Network service quality analysis method, system and storage medium

Abstract

The present disclosure provides a network service quality analysis method, system and storage medium, which relates to the field of communication technology, wherein the method includes: obtaining terminal quality data reported by the terminal; wherein the terminal quality data includes: first base station identification information; obtaining network management indicator data from the network management system, and based on the second base station identification information and the first base station identification information in the network management indicator data, performing association processing on the terminal quality data and the network management indicator data to generate base station association data corresponding to the base station; training a quality analysis model according to the base station association data, and using the trained quality analysis model to perform quality analysis. The method, system and storage medium disclosed in the present disclosure, by adding multiple indicators such as terminal delay, take the fluctuation of terminal delay within unit time as a new feature, perform problem diagnosis, enhance the individual weight of user services, and improve the accuracy and sensitivity of analysis.

Description

Technical Field

[0001] The present disclosure relates to the field of communication technology, and in particular to a network service quality analysis method, system and storage medium. Background Art

[0002] With the popularization of 4G networks, the data traffic in mobile networks has experienced explosive growth. Mobile communication operators around the world are facing huge pressures from network operations and business revenue. In the actual work of operators, mobile network problems are currently solved in the following ways: First, the intelligent operation system uses big data and AI technology to actively detect problems based on network management indicators (network operation quality indicators). However, since network management indicators cover up information such as individual and real-time network quality, analyzing problems through network management indicator data has problems such as insufficient accuracy and low sensitivity. Summary of the invention

[0003] In view of this, a technical problem to be solved by the present invention is to provide a network service quality analysis method, system and storage medium.

[0004] According to a first aspect of the present disclosure, a network service quality analysis method is provided, which is applied to a network service quality analysis system, and the method includes: obtaining terminal quality data reported by a terminal; wherein the terminal quality data includes: first base station identification information; obtaining network management indicator data from a network management system, and based on second base station identification information and the first base station identification information in the network management indicator data, associating the terminal quality data with the network management indicator data to generate base station associated data corresponding to the base station; training a quality analysis model according to the base station associated data, and using the trained quality analysis model to perform quality analysis.

[0005] Optionally, the terminal quality data further includes: delay, jitter, and packet loss information; the wireless access gateway receives the terminal quality data and session ID sent by the terminal, and adds the first base station identification information to the terminal quality data.

[0006] Optionally, training the quality analysis model according to the base station associated data includes: analyzing and processing the base station associated data to obtain label information corresponding to the base station associated data; and training the quality analysis model based on the base station associated data and the corresponding label information.

[0007] Optionally, the analyzing and processing the base station associated data to obtain label information corresponding to the base station associated data includes: screening and processing the base station associated data according to the collection time to distinguish between busy and idle service periods; selecting latency and PRB resource utilization as target assessment indicators; analyzing and processing the base station associated data based on the latency and the PRB resource utilization to determine the quality label information.

[0008] Optionally, the analyzing and processing of the base station associated data based on the delay and the PRB resource utilization includes: setting a horizontal comparison time period, performing statistical processing on the base station associated data of different base stations within the horizontal comparison time period, and obtaining statistical data corresponding to the delay and the PRB resource utilization; determining a threshold value of the delay and a threshold value of the PRB resource utilization based on the statistical data; and determining quality label information corresponding to the base station associated data according to the threshold value of the delay and the threshold value of the PRB resource utilization.

[0009] Optionally, the analyzing and processing of the base station associated data based on the delay and the PRB resource utilization includes: setting a longitudinal comparison time period, performing statistical processing on the base station associated data of the same base station within the longitudinal comparison time period, and obtaining statistical data corresponding to the delay and the PRB resource utilization; and adding corresponding feature information to the base station associated data based on the statistical data.

[0010] Optionally, the first base station identification information and the second base station identification information include: base station ECGI information.

