Method and apparatus for evaluating functional characteristics simulation of mobile networks

Abstract

The application provides a function characteristic simulation evaluation method and device for a mobile network, which comprises the following steps: obtaining a first sub-period KPI index matrix before single function characteristic application of a target pilot application area and a second sub-period KPI index matrix after the single function characteristic application; determining whether an index in a difference value matrix appears significant change based on a preset confidence level and a difference value matrix of the first sub-period KPI index matrix and the second sub-period KPI index matrix; performing sparse processing and / or median processing on the difference value matrix based on a judgment result, to obtain a sparse median matrix; and obtaining a function evaluation result of single function characteristic application of a to-be-applied area based on the sparse median matrix, to guide corresponding function characteristic application and simulation evaluation of other areas. The whole method is used for effectively improving function characteristic utilization efficiency, realizing effective simulation application and accurate evaluation of 4G or 5G cell function characteristics, and guiding reasonable and effective application of function characteristics.

Description

Technical Field

[0001] This invention relates to the field of communication technology, and in particular to a method and apparatus for simulating and evaluating the functional characteristics of mobile networks. Background Technology

[0002] LTE (Long Term Evolution) is a long-term evolution of technical standards developed by the 3GPP (3rd Generation Partnership Project) organization. Through the application of LTE's functional characteristics, LTE networks are capable of providing download speeds of 300 Mbit / s and upload speeds of 75 Mbit / s. 5G (5th generation mobile networks) is the fifth generation of mobile communication technology network developed by the 3GPP organization. The ITU IMT-2020 specification requires speeds of up to 20 Gbit / s, enabling wide channel bandwidth and high-capacity MIMO. It mainly includes three major application scenarios: eMBB, uRLLC, and mMTC.

[0003] The proper application of existing 4G / 5G functionalities can effectively improve network awareness. Before these functionalities are implemented on the network, pilot applications will be conducted to determine their characteristics, including both optimization effects and negative impacts. After the pilot programs are completed, based on the actual application results at each stage, grid-level applications and network-wide applications and evaluations will be carried out sequentially.

[0004] The current 4 / 5G functional applications are mainly based on cell traffic or cell scenario applications. The application strategy planning is not refined enough, the overall adjustment process is repetitive with many steps and long cycles, and the evaluation of the combined application of 4 / 5G multi-functional features is lacking.

[0005] Current evaluation methods are relatively crude, requiring multiple rounds of feature verification before definitive application effects can be obtained to guide subsequent deployment, evaluation, and rollback. The overall adjustment cycle is too long, and given the backdrop of 4G frequency decommissioning and the massive user growth and conversion during 5G deployment, it's impossible to accurately guarantee the timely effectiveness of application strategies in relevant scenarios. Given the ineffectiveness of application strategies, significant resources and manpower will be needed for secondary optimization of 4G / 5G functionalities.

[0006] Meanwhile, in the current network, there are situations where multiple functional resources are used simultaneously in the same scenario. Since it is impossible to evaluate the application effectiveness of a single feature in a scenario where multiple features are shared under the current application conditions, the actual evaluation results may be inaccurate.

[0007] Therefore, how to provide a functional characteristic simulation and evaluation scheme for mobile networks that can effectively simulate and accurately evaluate the functional characteristics of 4 / 5G cells and guide the reasonable and effective application of functional characteristics is a technical problem that urgently needs to be solved by those skilled in the art. Summary of the Invention

[0008] This invention provides a method and apparatus for simulating and evaluating the functional characteristics of mobile networks, which can effectively simulate and accurately evaluate the functional characteristics of 4 / 5G cells, and guide the reasonable and effective application of functional characteristics.

[0009] In a first aspect, the present invention provides a method for simulating and evaluating the functional characteristics of a mobile network, comprising:

[0010] Obtain the KPI indicator matrix for the first time period before the application of the single functional feature in the target pilot application area and the KPI indicator matrix for the second time period after the application of the single functional feature.

[0011] The judgment result of whether there is a significant change in the indicators in the difference matrix is ​​determined based on the preset credit level and the difference matrix between the first time period KPI indicator matrix and the second time period KPI indicator matrix;

[0012] Based on the judgment result, the difference matrix is ​​subjected to sparsification and / or medianization to obtain a sparse median matrix.

[0013] The functional evaluation results of applying individual functional characteristics to the region to be applied are based on the sparse median matrix.

[0014] Furthermore, the acquisition of the first time-segmented KPI indicator matrix of the target pilot application area before the application of the single functional feature and the second time-segmented KPI indicator matrix after the application of the single functional feature includes:

[0015] Based on daily network optimization needs and equipment support capabilities, a list of 4G or 5G functional features to be evaluated for application assessment is determined.

