A method and system for detecting network quality degradation

Abstract

The present invention provides a method and system for detecting network quality degradation, including: obtaining the operation data of the network to be measured; inputting the operation data of the network to be measured into a pre-trained network health detection model to obtain a network degradation detection result; wherein, the network health detection model is obtained by detecting based on network historical basic data samples in a preset dimension and analyzing and integrating multiple scenario traffic models. The present invention automatically analyzes the health data model of the whole network cells, forms a network health detection model with multiple traffic scenarios and multiple models, realizes the automatic analysis and positioning of problem cells, and can predict the probability of network failures in a region under normal future conditions.

Description

Technical Field

[0001] The present invention relates to the technical field of network optimization, and in particular to a method and system for detecting network quality degradation. Background Art

[0002] At present, the optimization work in wireless networks is mainly based on backward analysis, and the optimization of the business level requires a large amount of manpower, and the optimization implementation of sudden quality degradation problems lags behind.

[0003] Especially in emergency scenarios, optimization work is required to adapt to network changes in quasi-real time. For example, in scenarios such as sudden large-scale interference, continuous outages, and major event support, when network problems occur, it is necessary to continuously follow up on quasi-real-time network performance changes and adopt appropriate optimization strategies to improve real-time performance and perception, and try to reduce the adverse effects of network problems on user perception. The collaborative intelligent optimization support system uses automatic analysis and automatic optimization execution (parameter configuration) as a means, while making full use of the advantages of 4G and 5G multi-layer heterogeneous networks to try to reduce the adverse effects of network problems on user perception. With active optimization and closed-loop adjustment as the core, establish automated analysis and optimization methods for performance degradation problems, and gradually evolve towards intelligence. Summary of the invention

[0004] The present invention provides a network quality degradation detection method and system, which are used to solve the defect of determining network quality only by a fixed threshold value in the prior art.

[0005] In a first aspect, the present invention provides a method for detecting network quality degradation, comprising:

[0006] Obtain the network operation data to be tested;

[0007] The network operation data to be tested is input into a pre-trained network health detection model to obtain a network degradation detection result; wherein, the network health detection model is based on the network's historical basic data samples, performs detection under preset dimensions, and combines analysis and integration with multiple scenario traffic models.

[0008] In one embodiment, the network health detection model is obtained by the following steps:

[0009] Obtaining the network historical basic data sample;

[0010] Based on the preset dimension, the network historical basic data sample is detected to obtain a cell problem detection result;

[0011] Based on the cell problem detection results, a problem cell data tracking table is obtained, data of cells with potential problems in the existing network is tracked, and an initial network health detection model is determined;

[0012] Obtain the multiple scenario traffic models, and analyze and correct the initial network health detection model in combination with the multiple scenario traffic models to obtain the network health detection model.

[0013] In one embodiment, the obtaining of the network historical basic data samples includes:

[0014] Obtain data of different maintenance types through the northbound interface of the network management;

[0015] Intercept the data samples within a preset normal time range from the data of different maintenance types to obtain the basic data samples; wherein, the basic data samples include normal sampling data and abnormal sampling data;

[0016] Perform data cleaning on the basic data samples to obtain the network historical basic data samples.

[0017] In one embodiment, the detecting of the network historical basic data samples based on the preset dimension to obtain the cell problem detection result includes:

[0018] Build a model based on the network historical basic data samples, and use the time series analysis algorithm to determine the characteristics of the change of KPI over time;

[0019] Based on the geographical location continuity feature, determine the contiguous network problems, and obtain regional problems and single-point problems;

[0020] Regularly perform statistical analysis on the cells and regions with frequent problems to determine the frequent problems;

[0021] Determine the cell health model, and detect whether the cell health model deviates from the campus model based on the frequent problems to obtain the cell problem detection result.

[0022] In one embodiment, the obtaining of the problem cell data tracking table based on the cell problem detection result, tracking the data of the potential risk cells in the existing network to determine the initial network health detection model includes:

[0023] Integrate the cell problem detection result to obtain the problem cell data tracking table;

[0024] Based on the problem cell data tracking table, track the optimization progress of the network degradation problem and the network self-healing result until the data curve of the problem cell returns to the historical normal curve, and construct the initial network health detection model.

