Repeater monitoring method and system and storage medium

Abstract

The invention relates to the field of communication equipment management, in particular to a repeater monitoring method and system and a storage medium, and the scheme comprises the steps: obtaining operation data of a repeater subsystem during operation; performing data processing on the operation data to obtain processed target operation data, and inputting the target operation data into a target decision tree forest for processing to obtain an abnormal value and an abnormal response decision of the current repeater subsystem; and transmitting the abnormal value and the abnormal response decision to the embedded operation subsystem, so that the embedded operation subsystem performs alarm operation according to the abnormal value. According to the invention, real-time monitoring and intelligent management of the repeater subsystem are realized, abnormal conditions can be rapidly identified and transmitted to the embedded operation subsystem, and then alarm operation is triggered, so that the reliability and the safety of the method are improved, the maintenance cost and the downtime are reduced, and the working efficiency is improved. And the continuous and stable operation of the repeater subsystem is ensured.

Description

Technical Field

[0001] The embodiments of the present application relate to the field of communication equipment management, and in particular to a repeater monitoring method, system and storage medium. Background Art

[0002] With the continuous expansion and complexity of mobile communication networks, repeaters, as key equipment for enhancing network coverage and signal quality, are widely used in weak signal areas such as urban blind spots, indoor environments, tunnels, and rural areas. However, the widespread deployment of repeaters also brings new challenges. Since repeaters are responsible for relaying and amplifying base station signals, if they are not effectively monitored and managed, they may cause signal interference, uneven coverage, improper power regulation, and other problems, affecting the overall network performance and user experience. And with the advancement of modern communication, computing, network and microelectronics technology, embedded Linux systems have gradually met the development needs of a new generation of communication equipment. The Linux kernel has been widely used in various communication products with its rich network protocols and TCP / IP protocol stack as well as DNS and HTTP.

[0003] With the rapid development of computer technology and the popularization of the Internet, Web technology has begun to be widely used. Embedded Web servers are the result of the combination of embedded technology and network technology, and have proposed a new method for remote monitoring systems. At present, in traditional system design, if embedded Web technology is used to monitor the micro repeater equipment system, two motherboards are generally designed, and two MCUs are installed on the two motherboards to complete the repeater system and the monitoring system respectively. In this way, the complexity and power consumption of BOM will inevitably increase, especially in mass production, the cost will also increase greatly. Summary of the invention

[0004] One purpose of the embodiments of the present application is to provide a repeater monitoring method, system and storage medium to solve the technical problem that abnormalities cannot be discovered in time when monitoring repeaters and the cost is high.

[0005] In a first aspect, a repeater monitoring method is provided, which is applied to a repeater subsystem in a monitoring system, wherein the monitoring system further includes an embedded operation subsystem, and the method includes:

[0006] Acquiring operation data of the repeater subsystem during operation;

[0007] Performing data processing on the operation data to obtain processed target operation data;

[0008] Input the target operation data into the target decision tree forest for processing to obtain abnormal values ​​and abnormal response decisions of the current repeater subsystem;

[0009] The abnormal value and the abnormal response decision are transmitted to the embedded operating subsystem, so that the embedded operating subsystem performs an alarm operation according to the abnormal value.

[0010] In a second aspect, a monitoring system includes a repeater subsystem and an embedded operation subsystem, wherein the repeater subsystem is used to execute the method described in the first aspect.

[0011] In a third aspect, a repeater monitoring device is provided, which is applied to a repeater subsystem in a monitoring system, wherein the monitoring system further includes an embedded operation subsystem, and the device includes:

[0012] A sending unit, used for acquiring the operation data of the repeater subsystem when it is running;

[0013] A processing unit, used to process the operation data to obtain processed target operation data;

[0014] The processing unit is further used to input the target operation data into the target decision tree forest for processing, and obtain the abnormal value and abnormal response decision of the current repeater subsystem;

[0015] A transmission unit is used to transmit the abnormal value and the abnormal response decision to the embedded operating subsystem, so that the embedded operating subsystem performs an alarm operation according to the abnormal value.

[0016] In a fourth aspect, an embodiment of the present invention provides a computer device, including:

[0017] at least one processor; and,

[0018] a memory communicatively connected to the at least one processor; wherein,

[0019] The memory stores instructions that can be executed by the at least one processor, and the instructions are executed by the at least one processor to enable the at least one processor to perform the method according to the first aspect.

[0020] In a fifth aspect, a computer-readable storage medium is provided, wherein the computer-readable storage medium stores a computer program, wherein the computer program includes program instructions, and when the program instructions are executed by a processor, the processor executes the method as described in the first aspect.

[0021] In the scheme implemented by the above-mentioned repeater monitoring method, system, device, equipment and storage medium, first, the operating data of the repeater subsystem during operation is obtained, and secondly, the operating data is processed to obtain the processed target operating data, and then the target operating data is input into the target decision tree forest for processing to obtain the abnormal value and abnormal response decision of the current repeater subsystem, and finally, the abnormal value and the abnormal response decision are transmitted to the embedded operation subsystem, so that the embedded operation subsystem performs an alarm operation according to the abnormal value. This method obtains the operation data of the repeater subsystem in real time, and then fully understands the current state and performance of the system, providing basic data support for subsequent data processing and anomaly detection; then, by processing the original operation data, it can filter out noise and irrelevant information, extract valuable target operation data, improve the accuracy and reliability of the data, and provide a more accurate data basis for subsequent decision analysis; then, by using the target decision tree forest to analyze the target operation data, it can efficiently identify the abnormal situation in the system operation, and generate corresponding abnormal response decisions, improve the accuracy and response speed of anomaly detection; finally, by transmitting the detected abnormal values ​​and response decisions to the embedded operation subsystem, real-time alarm operations can be realized, and relevant personnel can be reminded to handle them in time, reducing the impact of the repeater subsystem failure on the overall operation, and improving the stability and safety of the repeater subsystem. Therefore, intelligent monitoring and abnormal management of the repeater subsystem are realized, and the operation efficiency and reliability are improved. BRIEF DESCRIPTION OF THE DRAWINGS

[0022] In order to more clearly illustrate the technical solutions of the embodiments of the present application, the drawings required for use in the description of the embodiments of the present application will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present application. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying any creative labor.