[0011] According to a second aspect of the present disclosure, a network service quality analysis system is provided, comprising: a terminal data acquisition module, used to acquire terminal quality data reported by a terminal; wherein the terminal quality data comprises: first base station identification information; a network management data acquisition module, used to acquire network management indicator data from a network management system; a data association processing module, used to associate the terminal quality data with the network management indicator data based on the second base station identification information and the first base station identification information in the network management indicator data, and generate base station association data corresponding to the base station; an association data use module, used to train a quality analysis model according to the base station association data, and use the trained quality analysis model to perform quality analysis.

[0012] Optionally, the terminal quality data further includes: delay, jitter, and packet loss information; the wireless access gateway receives the terminal quality data and session ID sent by the terminal, and adds the first base station identification information to the terminal quality data.

[0013] Optionally, the associated data usage module includes: a label acquisition unit, used to analyze and process the base station associated data to obtain label information corresponding to the base station associated data; and a model training unit, used to train the quality analysis model based on the base station associated data and the corresponding label information.

[0014] Optionally, the label acquisition unit is also used to filter and process the base station associated data according to the collection time to distinguish between busy and idle service periods; select latency and PRB resource utilization as target assessment indicators; analyze and process the base station associated data based on the latency and the PRB resource utilization to determine the quality label information.

[0015] Optionally, the label acquisition unit is specifically used to set a horizontal comparison time period, perform statistical processing on the base station associated data of different base stations within the horizontal comparison time period, and obtain statistical data corresponding to the delay and the PRB resource utilization rate; determine the threshold value of the delay and the threshold value of the PRB resource utilization rate based on the statistical data, and determine the quality label information corresponding to the base station associated data according to the threshold value of the delay and the threshold value of the PRB resource utilization rate.

[0016] Optionally, the label acquisition unit is specifically used to set a longitudinal comparison time period, perform statistical processing on the base station associated data of the same base station within the longitudinal comparison time period, and obtain statistical data corresponding to the delay and the PRB resource utilization rate; based on the statistical data, corresponding feature information is added to the base station associated data.

[0017] According to a third aspect of the present disclosure, a network service quality analysis system is provided, comprising: a memory; and a processor coupled to the memory, wherein the processor is configured to execute the method described above based on instructions stored in the memory.

[0018] According to a fourth aspect of the present disclosure, a computer-readable storage medium is provided, wherein the computer-readable storage medium stores computer instructions, and the instructions are executed by a processor to perform the above network service quality analysis method.

[0019] The network service quality analysis method, system and storage medium disclosed in the present invention associate terminal quality data with network management indicator data based on the second base station identification information in the network management indicator data and the first base station identification information in the terminal quality data to generate base station associated data corresponding to the base station; train a quality analysis model based on the base station associated data, and use the trained quality analysis model to perform quality analysis; by adding multiple indicators such as terminal delay, the fluctuation of terminal delay within a unit time is used as a new feature to perform problem diagnosis, thereby enhancing the individual weight of user services and improving the accuracy and sensitivity of analysis. BRIEF DESCRIPTION OF THE DRAWINGS

[0020] In order to more clearly illustrate the embodiments of the present disclosure or the technical solutions in the prior art, the drawings required for use in the embodiments or the description of the prior art will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present disclosure. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying creative labor.

[0021] Figure 1 A schematic diagram of network management indicator data in the prior art;

[0022] Figure 2 A flowchart of an embodiment of a method for analyzing network service quality according to the present disclosure;

[0023] Figure 3 It is a schematic diagram of a flow chart of analyzing and processing base station associated data in one embodiment of a network service quality analysis method according to the present disclosure;

[0024] Figure 4 It is a schematic diagram of a flow chart of analyzing and processing base station associated data in another embodiment of the network service quality analysis method disclosed in the present invention;

[0025] Figure 5 A schematic diagram of modules of an embodiment of a network service quality analysis system according to the present disclosure;

[0026] Figure 6 A schematic diagram of a module of a correlation data use module in an embodiment of a network service quality analysis system according to the present disclosure;

[0027] Figure 7 FIG. 4 is a module diagram of another embodiment of a network service quality analysis system according to the present disclosure. DETAILED DESCRIPTION

[0028] The present disclosure is described more fully below with reference to the accompanying drawings, in which exemplary embodiments of the present disclosure are described. The technical solutions in the embodiments of the present disclosure will be clearly and completely described below in conjunction with the drawings in the embodiments of the present disclosure. Obviously, the described embodiments are only part of the embodiments of the present disclosure, rather than all of the embodiments. Based on the embodiments in the present disclosure, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present disclosure. The technical solutions of the present disclosure are described in many aspects below in conjunction with various figures and embodiments.

