Abnormal cell identification method, electronic device, storage medium and program product

By acquiring and analyzing the index data of wireless network cells, using threshold prediction models and clustering algorithms to identify abnormal cells, the problem of identification errors caused by relying on manual experience in the prior art is solved, and the accuracy and flexibility of identification are improved, and adapted to different The actual situation of a type of cell.

CN119095087APending Publication Date: 2024-12-06CHINA UNITED NETWORK COMM GRP CO LTD +1
View PDF 0 Cites 2 Cited by

Patent Information

Application Number
CN202411337382.7
Authority / Receiving Office
CN · China
Patent Type
Applications(China)
Current Assignee / Owner
Filing Date
2024-09-24
Publication Date
2024-12-06

AI Technical Summary

Technical Problem

In the prior art, abnormal cell identification relies on manual experience, resulting in insufficient judgment errors, accuracy and flexibility, and it is impossible to effectively adapt to the actual situation of different types of cells.

Method used

By obtaining network index data of cells in the target area, using threshold prediction model and K-mean clustering algorithm to divide cell categories, determine key index data and graph trends, and identify problems in abnormal cells in combination with positioning delimited analysis algorithm to improve identification of Accuracy and flexibility.

Benefits of technology

It improves the accuracy and flexibility of abnormal cell recognition, reduces errors in manual judgment, can adapt to complex cell network conditions, and enhances the accuracy and generalization ability of network index threshold prediction of the model.

✦ Generated by Eureka AI based on patent content.

Smart Images

  • Figure CN119095087A_ABST
    Figure CN119095087A_ABST
Patent Text Reader

Abstract

The invention provides an abnormal cell identification method, electronic equipment, a storage medium and a program product, relates to the technical field of wireless networks, and is used for identifying cells with poor quality problems. The method comprises the following steps: acquiring network index data of a plurality of cells in a target area in a preset time period; obtaining an abnormal threshold based on the network index data and a threshold prediction model; the threshold prediction model is used for analyzing the network index data to obtain an abnormal threshold for identifying an abnormal cell; and determining an abnormal cell in the plurality of cells based on the network index data and the abnormal threshold.
Need to check novelty before this filing date? Find Prior Art

Description

Technical Field

[0001] The present application relates to the field of wireless network technology, and in particular to an abnormal cell identification method, electronic equipment, storage medium and program product. Background Art

[0002] With the rapid development of mobile communication technology, users' requirements for communication quality and service experience are increasing day by day. Therefore, ensuring the communication quality of each cell in the network has become a problem that operators need to face and solve. In this process, the identification of cells with abnormal communication has become a crucial link in the optimization and maintenance of communication networks.

[0003] At present, the identification method of abnormal cells is mainly based on manual experience, and the current cell situation is judged according to the detection data. This method that relies on manual experience is subjective and lacks clear basis, resulting in low accuracy. In addition, it is not flexible when facing different types of cells, and the analysis results cannot accurately fit the actual situation of the current cell. Summary of the invention

[0004] The present application provides an abnormal cell identification method, electronic device, storage medium and program product, which are used to improve the accuracy and flexibility of identifying abnormal cells.

[0005] In a first aspect, the present application provides a method for identifying abnormal cells, the method comprising: obtaining network indicator data of multiple cells in a target area in a preset time period; obtaining an abnormal threshold based on the network indicator data and a threshold prediction model; the threshold prediction model is used to analyze the network indicator data to obtain an abnormal threshold for identifying abnormal cells; based on the network indicator data and the abnormal threshold, determining abnormal cells among multiple cells.

[0006] The technical solution provided by the present application brings at least the following beneficial effects: the present application utilizes the cell network indicator data and the threshold prediction model to obtain the abnormal threshold of the network indicator data, and determines the abnormal threshold of multiple cell network indicator data based on the network indicator data and the abnormal threshold, thereby avoiding the misjudgment that may be caused by relying on manual experience for abnormal judgment in the prior art, and can improve the accuracy and flexibility of abnormal cell identification.

[0007] In combination with the first aspect above, as a possible implementation method, an abnormal threshold is obtained based on network indicator data and a threshold prediction model, including: based on the network indicator data and the K-means clustering algorithm, the network indicator data with a similarity greater than a preset threshold is divided into the same type of cells, and at least one category corresponding to multiple cells is determined; for each type of cell in at least one category, at least one key indicator data of each type of cell is determined based on the network indicator data of each type of cell; the key indicator data is used to reflect the voice connection status of each type of cell; based on at least one key indicator data of each type of cell, a graph trend of the key indicator of each type of cell is determined; the graph trend of the key indicator of each type of cell is input into the threshold prediction model to obtain the abnormal threshold of each type of cell, and the abnormal threshold of each type of cell is used to identify abnormal cells of each type of cell.

[0008] In combination with the above-mentioned first aspect, as a possible implementation method, before determining the graph trend of the key indicators of each type of cell based on multiple key indicator data of each type of cell, the method also includes: removing extreme values ​​in at least one key indicator data of each type of cell, and performing value filling processing on blank data and missed data.

[0009] In combination with the first aspect above, as a possible implementation manner, the method further includes: analyzing problems existing in the abnormal cell using a positioning and delimitation analysis algorithm based on network indicator data of the abnormal cell.

[0010] In combination with the first aspect above, as a possible implementation method, a positioning and delimitation analysis algorithm is used according to the network indicator data of the abnormal cell to analyze the problems existing in the abnormal cell, including: determining whether there is an equipment failure problem in the abnormal cell according to the network indicator data of the abnormal cell; in the case that there is no equipment failure problem in the abnormal cell, determining whether there is a network coverage problem in the abnormal cell; in the case that there is no network coverage problem in the abnormal cell, determining whether there is an interference problem in the abnormal cell; in the case that there is no interference problem in the abnormal cell, determining whether there is a capacity problem in the abnormal cell; in the case that there is no capacity problem in the abnormal cell, determining whether there is a parameter configuration problem in the abnormal cell.