[0023] Figure 1A is a schematic diagram of a monitoring system in one embodiment of the present invention;

[0024] Figure 1B is a schematic diagram of the internal interaction flow of the embedded operating subsystem in one embodiment of the present invention;

[0025] Figure 2 It is a flow chart of a repeater monitoring method in one embodiment of the present invention;

[0026] Figure 3 A schematic structural diagram of a repeater monitoring device in one embodiment of the present invention;

[0027] Figure 4 It is a schematic diagram of the structure of a computer device in one embodiment of the present invention. DETAILED DESCRIPTION

[0028] In order to make the purpose, technical solutions and advantages of the present application more clearly understood, the present application is further described in detail below in conjunction with the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are only used to explain the present application and are not intended to limit the present application. Based on the embodiments in the present application, all other embodiments obtained by ordinary technicians in the field without making creative work are within the scope of protection of the present application.

[0029] It should be noted that, if there is no conflict, the various features in the embodiments of the present application can be combined with each other, all within the scope of protection of the present application. In addition, although the functional module division is performed in the device schematic diagram and the logical order is shown in the flow chart, in some cases, the steps shown or described can be performed in a sequence different from the module division in the device or the flow chart. Furthermore, the words "first", "second", "third", etc. used in this application do not limit the data and execution order, but only distinguish the same items or similar items with basically the same functions and effects.

[0030] The present invention is described in detail below through specific embodiments.

[0031] See also Figure 1A , Figure 1A The monitoring system 10 of the present invention can be as follows: Figure 1A As shown, the monitoring system 10 includes a repeater subsystem 20 and an embedded operating subsystem 30 .

[0032] Among them, the above monitoring system uses the asynchronous processing dual-core ARM of the Zynq Ultralscale+MPSoC architecture as the core controller of the entire system. The dual cores inside the ARM are later replaced by ARM0 and ARM1. The Linux operating subsystem is transplanted on ARM0 to run network services, and a suitable web server is transplanted on the Linux operating subsystem. In addition, ARM1 is used to complete the implementation of the functional items of the entire repeater subsystem, and key data indicators such as gain, power, attenuation, RF switch, sub-band frequency band, etc. are transmitted to ARM0 in real time for real-time data interaction. That is, in this solution, the repeater subsystem is processed according to ARM1, and the embedded operating subsystem is processed according to ARM0.

[0033] Specifically, the Zynq Ultralscale+MPSoC architecture: This is a highly integrated multi-processor system-on-chip (SoC). This architecture is suitable for applications that require high-performance processing and flexible hardware configuration.

[0034] Specifically, asynchronous processing dual-core ARM: In the Zynq Ultralscale+ MPSoC, there are two ARM cores that can run independently and perform different tasks. This design can improve the processing power and efficiency of the system.

[0035] Specifically, the repeater subsystem (based on ARM1): responsible for the actual operation of the repeater, including signal amplification, frequency adjustment, power control, etc. Monitor the operating status of the repeater in real time and collect key performance indicators. The collected data is transmitted to ARM0 for further processing and monitoring.

[0036] Specifically, the embedded operating subsystem (based on ARM0): runs the Linux operating system and provides a network service interface; through the transplanted Web server, it can remotely access monitoring data and issue control commands; receive data from ARM1 and perform real-time data interaction, which may include data storage, analysis and user interface updates.

[0037] For reference Figure 1B , Figure 1B This is a schematic diagram of the internal interaction process of the embedded operating subsystem. The process described in the figure is about the interaction between the client and the server, especially the configuration and operation of the BOA server and the PHP interpreter.

[0038] Specifically: the client sends a request and connects to the server; the BOA server receives the client's request, and uses the configuration file to set the MIME type and the project file root directory, etc.; the BOA server communicates with the PHP interpreter through the CGI interface. The PHP interpreter compiles and processes secure sockets, databases, CGI interfaces, and web page preloading strategies. The PHP interpreter parses the static PHP file requested by the client (the location of the PHP interpreter is specified through the shebang line). The Config.txt configuration file contains the device type, model, system parameters, subband information, and descriptions of the display control switches of various functions.

[0039] The entire process describes the process of client requests being received and processed by the BOA server, the communication between the BOA server and the PHP interpreter, and how to use various configuration files to configure and control the behavior of the server and interpreter.

[0040] In view of this, the present application proposes a repeater monitoring method to solve the above problems, which is described in detail below.

[0041] See also Figure 2 , Figure 2A flow chart of a repeater monitoring method provided by an embodiment of the present invention is applied to a repeater subsystem in a monitoring system, wherein the monitoring system further includes an embedded operating subsystem, and the method includes the following steps:

[0042] S10, obtaining the operation data of the repeater subsystem during operation.

[0043] The operation data refers to the information collected and recorded on the performance indicators and operation status of the repeater under normal working conditions. The operation data includes but is not limited to signal strength, output power, receiving sensitivity, temperature, voltage, current, frequency offset, bit error rate, equipment status and other parameters.