[0025] In one embodiment, the obtaining of the multiple scenario traffic models, and analyzing and correcting the initial network health detection model in combination with the multiple scenario traffic models to obtain the network health detection model includes:

[0026] Determine the multiple scenario traffic models based on the traffic historical data of multiple scenarios;

[0027] According to multiple traffic scenarios in the multiple scenario traffic models, perform data screening on the initial network health detection model, remove historical data exceeding the difference threshold, and perform duplicate removal on the newly added operation data of newly connected cells to obtain optimized data;

[0028] Perform health analysis on the optimized data and perform cell-level aggregation to obtain the network health detection model.

[0029] In a second aspect, the present invention further provides a network quality degradation detection system, including:

[0030] An acquisition module for acquiring the operation data of the network to be measured;

[0031] A detection module for inputting the operation data of the network to be measured into a pre-trained network health detection model to obtain a network degradation detection result; wherein, the network health detection model is obtained by detecting based on the basic network historical data samples in a preset dimension and analyzing and integrating in combination with multiple scenario traffic models.

[0032] In a third aspect, the present invention further provides an electronic device, including a memory, a processor, and a computer program stored on the memory and executable on the processor, and when the processor executes the program, it implements the steps of the network quality degradation detection method as described in any one of the above.

[0033] In a fourth aspect, the present invention further provides a non-transitory computer-readable storage medium, on which a computer program is stored, and when the computer program is executed by a processor, it implements the steps of the network quality degradation detection method as described in any one of the above.

[0034] In a fifth aspect, the present invention further provides a computer program product, including a computer program, and when the computer program is executed by a processor, it implements the steps of the network quality degradation detection method as described in any one of the above.

[0035] The network quality degradation detection method and system provided by the present invention, by automatically analyzing the health data model of all cells in the network, form a network health detection model with multiple traffic scenarios and multiple models, realize the automatic analysis and positioning of problem cells, and can predict the probability of network failures in the region under normal future conditions. Description of the Drawings

[0036] To more clearly illustrate the technical solutions in the present invention or the prior art, the following will briefly introduce the drawings required in the description of the embodiments or the prior art. Obviously, the drawings in the following description are some embodiments of the present invention. For those of ordinary skill in the art, without creative efforts, other drawings can also be obtained based on these drawings.

[0037] Figure 1 It is a schematic flowchart of the network quality degradation detection method provided by the present invention;

[0038] Figure 2 It is a schematic structural diagram of the network quality degradation detection system provided by the present invention;

[0039] Figure 3 It is a schematic structural diagram of the electronic device provided by the present invention. Detailed implementation manners

[0040] To make the objectives, technical solutions and advantages of the present invention clearer, the following will clearly and completely describe the technical solutions in the present invention in conjunction with the drawings in the present invention. Obviously, the described embodiments are some but not all of the embodiments of the present invention. All other embodiments obtained by those of ordinary skill in the art without creative efforts based on the embodiments in the present invention belong to the scope of protection of the present invention.

[0041] In view of the limitations in network quality detection in the prior art, the present invention proposes a network quality degradation detection and analysis method based on the elasticity of mobile network performance KPIs, abandoning the traditional threshold method for defining poor quality problems, detecting network problems in real time, and improving the automatic optimization ability.

[0042] Figure 1 It is a schematic flowchart of the network quality degradation detection method provided by the present invention, as Figure 1 shown, including:

[0043] S1. Obtain the operation data of the network to be measured;

[0044] S2. Input the operation data of the network to be measured into a pre-trained network health detection model to obtain a network degradation detection result; wherein, the network health detection model is obtained by detecting in a preset dimension based on network historical basic data samples and integrating and analyzing in combination with various scenario traffic models.

[0045] Specifically, the core idea of the present invention is to establish a quasi-real-time closed-loop analysis function for performance degradation detection with automatic optimization for performance degradation problems as the core, aiming to improve the real-time performance of optimizing sudden performance degradation problems and improve the automatic optimization ability.

