User experience rate evaluation and optimization method and device, equipment and storage medium

Abstract

The invention provides a user experience rate evaluation method and device, an optimization method and device, electronic equipment and a computer readable storage medium, and relates to the technical field of communication quality evaluation. The evaluation method comprises the steps that measurement report MR data of a target area in a preset time period are collected, and the MR data comprise field information of a plurality of users; n-dimensional rasterization is carried out on the MR data of the target area to obtain field information of each N-dimensional grid, and N comprises one of 2, 3 and 4; and evaluating the user experience rate of each N-dimensional grid based on the field information of each N-dimensional grid. The method and the device are used for at least solving the problems of inaccurate division of network performance data, poor geographic presentation effect of the network performance data and low evaluation accuracy of the user experience rate in related technologies. And the method is suitable for user experience evaluation and optimization scenes.

Description

Technical Field

[0001] The present invention relates to the technical field of communication quality assessment, and in particular to an assessment method, optimization method, device, electronic device and computer-readable storage medium for user experience rate. Background Art

[0002] User experience assurance in key scenarios is an important part of communication services. An effective means of evaluating user experience is usually to measure it using user experience rate as a standard, and the accuracy of user rate capability assessment is a key factor affecting user experience assurance.

[0003] However, although the user experience rate evaluation methods in the prior art have achieved the geographical presentation of network performance data to a certain extent, the evaluation accuracy of the user experience rate still needs to be improved. Specifically: (1) Most of the user experience rates in the prior art are statistically analyzed at the cell level and regional level, which cannot effectively support the user experience and customer satisfaction of different services. (2) In the few prior arts that introduce geographic grids, the grid size and resolution are often relatively rough, which cannot fully reflect the subtle geographical differences, which may lead to inaccurate division of network performance data, poor geographical presentation of network performance data, and low accuracy of user experience rate evaluation. Summary of the invention

[0004] The technical problem to be solved by the present invention is to provide a method for evaluating user experience rate, an optimization method, a device, an electronic device and a computer-readable storage medium in response to the above-mentioned deficiencies in the prior art. The method can reduce the rasterization granularity of network performance data, improve the geographic presentation effect of network performance data, expand the application of user experience and customer satisfaction evaluation in different business scenarios, and improve the accuracy, efficiency and applicability of user experience rate evaluation.

[0005] In a first aspect, the present invention provides a method for evaluating user experience rate, comprising: collecting measurement report MR data of a target area within a preset time period, wherein the MR data includes field information of several users; performing N-dimensional rasterization on the MR data of the target area to obtain field information of each N-dimensional grid, wherein N includes one of 2, 3, and 4; and evaluating the user experience rate of each N-dimensional grid based on the field information of each N-dimensional grid.

[0006] Preferably, the MR data of the target area is N-dimensionally rasterized to obtain field information of each N-dimensional grid, specifically including: acquiring geographic location information of the target area, and rasterizing and rasterizing the target area to obtain geographic location information of each N-dimensional grid, wherein the N-dimensional grid is used to characterize grids at different levels; locating the user position of the MR data to obtain N-dimensional location information of several users; based on the geographic location information of each N-dimensional grid, the N-dimensional location information of several users and a clustering algorithm, backfilling the field information of several users to obtain field information of each N-dimensional grid, wherein the clustering algorithm includes any one of the following: K-means algorithm, density-based spatial clustering with noise DBSCAN algorithm and Mean Shift algorithm.

[0007] Preferably, the user position positioning of the MR data to obtain N-dimensional position information of several users specifically includes: performing MR aggregation on the MR data to obtain several MR groups; performing arrival angle ranging AOA estimation on the MR data to obtain arrival angle information of several users; performing topological relationship-based ray intersection on the arrival angle information of the same MR group to obtain initial N-dimensional position information of several users; performing MR calibration on the initial N-dimensional position information of several users based on a ray tracing model to obtain N-dimensional position information of several users.

[0008] Preferably, the field information includes modulation and coding MCS field information, rank RANK field information and resource block RB field information. The evaluating the user experience rate of each N-dimensional grid based on the field information of each N-dimensional grid specifically includes: performing mean regression calculation on the field information of each N-dimensional grid to obtain the regression MCS field information, regression RANK field information and regression RB field information of each N-dimensional grid; calculating the user experience rate of each N-dimensional grid based on the regression MCS field information, regression RANK field information and regression RB field information of each N-dimensional grid.

[0009] Preferably, the step of performing mean regression calculation on the field information of each N-dimensional grid to obtain the mean regression field information of each N-dimensional grid specifically includes: performing mean regression calculation on the field information of each N-dimensional grid according to formula (1):

[0010] Yt=α+βXt+εt(1), where Yt represents the regression field information at time t, Xt represents the field information at time t-1, α and β represent constants, and εt represents the mean regression error term at time t.

[0011] Preferably, the calculating the user experience rate of each N-dimensional grid based on the regression MCS field information, the regression RANK field information and the regression RB field information of each N-dimensional grid specifically includes: calculating the user experience rate of each N-dimensional grid according to formula (2):

[0012] GridRate=

[0013] AVERAGE(MCS)×AVERAGE(Rank)×AVERAGE(RNNum) (2),

[0014] Among them, Grid Rate represents the user experience rate of each N-dimensional grid, MCS represents the regression MCS field information of each N-dimensional grid, Rank represents the regression RANK field information of each N-dimensional grid, and RB Num represents the regression RB field information of each N-dimensional grid.

[0015] In a second aspect, the present invention also provides a method for optimizing user experience rate, comprising: obtaining a grid-level user experience rate evaluated by the user experience rate evaluation method provided by the first aspect above, wherein the grid-level user experience rate refers to the user experience rate of each N-dimensional grid; clustering all grid-level user experience rates in the same scenario, and calculating the scene-level user experience rate, wherein the scene-level user experience rate refers to the average of all grid-level user experience rates in each scenario; visually presenting the problem area, wherein the problem area refers to an area where the grid-level user experience rate or the scene-level user experience rate is lower than a preset threshold; based on anomaly detection algorithm and / or path loss and interference analysis model, root cause location is performed on the problem area to optimize the user experience rate of the problem area based on the root cause location result.

[0016] In a third aspect, the present invention also provides an evaluation device for user experience rate, comprising: an acquisition module, a rasterization module and an evaluation module, the acquisition module being used to acquire measurement report MR data of a target area within a preset time period, wherein the MR data includes location information and field information of several users, the rasterization module being connected to the acquisition module, and being used to perform N-dimensional rasterization on the MR data of the target N-dimensional spatial area to obtain field information of each N-dimensional grid, wherein N includes one of 2, 3, and 4, and the evaluation module being connected to the rasterization module, and being used to evaluate the user experience rate of each N-dimensional grid based on the field information of each N-dimensional grid.

[0017] In a fourth aspect, the present invention also provides a device for optimizing user experience rate, comprising: an acquisition module, a clustering module, a visualization module and a root cause location module, the acquisition module being used to acquire the grid-level user experience rate evaluated by the user experience rate evaluation device provided in the third aspect, wherein the grid-level user experience rate refers to the user experience rate of each N-dimensional grid; the clustering module being connected to the acquisition module, being used to cluster all grid-level user experience rates in the same scene, and calculating the scene-level user experience rate; the visualization module being connected to the clustering module, being used to visualize problem areas where the grid-level user experience rate or the scene-level user experience rate is lower than a preset threshold; the root cause location module being connected to the visualization module, being used to locate the root cause of the problem area based on anomaly detection algorithm and / or path loss and interference analysis model, so as to optimize the user experience rate of the problem area based on the root cause location result.

[0018] In a fifth aspect, the present invention also provides an electronic device comprising a memory and a processor, wherein the memory stores a computer program, and the processor is configured to run the computer program to implement the method for evaluating the user experience rate provided in the first aspect or the method for optimizing the user experience rate provided in the second aspect.

[0019] In a sixth aspect, the present invention further provides a computer-readable storage medium having a computer program stored thereon, which, when executed by a processor, implements the method for evaluating the user experience rate provided in the first aspect or the method for optimizing the user experience rate provided in the second aspect.