[0044] Specifically, signal strength is the received signal strength indicator (such as RSSI, RSRP, etc.); signal quality is the signal quality indicator (such as SINR, SNR, etc.); device status is the operating status information of the device (such as temperature, voltage, current, etc.); error log is the error and warning log generated during the operation of the device; traffic statistics are traffic statistics information such as data transmission rate and number of user connections.

[0045] The acquisition process is real-time, that is, data is collected continuously to detect any signs of performance degradation or potential failure in a timely manner. Further operating data can be obtained through sensors, analog-to-digital converters (ADCs), digital signal processors (DSPs) or other monitoring devices, which are placed at the "edge" where the data is generated, that is, close to the data source.

[0046] During the acquisition process, different frequencies can be set according to monitoring needs, ranging from several times to thousands of times per second, to ensure the timeliness and accuracy of the data.

[0047] Optionally, the acquired data needs to be stored in memory, database or log files for subsequent analysis and processing. The stored data can be used for historical trend analysis, troubleshooting and performance optimization.

[0048] It can be seen that in this embodiment, accurate data support is provided for monitoring by acquiring the operating data of the repeater subsystem during operation.

[0049] S20: Process the operation data to obtain processed target operation data.

[0050] Among them, the data processing refers to the process of cleaning, converting, normalizing, aggregating and other operations on the operating data to eliminate noise, correct errors and extract the target operating data.

[0051] Among them, target operating data: processed data, which are usually more accurate, more consistent and more suitable for subsequent analysis (such as statistical analysis, machine learning model training, etc.).

[0052] In a possible example, the data processing of the operating data to obtain processed target operating data includes: acquiring the data type of each data in the operating data; calculating the data type of each data respectively to obtain the first operating data corresponding to the data type of each data; and aggregating all the first operating data to obtain the target operating data.

[0053] It is necessary to identify and classify the data type of each operation data in the operation data, for example, signal strength data, temperature data, flow data, etc. Optionally, each data type is marked so that subsequent processing steps can perform corresponding processing on different types of data.

[0054] Specifically, in the process of calculating the data type of each data and obtaining the first operation data corresponding to the data type of each data, data cleaning can be performed first to remove noise and abnormal values ​​in the operation data to ensure the accuracy and reliability of the data; then data conversion can be performed to convert different types of data into a unified format or unit. For example, temperature data can be uniformly converted into degrees Celsius, and signal strength data can be converted into dBm, etc.; then data calculation can be performed: corresponding calculation processing can be performed for different data types. For example, the average value, maximum value, and minimum value of the signal strength can be calculated; the temperature change trend can be calculated; the total amount and peak value of the flow can be calculated, etc.

[0055] Among them, the calculation results of different types of data are summarized to form a comprehensive target operation data.

[0056] It can be seen that in this embodiment, by processing the operation data of the repeater subsystem, the processed target operation data can be obtained. The target operation data is more accurate and reliable, and provides valuable information support for subsequent decision analysis.

[0057] S30, inputting the target operation data into the target decision tree forest for processing, and obtaining abnormal values ​​and abnormal response decisions of the current repeater subsystem.

[0058] Among them, the target decision tree forest (Random Forest) is an integrated learning algorithm that improves prediction accuracy and stability by building multiple decision trees and combining their output results. Inputting the target operation data into the decision tree forest model can automatically analyze and identify outliers in the data and generate corresponding abnormal response decisions. Specifically, in this solution, the target decision tree can be composed of the repeater operation decision tree and the environment operation decision tree.

[0059] Outliers are data points that are significantly different from other data and may indicate some kind of error or anomaly.

[0060] Among them, abnormal response decision refers to the action taken according to preset rules or strategies when abnormal values ​​are detected, such as issuing alarms, recording events, and automatically adjusting system settings.

[0061] Specifically, the target operating data is input into the target decision tree forest, where each decision tree performs classification or regression analysis on the target operating data to identify patterns and relationships in the data; the target decision tree forest identifies outliers by comparing the characteristics of the data points with the normal behavior patterns learned in the model. If the characteristics of the data point are significantly different from the normal pattern, it may be marked as an anomaly; once an outlier is detected, the target decision tree forest determines how to respond based on preset rules or algorithms, which may include threshold setting, probability calculation, and conditional logic; each decision tree independently evaluates the target operating data and gives its own abnormal judgment, and the decision tree forest aggregates the results of all decision trees, and determines the final outlier and response decision by majority voting or average prediction value. If multiple decision trees indicate an anomaly, the system will take corresponding abnormal response decisions.

[0062] For example, the target decision forest model identifies that the average signal strength of -67.67dBm is low, which may indicate a signal coverage problem; the model identifies that the average temperature of 26°C is within the normal range and there is no abnormality; the model identifies that the total traffic volume of 450MB is within the normal range and there is no abnormality. At this time, for the identified signal strength abnormality, the model generates a response decision, such as: sending an alarm to notify the operator; recommending to check and adjust the antenna direction or position; suggesting that the number of repeaters in the coverage area may need to be increased.

[0063] It can be seen that in this embodiment, abnormal conditions of the repeater subsystem can be automatically identified and actions can be taken quickly, thereby improving the autonomy and reliability of the system.

[0064] S40: Transmit the abnormal value and the abnormal response decision to the embedded operating subsystem, so that the embedded operating subsystem performs an alarm operation according to the abnormal value.