[0046] Input the operation data of the network to be measured into the pre-established and trained network health detection model, and output the network degradation detection result. The network health detection model here is obtained by detecting based on the network historical basic data samples under the preset dimensions and integrating and analyzing in combination with various scenario traffic models.

[0047] The present invention forms a network health detection model with multiple traffic scenarios and multiple models by automatically analyzing the health data model of the whole network cells, realizes the automatic analysis and positioning of problem cells, and can predict the probability of network failures in the regional network under normal conditions in the future.

[0048] Based on the above embodiments, the network health detection model is obtained through the following steps:

[0049] Obtain the network historical basic data samples;

[0050] Taking the preset dimension as a benchmark, detect the network historical basic data samples to obtain the cell problem detection result;

[0051] Based on the cell problem detection result, obtain the problem cell data tracking table, track the data of the potential problem cells in the existing network, and determine the initial network health detection model;

[0052] Obtain the various scenario traffic models, and analyze and correct the initial network health detection model in combination with the various scenario traffic models to obtain the network health detection model.

[0053] Specifically, the network health detection model proposed by the present invention is obtained through the following four steps:

[0054] 1) Based on the existing network, obtain the network historical basic data samples;

[0055] 2) Taking multiple preset dimensions as a benchmark, detect the above network historical basic data samples to obtain the cell problem detection result;

[0056] 3) Further construct a problem cell data tracking table according to the cell problem detection result, track the data of the potential problem cells in the existing network, and obtain the initial network health detection model;

[0057] 4) Obtain the various scenario traffic models of the existing network, and analyze and correct the initial network health detection model in combination with the various scenario traffic models to obtain the final network health detection model.

[0058] The present invention constructs a model based on multiple historical data of the existing network, and continuously optimizes and corrects it to obtain a network health detection model that can adapt to the elasticity of network performance KPIs.

[0059] Based on any of the above embodiments, the obtaining of the network historical basic data samples includes:

[0060] Obtain data of different maintenance types through the northbound interface of the network management;

[0061] Intercept the data samples within a preset normal time range from the data of different maintenance types to obtain basic data samples; wherein, the basic data samples include normal sampling data and abnormal sampling data;

[0062] Perform data cleaning on the basic data samples to obtain the network historical basic data samples.

[0063] Specifically, collect basic data samples within the normal time of cell-related information including but not limited to performance, parameters, measurements, alarms, etc. from the northbound interfaces of network management systems of various manufacturers through a collection program. Here, performance, parameters, measurements, and alarms only reflect one aspect of network performance, and the quality of network performance can also be reflected from other dimensions. The present invention does not make specific limitations.

[0064] Then perform data cleaning and cell conventional data modeling on the above basic data samples to obtain network historical basic data samples.

[0065] The present invention obtains basic data samples from multiple dimensions of the existing network, which can comprehensively cover various performance indicators and make the subsequent model detection performance more complete.

[0066] Based on any of the above embodiments, the detecting of the network historical basic data samples with the preset dimension as the benchmark to obtain the cell problem detection result includes:

[0067] Perform modeling based on the network historical basic data samples, and use the time series analysis algorithm to determine the characteristics of KPI changes over time;

[0068] Based on the geographical location continuity characteristics, determine continuous network problems to obtain regional problems and single-point problems;

[0069] Regularly conduct statistical analysis on frequently occurring problem cells and regions to determine frequently occurring problems;

[0070] Determine the cell health model, and detect whether the cell health model deviates from the campus model based on the frequently occurring problems to obtain the cell problem detection result.

[0071] Specifically, detect the cell through historical normal sample data and abnormal sample data of several categories. The present invention takes four main dimensions of coverage, connection, disconnection, and handover as the benchmark and performs detection in the following four ways;

[0072] 1) Use historical data modeling and time series analysis technology to analyze the time-varying characteristics of KPIs and discover potential network problems;

[0073] 2) Based on the continuity of geographical locations, we aggregate the severe network problems in contiguous areas and obtain regional problems as well as single-site and single-cell problems;

[0074] 3) Regularly conduct statistical analysis on frequently occurring communities and regions to locate frequently occurring problems;

[0075] 4) Health model detection: locate the health model of the cell and whether it deviates from the healthy cell model.