[0016] Determine the call traffic statistics indicators, which include, but are not limited to: access capacity, hold capacity, call setup capacity, and carrying capacity;

[0017] Obtain the first time-segment KPI indicators for the target pilot application area before the application of a single functional feature. The sum of the individual indicators for each functional feature yields p rows of data. After removing blank items, sort the statistical indicators in ascending order and divide the statistical area into equal grids to obtain the first time-segment KPI indicator matrix M. abc ;

[0018] Obtain the second-period KPI indicators for the target pilot application area after the application of a single functional feature. The sum of the individual indicators for each functional feature yields p rows of data. After removing blank items, sort the statistical indicators in ascending order and divide the statistical area into equal grids to obtain the second-period KPI indicator matrix N. abc ;

[0019] Where M represents all time-segmented KPI information before the application of all features to be evaluated, and N represents all time-segmented KPI information after the application of all features to be evaluated; a, b, and c represent two matrix dimensions, where a represents the number of all features, b represents the number of all KPI statistical indicator items, and c represents the number of indicator segment grids.

[0020] Furthermore, the determination of whether the indicators in the difference matrix have changed significantly based on the preset confidence level and the difference matrix between the first time-segmented KPI indicator matrix and the second time-segmented KPI indicator matrix includes:

[0021] For a sample set M before adjustment, and N after adjustment, there are a × b × c sample groups, where M... abc and N abc These represent individual raster sample groups before and after adjustment, respectively. Each sample group contains d data points, and the number d is obtained using the following formula:

[0022] d = Ceiling(p / c), where Ceiling represents rounding up;

[0023] Each raster contains d data samples, M. abci N represents the individual unadjusted data in each sample group. abci Represents a single adjusted data point in each sample group, where i∈(0,d]; N abci -M abci This represents the fluctuation of the indicator before and after the adjustment; X represents N. abci -M abci The difference, Let X represent the mean, and S represent the standard deviation of the sample X. After obtaining the above data, construct a single statistic t according to the central limit theorem, such that t:

[0024]

[0025] Using a 95% confidence level, the hypothesis test is used to determine whether the p-value corresponding to t is significantly different from 0, thus obtaining the judgment result on whether the indicators in the difference matrix have changed significantly.

[0026] Further, the step of performing sparsification and / or medianization on the difference matrix based on the judgment result to obtain a sparse median matrix includes:

[0027] Retain the grid indicators that have reached the preset information level;

[0028] All invalid indicator data that do not reach the preset confidence level will be sparsified.

[0029] The grid indicators that have reached the preset confidence level are sorted in ascending order, and the median value is selected and filled into the corresponding grid to obtain a sparse median matrix, which is used to characterize the improvement of indicators in the corresponding grid segment after the application of the corresponding functional characteristics.

[0030] Furthermore, the functional evaluation results of applying single functional characteristics to the region to be applied by the sparse median matrix include:

[0031] Determine the application area and the expected thresholds for each indicator;

[0032] Extract user simulation evaluation data of community-level KPI indicators within the designated implementation area of ​​the application region;

[0033] Based on the sparse median matrix, a weighted method is used to determine the regional or cell-level KPI improvement data for the area to be applied.

[0034] Based on the comparison between the KPI improvement data and the expected thresholds of each indicator, corresponding preset functional feature application suggestions are output.

[0035] In a second aspect, embodiments of the present invention provide a functional characteristic simulation and evaluation apparatus for mobile networks, comprising:

[0036] The matrix acquisition module is used to acquire the first time-segment KPI indicator matrix of the target pilot application area before the application of a single functional feature and the second time-segment KPI indicator matrix after the application of the single functional feature.

[0037] The significant judgment module is used to determine whether there is a significant change in the indicators in the difference matrix based on the preset confidence level and the difference matrix between the first time period KPI indicator matrix and the second time period KPI indicator matrix;

[0038] The sparse median module is used to perform sparsification and / or medianization on the difference matrix based on the judgment result to obtain a sparse median matrix.

[0039] The evaluation results module is used to evaluate the functional characteristics of the application area based on the sparse median matrix.

[0040] Furthermore, the matrix acquisition module includes:

[0041] The first unit is used to determine the list of 4G or 5G functional features to be evaluated for application assessment based on daily network optimization needs and equipment support capabilities.

[0042] The second unit is used to determine call traffic statistics indicators, which include, but are not limited to, access capacity, hold capacity, connection capacity, and carrying capacity.