[0065] The embedded operation subsystem is an operating system integrated into the hardware device, which is responsible for real-time monitoring and management of the device's operating status. Transmitting abnormal values ​​and abnormal response decisions to the embedded operation subsystem allows the system to identify abnormal situations in a timely manner and perform corresponding alarm operations to ensure the safe and stable operation of the device.

[0066] Among them, transmission refers to sending the abnormal value and abnormal response decision to the embedded operation subsystem through the communication interface.

[0067] Optionally, before transmission, abnormal values ​​and abnormal response decisions can be packaged into data packets, and the data packets are sent via serial communication, network communication or other suitable communication interfaces; the embedded operating subsystem receives the data packets and parses the abnormal values ​​and response decisions therein; the embedded system performs corresponding alarm operations and / or automatic repair measures based on the parsed information, and the system records abnormal events and measures taken to facilitate subsequent analysis and auditing.

[0068] Specifically, when the target decision tree forest predicts the current anomaly in the repeater system, the embedded operation subsystem (Linux operating system) will perform the following subsequent steps: the embedded operation subsystem will immediately trigger an abnormal alarm, and send the alarm information to the computer web page display terminal through the Web server to remind the operation and maintenance personnel to pay attention; the abnormal situation and related data will be recorded in the log file for subsequent analysis and diagnosis, and these log files can be parsed and displayed by the PHP interpreter; if the system is configured with an automatic processing strategy, the embedded operation subsystem will automatically process according to the preset processing plan. For example, restart the relevant modules, adjust the parameters, or switch to the backup system, etc.; send the abnormal information to the remote operation and maintenance center through the network, and the operation and maintenance personnel can view the abnormal details through the remote monitoring system, and perform remote diagnosis and processing; after the abnormal processing is completed, the embedded operation subsystem will perform a system self-check to ensure that the system resumes normal operation, and record the self-check results in the log; feedback the processing results and current system status to the user through the web interface to ensure that the user understands the progress of the abnormal processing. Therefore, the embedded operation subsystem can effectively manage and handle abnormal situations in the repeater system, ensuring the stable operation and rapid recovery of the system.

[0069] It can be seen that this embodiment can quickly respond to potential problems, reduce the need for human intervention, improve the system's autonomous maintenance capability, and ensure that the repeater subsystem can be processed in a timely manner when an abnormality occurs, thereby ensuring the stability and reliability of the entire communication network.

[0070] This method obtains the operation data of the repeater subsystem in real time, and then fully understands the current state and performance of the system, providing basic data support for subsequent data processing and anomaly detection; then, by processing the original operation data, it can filter out noise and irrelevant information, extract valuable target operation data, improve the accuracy and reliability of the data, and provide a more accurate data basis for subsequent decision analysis; then, by using the target decision tree forest to analyze the target operation data, it can efficiently identify the abnormal situation in the system operation, and generate corresponding abnormal response decisions, improve the accuracy and response speed of anomaly detection; finally, by transmitting the detected abnormal values ​​and response decisions to the embedded operation subsystem, real-time alarm operations can be realized, and relevant personnel can be reminded to handle them in time, reducing the impact of the repeater subsystem failure on the overall operation, and improving the stability and safety of the repeater subsystem. Therefore, intelligent monitoring and abnormal management of the repeater subsystem are realized, and the operation efficiency and reliability are improved.

[0071] In a possible example, before inputting the target operation data into the target decision tree forest for processing to obtain the abnormal value and abnormal response decision of the current repeater subsystem, the method also includes: collecting historical operation data and environmental operation data corresponding to the repeater subsystem; performing data processing on the historical operation data and the environmental operation data to obtain a first operation decision tree corresponding to the historical operation data and a second operation decision tree corresponding to the environmental operation data; and obtaining a target decision tree forest based on the first operation decision tree and the second operation decision tree.

[0072] Among them, the historical operation data includes normal data and abnormal data. The historical operation data refers to the operation records of the repeater subsystem in the past period of time, including performance indicators such as signal strength, temperature, and flow.

[0073] Among them, environmental operation data includes normal data and abnormal data. Environmental operation data refers to data related to the operating environment of the repeater, such as weather, geographical location, surrounding buildings and other factors that affect signal propagation.

[0074] The first operation decision tree is a decision tree model that reflects the operation status of the repeater subsystem based on historical operation data.

[0075] The second operation decision tree is based on the environmental operation data to construct a decision tree model reflecting the operation environment of the repeater subsystem.

[0076] Among them, before training the decision tree, it is necessary to process the historical operation data and environmental operation data. The data processing includes but is not limited to cleaning, standardization and feature extraction, etc., to improve the prediction ability of the model.

[0077] Optionally, the processed data is used to train a decision tree using a machine learning algorithm such as CART (Classification and Regression Trees) or ID3 (Iterative Binary Tree 3).

[0078] In the process of obtaining the target decision tree forest based on the first operating decision tree and the second operating decision tree, the first operating decision tree and the second operating decision tree are combined to form a comprehensive decision tree forest model for more accurately analyzing and predicting the operation of the repeater subsystem.

[0079] It can be seen that the target decision tree forest obtained by the two types of data in this embodiment can provide a more comprehensive and accurate model for the monitoring and prediction of the repeater subsystem, thereby improving the fault detection and prevention capabilities of the system.