[0076] The present invention models common states of network operation and performs model training on basic sample data, thereby comprehensively covering fault data of the network in various states and enhancing the accuracy of the model.

[0077] Based on any of the above embodiments, obtaining a problem cell data tracking table based on the cell problem detection result, tracking data on cells with hidden dangers in the existing network, and determining an initial network health detection model include:

[0078] Based on the cell problem detection results, a data tracking table of the problem cell is obtained;

[0079] Based on the problem cell data tracking table, the optimization progress of the network degradation problem and the network self-healing result are tracked until the data curve of the problem cell recovers to the historical normal curve, and the tracking is stopped to build the initial network health detection model.

[0080] Specifically, the cell problem detection results generated by the aforementioned embodiment construct a problem cell data tracking table, which is used to track the optimization progress of subsequent degradation problems and possible network self-healing conditions, until the data curve of the problem cell returns to the historical normal curve, the tracking of the problem cell is stopped, and an initial network health detection model is constructed.

[0081] The present invention constructs a problem cell data tracking table through cell problem detection results, thereby realizing tracking of cells with hidden dangers in the existing network.

[0082] Based on any of the above embodiments, the acquiring of the multiple scenario traffic models, analyzing and correcting the initial network health detection model in combination with the multiple scenario traffic models to obtain the network health detection model includes:

[0083] Determine the multiple scenario traffic models based on multiple scenario traffic history data;

[0084] Screen the data of the initial network health detection model according to multiple traffic scenarios in the multiple scenario traffic models, remove historical data exceeding the difference threshold, and perform duplicate removal on the newly added operation data of newly connected cells to obtain optimized data;

[0085] Perform health analysis on the optimized data and perform cell-level aggregation to obtain the network health detection model.

[0086] Specifically, after constructing the initial network health detection model, obtain a large number of traffic models covering various scenarios from the existing network, automatically generate a health model in dimensions such as macro stations, in-building distribution, newly built stations, engineering stations, urban areas, rural areas, and commercial areas, discard the parts with large differences among them, and determine whether the newly connected cells are classified into the existing model or a newly created model based on data analysis. With the continuous accumulation of historical data, continuously optimize the health model and type of cells. Aggregate the health analysis results of the saved massive data reports at the cell level and count the compliance with the health degree to obtain the final network health detection model.

[0087] Finally, reverse-analyze whether there are hidden faults and problem cells in the wireless network according to the established network health detection model.

[0088] The present invention realizes the rapid positioning of the obvious and hidden problems of the wireless network and improves the network quality by analyzing the cell health model and performing model correction.

[0089] Next, the network quality degradation detection system provided by the present invention will be described. The network quality degradation detection system described below can be mutually corresponding and referred to the network quality degradation detection method described above.

[0090] Figure 2 It is a schematic structural diagram of the network quality degradation detection system provided by the present invention, as Figure 2 shown, including: an acquisition module 21 and a detection module 22, where:

[0091] The acquisition module 21 is used to acquire the operation data of the network to be tested; the detection module 22 is used to input the operation data of the network to be tested into a pre-trained network health detection model to obtain a network degradation detection result; wherein, the network health detection model is obtained by detecting according to the basic network historical data samples in a preset dimension and analyzing and integrating in combination with multiple scenario traffic models.

[0092] The present invention realizes the automatic analysis and positioning of problem cells by automatically analyzing the health data model of all cells in the network, and can predict the probability of network failures in regions under normal future conditions by forming a network health detection model with multiple traffic scenarios and multiple models.