[0029] At present, mobile communication operators around the world are facing huge dual pressures of network operation and business revenue. Mobile users are not only concerned about speed increase and fee reduction, but also about network quality and service level. Therefore, mobile user perception and satisfaction have become the focus of operators. In the actual work of operators, mobile network problems are currently solved in the following ways: first, the intelligent operation system, based on network management indicators (network operation quality indicators), uses big data and AI technology to actively detect problems; in addition, the wireless network optimization department obtains network problems and solves them through active route testing, combined with monitoring alarms issued by the network management system, and user complaints.

[0030] In actual work, the mainstream business model is for the intelligent operation system to identify problems and issue work orders. However, network management indicators have problems such as "lack of accuracy", "low sensitivity" and "loss of focus on individual users", which leads to the shortcomings of the intelligent operation system based on it.

[0031] like Figure 1 As shown in the figure, the network management indicators are the average values ​​of the indicators within the collection period and cannot reflect the quality situation at a certain point in time: Figure 1 The dotted line in the figure is the latency indicator in the network management indicator. Figure 1 The figure shows that the fluctuation from 0 to 5 is very stable and the trend is good; the solid line is the instantaneous value of the sample users monitored from 0 to 5, which fluctuates greatly.

[0032] The network management indicators are the mathematical averages of many collected users, and cannot focus on the quality of individual users: the PRB (Physical Resource Block) indicator in the KPI (Key Performance Indicator) is an idle station feature and is the average of many access users. However, the actual experience of a single user is high latency and lag.

[0033] At present, since network management indicators cover up individuals and the real-time network quality, the monitoring and alarm system based on this part of data has natural defects. The existing technology directly analyzes problems through network management indicator data, which has problems such as insufficient accuracy, low sensitivity, and loss of focus on individual users.

[0034] Figure 2 FIG. 1 is a flow chart of an embodiment of a network service quality analysis method according to the present disclosure. The network service quality analysis method is applied to a network service quality analysis system, such as Figure 1 As shown:

[0035] Step 201: Acquire terminal quality data reported by a terminal; wherein the terminal quality data includes first base station identification information.

[0036] Step 202, obtain network management indicator data from the network management system, and associate the terminal quality data with the network management indicator data based on the second base station identification information and the first base station identification information in the network management indicator data to generate base station association data corresponding to the base station.

[0037] In one embodiment, the network management indicator data may be a variety of indicator data corresponding to the base station, such as delay, connection rate, bandwidth, PRB resources and other indicator data, which can reflect the service experience quality. The first base station identification information and the second base station identification information include base station ECGI (E-UTRAN Cell Global Identifier, E-UTRAN cell global identifier) ​​information, etc.

[0038] Step 203: training a quality analysis model according to base station association data, and performing quality analysis using the trained quality analysis model.

[0039] In one embodiment, general feature processing and AI modeling can be performed, and the modeling can be trained by enhancing the base station associated data of the features. The quality analysis model can be a variety of deep learning network models, such as CNN (Convolutional Neural Networks) model, LSTM (Long Short-Term Memory) model, etc.

[0040] A quality analysis model can be trained based on base station associated data using a variety of existing methods. For example, training samples are generated based on base station associated data, quality labels corresponding to the base station associated data are obtained, samples are labeled using the quality labels, training sample sets are generated, and the initial quality analysis model is trained using the training sample sets. When making predictions, the base station associated data is input into the trained quality analysis model to obtain prediction information such as network service quality and service experience quality.