[0080] In a possible example, the data processing of the historical operation data and the environmental operation data to obtain a first operation decision tree corresponding to the historical operation data and a second operation decision tree corresponding to the environmental operation data includes: performing feature processing on the historical operation data and the environmental operation data to obtain historical feature data corresponding to the historical operation data and environmental feature data corresponding to the environmental operation data; performing importance sorting processing on the historical feature data and the environmental feature data respectively to obtain a first sequence corresponding to the historical feature data and a second sequence corresponding to the environmental feature data; performing interaction effect processing according to the first sequence and the second sequence to obtain historical feature interaction effect data and environmental feature interaction effect data; and obtaining a first operation decision tree corresponding to the historical operation data and a second operation decision tree corresponding to the environmental operation data according to the historical feature interaction effect data and the environmental feature interaction effect data.

[0081] Feature processing may include but is not limited to feature extraction, feature selection, and feature conversion. The purpose of feature processing is to extract the most useful information for prediction tasks from the original data. Specifically, feature extraction is to use statistical methods, data mining techniques, or domain knowledge to identify and extract key features in the data; feature selection is to sort the features according to their importance and select the most useful features for model prediction, which can reduce the complexity of the model and the risk of overfitting.

[0082] Among them, importance ranking can determine the importance of features through feature importance scoring and model-based feature selection (such as feature importance of random forest).

[0083] Among them, interaction effect analysis is to examine the interaction between features, which can be achieved through statistical tests (such as analysis of variance) or machine learning methods (such as feature combination).

[0084] In the process of sorting the importance of historical feature data, algorithms (such as random forests, gradient boosting trees, etc.) are used to evaluate the importance of historical feature data to determine which features have the greatest impact on system operation. For example, signal strength and temperature may have a greater impact on system operation, while traffic has a relatively smaller impact. The first sequence generated may be: signal strength>temperature>traffic.

[0085] In the process of ranking the importance of environmental feature data, the same or similar algorithm is used to evaluate the importance of environmental feature data to determine which environmental features have the greatest impact on system operation. For example, weather and surrounding buildings may have a greater impact on system operation, while geographical location has a relatively smaller impact. The second sequence generated may be: weather>surrounding buildings>geographic location.

[0086] In the process of performing interaction effect processing according to the first sequence and the second sequence to obtain historical feature interaction effect data and environmental feature interaction effect data, the interaction effect between historical features and environmental features is analyzed according to the first sequence and the second sequence. For example, the relationship between signal strength and temperature under different weather conditions, or the relationship between traffic under different surrounding building densities, is analyzed. The interaction effect data reflects the mutual influence between features, forming historical feature interaction effect data and environmental feature interaction effect data. For example, the relationship between signal strength and temperature on a sunny day may be different from that on a rainy day.

[0087] It can be seen that more accurate and comprehensive decision tree models are constructed in this embodiment, which can better capture the comprehensive impact of historical and environmental factors on the performance of the repeater subsystem, thereby improving the prediction and diagnosis capabilities of the repeater subsystem.

[0088] In a possible example, the method of obtaining a first operation decision tree corresponding to the historical operation data and a second operation decision tree corresponding to the environmental operation data according to the historical feature interaction effect data and the environmental feature interaction effect data includes: performing node screening on the historical feature interaction effect data and the environmental feature interaction effect data respectively to obtain historical feature initial node data and environmental feature initial node data; performing splitting processing on the historical feature initial node data and the environmental feature initial node data respectively to obtain a historical feature classification candidate set and an environmental feature classification candidate set; performing tree depth calculation according to the historical feature classification candidate set to obtain historical feature initial tree depth data; performing tree depth calculation according to the environmental feature classification candidate set to obtain environmental feature initial tree depth data; performing recursive initialization processing according to the historical feature initial node data, the historical feature classification candidate set and the historical feature initial tree depth data to obtain the first operation decision tree corresponding to the historical operation data; performing recursive initialization processing according to the environmental feature initial node data, the environmental feature classification candidate set and the environmental feature initial tree depth data to obtain the second operation decision tree corresponding to the environmental operation data.

[0089] Among them, in the process of decision tree construction, node screening refers to selecting which features to use as tree nodes, which divide the data set into smaller subsets for classification or regression.

[0090] Among them, the historical feature initial node data and the environmental feature initial node data are the feature data selected as the starting node of the decision tree after screening.

[0091] Specifically, the historical feature initial node data is the initial node selected from the historical feature interaction effect data. For example, in the interaction effect data of signal strength, temperature and flow, signal strength and temperature are selected as the initial nodes.

[0092] Specifically, the initial node data of environmental features is the initial nodes selected from the interactive effect data of environmental features. For example, in the interactive effect data of weather and surrounding buildings, weather and surrounding buildings are selected as initial nodes.

[0093] Among them, in the decision tree, splitting refers to dividing the current node into two or more child nodes according to the value of a certain feature. This process is called splitting processing.

[0094] Among them, the classification candidate set is a set of possible splitting points in the splitting process, and each splitting point is an attempt to divide the data according to different values ​​of the feature.

[0095] Specifically, the historical feature initial node data is split, and each initial node is split into multiple sub-nodes according to a specific splitting standard (such as information gain, Gini index, etc.). For example, a signal strength node can be split into multiple sub-nodes according to different signal strength thresholds.

[0096] Specifically, the initial node data of the environmental feature is split to generate a candidate set of environmental feature classifications. For example, a weather node can be split into multiple child nodes according to different weather conditions (sunny, rainy, etc.).

[0097] The depth of a decision tree is the number of nodes on the longest path from the root node to the farthest leaf node. The depth of a decision tree is calculated to determine the complexity of the decision tree.

[0098] Specifically, the historical feature initial tree depth data is the initial tree depth of the decision tree calculated based on the historical feature classification candidate set. The tree depth represents the longest path length from the root node to the leaf node. For example, if the historical feature classification candidate set contains 3 layers of splits, the initial tree depth is 3.