[0093] Based on the above embodiments, the network health detection model of the detection module 22 in the system includes: an acquisition sub-module 221, a detection sub-module 222, a tracking sub-module 223, and a correction sub-module 224, where:

[0094] The acquisition sub-module 221 is used to acquire the network historical basic data samples; the detection sub-module 222 is used to detect the network historical basic data samples based on the preset dimension to obtain the cell problem detection results; the tracking sub-module 223 is used to obtain the problem cell data tracking table based on the cell problem detection results, track the data of the potential problem cells in the current network, and determine the initial network health detection model; the correction sub-module 224 is used to acquire the multiple scenario traffic models, and analyze and correct the initial network health detection model in combination with the multiple scenario traffic models to obtain the network health detection model.

[0095] The present invention constructs a model based on multiple historical data of the current network, and continuously optimizes and corrects it to obtain a network health detection model that can adapt to the elasticity of network performance KPIs.

[0096] Based on any of the above embodiments, the acquisition sub-module 221 is specifically used to: obtain data of different maintenance types through the northbound interface of the network management; intercept the data samples within the preset normal time range in the data of different maintenance types to obtain the basic data samples; where the basic data samples include normal sampling data and abnormal sampling data; perform data cleaning on the basic data samples to obtain the network historical basic data samples.

[0097] The present invention obtains basic data samples from multiple dimensions of the current network, which can comprehensively cover various performance indicators and make the subsequent model detection performance more complete.

[0098] Based on any of the above embodiments, the detection sub-module 222 is specifically used to: build a model based on the network historical basic data samples, and use the time series analysis algorithm to determine the characteristics of KPI changes over time; based on the geographical location continuity characteristics, determine the continuous network problems, obtain the regional problems and single-point problems; regularly perform statistical analysis on the frequently-occurring problem cells and regions to determine the frequently-occurring problems; determine the cell health model, and detect whether the cell health model deviates from the campus model based on the frequently-occurring problems to obtain the cell problem detection results.

[0099] The present invention builds models for the common states of network operation, trains the model with the basic sample data, comprehensively covers the fault data of the network in various states, and enhances the accuracy of the model.

[0100] Based on any of the above embodiments, the tracking submodule 223 is specifically used to: integrate the cell problem detection results to obtain the problem cell data tracking table; track the optimization progress of the network degradation problem and the network self-healing results based on the problem cell data tracking table, stop tracking until the data curve of the problem cell returns to the historical normal curve, and build the initial network health detection model.

[0101] The present invention constructs a problem cell data tracking table through cell problem detection results, thereby realizing tracking of cells with hidden dangers in the existing network.

[0102] Based on any of the above embodiments, the correction submodule 224 is specifically used to: determine the multiple scenario traffic models based on multiple scenario traffic history data; perform data screening on the initial network health detection model according to the multiple traffic scenarios in the multiple scenario traffic models, remove historical data exceeding the difference threshold, and deduplicate the added new cell operation data to obtain optimized data; perform health analysis on the optimized data, and perform cell-level aggregation to obtain the network health detection model.

[0103] The present invention realizes rapid positioning of explicit and implicit problems of wireless networks and improves network quality by analyzing a cell health model and performing model correction.

[0104] Figure 3 An example of a physical structure diagram of an electronic device is shown in FIG. Figure 3 As shown, the electronic device may include: a processor 310, a communication interface 320, a memory 330 and a communication bus 340, wherein the processor 310, the communication interface 320 and the memory 330 communicate with each other through the communication bus 340. The processor 310 may call the logic instructions in the memory 330 to execute the network quality degradation detection method, which includes: obtaining the network operation data to be tested; inputting the network operation data to be tested into a pre-trained network health detection model to obtain a network degradation detection result; wherein the network health detection model is obtained by performing detection under a preset dimension based on a network history basic data sample and combining multiple scenario traffic models for analysis and integration.

[0105] In addition, when the logical instructions in the above-mentioned memory 330 are implemented in the form of software functional units and sold or used as independent products, they can be stored in a computer-readable storage medium. Based on such an understanding, the technical solution of the present invention, in essence, or the part that contributes to the prior art, or a part of this technical solution, can be embodied in the form of a software product. This computer software product is stored in a storage medium and includes several instructions for causing a computer device (which may be a personal computer, a server, or a network device, etc.) to execute all or part of the steps of the methods described in various embodiments of the present invention. The aforementioned storage medium includes: various media such as USB flash drives, mobile hard disks, read-only memories (ROMs), random access memories (RAMs), magnetic disks, or optical discs that can store program codes.