[0041] In one embodiment, the terminal quality data may be service quality of experience (QOE) data, and the terminal quality data includes delay, jitter, packet loss information, etc.; the wireless access gateway receives the terminal quality data and session ID sent by the terminal, and adds the first base station identification information to the terminal quality data.

[0042] Predefine the terminal quality data fields, including delay, jitter, packet loss and other fields. The terminal can obtain the above field values ​​through public protocols such as Ping and Trace, or through custom protocols to measure the network speed. Add parameter configuration in the wireless access gateway, and when all requests pass through the wireless access gateway, the base station ECGI information of the access network is added by default. After the terminal collects the quality data, call the network interface provided by the network service quality analysis system (including but not limited to HTTP / HTTPS interface) to report the terminal quality data. The report content includes QOE data, the current session ID, and the automatically added base station ECGI information.

[0043] The network service quality analysis system obtains large network data information from the intelligent network management through SFTP and other methods, and the information includes base station ECGI information. The network service quality analysis system associates the terminal quality data and network management indicator data through the ECGI (for example, EnbID, CellID) in the multi-party information to form a complete data feature of the same base station information.

[0044] In one embodiment, the base station associated data is analyzed and processed to obtain label information corresponding to the base station associated data, and the quality analysis model is trained based on the base station associated data and the corresponding label information. Figure 3 FIG. 1 is a flow chart of analyzing and processing base station associated data in one embodiment of the network service quality analysis method disclosed in the present invention. Figure 3 As shown:

[0045] Step 301 , filtering and processing base station related data according to the collection time, so as to distinguish between the busy period and the idle period.

[0046] Step 302: Select delay and PRB resource utilization as target assessment indicators.

[0047] Step 303: Analyze and process the base station associated data based on the delay and PRB resource utilization to determine the quality label information.

[0048] There are many methods for analyzing and processing base station associated data based on delay and PRB resource utilization. For example, set a horizontal comparison time period, perform statistical processing on the base station associated data of different base stations within the horizontal comparison time period, and obtain statistical data corresponding to the delay and PRB resource utilization; determine the threshold value of the delay and the threshold value of the PRB resource utilization based on the statistical data; determine the quality label information corresponding to the base station associated data based on the threshold value of the delay and the threshold value of the PRB resource utilization. Quality labels include labels such as good quality and poor quality.

[0049] A longitudinal comparison time period is set, and statistical processing is performed on the base station associated data of the same base station within the longitudinal comparison time period to obtain statistical data corresponding to the delay and PRB resource utilization rate; based on the statistical data, corresponding characteristic information is added to the base station associated data, and the characteristic information includes characteristic information such as busy station, large fluctuation, and historical quality.

[0050] In one embodiment, the network management cloud data is the time average within the collection cycle, which cannot reflect the quality at a certain point in time. The network management cloud data is the mathematical average of many collected users and cannot focus on the quality of individual users. Although operators can also obtain indicators on the terminal side through technologies such as DPI, the DPI system still obtains the average value within a large granularity time and cannot accurately reflect fluctuations. The network service quality analysis method disclosed in the present invention adds multiple indicators such as terminal delay, and uses the fluctuation of terminal delay per unit time as a new feature of cloud data, which is input into the network service quality analysis system for problem diagnosis, thereby enhancing the individual weight of user services.

[0051] For example, the network service quality analysis system's interface for collecting and reporting Qoe data is as follows

[0052] As shown in Table 1:

[0053]

[0054] Table 1-Qoe data collection and reporting interface

[0055] The CSV file is uploaded in tar.gz format and named appid_yyyymmdd_province.csv.tar.gz. The fields in the CSV file are shown in Table 2 below:

[0056]

[0057]

[0058] Table 2 - Fields in CSV file

[0059] Figure 4FIG. 1 is a flow chart of analyzing and processing base station associated data in another embodiment of the network service quality analysis method disclosed in the present invention. Figure 4 As shown:

[0060] Step 401, polling by base station, querying QOE data and network management indicator data, and performing correlation matching.

[0061] Step 402, the associated data is divided into bins according to the collection time so as to accurately reflect the status of the business busy and idle periods.