[0020] The present invention provides a method for evaluating user experience rate, an optimization method, a device, an electronic device, and a computer-readable storage medium. By multi-dimensionally rasterizing the MR data in the target area, the field information in the MR data is geographically presented in a fine-grained grid, which effectively improves the geographical presentation effect of the network performance data. The field information under the fine-grained grid can be used to better evaluate the grid-level user experience rate, and then the grid-level user experience rate can be applied to the user experience and customer satisfaction evaluation in different business scenarios, thereby improving the evaluation accuracy and applicability of the user experience rate. Therefore, the present invention can reduce the rasterization granularity of network performance data, improve the geographical presentation effect of network performance data, expand the application of user experience and customer satisfaction evaluation in different business scenarios, and improve the evaluation accuracy, efficiency, and applicability of the user experience rate. BRIEF DESCRIPTION OF THE DRAWINGS

[0021] Figure 1 This is a flow chart of a method for evaluating user experience rate according to Embodiment 1 of the present invention;

[0022] Figure 2 This is an example diagram of locating a user's position using MR data in Embodiment 1 of the present invention;

[0023] Figure 3 This is an example diagram of performing arrival angle ranging AOA estimation on MR data in embodiment 1 of the present invention;

[0024] Figure 4 This is an example diagram of performing topological relationship-based ray intersection on the arrival angle information of the same MR group in Embodiment 1 of the present invention;

[0025] Figure 5 This is an example diagram of backfilling field information of several users in Embodiment 1 of the present invention;

[0026] Figure 6 This is an example diagram of field information obtained by occupying different service cells in the same grid in Embodiment 1 of the present invention;

[0027] Figure 7 This is an example diagram of a user experience rate visualization in Embodiment 1 of the present invention;

[0028] Figure 8 This is an example diagram of another user experience rate visualization in Embodiment 1 of the present invention;

[0029] Fig. 9 This is an example diagram of root cause positioning in scenario a in embodiment 1 of the present invention;

[0030] Fig.10 This is an example diagram of coverage and rate before optimization in Embodiment 1 of the present invention;

[0031] Fig.11 This is an example diagram of coverage and rate after optimization in Embodiment 1 of the present invention;

[0032] Fig.12 This is a flow chart of a method for optimizing user experience rate according to Embodiment 2 of the present invention;

[0033] Fig.13 This is a schematic diagram of the structure of a device for evaluating user experience rate according to Embodiment 3 of the present invention;

[0034] Fig.14 This is a structural diagram of a device for optimizing user experience rate according to Embodiment 4 of the present invention. DETAILED DESCRIPTION

[0035] In order to enable those skilled in the art to better understand the technical solution of the present invention, the embodiments of the present invention will be further described in detail below with reference to the accompanying drawings.

[0036] It should be understood that the specific embodiments and drawings described herein are only used to explain the present invention rather than to limit the present invention.

[0037] It can be understood that, in the absence of conflict, the various embodiments of the present invention and the various features in the embodiments can be combined with each other.

[0038] It can be understood that, for the convenience of description, the drawings of the present invention only show the parts related to the present invention, while the parts irrelevant to the present invention are not shown in the drawings.

[0039] It can be understood that each unit and module involved in the embodiments of the present invention may correspond to only one physical structure, or may be composed of multiple physical structures, or multiple units and modules may be integrated into one physical structure.

[0040] It can be understood that, without conflict, the functions and steps marked in the flowcharts and block diagrams of the present invention may occur in an order different from that marked in the drawings.

[0041] It is understood that the flowcharts and block diagrams of the present invention illustrate the possible architectures, functions, and operations of the systems, devices, equipment, and methods according to the various embodiments of the present invention. Each box in the flowchart or block diagram may represent a unit, module, program segment, or code, which contains executable instructions for implementing the specified functions. Moreover, each box or combination of boxes in the block diagram and flowchart may be implemented by a hardware-based system that implements the specified functions, or may be implemented by a combination of hardware and computer instructions.

[0042] It can be understood that the units and modules involved in the embodiments of the present invention can be implemented by software or hardware. For example, the units and modules can be located in a processor.

[0043] Embodiment 1:

[0044] like Figure 1 As shown, this embodiment provides a method for evaluating user experience rate.

[0045] Evaluation methods for user experience rate include:

[0046] S101, collecting measurement report MR data of a target area within a preset time period, wherein the MR data includes field information of several users.

[0047] In this embodiment, the network performance data is usually presented in the following two forms: MR (Measurement Report) data from the network management system and MDT (Multi-Domain Test) data collected through road testing. MR data is performance data generated by user equipment (UE) and reported to the network management system. MR data usually includes field information related to the user's signal quality, signal strength, and connection status. MDT data is performance data collected through road testing or special network testing equipment, and usually involves detailed testing of network performance in a specific area or route. Therefore, under normal circumstances, MR data itself does not include the specific geographical location information of the user. When MDT data is collected during testing at a specific location, MDT data may include location data to provide information about the specific location of the user in the network. In this embodiment, MR data of the target area within a preset time period can be directly collected from the network management system to evaluate the user's perceived rate. Compared with collecting regional MDT data through road testing, MDT data may have a certain delay characteristic. The collection of MR data is more real-time, which can reflect the immediate performance of the current network, save time and effort, improve the evaluation efficiency and accuracy of user experience rate, and identify network coverage and performance problems more timely.

[0048] S102, performing N-dimensional rasterization on the MR data of the target area to obtain field information of each N-dimensional raster, wherein N includes one of 2, 3, and 4.

[0049] Specifically, S102: performing N-dimensional rasterization on the MR data of the target area to obtain field information of each N-dimensional raster, including steps S1021 to S1023:

[0050] S1021, obtaining geographic location information of the target area, and performing rasterization and raster layering processing on the target area to obtain geographic location information of each N-dimensional grid, wherein the N-dimensional grid is used to represent grids at different levels.

[0051] In this embodiment, taking N=3 as an example, the geographical location information of the target area is obtained, and the target area is rasterized and grid layered to divide the target area into multiple 3D grid units (i.e., 3D grids), where 3 dimensions refer to longitude, latitude, and altitude. Each 3D grid is assigned a unique grid ID to identify its geographical location. The size of the grid can be flexibly set according to the needs of the scene (such as 10 meters × 10 meters or 100 meters × 100 meters) to achieve a suitable spatial resolution. Taking the urban scene as an example, due to the dense buildings in the urban area and the high user distribution density, in order to facilitate the subsequent acquisition of higher positioning accuracy, the target area is divided into 20 meters × 20 meters grid units. Taking the rural and suburban scenes as an example, due to the low user density and open environment, the rural and suburban areas can be divided into 50 meters × 50 meters grid units. Larger grid units can reduce the amount of calculation while maintaining a certain positioning accuracy.

[0052] It should be noted that the grid layering process includes a variety of different division methods, depending on the application scenario and data type, for example, the grid can be divided according to the indoor and outdoor or multi-story building height. In this embodiment, N=3 is taken as an example, and the 3D grid refers to the grid after the target area is rasterized and the grid is layered according to the preset height.

[0053] If the target area includes urban, rural and suburban scenes, the target area can be divided into different grids, such as 20m×20m and 50m×50m.

[0054] S1022, performing user position positioning on the MR data to obtain N-dimensional position information of a plurality of users.

[0055] Specifically, S1022: performing user position positioning on MR data to obtain N-dimensional position information of several users, including: performing MR aggregation on MR data to obtain several MR groups; performing arrival angle ranging AOA estimation on MR data to obtain arrival angle information of several users; performing ray intersection based on topological relationships on the arrival angle information of the same MR group to obtain initial N-dimensional position information of several users; performing MR calibration on the initial N-dimensional position information of several users based on a ray tracing model to obtain N-dimensional position information of several users.