[0106] On the other hand, the present invention also provides a computer program product. The computer program product includes a computer program that can be stored on a non-transitory computer-readable storage medium. When the computer program is executed by a processor, the computer can execute the network quality degradation detection method provided by the above-mentioned various methods. The method includes: obtaining the operation data of the network to be tested; inputting the operation data of the network to be tested into a pre-trained network health detection model to obtain a network degradation detection result; wherein, the network health detection model is obtained by detecting according to the basic network historical data samples in a preset dimension and analyzing and integrating in combination with various scenario traffic models.

[0107] On another aspect, the present invention also provides a non-transitory computer-readable storage medium, on which a computer program is stored. When the computer program is executed by a processor, it realizes the network quality degradation detection method provided by the above-mentioned various methods. The method includes: obtaining the operation data of the network to be tested; inputting the operation data of the network to be tested into a pre-trained network health detection model to obtain a network degradation detection result; wherein, the network health detection model is obtained by detecting according to the basic network historical data samples in a preset dimension and analyzing and integrating in combination with various scenario traffic models.

[0108] The device embodiments described above are merely illustrative. The units described as separate components may or may not be physically separated, and the components shown as units may or may not be physical units, that is, they may be located in one place or distributed to multiple network units. Some or all of the modules can be selected according to actual needs to achieve the purpose of the solution of this embodiment. A person of ordinary skill in the art can understand and implement it without creative labor.

[0109] Through the description of the above embodiments, those skilled in the art can clearly understand that each embodiment can be implemented by means of software plus a necessary general hardware platform, and of course, it can also be implemented by hardware. Based on such an understanding, the essence of the above technical solution, or the part that contributes to the prior art, can be embodied in the form of a software product. This computer software product can be stored in a computer-readable storage medium, such as ROM / RAM, magnetic disk, optical disk, etc., and includes several instructions to enable a computer device (which can be a personal computer, server, or network device, etc.) to execute the methods described in each embodiment or some parts of the embodiments.

[0110] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention, rather than to limit them; although the present invention has been described in detail with reference to the foregoing embodiments, those of ordinary skill in the art should understand that they can still modify the technical solutions described in the foregoing embodiments, or perform equivalent replacements for some of the technical features; and these modifications or replacements do not make the essence of the corresponding technical solutions deviate from the spirit and scope of the technical solutions of the embodiments of the present invention.

Claims

1. A method for detecting network quality degradation, characterized in that, Including: Obtain the operation data of the network to be tested; Input the operation data of the network to be tested into a pre-trained network health detection model to obtain a network degradation detection result; wherein, the network health detection model is obtained by detecting according to network historical basic data samples in a preset dimension and analyzing and integrating multiple scenario traffic models; The network health detection model is obtained through the following steps: Obtain the network historical basic data samples; Taking the preset dimension as a benchmark, detect the network historical basic data samples to obtain a cell problem detection result; Based on the cell problem detection result, obtain a problem cell data tracking table, track the data of potential problem cells in the existing network, and determine an initial network health detection model; Obtain the multiple scenario traffic models, and analyze and correct the initial network health detection model in combination with the multiple scenario traffic models to obtain the network health detection model; The step of taking the preset dimension as a benchmark, detecting the network historical basic data samples, and obtaining a cell problem detection result includes: Build a model based on the network historical basic data samples, and use a time series analysis algorithm to determine the characteristics of KPI changes over time; Based on the geographical location continuity feature, determine continuous network problems, and obtain regional problems and single-point problems; Regularly conduct statistical analysis on frequently-occurring problem cells and regions to determine frequently-occurring problems; Determine a cell health model, and based on the frequently-occurring problems, detect whether the cell health model deviates from the campus model to obtain a cell problem detection result; The step of obtaining a problem cell data tracking table based on the cell problem detection result, tracking the data of potential problem cells in the existing network, and determining an initial network health detection model includes: Integrate the cell problem detection results to obtain the problem cell data tracking table; Based on the problem cell data tracking table, track the optimization progress of network degradation problems and the network self-healing results until the data curve of the problem cell returns to the historical normal curve, and then stop tracking to construct the initial network health detection model; The step of obtaining the multiple scenario traffic models, analyzing and correcting the initial network health detection model in combination with the multiple scenario traffic models to obtain the network health detection model includes: Determine the multiple scenario traffic models based on the traffic historical data of multiple scenarios; According to multiple traffic scenarios in the multiple scenario traffic models, perform data screening on the initial network health detection model, remove historical data exceeding the difference threshold, and perform duplicate removal on the operation data of newly added network cells to obtain optimized data; Conduct health analysis on the optimized data and perform aggregation at the cell level to obtain the network health detection model.