[0011] In a second aspect, the present application provides a model training method, the method comprising: obtaining historical network indicator data of multiple cells in a target area; constructing training samples based on the historical network indicator data and at least one category corresponding to the multiple cells; based on the training samples, training a threshold prediction model to be trained to obtain a trained threshold prediction model, the threshold prediction model is used to analyze the network indicator data to obtain an abnormal threshold for identifying abnormal cells.

[0012] The technical solution provided by the present application brings at least the following beneficial effects: for different types of cells, the present application uses the cell type and cell network indicator data as training samples to train the model, and obtains a threshold prediction model that can predict the abnormal threshold of the cell network indicator data, which can enable the model to adapt to complex cell network conditions and improve the accuracy and generalization ability of the model's network indicator threshold prediction.

[0013] In combination with the above second aspect, as a possible implementation method, constructing a training sample based on historical network indicator data and at least one category corresponding to multiple cells includes: based on a K-means clustering algorithm and historical network indicator data, classifying cells whose network indicator data similarity is greater than a preset threshold into the same type of cells, and determining at least one category corresponding to the multiple cells;

[0014] For each type of cell in at least one category, at least one key indicator data of each type of cell is determined based on the network indicator data of each type of cell; the key indicator data is used to reflect the voice connection status of each type of cell; the extreme values ​​in at least one key indicator data of each type of cell are removed, and blank data and missed data are supplemented. Based on at least one key indicator data of each type of cell, the graph trend of the key indicator of each type of cell is determined; at least one category corresponding to multiple cells and the graph trend of the key indicator data of each type of cell are used as sample data to construct training samples.

[0015] In a third aspect, the present application provides an abnormal cell identification device, which includes: an acquisition module and a processing module. In some embodiments, the abnormal cell identification device also includes an analysis module.

[0016] An acquisition module, used to acquire network indicator data of multiple cells in a target area during a preset time period;

[0017] The processing module is used to obtain an abnormal threshold based on network indicator data and a threshold prediction model; the threshold prediction model is used to analyze the network indicator data to obtain an abnormal threshold for identifying abnormal cells; based on the network indicator data and the abnormal threshold, the abnormal cell among multiple cells is determined.

[0018] A possible implementation method is a processing module, which is specifically used to classify cells with similarity of network indicator data greater than a preset threshold into the same type of cells based on network indicator data and a K-means clustering algorithm, and determine at least one category corresponding to multiple cells; for each type of cell in at least one category, determine at least one key indicator data of each type of cell based on the network indicator data of each type of cell; the key indicator data is used to reflect the voice connection status of each type of cell; based on at least one key indicator data of each type of cell, determine the graph trend of the key indicator of each type of cell; input the graph trend of the key indicator of each type of cell into a threshold prediction model to obtain an abnormal threshold value for each type of cell, and the abnormal threshold value of each type of cell is used to identify abnormal cells of each type of cell.

[0019] In another possible implementation manner, the processing module is further specifically configured to remove extreme values ​​in at least one key indicator data of each type of cell, and to perform value supplementation processing on blank data and missed data.

[0020] A possible implementation method is an analysis module, which is used to analyze the problems existing in the abnormal cell by using a positioning and delimitation analysis algorithm based on the network indicator data of the abnormal cell.

[0021] Another possible implementation method is that the analysis module is specifically used to determine whether there is an equipment failure problem in the abnormal cell based on the network indicator data of the abnormal cell; if there is no equipment failure problem in the abnormal cell, determine whether there is a network coverage problem in the abnormal cell; if there is no network coverage problem in the abnormal cell, determine whether there is an interference problem in the abnormal cell; if there is no interference problem in the abnormal cell, determine whether there is a capacity problem in the abnormal cell; if there is no capacity problem in the abnormal cell, determine whether there is a parameter configuration problem in the abnormal cell.

[0022] In a fourth aspect, the present application provides a model training device, which includes: an acquisition module, a processing module and a training module.

[0023] An acquisition module, used to acquire historical network indicator data of multiple cells in a target area;

[0024] A processing module, configured to construct a training sample based on historical network indicator data and at least one category corresponding to a plurality of cells;

[0025] The training module trains the threshold prediction model to be trained based on the training samples to obtain a trained threshold prediction model. The threshold prediction model is used to analyze network indicator data to obtain an abnormal threshold for identifying abnormal cells.

[0026] A possible implementation method is a processing module, which is specifically used to classify cells whose network indicator data similarity is greater than a preset threshold into the same type of cells based on the K-means clustering algorithm and historical network indicator data, and determine at least one category corresponding to multiple cells; for each type of cell in at least one category, determine at least one key indicator data of each type of cell based on the network indicator data of each type of cell; the key indicator data is used to reflect the voice connection status of each type of cell; remove extreme values ​​in at least one key indicator data of each type of cell, and perform value supplementation processing on blank data and missed data. Based on at least one key indicator data of each type of cell, determine the graph trend of the key indicator of each type of cell; use at least one category corresponding to multiple cells and the graph trend of the key indicator data of each type of cell as sample data to construct a training sample.

[0027] In a fifth aspect, the present application provides an electronic device, comprising: a processor and a memory; the memory stores instructions executable by the processor; when the processor is configured to execute the instructions, the electronic device implements the methods of the first aspect to the second aspect above.

[0028] In a sixth aspect, the present application provides a computer-readable storage medium, which includes: computer software instructions; when the computer software instructions are executed in an electronic device, the electronic device implements the methods of the first to second aspects above.

[0029] In a seventh aspect, the present application provides a computer program product, which includes a computer program; when the computer program runs in an electronic device, the electronic device implements the methods of the first to second aspects above.

[0030] The beneficial effects of the third to seventh aspects mentioned above refer to the corresponding descriptions of the first and second aspects and will not be repeated here. BRIEF DESCRIPTION OF THE DRAWINGS

[0031] Figure 1 A schematic diagram of the composition of a hardware device provided for this application;

[0032] Figure 2 A flowchart of an abnormal cell identification method provided by this application;

[0033] Figure 3 A flowchart of another abnormal cell identification method provided by this application;

[0034] Figure 4 A flowchart of another abnormal cell identification method provided by the present application;

[0035] Figure 5 A flowchart of another abnormal cell identification method provided by the present application;

[0036] Figure 6 A flowchart of a model training method provided for this application;

[0037] Figure 7 A flowchart of another model training method provided for this application;

[0038] Figure 8 A schematic diagram of the composition of an abnormal cell identification device provided in this application;

[0039] Fig. 9 A schematic diagram of the composition of a model training device provided in this application;

[0040] Fig.10 The present application provides a schematic structural diagram of an electronic device. DETAILED DESCRIPTION

[0041] A call bill data recording method provided by the present application will be described in detail below with reference to the accompanying drawings.