[0056] In this embodiment, the user position is located on the MR data to obtain the 3D position information of several users, such as Figure 2As shown, it specifically includes: ① Using MR aggregation to aggregate data features in MR data, complete MR coarse positioning, and obtain several MR groups. MR groups can represent user groups, and the coarse positioning results are used to characterize the groups to which users belong, wherein the data features include but are not limited to: user features, call features, cell features, and level features. ② Using user azimuth AOA (Angle-of-Arrival) estimation, perform the first precise correction of user positioning on several MR groups to obtain the angle of arrival information of several users. The first precise correction result is used to characterize the angle of arrival information of each user in the MR group. ③ Based on the topological relationship ray intersection, perform the secondary correction of user positioning on several MR groups to obtain the initial 3D position information of several users. The secondary correction result is used to characterize the initial 3D position information of each user in the MR group. ④ After using MR aggregation to aggregate data features in MR data to obtain several MR groups, several MR groups are also used to build indoor and outdoor models of MR; based on the positioning results (i.e. Figure 2 The initial 3D position information of several users (i.e., the initial 3D position information of all users in all MR groups) is corrected three times based on the map information in the positioning result and the indoor and outdoor model of MR) and the indoor and outdoor model of MR to obtain the 3D position information of several users. The three correction results are the 3D position information of several users, wherein the indoor and outdoor model includes but is not limited to the ray tracing model; in addition, in addition to correcting the initial 3D position information of several users three times based on the positioning result and the indoor and outdoor model of MR, this embodiment can also use Markov chain, Kalman filter or time association method to correct the initial 3D position information of several users three times. This embodiment realizes indoor and outdoor separation and 3D positioning by rasterizing and grid layering the target area and locating the user position of MR data, thereby improving the positioning accuracy of MR data, improving the matching accuracy of field information under the grid, and improving the distribution accuracy of network performance data.

[0057] MR aggregation is used to aggregate the data features in the MR data to complete the rough positioning of MR, which specifically includes: modeling the path and data features of each user in the MR data. The purpose is to simulate the level performance and abnormal event performance of the user when indoors and outdoors. When the level features and call features are the same, the user's location (i.e. indoor or outdoor) can be roughly determined to obtain indoor MR grouping and outdoor MR grouping. For the superposition of level features, coverage modeling can also be performed based on existing models, such as the level coverage model function, while superimposing call features and many data features of users. The formula of the level coverage model function is:

[0058] Using the user's AOA (Angle-of-Arrival) estimation, the first accurate correction of user positioning is performed on several MR groups, such as Figure 3 As shown, specifically, in the MR of the actual user, the private field AOA can be used to estimate the user's angle position. When the user has multiple service cells (also regarded as base stations) as neighboring cells, the network measurement, network structure and user behavior are used to roughly and quickly locate the calling user in the network, and a multi-dimensional equation group is constructed according to the connection status of the user and the base station in the network. The initial positioning result (i.e., the arrival angle information of several users) is calculated by simultaneous solution, and the initial positioning result after MR aggregation (i.e., several MR groups) is corrected.

[0059] Based on the topological relationship ray intersection, the user positioning is corrected for several MR groups, such as Figure 4 As shown, it specifically includes: splicing the service cell's work parameter related information identified by the user from the MR data, the work parameter related information includes the service cell ID (i.e. Figure 4 CELL1, CELL2, CELL3 in the cell), service cell location (i.e. Figure 4 (x1, y1), (x2, y2), (x3, y3)), antenna height, azimuth, downtilt, beamwidth, pilot power, and serving cell marker. Calculate the user's arrival angle information delay from MR (i.e. Figure 4 The time delay and level information in the 3D algorithm of line-circle intersection are used to transform the arrival angle information of several users (i.e. Figure 4 F(RSCP1, Araval1, ...), F(RSCP2, Araval2, ...), F(RSCP2, Araval2, ...)) in the 3D position information are further calibrated to obtain the initial 3D position information of several users (i.e. Figure 4 In this embodiment, the 3D algorithm of line-circle intersection is taken as WCCL (Weighted Centroid Correction Localization) fast positioning algorithm as an example.

[0060] Taking the indoor and outdoor model of MR as a ray tracing model as an example, based on the positioning results and the indoor and outdoor model of MR, the initial 3D position information of several users is corrected three times, specifically including: ① According to the 3D map parameters (i.e., positioning results), the ray tracing model (Volcano) forms a basic 3D library to strictly correspond the initial N-dimensional position information of several users to the map trajectory. ② Then use the historical positioning data feature library and the initial 3D position information of several users for feature matching to correct the initial 3D position information of several users (including indoor and outdoor, altitude, longitude and latitude). ③ Import the corrected results (i.e., the N-dimensional position information of several users) as 3D positions into the map for final calibration, and import the calibrated results into the historical positioning data feature library, and iteratively update the historical positioning data feature library for the next user position positioning of MR data.

[0061] S1023, based on the geographic location information of each N-dimensional grid, the N-dimensional location information of a number of users and a clustering algorithm, backfill the field information of the number of users to obtain the field information of each N-dimensional grid, wherein the clustering algorithm includes any one of the following: K-means algorithm, density-based spatial clustering with noise application DBSCAN algorithm and Mean Shift algorithm.

[0062] In this embodiment, Figure 5 As shown, the field information of several users is backfilled, specifically including: creating a data set based on the field information and N-dimensional location information of several users of MR data. The massive data set and the grid layer location (i.e., the geographic location information of each N-dimensional grid) are grid clustered and matched, and all data information of the grid corresponding to the position grid belonging to the grid ID is clustered based on the grid ID, and the field information based on the grid ID is output, as shown in Table 1. This embodiment can match the field information of massive MR data with the grid through a clustering algorithm, provide accurate, grid-level network performance data distribution, and thereby improve the accuracy of subsequent user experience rate evaluation.

[0063] Table 1 Field information based on grid ID

[0064]

[0065] It should be noted that the K-means algorithm is a common unsupervised clustering algorithm that iteratively assigns data to K cluster centers so that the distance between each data point and the nearest cluster center is minimized. The DBSCAN algorithm (Density-Based Spatial Clustering of Applications with Noise) is a density-based clustering algorithm that is suitable for spatial data clustering and can identify noise data well. The DBSCAN algorithm is particularly suitable for processing MR data with uneven distribution. It can identify high-density areas and isolated points during raster clustering. The Mean Shift clustering algorithm is a clustering method based on kernel density estimation. By moving data points, it gradually moves closer to the position with the highest density and finally forms a cluster. It is suitable for situations where the density distribution of data points is uneven.

[0066] Based on the geographic location information of each 3D grid, the 3D location information of several users and the K-means algorithm, the field information of several users is backfilled to obtain the field information of each 3D grid, specifically including: ① The 3D location information (longitude, latitude and altitude) of several users and the field information of several users are used as input feature vectors. ② According to the density of grid division and the number of grids in the target area, the number of clusters K is set; the geographic location information of K cluster centers is determined according to the geographic location information of each 3D grid. ③ The distance between the 3D location information of several users and the geographic location information of K cluster centers is calculated through multiple iterations, and several users are assigned to the closest cluster center, and the geographic location information of the cluster center is continuously updated. ④ Finally, the field information of several users is assigned to a grid ID, and the average value or other statistical features of each field information in the grid are calculated as the clustering result of the grid.

[0067] Based on the geographic location information of each 3D grid, the 3D location information of several users and the DBSCAN algorithm, the field information of several users is backfilled to obtain the field information of each 3D grid, including: ① The input data is also the 3D location information (longitude, latitude and altitude) of several users and the field information of several users. ② According to the geographic location information of each 3D grid, the neighborhood radius (eps) and the minimum number of points (minPts) are set. The neighborhood radius and the minimum number of points are used to determine whether the field information of several users belongs to an area with high enough density. ③ The DBSCAN algorithm traverses each point (i.e. user) in the data set to determine whether the number of points in its neighborhood is greater than minPts. If so, it is considered that all points in the neighborhood greater than minPts belong to the same cluster, otherwise they are marked as noise points. ④ Output results: A series of grid IDs and clustering results are obtained, and the field information in the grid is statistically summarized.

[0068] Based on the geographic location information of each 3D grid, the 3D location information of several users and the Mean Shift clustering algorithm, the field information of several users is backfilled to obtain the field information of each 3D grid, including: ① The input data is also the 3D location information (longitude, latitude and altitude) of several users and the field information of several users. ② Select an appropriate kernel function (such as a Gaussian kernel) and bandwidth parameter to determine the direction and distance of movement of the data point (that is, the field information and 3D location information of any user in the data set). ③ Continuously move the 3D location information of the data point until all data points gather in a high-density area to form a cluster. ④ Output results: According to the final 3D location information of the data point, assign it to the corresponding grid and output the statistical results after clustering.