[0062] Step 403, select delay and PRB resource utilization as target assessment indicators, and find out the data distribution pattern by horizontal comparison and vertical analysis methods.

[0063] Step 404, the horizontal comparison is calculated based on one week, and the data divided into the same time period within one week are statistically analyzed to find the mean, variance, quantile, extreme value, etc., and observe the data distribution pattern.

[0064] Step 405: According to horizontal statistics, combined with the quartile values ​​or the top 5% in descending order, find the threshold values ​​of the delay and PRB indicators. According to the threshold values, label the samples with "good or bad quality", where 0 is good and 1 is bad.

[0065] Step 406, the longitudinal analysis is calculated based on 2 weeks, and the data of the same base station in the same time period within 2 weeks are counted to find the mean, variance, quantile, extreme value, etc., and observe the data distribution pattern.

[0066] Step 407: Based on the longitudinal statistics, determine whether the base station is a daily busy station. And add "busy station" as a new feature, 0 for no and 1 for yes. Similarly, based on the variance fluctuation of each indicator, add features such as "large fluctuation" and "high historical quality difference".

[0067] In one embodiment, after merging and enhancing Qoe and large network data, network problems are diagnosed and analyzed through data analysis and AI modeling. In actual applications, PRB utilization is selected as the evaluation indicator of wireless network quality, and system functions are implemented through Python. Some code snippets are as follows:

[0068] #Problem Base Station Screening

[0069] trouble_enb_ids=find_trouble_enb();

[0070] #Divided according to whether there is corresponding Qoe data

[0071] trouble_enb_with_qoe_ids,trouble_enb_without_qoe_ids=qoe_relation_divide(trouble_enb_ids);

[0072] #First check the base stations with Qoe data and get:

[0073] #Problem base stations on the wireless side, suspected problem base stations on the transmission side

[0074] wireless_trouble_ids, trans_possible_trouble_ids=qoe_effect_filter(trouble_enb_with_qoe_ids);

[0075] #Merge the base station set with no Qoe data and the base station set with suspected transmission side problems

[0076] trans_possible_trouble_ids=trouble_enb_without_qoe_ids.

[0077] append(trans_possible_trouble_ids).drop_duplicates();

[0078] #First check whether it is a problem on the transmission side, and make a rough judgment on the transmission side

[0079] #Here, the transmission side can be segmented into three categories:

[0080] #1. Set of base stations associated with ab device issues

[0081] #2. Base station collection associated with administrative area issues

[0082] #3. Base station collection associated with geographic location issues

[0083] ab_device_trouble_ids,administrate_trouble_ids,geographical_trouble_ids,unresolved_trouble_ids=transmission_stab_location(trans_trouble_enb_ids);

[0084] #The problem base stations that cannot be located on the transmission side and the problem base stations on the wireless side determined by Qoe are further subdivided on the wireless side.

[0085] #Here, the segmentation on the wireless side can be used to locate:

[0086] #1. Coverage Issue

[0087] #2. Interference Problem

[0088] #3. Load Issues

[0089] wireless_trouble_enb_ids=wireless_trouble_enb_ids.

[0090] append(unresolved_trouble_ids).

[0091] drop_duplicates()

[0092] mr_trouble_ids, interfer_trouble_ids, load_trouble_ids=

[0093] wireless_stab_location(wireless_trouble_enb_ids).

[0094] In one embodiment, Figure 5 As shown, the present disclosure provides a network service quality analysis system 50, including a terminal data acquisition module 51, a network management data acquisition module 52, a data association processing module 53 and an association data use module 54. The terminal data acquisition module 51 acquires terminal quality data reported by the terminal; wherein the terminal quality data includes the first base station identification information.

[0095] The network management data acquisition module 52 acquires the network management indicator data from the network management system. The data association processing module 53 performs association processing on the terminal quality data and the network management indicator data based on the second base station identification information and the first base station identification information in the network management indicator data, and generates base station association data corresponding to the base station. The association data use module 54 trains a quality analysis model according to the base station association data, and uses the trained quality analysis model to perform quality analysis.