[0042] The term "and / or" in this article is merely a description of the association relationship of associated objects, indicating that three relationships may exist. For example, A and / or B can mean: A exists alone, A and B exist at the same time, and B exists alone.

[0043] The terms "first" and "second" and the like in the specification and drawings of this application are used to distinguish different objects, or to distinguish different processing of the same object, rather than to describe a specific order of objects.

[0044] In addition, the terms "including" and "having" and any variations thereof mentioned in the description of the present application are intended to cover non-exclusive inclusions. For example, a process, method, system, product or device that includes a series of steps or units is not limited to the listed steps or units, but may optionally include other steps or units that are not listed, or may optionally include other steps or units that are inherent to these processes, methods, products or devices.

[0045] It should be noted that, in the embodiments of the present application, words such as "exemplary" or "for example" are used to indicate examples, illustrations or descriptions. Any embodiment or design described as "exemplary" or "for example" in the embodiments of the present application should not be interpreted as being more preferred or more advantageous than other embodiments or designs. Specifically, the use of words such as "exemplary" or "for example" is intended to present related concepts in a specific way.

[0046] In order to facilitate the clear description of the technical solutions of the embodiments of the present application, in the embodiments of the present application, the words "first", "second", etc. are used to distinguish the same items or similar items with basically the same functions and effects. Those skilled in the art can understand that the words "first", "second", etc. are not limiting the quantity and execution order.

[0047] In the description of the present application, unless otherwise specified, “plurality” means two or more.

[0048] In the current wireless communication network management, anomaly detection of wireless cells is a key link to ensure stable network operation and service quality. However, the existing anomaly detection methods mainly rely on manual experience and make judgments by setting complex rule thresholds. The operation process of this method is as follows: First, based on long-term accumulated manual experience, a set of anomaly judgment rules are formulated. These rules cover various key performance indicator (KPI) thresholds that may affect the performance of wireless cells; then, in a huge database system, these rules are used to screen and compare massive data to identify wireless cells that meet the abnormal conditions; finally, professional technicians are dispatched to the site to conduct further detection, confirmation and maintenance work on these wireless cells that are judged to be abnormal.

[0049] Although this detection method based on manual experience can identify and solve abnormal problems in some wireless cells to a certain extent, its limitations are becoming increasingly prominent. Specifically, since the application of new technologies is often accompanied by new performance characteristics and challenges, detection based on manual experience often finds it difficult to quickly adapt to these changes. When new equipment or new technologies are introduced, it is difficult to quickly formulate accurate detection plans due to the lack of targeted experience accumulation, resulting in the update speed of detection rules lagging behind technological development and being unable to effectively match the actual situation of the current wireless cells. In addition, the manually formulated detection rules often rely on human intuition and experience, lack clear quantitative standards, and easily bring inconvenience to subsequent abnormality handling and maintenance work.

[0050] In response to the above technical problems, the present application provides a method for identifying abnormal cells, the idea of ​​which is to: obtain network indicator data of multiple cells in a target area in a preset time period; obtain an abnormal threshold based on the network indicator data and a threshold prediction model; the threshold prediction model is used to analyze the network indicator data to obtain an abnormal threshold for identifying abnormal cells; and determine abnormal cells among multiple cells based on the network indicator data and the abnormal threshold. This method can avoid the misjudgment that may be caused by relying on manual experience for abnormal judgment in the prior art, simplify the judgment process, and thus improve the accuracy and flexibility of abnormal cell identification.

[0051] An abnormal cell identification method provided in an embodiment of the present application is described in detail below in conjunction with the accompanying drawings of the specification.

[0052] Figure 1 A schematic diagram of the composition of a hardware device provided in an embodiment of the present application. Figure 1 As shown, it includes: a collection device 101, a processing device 102 and an analysis device 103. The collection device 101 and the processing device 102 are connected to each other, and the processing device 102 and the analysis device 103 are connected to each other.

[0053] In some embodiments, the collection device 101 is used to obtain network indicator data of multiple cells in the target area in a preset time period, and send the collected network indicator data to the processing device 102.

[0054] In some embodiments, the processing device 102 is used to classify the network indicator data with a similarity greater than a preset threshold into the same type of cells based on the network indicator data and the K-means clustering algorithm, and determine at least one category corresponding to multiple cells; for each type of cell in at least one category, perform dimensionality reduction processing on the network indicator data of each type of cell to obtain multiple key indicator data; based on the multiple key indicator data of each type of cell, determine the graph trend of the key indicator of each type of cell; input the graph trend of the key indicator of each type of cell into the threshold prediction model to obtain the abnormal threshold of the network indicator data of each type of cell, and send the abnormal threshold to the analysis device 103.

[0055] In some embodiments, the processing device 102 is further used to train the threshold prediction model to obtain a trained threshold prediction model.

[0056] In some embodiments, the processing device 102 is further configured to send the trained threshold prediction model to the analysis device 103 .

[0057] Exemplarily, the processing device 102 may be a server cluster composed of multiple servers, or a single server, or a computer, or a processor or processing chip in a server or computer, etc. The embodiment of the present application does not limit the specific device form of the processing device 102.

[0058] In some embodiments, the analysis device 103 is used to obtain an abnormal threshold based on the network indicator data and the threshold prediction model, and then determine the abnormal cell among the multiple cells based on the network indicator data and the abnormal threshold.

[0059] Exemplarily, the analysis device 103 may be a server cluster composed of multiple servers, or a single server, or a computer, or a processor or processing chip in a server or computer, etc. The embodiment of the present application does not limit the specific device form of the analysis device 103.