[0099] Specifically, the environmental feature initial tree depth data is the initial tree depth of the decision tree calculated based on the environmental feature classification candidate set. For example, if the environmental feature classification candidate set contains 4 layers of splits, the initial tree depth is 4. The recursive initialization process is to recursively apply the above steps to create each node and child node of the tree when constructing the decision tree until the stopping condition (such as the maximum depth of the tree, the minimum number of samples of node data, etc.) is met.

[0100] Among them, in the process of recursively initializing the historical feature initial node data, the historical feature classification candidate set and the historical feature initial tree depth data to obtain the first running decision tree corresponding to the historical running data, the historical feature initial node data, the historical feature classification candidate set and the historical feature initial tree depth data are used to recursively construct a decision tree. In the recursive process, the best splitting point is continuously selected to split the node into child nodes until the stopping condition is met (such as the tree depth reaches the initial tree depth data or the node purity reaches a certain threshold). For example, starting from the signal strength node, it is split into child nodes according to different thresholds of signal strength, and then each child node is further split until a complete decision tree is constructed.

[0101] Among them, recursive initialization processing is performed according to the environmental feature initial node data, the environmental feature classification candidate set and the environmental feature initial tree depth data to obtain the second operation decision tree corresponding to the environmental operation data, and the environmental feature initial node data, environmental feature classification candidate set and environmental feature initial tree depth data are used to recursively construct a decision tree. The recursive process is similar to the construction of the first operation decision tree, and finally a decision tree reflecting the environmental operation status is obtained. For example, starting from the weather node, it is split into child nodes according to different weather conditions, and then each child node is further split until a complete decision tree is constructed.

[0102] In the recursive process, each node will be split according to its classification candidate set until the predetermined stopping condition is reached.

[0103] It can be seen that in this embodiment, two decision trees can be obtained through the above process, one for the historical operation data and the other for the environmental operation data. The two decision trees can be used for classification or regression tasks to help the monitoring system predict the future state of the repeater subsystem or diagnose potential problems.

[0104] In a possible example, obtaining a target decision tree forest based on the first running decision tree and the second running decision tree includes: training the first running decision tree and the second running decision tree respectively to obtain a first prediction value of the first running decision tree and a second prediction value of the second running decision tree; and integrating the first prediction value and the second prediction value to obtain the target decision tree forest.

[0105] In the process of training the first running decision tree and obtaining the first predicted value of the first running decision tree, the first running decision tree constructed with historical running data is trained, and the decision tree model is adjusted and optimized using the training data set. During the training process, the decision tree will learn how to classify or regress according to the input features (such as signal strength, temperature, etc.) to obtain the corresponding predicted value, and finally obtain the first predicted value of the first running decision tree.

[0106] In the process of training the second running decision tree and obtaining the second predicted value of the second running decision tree, the second running decision tree constructed by the environmental running data is trained, and the decision tree model is adjusted and optimized using the training data set. During the training process, the decision tree will learn how to classify or regress according to the input features (such as weather, surrounding buildings, etc.) to obtain the corresponding predicted value, and finally, the second predicted value of the second running decision tree is obtained.

[0107] Among them, ensemble processing is to combine the prediction results of multiple running decision trees to improve the overall prediction performance and robustness. The ensemble method can be simple voting, averaging or more complex algorithms such as stacking.

[0108] The first prediction value is obtained by applying the trained first-run decision tree to new historical run data.

[0109] The second prediction value is obtained by applying the trained second operation decision tree to new environment operation data.

[0110] Specifically, the integration method can be weighted average, that is, different weights are assigned according to the importance of the two predicted values, and then the weighted average is calculated as the final prediction result; the integration method can be a voting mechanism, that is, for classification tasks, a voting mechanism can be used to vote on the prediction results of the two decision trees, and the category with the most votes is selected as the final prediction result; the integration method can be stacking, that is, the first prediction value and the second prediction value are used as new features and input into a higher-level model (such as linear regression, neural network, etc.) for further training and prediction. Therefore, through the integration method, the advantages of the two decision tree models can be comprehensively utilized to improve the overall prediction performance.

[0111] It can be seen that in this embodiment, the first operating decision tree and the second operating decision tree are integrated to obtain the final target decision tree forest. The target decision tree forest can integrate the information of multiple decision trees, thereby providing more accurate and stable results when monitoring and predicting the operating status of the repeater subsystem.

[0112] In one possible example, the integrating processing of the first prediction value and the second prediction value to obtain the target decision tree forest includes: taking a weighted average of the first prediction value and the second prediction value to obtain a first weight of the first running decision tree and a second weight of the second running decision tree; and obtaining the target decision tree forest according to the first weight of the first running decision tree and the second weight of the second running decision tree.

[0113] Among them, weighted average is a method of combining multiple predictions, where each prediction is given a weight based on its importance or reliability. The weight is a number between 0 and 1 that indicates the relative importance of each prediction in the final result.

[0114] Among them, the first weight corresponds to the contribution degree of the first running decision tree in the integrated model; the second weight corresponds to the contribution degree of the second running decision tree in the integrated model.

[0115] Among them, weights can be determined by various methods, such as cross-validation, model performance evaluation (such as accuracy, mean square error, etc.) or using specific algorithms (such as gradient boosting). The size of the weight reflects the influence of the corresponding decision tree in the integrated model. Models with better performance are usually given higher weights.

[0116] The target decision tree forest is the final model obtained by integrating multiple decision trees by weighted average. In this forest, the prediction of each tree contributes to the final prediction result according to its weight.