[0069] In addition, if Figure 6 As shown, in the same grid, the service cell occupied by the user (i.e. Figure 6 There may be more than one cell 1, cell 2, and cell 3 in the grid. Since the distances from the serving cell to the current grid are different, the field information (i.e. Figure 6 There may be large differences in MCS1, RANK1, RB1, MCS2, RANK2, RB2, MCS3, RANK3, RB3). Therefore, no matter which clustering algorithm is used, it is also necessary to perform raster data processing on the statistical results after clustering, where raster data processing includes but is not limited to field statistical calculations, data screening, and outlier processing. Field statistical calculation: Perform statistical calculations on the field information of each raster, and the output includes at least one of the following: mean, standard deviation, maximum, and minimum. Data screening and outlier processing: According to the distribution characteristics of the field information of all rasters, screen out abnormal rasters, or further process the noise field information.

[0070] Specifically, the field information includes modulation and coding MCS field information, rank RANK field information and resource block RB field information.

[0071] It should be noted that MCS (Modulation and Coding Scheme, used to characterize air interface quality) defines a modulation scheme (Modulation) and redundant coding scheme (Code Rate) used when a time-frequency resource carries binary data. The higher the MCS, the higher the modulation mode and the corresponding transmission code rate. If other conditions (including but not limited to RANK and RB) remain unchanged, the higher the air interface requirements will be if you want to maintain the same bit error rate. In low-order MCS, the quantization error of the 256QAM table is larger than that of the 64QAM table, so the 64QAM table is preferred for the mid / far point in the scenario where the low-order MCS is required. For users covered by the same cell, the near point means that the probability of obtaining network resource capabilities and user experience rate is relatively high. The far point is affected by signal coverage, interference, and attenuation, and the user experience rate is relatively reduced. In order to better reflect the user experience rate capability, the user experience rate of edge users is usually evaluated as the user experience rate capability. Therefore, in order to restore the cell network capability rate, the MCS of the 64QAM table is used to evaluate the user experience rate, and the spectrum efficiency is determined according to the MCS selection order corresponding table. To sum up, when the base station makes modulation selection, the 256QAM table is selected for the mid / near point users, and the 64QAM table is selected for the mid / far point users. The base station periodically determines the switching of the two tables and selects MCS / CQI (Channel Quality Indication) / spectral efficiency threshold.

[0072] RANK (multipath environment or rank) refers to the rank of the MIMO (Multipe Input Multiple Output) channel matrix, which indicates the maximum number of layers that can transmit data simultaneously. The maximum RANK does not exceed the size of the short side of the MIMO channel matrix. That is, in the case of Massive MIMO, the maximum RANK value depends on the number of receiving antennas on the UE side, which is usually 4, indicating 1 to 4 channels. Simply put, the same time-frequency resources are divided into several parts in space for simultaneous transmission. When the time-frequency resources remain unchanged, the higher the RANK, the higher the actual throughput and the greater the rate.

[0073] RB (Resource Block) number: 5G (5th Generation Mobile Communication Technology) adopts TDD (Time Division Duplexing) networking, using 100M bandwidth (i.e. 273 RBs) as resources to provide user network services. According to the time-frequency distribution, in the frequency domain, 1 RB is composed of 12 subcarriers (Resource Element, RE), and in the time domain, each time slot has 14 symbols. Taking the 7:3 time slot ratio as an example, according to the uplink and downlink allocation, the downlink accounts for 70% and the uplink accounts for 30%. The network capacity of each cell remains consistent, that is, the total number of RBs and the number of RE / RB in the cell under the network are basically consistent.

[0074] S103: Evaluate the user experience rate of each N-dimensional grid based on the field information of each N-dimensional grid.

[0075] In this embodiment, the CQI Table is introduced and explained in detail in 3GPP (3rd Generation Partnership Project). Each equipment manufacturer uses the CQI reported by the terminal to map the wireless environment and finally obtain the user MCS. The base station determines the transmission efficiency and transmission quality of the user service through the MCS, that is, the user's rate capability is determined at the same time. Therefore, the air interface rate is mainly affected by MCS, RANK and the number of RBs. Usually, the air interface rate capability of the user's location is calculated based on the MCS of the main service cell at the user's location, the RANK at the user's location and the number of RBs of the user, specifically including: the air interface rate capability of the user's location = (X = MCS) × (Z = RB number) × (Y = Rank), where X is the MCS value of ≥ 90% of the sampling points. Since X evaluates the situation under full filling packets, RB corresponds to bandwidth, such as the number of RBs supported by TDD 100MHz carrier bandwidth is 273, and Y is the average rank value of the user.

[0076] The air interface rate is used to evaluate the maximum carrying capacity of the network and help operators understand and optimize the network architecture, while the user experience rate is used to measure the real user experience, analyze the service quality, and guide the improvement of user satisfaction. The air interface rate and the user experience rate are interdependent to a certain extent. A higher air interface rate can provide users with a better experience rate under ideal conditions. However, since the user experience rate is subject to end-to-end restrictions, it is not purely a reflection of the air interface rate capability of the wireless network. For example, when a user sends an instant message, the user experience rate may be lower than 10Mbps, but the actual network air interface has no interference and good coverage, and the actual air interface can reach an air interface rate of 1Gbps. Therefore, in order to isolate the user experience rate from the air interface rate capability, it is necessary to convert the air interface rate capability of the user's location to obtain the user experience rate. The conversion means include but are not limited to protocol tail packet removal.

[0077] To summarize, this embodiment can utilize the indicator field information used to calculate the air interface rate to calculate the user experience rate, that is, utilize the MCS field information, RANK field information and RB field information to calculate the user experience rate, comprehensively considering the efficiency of MCS, the effect of MIMO technology and the use of resource blocks, so that the evaluation of the user experience rate is more accurate and comprehensive.

[0078] Specifically, S103: based on the field information of each N-dimensional grid, evaluating the user experience rate of each N-dimensional grid, including steps S1031 and S1032:

[0079] S1031, performing mean regression calculation on the field information of each N-dimensional grid to obtain regression MCS field information, regression RANK field information and regression RB field information of each N-dimensional grid.

[0080] Specifically, S1031: performing mean regression calculation on the field information of each N-dimensional grid to obtain the regression MCS field information, regression RANK field information and regression RB field information of each N-dimensional grid, including: performing mean regression calculation on the field information of each N-dimensional grid according to formula (1):

[0081] Yt=α+βXt+εt(1), where Yt represents the regression field information at time t, Xt represents the field information at time t-1, α and β represent constants, and εt represents the mean regression error term at time t.

[0082] In this embodiment, since the user's field information is constantly changing, the user experience rate usually changes multiple times from the field information collected at time T to the time T+1 when the user experience rate is evaluated. Therefore, the user experience rate evaluated at time T+1 can often only represent the user experience rate at time T, and cannot achieve real-time user experience rate evaluation. Mean regression is a data method that first determines a statistical range (grid ID) and calculates the average value within the statistical range. Regardless of whether any field information within the statistical range is higher or lower than the value center (or mean), any field information within the statistical range will regress to the value center with a high probability. Therefore, this embodiment performs mean regression calculation on the field information of each N-dimensional grid at time T, such as CellDLInitMcs, Cell DL Rank and Cell DL CP Used RB Num in Table 1, and predicts the field information of each N-dimensional grid at time T+1 in advance to evaluate the user experience rate at time T+1. This can effectively improve the accuracy and stability of the user experience rate evaluation, provide real-time and reliable user experience insights in a dynamic environment, and help network operators adjust strategies in a timely manner and optimize resource allocation, thereby maximizing user satisfaction and service quality.

[0083] S1032, calculating the user experience rate of each N-dimensional grid based on the regression MCS field information, the regression RANK field information, and the regression RB field information of each N-dimensional grid.

[0084] Specifically, S1032: calculating the user experience rate of each N-dimensional grid based on the regression MCS field information, the regression RANK field information and the regression RB field information of each N-dimensional grid, including: calculating the user experience rate of each N-dimensional grid according to formula (2):

[0085] Grid Rate=

[0086] AVERAGE(MCS)×AVERAGE(Rank)×AVERAGE(RB Num) (2),

[0087] Among them, Grid Rate represents the user experience rate of each N-dimensional grid, MCS represents the regression MCS field information of each N-dimensional grid, Rank represents the regression RANK field information of each N-dimensional grid, and RB Num represents the regression RB field information of each N-dimensional grid.