[0060] In some embodiments, the processing device 102 may exist independently of the analyzing device 103 . In this case, the processing device 102 may be connected to the analyzing device 103 via a communication link or a wireless network. The processing device 102 may also be integrated with the analyzing device 103 .

[0061] It should be noted that the application scenarios of the embodiments of the present disclosure are not limited. The system architecture and business scenarios described in the embodiments of the present disclosure are intended to more clearly illustrate the technical solutions of the embodiments of the present disclosure, and do not constitute a limitation on the technical solutions provided by the embodiments of the present disclosure. It is known to those skilled in the art that with the evolution of network architecture and the emergence of new business scenarios, the technical solutions provided by the embodiments of the present disclosure are also applicable to similar technical problems.

[0062] The abnormal cell identification method provided in the embodiment of the present application is specifically introduced below.

[0063] Figure 2 Flow chart of an abnormal cell identification method provided in an embodiment of the present application. Exemplarily, the abnormal cell identification method provided in an embodiment of the present application can be applied to Figure 1 The analysis device 103 shown in FIG. Figure 2 As shown, the abnormal cell identification method provided in the embodiment of the present application includes:

[0064] S10. Obtain network indicator data of multiple cells in the target area during a preset time period.

[0065] In some embodiments, the network indicator data of multiple cells in the target area includes KPI data.

[0066] Exemplarily, the network indicator data may be data uplink and downlink traffic, physical resource block (PRB) utilization, and the like.

[0067] For example, the preset time period can be set according to actual needs, and the present application embodiment does not limit this. For example, the preset time period can be one day, one week, or one month.

[0068] S20. Obtain an abnormal threshold based on the network indicator data and the threshold prediction model.

[0069] Among them, the threshold prediction model is used to analyze network indicator data to obtain the abnormal threshold used to identify abnormal cells.

[0070] In some embodiments, Figure 3 As shown, the above step S20 can be specifically implemented as the following steps S201-S204:

[0071] S201. Based on network indicator data and a K-means clustering algorithm, cells with similarity of network indicator data greater than a preset threshold are classified into the same category, and at least one category corresponding to multiple cells is determined.

[0072] S202: For each type of cell in at least one category, determine at least one key indicator data of each type of cell according to the network indicator data of each type of cell.

[0073] The key indicator data is used to reflect the voice connection status of each type of cell. Exemplarily, the at least one key indicator includes a connection rate and a call drop rate.

[0074] In some embodiments, the present application uses principal component analysis (PCA) to perform dimensionality reduction processing on the network indicator data of each type of cell, and screens out network indicator data related to the calculation of at least one key indicator data. At the same time, PCA can reduce the data characteristics of these network indicator data, so that the network indicator data after dimensionality reduction processing can be calculated to obtain key indicator data.

[0075] S203: Determine a graph trend of the key indicator of each type of cell based on at least one key indicator data of each type of cell.

[0076] Among them, the graph trend of the key indicator represents the numerical distribution of the key indicator.

[0077] In some embodiments, a graph analysis tool is used to analyze a plurality of key indicator data of each type of cell to obtain a graph trend of the key indicator data of each type of cell.

[0078] In some embodiments, before the above step S203, the method further includes: removing extreme values ​​in a plurality of key indicator data of each type of cell, and performing value supplementation processing on blank data and missed data.

[0079] It is understandable that there may be some abnormally large or small data points in the data, and these extreme values ​​may have an adverse effect on subsequent analysis. Therefore, these extreme values ​​need to be removed. At the same time, due to various reasons (such as equipment failure, test interruption, etc.), there may be blanks or missed measurements in the data, and value filling processing is required to ensure the integrity and consistency of the data.

[0080] S204 , respectively inputting the graph trends of the key indicators of each type of cell into the threshold prediction model to obtain the abnormal threshold of each type of cell.

[0081] The abnormal threshold of each type of cell is used to identify abnormal cells in each type of cell.

[0082] It can be understood that through the collection of network indicator data, the K-means clustering algorithm is used to divide cells with similar performance into different categories. Then, for each category of cells, the key indicator data (such as connection rate, call drop rate) is determined by methods such as PCA, and the graph trends of these key indicators are analyzed to understand the distribution of values. After removing extreme values ​​and completing data, the graph trends are input into the threshold prediction model to obtain the abnormal threshold for identifying abnormal cells in each category of cells. This process can cope with complex cell situations and enhance the accuracy and flexibility of identifying abnormal cells.

[0083] S30: Determine an abnormal cell among the multiple cells based on the network indicator data and the abnormal threshold.

[0084] In some embodiments, the network indicator data of the cell is compared with the abnormal threshold, and the cell whose network indicator data exceeds the abnormal threshold is determined to be an abnormal cell, that is, a cell with poor voice quality problems, and the period during which the abnormal cell has the poor voice quality problem (that is, the period during which the abnormal cell has the poor voice quality problem) is obtained.

[0085] In some embodiments, in addition to the above method of determining abnormal cells using network indicator data and abnormal thresholds, the present application may also determine abnormal cells in the following manner:

[0086] Method 1: Collect drive test (DT) data and call quality test (CQT) data from terminal devices of people of various occupations, and use analysis software to identify abnormal cells with abnormal key indicator data.

[0087] Exemplarily, various occupational types include high-mobility jobs, such as food deliverymen, taxi drivers, appliance installers, etc.

[0088] In some embodiments, the built-in or external testing software of the terminal device (such as a smart phone, vehicle-mounted equipment, etc.) will automatically start after receiving the test instructions to perform real-time DT and CQT tests. After the test is completed, the collected data will be uploaded to the operator's data analysis platform. The data analysis platform analyzes the test data through the corresponding software and then identifies and outputs abnormal cells.

[0089] Understandably, the daily work trajectory of highly mobile work widely covers all corners of the city, especially those areas that are difficult to be covered by conventional monitoring methods in daily operations, which can capture the performance status of wireless networks more comprehensively.

[0090] Method 2: Use cluster analysis algorithm to process customer service complaint data and identify abnormal cells with abnormal key indicator data.