[0117] Among them, for classification problems, weighted averaging may involve weighting the predicted probability of each category and then selecting the category with the highest weighted average probability as the final prediction; for regression problems, weighted averaging directly weights the predicted values ​​of each decision tree and then sums them up to get the final prediction result.

[0118] Specifically, once the weights of each decision tree are determined, they can be combined into an integrated model, namely the target decision tree forest; when making predictions, the prediction of each tree is multiplied by its corresponding weight, and then all weighted prediction values ​​are added together to form the final prediction result.

[0119] It can be seen that the target decision tree forest in this embodiment can more flexibly and effectively utilize the information provided by different decision trees, thereby achieving improvements in prediction accuracy and model stability. It is therefore suitable for when different decision tree models perform well in different situations. Through weighted averaging, the advantages of each model can be combined to obtain a more reliable prediction result.

[0120] It should be noted that, in each of the above-mentioned embodiments, there is not necessarily a certain order between the above-mentioned steps. A person skilled in the art can understand, based on the description of the embodiments of the present application, that in different embodiments, the above-mentioned steps may have different execution orders, that is, they may be executed in parallel, may be executed interchangeably, and so on.

[0121] As another aspect of the embodiment of the present application, the embodiment of the present application provides a repeater monitoring device, wherein the repeater monitoring device can be a software module, the software module includes a plurality of instructions, which are stored in a memory, and the processor can access the memory and call the instructions for execution to complete the repeater monitoring method described in the above-mentioned various embodiments.

[0122] See also Figure 3 , Figure 3 Schematic diagram of the structure of a repeater monitoring device provided in an embodiment of the present application. Figure 3 As shown, a repeater subsystem is applied to a monitoring system, and the monitoring system further includes an embedded operation subsystem, characterized in that the repeater monitoring device includes:

[0123] An acquisition unit 301 is used to acquire operation data of the repeater subsystem when it is running;

[0124] The processing unit 302 is used to process the operation data to obtain processed target operation data;

[0125] The processing unit 302 is further used to input the target operation data into the target decision tree forest for processing, and obtain the abnormal value and abnormal response decision of the current repeater subsystem;

[0126] The transmission unit 303 is used to transmit the abnormal value and the abnormal response decision to the embedded operating subsystem, so that the embedded operating subsystem performs an alarm operation according to the abnormal value.

[0127] This method obtains the operation data of the repeater subsystem in real time, and then fully understands the current state and performance of the system, providing basic data support for subsequent data processing and anomaly detection; then, by processing the original operation data, it can filter out noise and irrelevant information, extract valuable target operation data, improve the accuracy and reliability of the data, and provide a more accurate data basis for subsequent decision analysis; then, by using the target decision tree forest to analyze the target operation data, it can efficiently identify the abnormal situation in the system operation, and generate corresponding abnormal response decisions, improve the accuracy and response speed of anomaly detection; finally, by transmitting the detected abnormal values ​​and response decisions to the embedded operation subsystem, real-time alarm operations can be realized, and relevant personnel can be reminded to handle them in time, reducing the impact of the repeater subsystem failure on the overall operation, and improving the stability and safety of the repeater subsystem. Therefore, intelligent monitoring and abnormal management of the repeater subsystem are realized, and the operation efficiency and reliability are improved.

[0128] It should be noted that the above repeater monitoring device can execute the repeater monitoring method provided in the embodiment of the present application, and has the corresponding functional modules and beneficial effects of the execution method. For technical details not fully described in the repeater monitoring device embodiment, please refer to the repeater monitoring method provided in the embodiment of the present application.

[0129] See also Figure 4 , Figure 4 1 is a schematic diagram of a computer device provided in an embodiment of the present application. The computer device includes one or more processors and a memory. The memory is connected to the one or more processors, for example, connected to the processor via a bus.

[0130] The processor is configured to support the computer device to perform the corresponding functions in the method in the above method embodiment. The processor can be a central processing unit (CPU), a network processor (NP), a hardware chip or any combination thereof. The above hardware chip can be an application specific integrated circuit (ASIC), a programmable logic device (PLD) or a combination thereof. The above PLD can be a complex programmable logic device (CPLD), a field-programmable gate array (FPGA), a generic array logic (GAL) or any combination thereof.

[0131] The memory is used to store program codes, etc. The memory may include volatile memory (VM), such as random access memory (RAM); the memory may also include non-volatile memory (NVM), such as read-only memory (ROM), flash memory, hard disk drive (HDD) or solid-state drive (SSD); the memory may also include a combination of the above types of memory.

[0132] The memory can be used to store non-volatile software programs, non-volatile computer executable programs and modules, such as program instructions / modules corresponding to the repeater monitoring method in the embodiment of the present application. The processor executes various functional applications and data processing of the repeater monitoring method and the repeater monitoring device by running the non-volatile software programs, instructions and modules stored in the memory, that is, realizes the functions of the various modules or units of the repeater monitoring method and the repeater monitoring device provided in the above method embodiment.

[0133] The memory may include a program storage area and a data storage area, wherein the program storage area may store an operating system and an application required for at least one function. The data storage area may store data created according to the use of the repeater monitoring device, etc. In some embodiments, the memory may optionally include a memory remotely arranged relative to the processor, and these remote memories may be connected to the repeater monitoring device via a network. Examples of the above-mentioned network include, but are not limited to, the Internet, an intranet, a local area network, a mobile communication network, and combinations thereof.