[0088] In this embodiment, according to formula (2), the user experience rate of each 3-dimensional grid in Table 1 is calculated as AVERAGE (Cell DL Init Mcs) × AVERAGE (Cell DL Rank) × AVERAGE (Cell DL CP Used RB Num).

[0089] It should be noted that after evaluating the user experience rate, it is usually accompanied by the application and analysis of the user experience rate to solve different scenario requirements. The application and analysis of the user experience rate usually includes the geographical presentation of the user experience rate, the aggregation of the scenario-level user experience rate and the optimization of the user experience rate. Through the application and analysis of the user experience rate, after identifying the problem, measures are taken to optimize it, such as adjusting bandwidth allocation, improving network equipment or adjusting service strategies to improve the overall user experience. That is to say, after implementing the analysis of network performance data through steps S101 and S102, and implementing the calculation of the user experience rate through step S103, this embodiment can also geographically present the evaluated user experience rate, aggregate and optimize the scenario-level user experience rate to identify areas or scenarios with low user experience rates and improve low user experience rates.

[0090] The geographical presentation of the user experience rate refers to displaying the calculation results of the user experience rate on a map in an intuitive way. Different ranges of the user experience rate can be used to represent the impact of network speed on user experience. Therefore, different ranges of user experience rates are displayed with different symbols. Taking N=2 as an example, the geographical presentation of the user experience rate is as follows: Figure 7 As shown, the user experience rate of each 2D grid is Figure 7 The 2D grid capabilities are as follows: a 2D grid with a user experience rate of less than 10 Mbps indicates a very poor network experience, which may result in slow web page loading, frequent video buffering, and high latency in online games; a 2D grid with a user experience rate in the range of [10, 20) Mbps indicates that the user experience is starting to improve, but buffering and slow loading speeds may still be encountered, especially when using multiple devices at the same time; a 2D grid with a user experience rate in the range of [20, 50) Mbps indicates that it is sufficient to support streaming services (such as 1080p video) and basic multi-device operations, but problems may occur during peak hours; a 2D grid with a user experience rate in the range of [50, 100) Mbps indicates that the user experience is relatively good and can support 4K video streaming, online games, and other bandwidth-intensive activities; a 2D grid with a user experience rate above 100 Mbps indicates that the network connection is very good and can support multiple devices at the same time, smooth high-resolution video playback, large file downloads, and other applications that are highly dependent on bandwidth.

[0091] like Figure 8 As shown, this embodiment can also use color coding (such as red for low rate, green for high rate) to visualize the user experience rate distribution of each 3D grid, so that network optimization personnel can quickly identify low-rate areas. By overlaying the user experience rate of each 3D grid on the electronic map, the spatial distribution of network performance and problem areas can be clearly seen, which can greatly improve the efficiency of optimization; it can also be combined with the actual terrain and building distribution to assist in evaluating network performance issues in coverage blind areas and complex scenarios.

[0092] The aggregation of scene-level user experience rate refers to the aggregation and analysis of the user experience rate of each 3D grid in different geographical scenes (such as commercial areas, residential areas, and transportation hubs) in order to optimize network performance in a more targeted manner. The aggregation of scene-level user experience rate specifically includes: ① According to the Geographic Information System (GIS) data, the user experience rate of each 3D grid is assigned to different scene categories (such as outdoor urban areas, indoor commercial centers, subway stations, and tunnels). ② Attach a scene label to each grid so that it can be classified and aggregated according to the scene label in the aggregation analysis. ③ Use K-means or DBSCAN algorithm to group the grid-level user experience rate of the same scene label, where the grid-level experience rate refers to the user experience rate of each 3D grid. Through the aggregation of scene-level user experience rate, common problems of each scene can be identified and the performance bottleneck of the network in a specific scene can be revealed; through the aggregation and comparison of scene-level problems, network optimization personnel can formulate more refined optimization strategies to meet the needs of users in different scenes.

[0093] Use K-means or DBSCAN algorithm to group the grid-level user experience rate of the same scene label, specifically including: ① Select the feature vector of each 3D grid: grid-level user experience rate, interference level and number of users. The interference level includes but is not limited to RSRP (Reference Signal Receiving Power) and SINR (Signal to Interference plus Noise Ratio). ② Set the initial parameters of clustering, such as the number of clusters K or the minimum cluster density (for DBSCAN). ③ Execute the clustering algorithm and output the center point and range of each cluster group. ④ Scenario summary formula: Summarize the clustering results in the same scene and use the formula: Calculate the scene-level user experience rate and the distribution characteristics of the scene-level user experience rate, where N is the total number of 3D grids in the scene, Grid Rate i is the user experience rate of the i-th 3D grid.

[0094] The optimization of user experience rate includes root cause location and formulation of targeted optimization solutions. Root cause location aims to find the root cause behind the grid of low user experience rate through in-depth analysis. The optimization of user experience rate combines multiple algorithms, such as multivariate linear regression, decision tree, and anomaly detection, to investigate the key factors affecting network performance. Common root causes include severe signal interference, uneven resource allocation, and poor coverage. Root cause location can identify the main reasons for the rate drop through multi-dimensional analysis of MR data (such as signal strength, signal-to-noise ratio, and resource block occupancy). Optimization solutions include adjusting antenna direction, optimizing resource scheduling, or increasing network coverage.

[0095] Multidimensional analysis of MR data, specifically including: combining multiple indicators of MR data, such as RSRP, SINR, CQI, resource block occupancy (RB Utilization), and analyzing the impact of each indicator on user experience rate through a multidimensional regression model. For example: using a multivariate linear regression or decision tree model, that is, the formula user experience rate = β 0 +β 1 ×RSRP+β 2 ×SINR+β 3 ×CQI, we can find the indicator that has the greatest impact on the user experience rate, where β i is the regression coefficient of each indicator, which is determined by training the model.

[0096] In addition to combining multiple indicators of MR data and analyzing the impact of each indicator on the user experience rate through a multidimensional regression model, this embodiment can also locate the root cause of the problem area based on anomaly detection algorithms and / or path loss and interference analysis models to optimize the user experience rate of the problem area based on the root cause location results.

[0097] Based on the anomaly detection algorithm, the root cause of the problem area is located, including: using statistical methods (such as Z-score or Tukey's fences) or machine learning anomaly detection algorithms (such as Isolation Forest, LOF (Local Outliers Factor)) to identify outliers in MR data and analyze the impact of outliers on user experience rate.

[0098] Based on the path loss and interference analysis model, the root cause of the problem area is located, specifically including: calculating the path loss of each 3D grid according to the wireless propagation model, and locating possible interference sources in combination with interference analysis, where the wireless propagation model includes but is not limited to: COST231-Hata model and Okumura-Hata model. Locating possible interference sources in combination with interference analysis, specifically including: finding possible interfering cells or user equipment by analyzing the RSRP and SINR distribution of cells within and around the grid.

[0099] Taking N=2 and the target area as the urban area around Lake A and Scenic Area B, after evaluating the grid-level user experience rate and calculating the scene-level user experience rate, there are 321 scenarios in the urban area around Lake A and Scenic Area B where the user experience rate is at risk, accounting for 4.4% of the problem scenarios. Through TOP scenario screening, that is, sorting the scenarios with insufficient user experience rate, the TOP k scenarios in the sorting results are screened out to determine as the target scenarios, and scenario a is the target scenario among the scenarios with insufficient user experience rate. The scene-level user experience rate of scenario a is geographically presented and the root cause is located, and it is determined that scenario a is mainly a weak coverage problem. The problem is characterized by the 5G NR downlink rate capability being less than 100Mbps, accounting for 21%, such as Fig. 9 shown.