[0091] In some embodiments, the complaint data is analyzed based on a cluster analysis algorithm to identify abnormal cells.

[0092] Figure 4 A flowchart of another abnormal cell identification method provided in an embodiment of the present application is as follows: Figure 4 As shown, after the above step S30, the abnormal cell identification method provided by the present application further includes:

[0093] S40. Analyze problems existing in abnormal cells using a positioning and delimitation analysis algorithm based on network indicator data of abnormal cells.

[0094] Among them, the positioning and demarcation analysis algorithm is a comprehensive process that aims to determine the root cause of the poor quality problem in the cell and define the scope and boundaries of the problem.

[0095] Exemplarily, the abnormal cell may be a cell with poor voice quality problems. Exemplarily, the poor voice quality problems include a high call drop rate and / or a low connection rate. The embodiment of the present application may also use a positioning and delimitation analysis algorithm based on the network indicator data of the abnormal cell to analyze the reasons for the high call drop rate in the abnormal cell and / or the reasons for the low connection rate in the abnormal cell.

[0096] In some embodiments, the above step S40 is specifically implemented as the following steps S401-S405:

[0097] S401. Determine whether there is an equipment failure problem in the abnormal cell based on network indicator data of the abnormal cell.

[0098] In some embodiments, when an alarm message appears at a base station site during a period in which the abnormal cell has poor voice quality, it is determined that the abnormal cell has an equipment failure problem.

[0099] Exemplarily, the equipment failure problems indicated by the alarm information include but are not limited to at least one of the following: base station unavailable, cell unavailable, active antenna unit (Active Antenna Unit, AUU) unavailable, clock alarm, hub abnormality, temperature abnormality, common public radio interface (Common Public Radio Interface, CPRI) optical path abnormality.

[0100] S402: When there is no equipment failure problem in the abnormal cell, determine whether there is a network coverage problem in the abnormal cell.

[0101] Exemplarily, network coverage problems may be weak coverage, over-coverage, and overlapping coverage.

[0102] In some embodiments, during the period when the abnormal cell voice quality problem occurs, if the proportion of sampling points where the measured abnormal cell signal power is greater than or equal to -105dBm accounts for less than 90% of the total sampling points, or if the signal power of the cell in the drive test data is less than -110dBm, it is determined that the abnormal cell has a weak coverage problem. Exemplarily, the signal power may be the power of the secondary synchronization signal (SS-RSRP).

[0103] In some embodiments, during the period when the poor voice quality problem occurs in the abnormal cell, when the coverage radius of the base station coverage area is greater than or equal to a preset distance threshold, or when the number of cell coverage layers at the road test sampling point is greater than a preset layer threshold, it is determined that the abnormal cell has an over-coverage problem.

[0104] The coverage radius of the base station coverage area may be determined based on a preset time advance (TA) value. Exemplarily, the preset distance threshold may be a maximum coverage radius of the base station coverage area.

[0105] It should be noted that in the field of communications, TA refers to the difference between the actual time when the terminal signal reaches the base station within the coverage area of ​​the base station and the time when the terminal signal reaches the base station assuming that the distance between the terminal and the base station is 0. It is generally believed that if the TA value of the terminal is 1, the distance between the terminal and the base station is 78 meters. Based on this, the maximum TA value indirectly indicates the maximum coverage radius of the base station coverage area.

[0106] Exemplarily, assuming that the preset TA value is 8, if the terminal TA is greater than or equal to 8, it means that the coverage radius of the base station coverage area is greater than or equal to the preset distance threshold (ie, 624 meters), and then it is determined that the abnormal cell has an over-coverage problem.

[0107] In some embodiments, during the period when the poor voice quality problem in abnormal cells occurs, when the number of cells (including abnormal cells) whose signal strength differs from that of abnormal cells by less than a certain threshold (such as 6dBm) is greater than a certain value (such as 3), it is determined that the abnormal cells have overlapping coverage problems.

[0108] S403: When there is no network coverage problem in the abnormal cell, determine whether there is an interference problem in the abnormal cell.

[0109] In some embodiments, when the noise floor values ​​of the abnormal cell and the neighboring cells are greater than -112.2 dBm during the period when the abnormal cell voice quality problem occurs, it is determined that the abnormal cell has an interference problem.

[0110] It should be noted that, usually, the calculation of the background noise satisfies the following formula:

[0111] N0=-174+10log(BW)+NF,

[0112] Where N0 is the noise floor power spectral density (in dBm / Hz), -174 is the thermal noise power spectral density (at room temperature, approximately 290K), BW is the system bandwidth (in Hz), and NF is the system noise figure (in dB).

[0113] For example, for a 5G system, if a 360KHz resource block (RB) and a typical system noise factor of 5dB are considered, the noise floor is:

[0114] N0=-174+10log(360×10 3 )+5=-113dBm / RB,

[0115] Or, if we consider a bandwidth of 100MHz, the noise floor is:

[0116] N0=-174+10log(100×10 6 )+5=-87dBm / 100MHz.

[0117] In some embodiments, when external signals (such as signals from other wireless communication systems, electromagnetic interference, etc.) enter the receiver, the noise level of the receiver will be increased, which is called interference. Interference will cause the sensitivity of the receiver to decrease (i.e., desensitization). Based on this, if desensitization of 0.8dB is allowed, the corresponding average noise intensity of the 100MHz bandwidth cell is higher than -112.2dBm, and the cell is considered to be interfered.

[0118] S404: When there is no interference problem in the abnormal cell, determine whether there is a capacity problem in the abnormal cell.

[0119] In some embodiments, when the uplink and downlink physical resource block (PRB) utilization rate of the abnormal cell and the neighboring cell per hour is greater than or equal to 60% during the period when the poor voice quality problem occurs in the abnormal cell, it is determined that the abnormal cell has a capacity problem.

[0120] Among them, the uplink PRB utilization is calculated based on the ratio between the total number of PRBs occupied by the physical uplink control channel (PUCCH), the physical random access channel (PRACH) and the physical uplink shared channel (PUSCH) and the total number of PRBs available for uplink; the downlink PRB utilization is calculated based on the ratio between the number of PRBs actually used in the downlink and the total number of PRBs available for downlink.

