference. At the same time, the model results are linked with the dispatch system to form a closed loop of handling, which effectively realizes cloud-edge collaboration and significantly improves the deployment rationality and intelligence level of home broadband networks.

[0073] Optionally, the training process for the abnormal placement model is as follows: Historical sample data was collected from the actual operating network. The samples consisted of edge detection results, device operation data, terminal behavior logs, and manual work order rectification results. The collected raw features were transformed and enhanced, and the training set was used as input to the XGBoost classification model. A multi-round iterative boosting tree approach was employed to optimize the objective function. Performance verification and model evaluation were performed using an independent test set, and the model was deployed online. Based on the collected device MAC addresses and access times, a list of access terminals for each master and slave device was generated. For example, the data structure for accessing slave devices was [(Device MAC1, Access Device MAC, Access Time time1, Access Action), (Device MAC2, Access Device MAC, Access Time time2, Access Action), ...], and stored in a local temporary data storage area to provide basic data for subsequent model analysis.

[0074] The method described above, based on the existing FTTR networking architecture, fully utilizes the local computing capabilities of master and slave ONU devices and the intelligent analysis capabilities of the cloud side. Without adding new hardware, it can achieve large-scale deployment simply by upgrading edge software modules and deploying cloud-side models. This reduces network resource waste, user experience degradation, and after-sales complaints caused by improper device deployment, thereby helping operators effectively control network maintenance costs and improve service reputation and user satisfaction.

[0075] The following describes the FTTR edge device deployment anomaly identification device provided by the present invention. The FTTR edge device deployment anomaly identification device described below corresponds to the FTTR edge device deployment anomaly identification method described above. The FTTR edge device deployment anomaly identification device of this application embodiment is as follows: Figure 5 As shown, it includes: The acquisition module 510 is used to acquire signal strength data of edge devices and behavioral characteristic data of terminals; The determination module 520 is used to determine whether the deployment of the edge device is abnormal based on the signal strength of the edge device and the behavioral characteristic data of the terminal.

[0076] Figure 6 A schematic diagram of the physical structure of an electronic device is provided. This electronic device may include a processor 610, a communications interface 620, a memory 630, and a communication bus 640. The processor 610, communications interface 620, and memory 630 communicate with each other via the communication bus 640. The processor 610 can invoke logical instructions in the memory 630 to execute a method for identifying abnormal deployments of TTR edge devices. This method includes: acquiring signal strength data of the edge device and behavioral characteristic data of the terminal; and determining whether the deployment of the edge device is abnormal based on the signal strength data of the edge device and the behavioral characteristic data of the terminal.

[0077] Furthermore, the logical instructions in the aforementioned memory 630 can be implemented as software functional units and, when sold or used as independent products, can be stored in a computer-readable storage medium. Based on this understanding, the technical solution of the present invention, in essence, or the part that contributes to the prior art, or a part of the technical solution, can be embodied in the form of a software product. This computer software product is stored in a storage medium and includes several instructions to cause a computer device (which may be a personal computer, server, or network device, etc.) to execute all or part of the steps of the methods described in the various embodiments of the present invention. The aforementioned storage medium includes various media capable of storing program code, such as USB flash drives, portable hard drives, read-only memory (ROM), random access memory (RAM), magnetic disks, or optical disks.

[0078] On the other hand, the present invention also provides a computer program product, the computer program product including a computer program, the computer program being stored on a non-transitory computer-readable storage medium, the computer program being executed by a processor, the computer being able to execute the TTR edge device deployment anomaly identification method provided by the above methods, the method including: acquiring the signal strength of the edge device and the behavioral characteristic data of the terminal; determining whether the deployment of the edge device is abnormal based on the signal strength of the edge device and the behavioral characteristic data of the terminal.

[0079] In another aspect, the present invention also provides a non-transitory computer-readable storage medium having a computer program stored thereon, which, when executed by a processor, implements a method for identifying TTR edge device deployment anomalies provided by the methods described above. The method includes: acquiring signal strength data of the edge device and behavioral characteristic data of the terminal; and determining whether the deployment of the edge device is abnormal based on the signal strength data of the edge device and the behavioral characteristic data of the terminal.

[0080] The device embodiments described above are merely illustrative. The units described as separate components may or may not be physically separate. The components shown as units may or may not be physical units; that is, they may be located in one place or distributed across multiple network units. Some or all of the modules can be selected to achieve the purpose of this embodiment according to actual needs. Those skilled in the art can understand and implement this without any creative effort.