[0100] The scenario-level user experience rate of scenario a is geographically presented and the root cause is located, including: by geographically presenting all grid-level user experience rates in scenario a, the problem area in scenario a can be directly presented, and it is determined that there is a 5G disconnection problem in the problem area, and the root cause of the problem area is located. The average RSRP, SINR, downlink rate and comprehensive coverage of the 4G network in the problem area are -66dBm, 39db, 35Mbps, and 100%, respectively. The average RSRP, SINR, downlink rate and comprehensive coverage of the 5G network are -98dBm, 1.6db, 40Mbps, and 35.71%, respectively. The 4G coverage is good, and the 5G is disconnected internally. It is determined that there is 4G indoor distribution but no 5G indoor distribution in the problem area. The optimization plan is formulated as follows: share the common mode of indoor distribution, and open NR2.1 indoor distribution to fill the 5G coverage gap. After the opening of NR2.1G, the network coverage and user experience have been greatly improved. For example Fig.10 The coverage and rate before optimization shown and Fig.11 After the optimization of coverage and rate, the 5G network coverage changes from partial disconnection to continuous coverage, and the overall RSRP value (i.e. Fig.10 and Fig.11 The NR SS-RSRP in the network is improved from -98dBm to about -70dBm, the downlink rate is increased from 32Mbps to 113Mbps, and the uplink rate is increased from 5Mbps to 66Mbps, and the user perception experience is fully guaranteed.

[0101] The five links of network performance data analysis, user experience rate calculation, user experience rate geographic presentation, scenario-level user experience rate aggregation and user experience rate optimization are closely linked to form a complete closed-loop optimization process. Through the refined evaluation of network quality and rate capability, as well as geographic presentation and scenario-level aggregation, the fundamental problems of low-rate areas can be accurately located, the optimization process is gradually promoted, and the continuous improvement of network performance can be achieved, which can not only improve the coverage quality of the network, but also significantly improve the actual experience of users in different scenarios, providing strong technical support for network optimization and user experience assurance.

[0102] The present embodiment provides a method for evaluating user experience rate, which performs multi-dimensional rasterization on MR data in a target area, and then geographically presents the field information in the MR data in a fine-grained raster, thereby effectively improving the geographical presentation effect of network performance data. The field information under the fine-grained raster can be used to better evaluate the raster-level user experience rate, and then the raster-level user experience rate can be applied to the user experience and customer satisfaction evaluation in different business scenarios, thereby improving the evaluation accuracy and applicability of the user experience rate, achieving accurate evaluation of the user experience rate, reducing the rasterization granularity of the network performance data, improving the geographical presentation effect of the network performance data, expanding the application of the user experience and customer satisfaction evaluation in different business scenarios, and improving the evaluation accuracy, efficiency and applicability of the user experience rate.

[0103] Embodiment 2:

[0104] like Fig.12 As shown, this embodiment provides a method for optimizing user experience rate. The method for optimizing user experience rate includes:

[0105] S202 : Obtain a grid-level user experience rate, where the grid-level user experience rate refers to a user experience rate of each N-dimensional grid.

[0106] In this embodiment, after evaluating the user experience rate, it is usually accompanied by application and analysis of the user experience rate to solve different scenario requirements. The application and analysis of the user experience rate usually includes the geographical presentation of the user experience rate, the aggregation of the scenario-level user experience rate and the optimization of the user experience rate. Through the application and analysis of the user experience rate, after identifying the problem, measures are taken to optimize it, such as adjusting bandwidth allocation, improving network equipment or adjusting service policies to improve the overall user experience. That is to say, after implementing the analysis of network performance data through steps S101 and S102, and implementing the calculation of the user experience rate through step S103, this embodiment can also geographically present the evaluated user experience rate, aggregate and optimize the scenario-level user experience rate to identify areas or scenarios with low user experience rates and improve the low user experience rates.

[0107] S202, clustering all grid-level user experience rates in the same scenario, and calculating the scenario-level user experience rate, wherein the scenario-level user experience rate refers to the average of all grid-level user experience rates in each scenario.

[0108] In this embodiment, the aggregation of scene-level user experience rate refers to the aggregation and analysis of the user experience rate of each 3D grid in different geographical scenes (such as commercial areas, residential areas, and transportation hubs) in order to optimize network performance in a more targeted manner. The aggregation link of scene-level user experience rate is specifically included: ① According to the Geographic Information System (GIS) data, the user experience rate of each 3D grid is assigned to different scene categories (such as outdoor urban areas, indoor commercial centers, subway stations, and tunnels). ② Attach a scene label to each grid so that it can be classified and aggregated according to the scene label in the aggregation analysis. ③ Use K-means or DBSCAN algorithm to group the grid-level user experience rates of the same scene label, where the grid-level experience rate refers to the user experience rate of each 3D grid. Through the aggregation of scene-level user experience rate, common problems of each scene can be identified, revealing the performance bottleneck of the network in a specific scene; through the summary and comparison of problems at the scene level, network optimization personnel can formulate more refined optimization strategies to meet the needs of users in different scenes.

[0109] Use K-means or DBSCAN algorithm to group the grid-level user experience rate of the same scene label, specifically including: ① Select the feature vector of each 3D grid: grid-level user experience rate, interference level and number of users. The interference level includes but is not limited to RSRP (Reference Signal Receiving Power) and SINR (Signal to Interference plus Noise Ratio). ② Set the initial parameters of clustering, such as the number of clusters K or the minimum cluster density (for DBSCAN). ③ Execute the clustering algorithm and output the center point and range of each cluster group. ④ Scenario summary formula: Summarize the clustering results in the same scene and use the formula: Calculate the scene-level user experience rate and the distribution characteristics of the scene-level user experience rate, where N is the total number of 3D grids in the scene, Grid Rate i is the user experience rate of the i-th 3D grid.

[0110] S203, visually presenting the problem area, where the problem area refers to an area where the grid-level user experience rate or the scene-level user experience rate is lower than a preset threshold.

[0111] In this embodiment, the geographical presentation of the user experience rate refers to displaying the calculation result of the user experience rate on a map in an intuitive manner, wherein different ranges of the user experience rate can be used to represent the impact of network speed on user experience, and therefore, different ranges of user experience rates are displayed with different identifiers. Taking N=2 as an example, the geographical presentation of the user experience rate is as follows: Figure 7 As shown, the user experience rate of each 2D grid is Figure 7 The 2D grid capabilities are as follows: a 2D grid with a user experience rate of less than 10 Mbps indicates a very poor network experience, which may result in slow web page loading, frequent video buffering, and high latency in online games; a 2D grid with a user experience rate in the range of [10, 20) Mbps indicates that the user experience is starting to improve, but buffering and slow loading speeds may still be encountered, especially when using multiple devices at the same time; a 2D grid with a user experience rate in the range of [20, 50) Mbps indicates that it is sufficient to support streaming services (such as 1080p video) and basic multi-device operations, but problems may occur during peak hours; a 2D grid with a user experience rate in the range of [50, 100) Mbps indicates that the user experience is relatively good and can support 4K video streaming, online games, and other bandwidth-intensive activities; a 2D grid with a user experience rate above 100 Mbps indicates that the network connection is very good and can support multiple devices at the same time, smooth high-resolution video playback, large file downloads, and other applications that are highly dependent on bandwidth.

[0112] like Figure 8 As shown, this embodiment can also use color coding (such as red for low rate, green for high rate) to visualize the user experience rate distribution of each 3D grid, so that network optimization personnel can quickly identify low-rate areas. By overlaying the user experience rate of each 3D grid on the electronic map, the spatial distribution of network performance and problem areas can be clearly seen, which can greatly improve the efficiency of optimization; it can also be combined with the actual terrain and building distribution to assist in evaluating network performance issues in coverage blind areas and complex scenarios.

[0113] S204, based on the anomaly detection algorithm and / or the path loss and interference analysis model, root cause location is performed on the problem area to optimize the user experience rate of the problem area based on the root cause location result.

[0114] In this embodiment, the optimization link of the user experience rate includes root cause location and formulation of targeted optimization solutions. Root cause location aims to find the root cause behind the grid with low user experience rate through in-depth analysis. The optimization link of the user experience rate combines multiple algorithms, such as multivariate linear regression, decision tree, and anomaly detection, to investigate the key factors affecting network performance. Common root causes include severe signal interference, uneven resource allocation, and poor coverage. Root cause location can identify the main reasons for the rate drop through multi-dimensional analysis of MR data (such as signal strength, signal-to-noise ratio, and resource block occupancy). Optimization solutions include adjusting antenna direction, optimizing resource scheduling, or increasing network coverage.

[0115] Multidimensional analysis of MR data, specifically including: combining multiple indicators of MR data, such as RSRP, SINR, CQI, resource block occupancy (RB Utilization), and analyzing the impact of each indicator on user experience rate through a multidimensional regression model. For example: using a multivariate linear regression or decision tree model, that is, the formula user experience rate = β 0 +β 1 ×RSRP+β 2 ×SINR+β 3 ×CQI, we can find the indicator that has the greatest impact on the user experience rate, where β i is the regression coefficient of each indicator, which is determined by training the model.