[0121] S405: When there is no capacity problem in the abnormal cell, determine whether there is a parameter configuration problem in the abnormal cell.

[0122] In some embodiments, during the period when the abnormal cell voice quality problem occurs, when the following events occur, it is determined that there is a parameter configuration problem in the abnormal cell:

[0123] Event 1: The VoNR function switch of the base station or terminal is not turned on.

[0124] Event 2: A2 event RSRP threshold is less than -112dBm or greater than -100dBm.

[0125] Event 3: The connected user equipment inactivity timer threshold is less than 3s or greater than 20s.

[0126] Event 4: Minimum access level threshold cell -128dBm.

[0127] It can be understood that the abnormal cell identification method provided in the embodiment of the present application collects network indicator data of multiple cells in the target area in a preset time period, pre-processes the data based on K-means clustering and principal component analysis, classifies the cells and retains the target network indicator data, and then uses the threshold prediction model to analyze these data to identify abnormal cells.

[0128] In some embodiments, after determining that the abnormal cell has the above problems, a problem report is generated so that relevant staff can take corresponding measures according to the problem report, such as Figure 5 As shown, there are the following steps:

[0129] S1. Determine whether there is equipment failure in the abnormal cell.

[0130] If so, determine to take equipment maintenance measures for base station unavailability, cell unavailability and other faults; if not, execute step S2.

[0131] S2. Determine whether there is a network coverage problem in the abnormal cell.

[0132] If so, for the weak coverage caused by no base station construction, determine to take base station construction measures, for the weak coverage caused by too low power parameters, for the overcoverage when the power parameters are too low or the base station site is too high, and for the overlapping coverage caused by no main service area, determine to take corresponding optimization measures; if not, execute step S3.

[0133] S3. Determine whether there is interference problem in the abnormal cell.

[0134] If yes, determine to take corresponding optimization measures for external interference and system interference; if no, execute step S4.

[0135] S4. Determine whether there is a capacity problem in the abnormal cell.

[0136] If yes, determine to take corresponding optimization measures according to the specific situation; if no, execute step S5.

[0137] Exemplarily, capacity issues include insufficient site capacity, unbalanced load, insufficient bandwidth, license restrictions, and other issues.

[0138] S5. Determine whether there is a parameter configuration problem in the abnormal cell.

[0139] If so, determine the appropriate optimization measures based on the specific situation.

[0140] Exemplary parameter configuration problems include neighboring cell omissions, unreasonable switching threshold settings, 4G / 5G interoperability edge anomalies, etc., and corresponding optimization measures are determined.

[0141] like Figure 6 As shown, the embodiment of the present application also provides a model training method, which is applied to Figure 1 The processing device 102 shown in FIG. 1 , the model training method comprises the following steps:

[0142] S501. Obtain historical network indicator data for multiple cells in a target area.

[0143] S502: Construct training samples based on historical network indicator data and at least one category corresponding to multiple cells.

[0144] In some embodiments, Figure 7 As shown, the above step S502 can be specifically implemented as the following steps S521 to S525:

[0145] S521. Based on the K-means clustering algorithm and historical network indicator data, the cells whose network indicator data similarity is greater than a preset threshold are classified into the same type of cells, and at least one category corresponding to the multiple cells is determined.

[0146] It can be understood that the K-means clustering algorithm is used to analyze the historical network indicator data of multiple cells, and then relatively similar cells are classified into one category to obtain different cell categories, which is helpful for subsequent model training for different cell categories, thereby improving the accuracy of the model.

[0147] S522: For each type of cell in at least one category, determine at least one key indicator data of each type of cell according to the network indicator data of each type of cell.

[0148] The key indicator data is used to reflect the voice connection status of each type of cell. Exemplarily, the key indicator data includes connection rate data and call drop rate data.

[0149] In some embodiments, principal component analysis (PCA) is used to perform dimensionality reduction processing on the network indicator data of each type of cell, filter out network indicator data related to the calculation of at least one key indicator data, and use the filtered network indicator data to calculate at least one key indicator data.

[0150] S523: Remove extreme values ​​in at least one key indicator data of each type of cell, and perform value supplementation processing on blank data and missed data.

[0151] S524: Based on at least one key indicator data of each type of cell, determine a graph trend of the key indicator of each type of cell.

[0152] S525: Use at least one category corresponding to the multiple cells and the graph trend of the key indicator data of each category of cells as sample data to construct a training sample.

[0153] S503: Based on the training samples, the threshold prediction model to be trained is trained to obtain a trained threshold prediction model.

[0154] Among them, the threshold prediction model is used to analyze network indicator data to obtain the abnormal threshold used to identify abnormal cells.

[0155] It can be seen that the above mainly introduces the solution provided by the embodiment of the present application from the perspective of the method. In order to achieve the above functions, the embodiment of the present application provides a hardware structure and / or software module corresponding to each function. It should be easily appreciated by those skilled in the art that, in combination with the modules and algorithm steps of each example described in the embodiment disclosed herein, the embodiment of the present application can be implemented in the form of hardware or a combination of hardware and computer software. Whether a function is executed in the form of hardware or computer software driving hardware depends on the specific application and design constraints of the technical solution. Professional and technical personnel can use different methods to implement the described functions for each specific application, but such implementation should not be considered to exceed the scope of the present invention.

[0156] The embodiment of the present application can divide the functional modules of the abnormal cell identification device according to the above method example. For example, each functional module can be divided according to each function, or two or more functions can be integrated into one processing module. The above integrated module can be implemented in the form of hardware or in the form of software functional modules. Optionally, the division of modules in the embodiment of the present application is schematic and is only a logical functional division. There may be other division methods in actual implementation.

[0157] The embodiment of the present application provides an abnormal cell identification device, which is used to execute the method required to be executed by any device in the abnormal cell identification system. The abnormal cell identification device can be the abnormal cell identification device involved in the present application, or a module in the abnormal cell identification device; or a chip in the abnormal cell identification device, or other devices for executing the abnormal cell identification method, which is not limited in the present application.