[0081] Through the above description of the embodiments, those skilled in the art can clearly understand that each embodiment can be implemented by means of software plus necessary general-purpose hardware platforms, and of course, it can also be implemented by hardware. Based on this understanding, the above technical solutions, in essence or the part that contributes to the prior art, can be embodied in the form of a software product. This computer software product can be stored in a computer-readable storage medium, such as ROM / RAM, magnetic disk, optical disk, etc., and includes several instructions to cause a computer device (which may be a personal computer, server, or network device, etc.) to execute the methods described in the various embodiments or some parts of the embodiments.

[0082] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention, and not to limit them; although the present invention has been described in detail with reference to the foregoing embodiments, those skilled in the art should understand that modifications can still be made to the technical solutions described in the foregoing embodiments, or equivalent substitutions can be made to some of the technical features; and these modifications or substitutions do not cause the essence of the corresponding technical solutions to deviate from the spirit and scope of the technical solutions of the embodiments of the present invention.

Claims

1. A method for identifying abnormal deployments of FTTR edge devices, characterized in that, Edge devices used in FTTR systems include: Acquire signal strength data of the edge device and behavioral characteristic data of the terminals served by the edge device; Based on the signal strength of the edge device and the behavioral characteristics data of the terminal, it is determined whether the deployment of the edge device is abnormal.

2. The method for identifying abnormal deployment of FTTR edge devices according to claim 1, characterized in that, The terminal's behavioral characteristic data includes data on the terminal's access to edge devices and data on the terminal's switching between edge devices; determining whether the deployment of the edge devices is abnormal based on the signal strength of the edge devices and the terminal's behavioral characteristic data includes: Based on the data accessed by the terminal to the edge device, the overlap of terminals served by different edge devices is determined; The number of times the terminal switches between edge devices is determined based on the data of the terminal switching between edge devices; The deployment of the edge devices is determined to be abnormal based on the signal strength of the edge devices, the overlap of the terminals served by the different edge devices, and the number of times the terminals are switched.

3. The method for identifying abnormal FTTR edge device deployment according to claim 2, characterized in that, The step of determining whether the deployment of the edge devices is abnormal based on the signal strength of the edge devices, the overlap of terminals served by different edge devices, and the number of terminal handovers includes: If the signal strength of the edge device is greater than a first threshold, the overlap of terminals served by different edge devices is greater than a second threshold, and the number of handovers of the terminals is greater than a third threshold, then the edge device deployment is determined to be abnormal.

4. The method for identifying abnormal deployment of FTTR edge devices according to claim 3, characterized in that, In the event that the edge device deployment is abnormal, the method further includes: Determine the historical number of times the edge device has been tuned; Based on the historical number of optimizations, determine the handling method for abnormal edge device deployments.

5. The method for identifying abnormal deployment of FTTR edge devices according to claim 4, characterized in that, The step of determining the handling method for the edge device based on the historical optimization count includes: If the number of historical optimization attempts is less than the fourth threshold, adjust the power of the edge device and update the number of historical optimization attempts. If the number of historical optimizations is greater than or equal to the fourth threshold, cloud devices are triggered to collaboratively handle the deployment anomalies of the edge devices.

6. The method for identifying abnormal FTTR edge device deployment according to claim 4, characterized in that, The triggering of cloud devices to collaboratively handle the deployment anomalies of the edge devices includes: Send a first message to the cloud device; the first message is used to instruct the cloud device to determine whether the edge device is deployed abnormally based on the abnormal placement analysis model; the abnormal placement analysis model is trained based on sample data; the sample data includes the signal strength of the edge device, the behavioral characteristic data of the terminal, and the data of edge device deployment abnormal / normal confirmed by the operation and maintenance personnel on site.

7. A device for identifying abnormal deployment of FTTR edge devices, characterized in that, include: The acquisition module is used to acquire the signal strength of the edge device and the behavioral characteristic data of the terminal served by the edge device; The determination module is used to determine whether the deployment of the edge device is abnormal based on the signal strength of the edge device and the behavioral characteristic data of the terminal.

8. An edge device comprising a memory, a processor, and a computer program stored in the memory and executable on the processor, characterized in that, When the processor executes the computer program, it implements the method for identifying FTTR edge device deployment anomalies as described in any one of claims 1 to 6.

9. A non-transitory 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 method for identifying FTTR edge device deployment anomalies as described in any one of claims 1 to 6.

10. A computer program product, comprising a computer program, characterized in that, When the computer program is executed by the processor, it implements the method for identifying FTTR edge device deployment anomalies as described in any one of claims 1 to 6.

Images