[0096] In one embodiment, Figure 6 As shown, the associated data using module 54 includes: a label acquisition unit 541 and a model training unit 542. The label acquisition unit 541 analyzes and processes the base station associated data to obtain label information corresponding to the base station associated data; the model training unit 542 trains the quality analysis model based on the base station associated data and the corresponding label information.

[0097] The label acquisition unit 541 filters and processes the base station associated data according to the collection time to distinguish between busy and idle periods, and selects latency and PRB resource utilization as target assessment indicators. The label acquisition unit 541 analyzes and processes the base station associated data based on latency and PRB resource utilization to determine quality label information.

[0098] The label acquisition unit 541 sets a horizontal comparison time period, performs statistical processing on the base station associated data of different base stations within the horizontal comparison time period, and obtains statistical data corresponding to the delay and PRB resource utilization. The label acquisition unit 541 determines the threshold value of the delay and the threshold value of the PRB resource utilization based on the statistical data, and determines the quality label information corresponding to the base station associated data according to the threshold value of the delay and the threshold value of the PRB resource utilization.

[0099] The tag acquisition unit 541 sets a longitudinal comparison time period, performs statistical processing on the base station associated data of the same base station within the longitudinal comparison time period, and obtains statistical data corresponding to the delay and PRB resource utilization. The tag acquisition unit 541 adds corresponding feature information to the base station associated data based on the statistical data.

[0100] In one embodiment, Figure 7 FIG. 1 is a module diagram of another embodiment of a network service quality analysis system according to the present disclosure. Figure 7 As shown, the device may include a memory 701, a processor 702, a communication interface 703 and a bus 704. The memory 701 is used to store instructions, the processor 702 is coupled to the memory 701, and the processor 702 is configured to execute the above-mentioned network service quality analysis method based on the instructions stored in the memory 701.

[0101] The memory 701 may be a high-speed RAM memory, a non-volatile memory, etc. The memory 701 may also be a memory array. The memory 701 may also be divided into blocks, and the blocks may be combined into virtual volumes according to certain rules. The processor 702 may be a central processing unit CPU, or an application-specific integrated circuit ASIC (Application Specific Integrated Circuit), or one or more integrated circuits configured to implement the network service quality analysis method disclosed in the present invention.

[0102] In one embodiment, the present disclosure provides a computer-readable storage medium storing computer instructions, which, when executed by a processor, implement the network service quality analysis method in any of the above embodiments.

[0103] The network service quality analysis method, system and storage medium provided in the above-mentioned embodiments associate the terminal quality data and the network management indicator data based on the second base station identification information in the network management indicator data and the first base station identification information in the terminal quality data to generate base station associated data corresponding to the base station; train a quality analysis model according to the base station associated data, and use the trained quality analysis model to perform quality analysis; by adding multiple indicators such as terminal delay, the fluctuation of terminal delay per unit time is used as a new feature to perform problem diagnosis, thereby enhancing the individual weight of user services and improving the accuracy and sensitivity of the analysis.

[0104] The method and system of the present disclosure may be implemented in many ways. For example, the method and system of the present disclosure may be implemented by software, hardware, firmware, or any combination of software, hardware, and firmware. The above order of steps for the method is for illustration only, and the steps of the method of the present disclosure are not limited to the order specifically described above, unless otherwise specifically stated. In addition, in some embodiments, the present disclosure may also be implemented as a program recorded in a recording medium, which includes machine-readable instructions for implementing the method according to the present disclosure. Therefore, the present disclosure also covers a recording medium storing a program for executing the method according to the present disclosure.

[0105] The description of the present disclosure is given for the purpose of illustration and description, and is not intended to be exhaustive or to limit the present disclosure to the disclosed form. Many modifications and variations will be apparent to those of ordinary skill in the art. The embodiments are selected and described in order to better illustrate the principles and practical applications of the present disclosure, and to enable those of ordinary skill in the art to understand the present disclosure and thereby design various embodiments with various modifications suitable for specific uses.