[0158] Figure 8 The following is a schematic diagram of the composition of an abnormal cell identification device provided in an embodiment of the present application, which can implement the abnormal cell identification method provided in the above method embodiment. Figure 8 As shown, the abnormal cell identification device 600 includes: an acquisition module 601 and a processing module 602 . In some embodiments, the abnormal cell identification device 600 further includes an analysis module 603 .

[0159] An acquisition module 601 is used to acquire network indicator data of multiple cells in a target area in a preset time period;

[0160] Processing module 602 is used to obtain an abnormal threshold based on network indicator data and a threshold prediction model; the threshold prediction model is used to analyze the network indicator data to obtain an abnormal threshold for identifying abnormal cells; based on the network indicator data and the abnormal threshold, determine the abnormal cell among multiple cells.

[0161] A possible implementation method, processing module 602, is specifically used to classify cells with similarity of network indicator data greater than a preset threshold into the same type of cells based on network indicator data and K-means clustering algorithm, and determine at least one category corresponding to multiple cells; for each type of cell in at least one category, determine at least one key indicator data of each type of cell based on the network indicator data of each type of cell; the key indicator data is used to reflect the voice connection status of each type of cell; based on at least one key indicator data of each type of cell, determine the graph trend of the key indicator of each type of cell; input the graph trend of the key indicator of each type of cell into the threshold prediction model to obtain the abnormal threshold of each type of cell, and the abnormal threshold of each type of cell is used to identify abnormal cells of each type of cell.

[0162] In another possible implementation, the processing module 602 is further specifically configured to remove extreme values ​​in at least one key indicator data of each type of cell, and perform value supplementation processing on blank data and missed data.

[0163] In a possible implementation, the analysis module 603 is used to analyze the problems existing in the abnormal cell by using a positioning and delimitation analysis algorithm according to the network indicator data of the abnormal cell.

[0164] Another possible implementation method, the analysis module 603 is specifically used to determine whether there is an equipment failure problem in the abnormal cell based on the network indicator data of the abnormal cell; if there is no equipment failure problem in the abnormal cell, determine whether there is a network coverage problem in the abnormal cell; if there is no network coverage problem in the abnormal cell, determine whether there is an interference problem in the abnormal cell; if there is no interference problem in the abnormal cell, determine whether there is a capacity problem in the abnormal cell; if there is no capacity problem in the abnormal cell, determine whether there is a parameter configuration problem in the abnormal cell.

[0165] Fig. 9 The following is a schematic diagram of the composition of a model training device provided in an embodiment of the present application, which can implement the model training method provided in the above method embodiment. Fig. 9 As shown, the model training device 700 includes: an acquisition module 701, a processing module 702 and a training module 703.

[0166] An acquisition module 701 is used to acquire historical network indicator data of multiple cells in a target area;

[0167] A processing module 702, configured to construct a training sample based on historical network indicator data and at least one category corresponding to a plurality of cells;

[0168] The training module 703 trains the threshold prediction model to be trained based on the training samples to obtain a trained threshold prediction model. The threshold prediction model is used to analyze the network indicator data to obtain an abnormal threshold for identifying abnormal cells.

[0169] A possible implementation method, processing module 702, is specifically used to classify cells whose network indicator data similarity is greater than a preset threshold into the same type of cells based on the K-means clustering algorithm and historical network indicator data, and determine at least one category corresponding to multiple cells; for each type of cell in at least one category, determine at least one key indicator data of each type of cell based on the network indicator data of each type of cell; the key indicator data is used to reflect the voice connection status of each type of cell; remove the extreme value of at least one key indicator data of each type of cell, and perform value supplementation processing on blank data and missed data. Based on at least one key indicator data of each type of cell, determine the graph trend of the key indicator of each type of cell; use at least one category corresponding to multiple cells and the graph trend of the key indicator data of each type of cell as sample data to construct a training sample.

[0170] In the case of implementing the functions of the above-mentioned integrated modules in the form of hardware, the embodiment of the present invention provides a possible structural diagram of the electronic device involved in the above-mentioned embodiment. Fig.10 As shown, the electronic device 800 includes: a processor 802 , a communication interface 803 , and a bus 804 . Optionally, the electronic device 800 may further include a memory 801 .

[0171] The processor 802 may be a processor that implements or executes various exemplary logic blocks, modules, and circuits described in conjunction with the disclosure of the present application. The processor 802 may be a central processing unit, a general-purpose processor, a digital signal processor, an application-specific integrated circuit, a field programmable gate array, or other programmable logic devices, transistor logic devices, hardware components, or any combination thereof. It may implement or execute various exemplary logic blocks, modules, and circuits described in conjunction with the disclosure of the present application. The processor 802 may also be a combination that implements computing functions, such as a combination of one or more microprocessors, a combination of a DSP and a microprocessor, and the like.

[0172] The communication interface 803 is used to connect with other devices via a communication network, such as Ethernet, wireless access network, wireless local area network (WLAN), etc.

[0173] The memory 801 may be a read-only memory (ROM) or other types of static storage devices that can store static information and instructions, a random access memory (RAM) or other types of dynamic storage devices that can store information and instructions, or an electrically erasable programmable read-only memory (EEPROM), a disk storage medium or other magnetic storage device, or any other medium that can be used to carry or store desired program codes in the form of instructions or data structures and can be accessed by a computer, but is not limited thereto.

[0174] As a possible implementation, the memory 801 may exist independently of the processor 802, and the memory 801 may be connected to the processor 802 via a bus 804, and is used to store instructions or program codes. When the processor 802 calls and executes the instructions or program codes stored in the memory 801, the abnormal cell identification method provided in the embodiment of the present invention can be implemented.

[0175] In another possible implementation, the memory 801 may also be integrated with the processor 802 .

[0176] The bus 804 may be an extended industry standard architecture (EISA) bus, etc. The bus 804 may be divided into an address bus, a data bus, a control bus, etc. For ease of representation, Fig.10 Only one thick line is used in the diagram, but this does not mean that there is only one bus or only one type of bus.