2. The method for detecting network quality degradation according to claim 1, characterized in that, The step of obtaining the network historical basic data samples includes: Obtain data of different maintenance types through the northbound interface of the network management; Intercept the data samples within a preset normal time range from the data of different maintenance types to obtain basic data samples; wherein, the basic data samples include normal sampling data and abnormal sampling data; Clean the basic data samples to obtain the basic network historical data samples.

3. A system for detecting network quality degradation, characterized in that, Including: An acquisition module, configured to acquire the operation data of the network to be tested; A detection module, configured to input the operation data of the network to be tested into a pre-trained network health detection model to obtain a network degradation detection result; wherein, the network health detection model is obtained by detecting in a preset dimension based on the basic network historical data samples and analyzing and integrating multiple scenario traffic models; The detection module includes: An acquisition sub-module, configured to acquire the basic network historical data samples; A detection sub-module, configured to detect the basic network historical data samples based on the preset dimension to obtain a cell problem detection result; A tracking sub-module, configured to obtain a problem cell data tracking table based on the cell problem detection result, track the data of the potential problem cells in the existing network, and determine the initial network health detection model; A correction sub-module, configured to acquire the multiple scenario traffic models, analyze and correct the initial network health detection model in combination with the multiple scenario traffic models to obtain the network health detection model; The detection sub-module is specifically configured to: build a model based on the basic network historical data samples, use a time series analysis algorithm to determine the characteristics of the KPI changing over time; based on the geographical location continuity characteristics, determine the contiguous network problems, obtain the regional problems and single-point problems; regularly perform statistical analysis on the frequently occurring problem cells and regions to determine the frequently occurring problems; determine the cell health model, and detect whether the cell health model deviates from the campus model based on the frequently occurring problems to obtain the cell problem detection result; The tracking sub-module is specifically configured to: comprehensively obtain the problem cell data tracking table based on the cell problem detection result; track the optimization progress of the network degradation problem and the network self-healing result based on the problem cell data tracking table until the data curve of the problem cell returns to the historical normal curve, and construct the initial network health detection model; The correction sub-module is specifically configured to: determine the multiple scenario traffic models based on the historical data of multiple scenario traffic; perform data screening on the initial network health detection model according to the multiple traffic scenarios in the multiple scenario traffic models, remove the historical data exceeding the difference threshold, and perform duplicate removal operations on the operation data of the newly added network access cells to obtain the optimized data; perform health analysis on the optimized data and perform cell-level aggregation to obtain the network health detection model.

4. An electronic device, comprising a memory, a processor, and a computer program stored on the memory and executable on the processor, wherein, When the processor executes the program, it implements the steps of the network quality degradation detection method according to any one of claims 1 to 2.

5. A non-transitory computer-readable storage medium, having stored thereon a computer program, wherein, When the computer program is executed by the processor, it implements the steps of the network quality degradation detection method according to any one of claims 1 to 2.

6. A computer program product, comprising a computer program, wherein, When the computer program is executed by the processor, it implements the steps of the network quality degradation detection method according to any one of claims 1 to 2.

Images