[0043] The third unit is used to obtain the first time-segment KPI indicators of the target pilot application area before the application of a single functional feature. The sum of the single functional feature indicators yields p rows of data. After removing blank items, the statistical indicators are sorted in ascending order, and the statistical area is divided into equal parts to form a grid, resulting in the first time-segment KPI indicator matrix M. abc ;

[0044] The fourth unit is used to obtain the second-period KPI indicators of the target pilot application area after the application of a single functional feature. The sum of the individual indicators of the single functional feature yields p rows of data. After removing blank items, the statistical indicators are sorted in ascending order, and the statistical area is divided into equal parts to form a grid, resulting in the second-period KPI indicator matrix N. abc ;

[0045] Where M represents all time-segmented KPI information before the application of all features to be evaluated, and N represents all time-segmented KPI information after the application of all features to be evaluated; a, b, and c represent two matrix dimensions, where a represents the number of all features, b represents the number of all KPI statistical indicator items, and c represents the number of indicator segment grids.

[0046] Furthermore, the significant determination module includes:

[0047] The fifth unit is used for a sample set M before adjustment, and N after adjustment, containing a × b × c sample groups, where M... abc and N abc These represent individual raster sample groups before and after adjustment, respectively. Each sample group contains d data points, and the number d is obtained using the following formula:

[0048] d = Ceiling(p / c), where Ceiling represents rounding up;

[0049] The sixth unit is used when each raster contains d data points, M. abci N represents the individual unadjusted data in each sample group. abci Represents a single adjusted data point in each sample group, where i∈(0,d]; N abci -M abci This represents the fluctuation of the indicator before and after the adjustment; X represents N. abci -M abci The difference, Let X represent the mean, and S represent the standard deviation of the sample X. After obtaining the above data, construct a single statistic t according to the central limit theorem, such that t:

[0050]

[0051] Unit 7 is used to determine whether the p-value corresponding to t is significantly different from 0 based on hypothesis testing at a 95% confidence level, and to obtain the judgment result of whether the indicators in the difference matrix have changed significantly.

[0052] Furthermore, the sparse median module includes:

[0053] The first sparse unit is used to retain the raster index that has reached the preset information level;

[0054] The second sparse unit is used to sparse out all invalid indicator data that do not reach the preset confidence level.

[0055] The first median unit is used to sort the grid indicators that have reached the preset confidence level in ascending order, select the median and fill it into the corresponding grid to obtain a sparse median matrix, which is used to characterize the improvement of indicators in the corresponding grid segment after the application of the corresponding functional characteristics.

[0056] Furthermore, the evaluation result module includes:

[0057] The threshold determination unit is used to determine the expected thresholds for the application area and various indicators.

[0058] The data extraction unit is used to extract user simulation evaluation data of community-level KPI indicators within the designated implementation area of ​​the application region;

[0059] The data determination unit is used to determine the regional or cell-level KPI improvement data for the application area based on the sparse median matrix and using a weighted method.

[0060] The suggested output unit is used to compare the KPI improvement data with the expected thresholds of each indicator and output the corresponding preset functional feature application suggestion decision.

[0061] Thirdly, the present invention provides an electronic device, including a memory and a memory storing a computer program, wherein the processor executes the program to implement the steps of the functional characteristic simulation evaluation method for mobile networks described in the first aspect.

[0062] Fourthly, the present invention provides a processor-readable storage medium storing a computer program for causing the processor to perform the steps of the functional characteristic simulation evaluation method for mobile networks described in the first aspect.

[0063] This invention provides a method and apparatus for simulating and evaluating the functional characteristics of mobile networks. Based on gridded hypothesis testing and data sparsification, it simulates the application of 4 / 5G cell functional characteristics. After small-scale application of a single functional characteristic, it analyzes the results of call statistics data to simulate and evaluate the application effect of a large-scale mixed application of functional characteristics. Hypothesis testing is used to determine the effectiveness of the functional characteristics, and data sparsification and medianization are employed for data processing to derive an application effect evaluation. This guides the application and simulation evaluation of corresponding functional characteristics in other areas. The entire method effectively improves the utilization efficiency of functional characteristics, enabling effective simulation and accurate evaluation of 4 / 5G cell functional characteristics, and guiding the rational and effective application of functional characteristics. Attached Figure Description

[0064] To more clearly illustrate the technical solutions in this invention or the prior art, the drawings used in the description of the embodiments or the prior art will be briefly introduced below. Obviously, the drawings described below are some embodiments of this invention. For those skilled in the art, other drawings can be obtained from these drawings without creative effort.

[0065] Figure 1 This is one of the flowcharts for a method for simulating and evaluating the functional characteristics of a mobile network, provided in an embodiment of the present invention.