Claims

1. A network service quality analysis method, applied to a network service quality analysis system, the method comprising: Acquire terminal quality data reported by the terminal; wherein the terminal quality data includes: first base station identification information; Acquire network management indicator data from a network management system, and perform association processing on the terminal quality data and the network management indicator data based on the second base station identification information and the first base station identification information in the network management indicator data to generate base station association data corresponding to the base station; Training a quality analysis model according to the base station association data, and using the trained quality analysis model to perform quality analysis; Wherein, the training of the quality analysis model according to the base station association data comprises: The base station-related data is screened and processed according to the collection time to distinguish between busy and idle service periods; delay and PRB resource utilization are selected as target assessment indicators; the base station-related data is analyzed and processed based on the delay and the PRB resource utilization to determine quality label information; The analyzing and processing the base station associated data based on the time delay and the PRB resource utilization rate includes: Setting a horizontal comparison time period, performing statistical processing on base station associated data of different base stations within the horizontal comparison time period, and obtaining statistical data corresponding to the delay and the PRB resource utilization rate; determining a threshold value of the delay and a threshold value of the PRB resource utilization rate based on the statistical data; determining quality label information corresponding to the base station associated data according to the threshold value of the delay and the threshold value of the PRB resource utilization rate; Setting a longitudinal comparison time period, performing statistical processing on base station associated data of the same base station within the longitudinal comparison time period, and obtaining statistical data corresponding to the delay and the PRB resource utilization rate; adding corresponding feature information to the base station associated data based on the statistical data; The quality analysis model is trained based on the base station association data and the quality label information.

2. The method of claim 1, wherein: The terminal quality data also includes: delay, jitter, and packet loss information; The wireless access gateway receives the terminal quality data and the session ID sent by the terminal, and adds the first base station identification information to the terminal quality data.

3. The method according to claim 1 or 2, wherein: The first base station identification information and the second base station identification information include: base station ECGI information.

4. A network service quality analysis system, comprising: The terminal data acquisition module is used to acquire the terminal quality data reported by the terminal; wherein the terminal quality data includes: the first base station identification information; A network management data acquisition module is used to obtain network management indicator data from the network management system; A data association processing module, configured to perform association processing on the terminal quality data and the network management indicator data based on the second base station identification information and the first base station identification information in the network management indicator data, and generate base station association data corresponding to the base station; An associated data using module, used for training a quality analysis model according to the base station associated data, and performing quality analysis using the trained quality analysis model; Wherein, the associated data using module includes: A label acquisition unit is used to filter and process the base station associated data according to the collection time to distinguish between busy and idle service periods; select delay and PRB resource utilization as target assessment indicators; analyze and process the base station associated data based on the delay and the PRB resource utilization to determine quality label information; Among them, the label acquisition unit is specifically used to set a horizontal comparison time period, perform statistical processing on the base station associated data of different base stations within the horizontal comparison time period, and obtain statistical data corresponding to the delay and the PRB resource utilization rate; determine the threshold value of the delay and the threshold value of the PRB resource utilization rate based on the statistical data, and determine the quality label information corresponding to the base station associated data according to the threshold value of the delay and the threshold value of the PRB resource utilization rate; set a vertical comparison time period, perform statistical processing on the base station associated data of the same base station within the vertical comparison time period, and obtain statistical data corresponding to the delay and the PRB resource utilization rate; add corresponding feature information to the base station associated data based on the statistical data; A model training unit is used to train the quality analysis model based on the base station association data and the quality label information.

5. The system of claim 4, wherein: The terminal quality data also includes: delay, jitter, and packet loss information; The wireless access gateway receives the terminal quality data and the session ID sent by the terminal, and adds the first base station identification information to the terminal quality data.

6. A network service quality analysis system, comprising: Memory; and a processor coupled to the memory, wherein the processor is configured to execute the method according to any one of claims 1 to 3 based on instructions stored in the memory. 7 . A computer-readable storage medium, wherein the computer-readable storage medium non-transitorily stores computer instructions, wherein the instructions are executed by a processor to execute the method according to claim 1 .

Images