[0134] The one or more modules are stored in the memory, and when executed by the one or more processors, execute the repeater monitoring method in any of the above method embodiments, for example, execute the method steps described in the above method embodiments, and realize the functions of the modules described in the above device embodiments.

[0135] An embodiment of the present application further provides a computer-readable storage medium, wherein the computer-readable storage medium stores a computer program, wherein the computer program includes program instructions, and when the program instructions are executed by a computer, the computer executes the method described in the above embodiment.

[0136] A person skilled in the art can understand that all or part of the processes in the above-mentioned embodiments can be implemented by instructing the relevant hardware through a computer program, and the program can be stored in a computer-readable storage medium, and when the program is executed, it can include the processes of the embodiments of the above-mentioned methods. The storage medium can be a disk, an optical disk, a read-only memory (ROM) or a random access memory (RAM), etc.

[0137] The above disclosure is only the preferred embodiment of the present application, which certainly cannot be used to limit the scope of rights of the present application. Therefore, equivalent changes made according to the claims of the present application are still within the scope covered by the present application.

Claims

1. A repeater monitoring method, characterized in that: A repeater subsystem applied to a monitoring system, wherein the monitoring system further comprises an embedded operating subsystem, and the method comprises: Acquiring operation data of the repeater subsystem during operation; Performing data processing on the operation data to obtain processed target operation data; Input the target operation data into the target decision tree forest for processing to obtain abnormal values ​​and abnormal response decisions of the current repeater subsystem; The abnormal value and the abnormal response decision are transmitted to the embedded operating subsystem, so that the embedded operating subsystem performs an alarm operation according to the abnormal value.

2. The method according to claim 1, characterized in that The performing data processing on the operation data to obtain processed target operation data includes: Obtaining the data type of each data in the operation data; Calculate the data type of each data respectively to obtain the first operation data corresponding to the data type of each data; All the first operation data are collected to obtain the target operation data.

3. The method according to claim 1, characterized in that Before inputting the target operation data into the target decision tree forest for processing to obtain the abnormal value and abnormal response decision of the current repeater subsystem, the method further includes: Collecting historical operation data and environmental operation data corresponding to the repeater subsystem; Processing the historical operation data and the environmental operation data to obtain a first operation decision tree corresponding to the historical operation data and a second operation decision tree corresponding to the environmental operation data; A target decision tree forest is obtained according to the first running decision tree and the second running decision tree.

4. The method according to claim 3, characterized in that: The performing data processing on the historical operation data and the environmental operation data to obtain a first operation decision tree corresponding to the historical operation data and a second operation decision tree corresponding to the environmental operation data includes: Performing feature processing on the historical operation data and the environmental operation data to obtain historical feature data corresponding to the historical operation data and environmental feature data corresponding to the environmental operation data; Respectively performing importance sorting processing on the historical feature data and the environmental feature data to obtain a first sequence corresponding to the historical feature data and a second sequence corresponding to the environmental feature data; Performing interaction effect processing according to the first sequence and the second sequence to obtain historical feature interaction effect data and environmental feature interaction effect data; According to the historical feature interaction effect data and the environmental feature interaction effect data, a first operation decision tree corresponding to the historical operation data and a second operation decision tree corresponding to the environmental operation data are obtained.

5. The method according to claim 4, characterized in that The obtaining, according to the historical feature interaction effect data and the environmental feature interaction effect data, a first operation decision tree corresponding to the historical operation data and a second operation decision tree corresponding to the environmental operation data comprises: Performing node screening on the historical feature interaction effect data and the environmental feature interaction effect data respectively to obtain historical feature initial node data and environmental feature initial node data; Splitting the historical feature initial node data and the environmental feature initial node data respectively to obtain a historical feature classification candidate set and an environmental feature classification candidate set; Calculate the tree depth according to the candidate set of historical feature classification to obtain initial tree depth data of historical features; Perform the tree depth calculation according to the environmental feature classification candidate set to obtain initial tree depth data of environmental features; Recursively initializing the historical feature initial node data, the historical feature classification candidate set, and the historical feature initial tree depth data to obtain a first operation decision tree corresponding to the historical operation data; Recursive initialization processing is performed according to the environmental feature initial node data, the environmental feature classification candidate set and the environmental feature initial tree depth data to obtain a second operation decision tree corresponding to the environmental operation data.

6. The method according to claim 3, characterized in that The step of obtaining a target decision tree forest according to the first running decision tree and the second running decision tree comprises: Training the first running decision tree and the second running decision tree respectively to obtain a first prediction value of the first running decision tree and a second prediction value of the second running decision tree; The first prediction value and the second prediction value are integrated to obtain the target decision tree forest.

7. The method according to claim 6, characterized in that The integrating the first prediction value and the second prediction value to obtain the target decision tree forest includes: Taking a weighted average of the first prediction value and the second prediction value to obtain a first weight of the first running decision tree and a second weight of the second running decision tree; The target decision tree forest is obtained according to the first weight of the first running decision tree and the second weight of the second running decision tree.

8. A monitoring system, characterized in that: It comprises a repeater subsystem and an embedded operation subsystem, and the repeater subsystem is used to execute the method according to any one of claims 1-7.

9. A computer device, characterized in that: The method comprises a memory and a processor, wherein the memory is connected to the processor, and the processor is used to execute one or more computer programs stored in the memory. When the processor executes the one or more computer programs, the computer device implements the method according to any one of claims 1 to 7.

10. A computer-readable storage medium, characterized in that: The computer-readable storage medium stores a computer program, wherein the computer program includes program instructions, and when the program instructions are executed by a processor, the processor is caused to perform the method according to any one of claims 1 to 7.

Images