[0116] In addition to combining multiple indicators of MR data and analyzing the impact of each indicator on the user experience rate through a multidimensional regression model, this embodiment can also locate the root cause of the problem area based on anomaly detection algorithms and / or path loss and interference analysis models to optimize the user experience rate of the problem area based on the root cause location results.

[0117] Based on the anomaly detection algorithm, the root cause of the problem area is located, including: using statistical methods (such as Z-score or Tukey's fences) or machine learning anomaly detection algorithms (such as Isolation Forest, LOF (Local Outliers Factor)) to identify outliers in MR data and analyze the impact of outliers on user experience rate.

[0118] Based on the path loss and interference analysis model, the root cause of the problem area is located, specifically including: calculating the path loss of each 3D grid according to the wireless propagation model, and locating possible interference sources in combination with interference analysis, where the wireless propagation model includes but is not limited to: COST231-Hata model and Okumura-Hata model. Locating possible interference sources in combination with interference analysis, specifically including: finding possible interfering cells or user equipment by analyzing the RSRP and SINR distribution of cells within and around the grid.

[0119] Taking N=2 and the target area as the urban area around Lake A and Scenic Area B, after evaluating the grid-level user experience rate and calculating the scene-level user experience rate, there are 321 scenarios in the urban area around Lake A and Scenic Area B where the user experience rate is at risk, accounting for 4.4% of the problem scenarios. Through TOP scenario screening, that is, sorting the scenarios with insufficient user experience rate, the TOP k scenarios in the sorting results are screened out to determine as the target scenarios, and scenario a is the target scenario among the scenarios with insufficient user experience rate. The scene-level user experience rate of scenario a is geographically presented and the root cause is located, and it is determined that scenario a is mainly a weak coverage problem. The problem is characterized by the 5G NR downlink rate capability being less than 100Mbps, accounting for 21%, such as Fig. 9 shown.

[0120] The scenario-level user experience rate of scenario a is geographically presented and the root cause is located, including: by geographically presenting all grid-level user experience rates in scenario a, the problem area in scenario a can be directly presented, and it is determined that there is a 5G disconnection problem in the problem area, and the root cause of the problem area is located. The average RSRP, SINR, downlink rate and comprehensive coverage of the 4G network in the problem area are -66dBm, 39db, 35Mbps, and 100%, respectively. The average RSRP, SINR, downlink rate and comprehensive coverage of the 5G network are -98dBm, 1.6db, 40Mbps, and 35.71%, respectively. The 4G coverage is good, and the 5G is disconnected internally. It is determined that there is 4G indoor distribution but no 5G indoor distribution in the problem area. The optimization plan is formulated as follows: share the common mode of indoor distribution, and open NR2.1 indoor distribution to fill the 5G coverage gap. After the opening of NR2.1G, the network coverage and user experience have been greatly improved. For example Fig.10 The coverage and rate before optimization shown and Fig.11 After the optimization of coverage and rate, the 5G network coverage changes from partial disconnection to continuous coverage, and the overall RSRP value (i.e. Fig.10 and Fig.11 The NR SS-RSRP in the network is improved from -98dBm to about -70dBm, the downlink rate is increased from 32Mbps to 113Mbps, and the uplink rate is increased from 5Mbps to 66Mbps, and the user perception experience is fully guaranteed.

[0121] The five links of network performance data analysis, user experience rate calculation, user experience rate geographic presentation, scenario-level user experience rate aggregation and user experience rate optimization are closely linked to form a complete closed-loop optimization process. Through the refined evaluation of network quality and rate capability, as well as geographic presentation and scenario-level aggregation, the fundamental problems of low-rate areas can be accurately located, the optimization process is gradually promoted, and the continuous improvement of network performance can be achieved, which can not only improve the coverage quality of the network, but also significantly improve the actual experience of users in different scenarios, providing strong technical support for network optimization and user experience assurance.

[0122] The present embodiment provides a method for optimizing user experience rate. Through refined evaluation of network quality and rate capability, as well as geographic presentation and scenario-level convergence, the fundamental problems of low-rate areas can be accurately located. The optimization process is gradually advanced to achieve continuous improvement of network performance. This can not only improve the coverage quality of the network, but also significantly improve the actual experience of users in different scenarios. It provides strong technical support for network optimization and user experience assurance, realizes accurate evaluation of user experience rate, reduces the rasterization granularity of network performance data, improves the geographic presentation effect of network performance data, expands the application of user experience and customer satisfaction evaluation in different business scenarios, and improves the evaluation accuracy, efficiency and applicability of user experience rate.

[0123] Embodiment 3:

[0124] like Fig.13 As shown, this embodiment provides an evaluation device for user experience rate, including: an acquisition module 31, a rasterization module 32 and an evaluation module 33, the acquisition module 31 is used to collect measurement report MR data of a target area within a preset time period, wherein the MR data includes location information and field information of several users, the rasterization module 32 is connected to the acquisition module 31, and is used to perform N-dimensional rasterization on the MR data of the target area to obtain field information of each N-dimensional grid, wherein N includes one of 2, 3, and 4, and the evaluation module 33 is connected to the rasterization module 32, and is used to evaluate the user experience rate of each N-dimensional grid based on the field information of each N-dimensional grid.

[0125] Specifically, the rasterization module 32 includes: a rasterization unit 321, a user positioning unit 322 and a backfilling unit 323. The rasterization unit 321 is used to obtain the geographic location information of the target area, and to rasterize and layer the target area to obtain the geographic location information of each N-dimensional grid, wherein the N-dimensional grid is used to characterize grids at different levels. The user positioning unit 322 is used to perform user location positioning on MR data to obtain N-dimensional location information of several users. The backfilling unit 323 is used to backfill the field information of several users based on the geographic location information of each N-dimensional grid, the N-dimensional location information of several users and a clustering algorithm to obtain the field information of each N-dimensional grid, wherein the clustering algorithm includes any one of the following: K-means algorithm, density-based noisy application spatial clustering DBSCAN algorithm and Mean Shift algorithm.

[0126] Specifically, the user positioning unit 322 includes: an aggregation subunit, an estimation subunit, a ray intersection subunit and a calibration subunit. The aggregation subunit is used to perform MR aggregation on the MR data to obtain a number of MR groups. The estimation subunit is used to perform arrival angle ranging AOA estimation on the MR data to obtain arrival angle information of a number of users. The ray intersection subunit is used to perform ray intersection based on a topological relationship on the arrival angle information of the same MR group to obtain initial N-dimensional position information of a number of users. The calibration subunit is used to perform MR calibration on the initial N-dimensional position information of a number of users based on a ray tracing model to obtain N-dimensional position information of a number of users.

[0127] Specifically, the evaluation module 33 includes: a first calculation unit 331 and a second calculation unit 332, the first calculation unit 331 is used to perform mean regression calculation on the field information of each N-dimensional grid to obtain the regression MCS field information, regression RANK field information and regression RB field information of each N-dimensional grid, and the second calculation unit 332 is used to calculate the user experience rate of each N-dimensional grid based on the regression MCS field information, regression RANK field information and regression RB field information of each N-dimensional grid.

[0128] Specifically, the first calculation unit 331 includes: a first calculation subunit, configured to perform mean regression calculation on the field information of each N-dimensional grid according to formula (1):

[0129] Yt=α+βXt+εt(1), where Yt represents the regression field information at time t, Xt represents the field information at time t-1, α and β represent constants, and εt represents the mean regression error term at time t.

[0130] Specifically, the second calculation unit 332 includes: a second calculation subunit, configured to calculate the user experience rate of each N-dimensional grid according to formula (2):

[0131] Grid Rate=

[0132] AVERAGE(MCS)×AVERAGE(Rank)×AVERAGE(RB Num) (2),

[0133] Among them, Grid Rate represents the user experience rate of each N-dimensional grid, MCS represents the regression MCS field information of each N-dimensional grid, Rank represents the regression RANK field information of each N-dimensional grid, and RB Num represents the regression RB field information of each N-dimensional grid.