[0177] Through the description of the above implementation methods, technical personnel in the relevant field can clearly understand that for the convenience and simplicity of description, only the division of the above-mentioned functional modules is used as an example. In actual applications, the above-mentioned functions can be assigned to different functional modules as needed, that is, the internal structure of the service calling device can be divided into different functional modules to complete all or part of the functions described above.

[0178] The present application also provides a computer-readable storage medium. All or part of the processes in the above method embodiments can be completed by computer program instructions to instruct related hardware, and the program can be stored in the above computer-readable storage medium. When the computer program instructions are executed on a computer, the computer executes the abnormal cell identification method described in any of the above embodiments.

[0179] Exemplarily, the above-mentioned computer-readable storage media may include, but are not limited to: magnetic storage devices (e.g., hard disks, floppy disks or magnetic tapes, etc.), optical disks (e.g., compact disks (CD), digital versatile disks (DVD), etc.), smart cards and flash memory devices (e.g., erasable programmable read-only memory (EPROM), cards, sticks or key drives, etc.). The various computer-readable storage media described in the present disclosure may represent one or more devices and / or other machine-readable storage media for storing information. The term "machine-readable storage medium" may include, but is not limited to, wireless channels and various other media capable of storing, containing and / or carrying instructions and / or data.

[0180] An embodiment of the present application further provides a computer program product, which includes a computer program. When the computer program product is run on a computer, the computer is enabled to execute any abnormal cell identification method provided in the above embodiments.

[0181] The above are only specific implementations of the present application, but the protection scope of the present application is not limited thereto. Any changes or substitutions within the technical scope disclosed in the present application should be included in the protection scope of the present application. Therefore, the protection scope of the present application should be based on the protection scope of the claims.

Claims

1. A method for identifying abnormal cells, characterized in that: The method comprises: Obtain network indicator data for multiple cells in a target area during a preset time period; Based on the network indicator data and the threshold prediction model, an abnormal threshold is obtained; the threshold prediction model is used to analyze the network indicator data to obtain an abnormal threshold for identifying abnormal cells; Based on the network indicator data and the abnormal threshold, an abnormal cell among the multiple cells is determined.

2. The method according to claim 1, characterized in that The obtaining of an abnormal threshold based on the network indicator data and the threshold prediction model includes: Based on the network indicator data and the K-means clustering algorithm, classify the cells whose similarity of the network indicator data is greater than a preset threshold into the same type of cells, and determine at least one category corresponding to the multiple cells; For each type of cell in the at least one category, determining at least one key indicator data of each type of cell according to the network indicator data of each type of cell; the key indicator data is used to reflect the voice connection status of each type of cell; Determine a graph trend of the key indicator of each type of cell based on at least one key indicator data of each type of cell; The graph trends of the key indicators of each type of cells are respectively input into the threshold prediction model to obtain the abnormal threshold of each type of cells, and the abnormal threshold of each type of cells is used to identify abnormal cells of each type of cells.

3. The method according to claim 2, characterized in that Before determining the graph trend of the key indicator of each type of cell based on the multiple key indicator data of each type of cell, the method further includes: The extreme value in at least one key indicator data of each type of cell is removed, and the blank data and missed data are supplemented.

4. The method according to claim 1, characterized in that: The method further comprises: According to the network indicator data of the abnormal cell, a positioning and delimitation analysis algorithm is used to analyze the problems existing in the abnormal cell.

5. The method according to claim 4, characterized in that The method of analyzing problems existing in the abnormal cell by using a positioning and delimitation analysis algorithm according to the network indicator data of the abnormal cell includes: Determining whether there is a device failure problem in the abnormal cell according to the network indicator data of the abnormal cell; In the case that there is no equipment failure problem in the abnormal cell, determining whether there is a network coverage problem in the abnormal cell; In the case that the abnormal cell does not have a network coverage problem, determining whether the abnormal cell has an interference problem; In the case that the abnormal cell does not have an interference problem, determining whether the abnormal cell has a capacity problem; In the case that the abnormal cell does not have a capacity problem, it is determined whether the abnormal cell has a parameter configuration problem.

6. A model training method, characterized in that: The method comprises: Obtain historical network indicator data for multiple cells in the target area; Constructing a training sample based on the historical network indicator data and at least one category corresponding to the multiple cells; Based on the training samples, the threshold prediction model to be trained is trained to obtain a trained threshold prediction model, and the threshold prediction model is used to analyze network indicator data to obtain an abnormal threshold for identifying abnormal cells.

7. The method according to claim 6, characterized in that The constructing a training sample based on the historical network indicator data and at least one category corresponding to the multiple cells includes: Based on the K-means clustering algorithm and the historical network indicator data, the cells whose similarity of the network indicator data is greater than a preset threshold are classified into the same type of cells, and at least one category corresponding to the multiple cells is determined; For each type of cell in the at least one category, determining at least one key indicator data of each type of cell according to the network indicator data of each type of cell; the key indicator data is used to reflect the voice connection status of each type of cell; Removing extreme values ​​in at least one key indicator data of each type of cell, and performing value supplementation processing on blank data and missed data; Determine a graph trend of the key indicator of each type of cell based on at least one key indicator data of each type of cell; The training sample is constructed by taking at least one category corresponding to the multiple cells and the graph trend of the key indicator data of each category of cells as sample data.

8. An electronic device, characterized in that: The method comprises a processor and a memory, wherein the processor is coupled to the memory; the memory is used to store computer instructions, and the computer instructions are loaded and executed by the processor to enable a computer device to implement a method according to any one of claims 1 to 5, or a method according to any one of claims 6 to 7.

9. A computer-readable storage medium, characterized in that: The computer-readable storage medium includes computer-executable instructions. When the computer-executable instructions are executed on a computer, the computer executes the method according to any one of claims 1 to 5, or the method according to any one of claims 6 to 7.

10. A computer program product, characterized in that The computer program product comprises a computer program, and when the computer program is run on an electronic device, the electronic device executes the method according to any one of claims 1 to 5, or the method according to any one of claims 6 to 7.

Citation Information

Cited By

  • Combination optimization method and system for communication signals

    CN119562283A

  • Method for identifying different sites based on 5GMRO measurement

    CN119719917A