[0134] It can be understood that the above-mentioned user experience rate evaluation device executes the user experience rate evaluation method corresponding to the embodiment 1 provided above. Therefore, the beneficial effects that can be achieved can refer to the beneficial effects of the scheme corresponding to the user experience rate evaluation method of the embodiment 1 above, and will not be repeated here.

[0135] Embodiment 4:

[0136] like Fig.14 As shown, this embodiment also provides an optimization device for user experience rate, including: an acquisition module 41, a clustering module 42, a visualization module 43 and a root cause positioning module 44, the acquisition module 41 is used to acquire the grid-level user experience rate, wherein the grid-level user experience rate refers to the user experience rate of each N-dimensional grid, the clustering module 42 is connected to the acquisition module 41, and is used to cluster all grid-level user experience rates in the same scene and calculate the scene-level user experience rate, the visualization module 43 is connected to the clustering module 42, and is used to visualize the problem area where the grid-level user experience rate or the scene-level user experience rate is lower than a preset threshold, and the root cause positioning module 44 is connected to the visualization module 43, and is used to locate the root cause of the problem area based on the anomaly detection algorithm and / or the path loss and interference analysis model, so as to optimize the user experience rate of the problem area based on the root cause positioning result.

[0137] It can be understood that the above-mentioned user experience rate optimization device performs the user experience rate optimization method corresponding to the embodiment 2 provided above. Therefore, the beneficial effects that can be achieved can refer to the beneficial effects of the solution corresponding to the user experience rate optimization method of the embodiment 2 above. Embodiment 5 will not be repeated here:

[0138] This embodiment also provides an electronic device, including a memory and a processor, wherein the memory stores a computer program, and the processor is configured to run the computer program to implement the method for evaluating the user experience rate in the above-mentioned embodiment 1 or the method for optimizing the user experience rate in embodiment 2.

[0139] Embodiment 6:

[0140] This embodiment also provides a computer-readable storage medium on which a computer program is stored. When the computer program is executed by a processor, the method for evaluating the user experience rate in the above-mentioned embodiment 1 or the method for optimizing the user experience rate in embodiment 2 is implemented.

[0141] It is to be understood that the above embodiments are merely exemplary embodiments used to illustrate the principles of the present invention, but the present invention is not limited thereto. For those of ordinary skill in the art, various modifications and improvements can be made without departing from the spirit and essence of the present invention, and these modifications and improvements are also considered to be within the scope of protection of the present invention.

Claims

1. A method for evaluating user experience rate, characterized in that: include: Collecting measurement report MR data of the target area within a preset time period, wherein the MR data includes field information of several users; Performing N-dimensional rasterization on the MR data of the target area to obtain field information of each N-dimensional raster, wherein N includes one of 2, 3, and 4; Based on the field information of each N-dimensional grid, the user experience rate of each N-dimensional grid is evaluated.

2. The method for evaluating user experience rate according to claim 1, characterized in that: The MR data of the target area is gridded in N dimensions to obtain field information of each N-dimensional grid, specifically including: Obtaining geographic location information of the target area, and performing rasterization and raster layering processing on the target area to obtain geographic location information of each N-dimensional raster, wherein the N-dimensional raster is used to represent rasters at different levels; Perform user location positioning on MR data to obtain N-dimensional location information of several users; Based on the geographic location information of each N-dimensional grid, the N-dimensional location information of several users and a clustering algorithm, the field information of several users is backfilled to obtain the field information of each N-dimensional grid, wherein the clustering algorithm includes any one of the following: K-means algorithm, density-based spatial clustering with noise application DBSCAN algorithm and Mean Shift algorithm.

3. The method for evaluating user experience rate according to claim 2, characterized in that: The user position positioning is performed on the MR data to obtain N-dimensional position information of a plurality of users, specifically including: Perform MR aggregation on the MR data to obtain several MR groups; Perform AOA estimation on MR data to obtain the angle of arrival information of several users; Performing ray intersection based on topological relationship on the arrival angle information of the same MR group to obtain initial N-dimensional position information of several users; The initial N-dimensional position information of several users is MR calibrated based on the ray tracing model to obtain the N-dimensional position information of several users.

4. The method for evaluating user experience rate according to claim 1, characterized in that: The field information includes modulation and coding MCS field information, rank RANK field information and resource block RB field information. The step of evaluating the user experience rate of each N-dimensional grid based on the field information of each N-dimensional grid specifically includes: Perform mean regression calculation on the field information of each N-dimensional grid to obtain regression MCS field information, regression RANK field information and regression RB field information of each N-dimensional grid; Based on the regression MCS field information, regression RANK field information and regression RB field information of each N-dimensional grid, the user experience rate of each N-dimensional grid is calculated.

5. The method for evaluating user experience rate according to claim 4, characterized in that: The mean regression calculation is performed on the field information of each N-dimensional grid to obtain the mean regression field information of each N-dimensional grid, specifically including: According to formula (1), the mean regression calculation is performed on the field information of each N-dimensional grid: Yt=α+βXt+εt(1), where Yt represents the regression field information at time t, Xt represents the field information at time t-1, α and β represent constants, and εt represents the mean regression error term at time t.

6. The method for evaluating user experience rate according to claim 4, characterized in that: The calculating of the user experience rate of each N-dimensional grid based on the regression MCS field information, the regression RANK field information and the regression RB field information of each N-dimensional grid specifically includes: According to formula (2), the user experience rate of each N-dimensional grid is calculated: Grid Rate= AVERAGE(MCS)×AVERAGE(Rank)×AVERAGE(RB Num) (2), Among them, Grid Rate represents the user experience rate of each N-dimensional grid, MCS represents the regression MCS field information of each N-dimensional grid, Rank represents the regression RANK field information of each N-dimensional grid, and RB Num represents the regression RB field information of each N-dimensional grid.

7. A method for optimizing user experience rate, characterized in that: include: Obtaining a grid-level user experience rate evaluated by the method for evaluating user experience rate according to any one of claims 1 to 6, wherein the grid-level user experience rate refers to the user experience rate of each N-dimensional grid; Cluster all grid-level user experience rates in the same scenario and calculate the scenario-level user experience rate, where the scenario-level user experience rate refers to the average of all grid-level user experience rates in each scenario; Visually present the problem area, where the problem area refers to the area where the grid-level user experience rate or the scene-level user experience rate is lower than the preset threshold; Based on the anomaly detection algorithm and / or the path loss and interference analysis model, the root cause of the problem area is located to optimize the user experience rate of the problem area based on the root cause location result.

8. A device for evaluating user experience rate, characterized in that: include: Acquisition module, rasterization module and evaluation module, The acquisition module is used to collect the measurement report MR data of the target area within a preset time period, wherein the MR data includes the location information and field information of several users, The rasterization module is connected to the acquisition module and is used to perform N-dimensional rasterization on the MR data of the target area to obtain field information of each N-dimensional raster, where N includes one of 2, 3, and 4. The evaluation module is connected to the gridding module and is used to evaluate the user experience rate of each N-dimensional grid based on the field information of each N-dimensional grid.

9. A device for optimizing user experience rate, characterized in that: include: Acquisition module, clustering module, visualization module and root cause location module, an acquisition module, configured to acquire a grid-level user experience rate evaluated by the user experience rate evaluation device according to claim 8, wherein the grid-level user experience rate refers to the user experience rate of each N-dimensional grid, The clustering module is connected to the acquisition module to cluster all grid-level user experience rates in the same scene and calculate the scene-level user experience rate. The visualization module, connected with the clustering module, is used to visualize the problem areas where the grid-level user experience rate or scene-level user experience rate is lower than the preset threshold. The root cause location module is connected to the visualization module and is used to locate the root cause of the problem area based on the anomaly detection algorithm and / or the path loss and interference analysis model, so as to optimize the user experience rate of the problem area based on the root cause location result.

10. An electronic device, characterized in that: It includes a memory and a processor, wherein the memory stores a computer program, and the processor is configured to run the computer program to implement a method for evaluating a user experience rate as described in any one of claims 1 to 6, or a method for optimizing a user experience rate as described in claim 7.

11. A computer-readable storage medium having a computer program stored thereon, characterized in that: When the computer program is executed by a processor, it implements a method for evaluating a user experience rate as described in any one of claims 1 to 6, or a method for optimizing a user experience rate as described in claim 7.

Images