[0066] Figure 2 A second flowchart of a method for simulating and evaluating the functional characteristics of a mobile network, provided as an embodiment of the present invention;

[0067] Figure 3 A flowchart of a method for simulating and evaluating the functional characteristics of a mobile network, provided in an embodiment of the present invention;

[0068] Figure 4 A flowchart of a method for simulating and evaluating the functional characteristics of a mobile network, provided as an embodiment of the present invention;

[0069] Figure 5 The fifth flowchart of a method for simulating and evaluating the functional characteristics of a mobile network, provided as an embodiment of the present invention;

[0070] Figure 6 A flowchart of a method for simulating and evaluating the functional characteristics of a mobile network, provided as an embodiment of the present invention, is shown in Figure 6.

[0071] Figure 7 This is a schematic diagram of the composition structure of a functional characteristic simulation and evaluation device for mobile networks provided in an embodiment of the present invention;

[0072] Figure 8 This is a schematic diagram of the structure of an electronic device provided in an embodiment of the present invention. Detailed Implementation

[0073] To make the objectives, technical solutions, and advantages of this invention clearer, the technical solutions of this invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some, not all, of the embodiments of this invention. All other embodiments obtained by those skilled in the art based on the embodiments of this invention without creative effort are within the scope of protection of this invention.

[0074] The following is combined Figures 1-6 This invention describes a method for simulating and evaluating the functional characteristics of mobile networks.

[0075] Figure 1 This is one of the flowcharts for a method for simulating and evaluating the functional characteristics of a mobile network, provided in an embodiment of the present invention. Figure 2 A second flowchart of a method for simulating and evaluating the functional characteristics of a mobile network, provided as an embodiment of the present invention; Figure 3 A flowchart of a method for simulating and evaluating the functional characteristics of a mobile network, provided in an embodiment of the present invention; Figure 4 A flowchart of a method for simulating and evaluating the functional characteristics of a mobile network, provided as an embodiment of the present invention; Figure 5 The fifth flowchart of a method for simulating and evaluating the functional characteristics of a mobile network, provided as an embodiment of the present invention; Figure 6 This is the sixth flowchart of a method for simulating and evaluating the functional characteristics of a mobile network, provided as an embodiment of the present invention.

[0076] In one specific embodiment of the present invention, the present invention provides a method for simulating and evaluating the functional characteristics of a mobile network, comprising:

[0077] Step 110: Obtain the KPI indicator matrix for the first time period before the application of the single functional feature in the target pilot application area and the KPI indicator matrix for the second time period after the application of the single functional feature;

[0078] In this embodiment of the invention, to achieve the extraction of call statistics data and the calculation of rasterized hypothesis testing, this step enables the comparison of call statistics indicators after the application of 4 / 5G functional features, which is used to determine the improvement effect of the functional feature application on relevant statistical indicators. By rasterizing the hourly statistical indicators of the cell according to numerical intervals, hypothesis testing is used to determine whether the indicators within the corresponding raster segment are significantly improved after the application of the functional feature. Firstly, the first time-segment KPI indicator matrix before the application of a single functional feature and the second time-segment KPI indicator matrix after the application of a single functional feature can be obtained for the target pilot application area.

[0079] Step 120: Based on the preset confidence level and the difference matrix between the first time-segmented KPI index matrix and the second time-segmented KPI index matrix, determine the judgment result of whether the indicators in the difference matrix have changed significantly;

[0080] Specifically, in order to statistically compare the two sets of data before and after the application of functional characteristics in each grid, a comparative hypothesis test method is used to confirm the significance. The difference matrix is ​​obtained by subtracting the KPI index matrix of the first time period from the KPI index matrix of the second time period. Then, based on the calculated confidence level, a judgment on whether there is a significant change is made using a preset confidence level.

[0081] Step 130: Based on the judgment result, perform sparsification and / or medianization on the difference matrix to obtain a sparse median matrix;

[0082] This step enables sparsification and medianization of the functional feature application effect data, improving the computational efficiency of application evaluation. Specifically, based on the output of step A4, the effectiveness of the relevant functional features in improving indicators within each raster segment interval can be determined. To reduce subsequent computation, the data is sparsified, retaining raster indicators that reach the 95% confidence level threshold, while all invalid indicator data that do not reach the confidence level threshold are sparsified. After sparsifying the invalid raster indicator data, medianization is used to fill in the valid raster data.

[0083] Step 140: Based on the sparse median matrix, evaluate the functional characteristics of the application area.

[0084] This step simulates and evaluates the application effects of corresponding functional characteristics, and outputs application decision suggestions to assist in the application of functional characteristics. Since multiple functional characteristics are often used in the existing network area to achieve network optimization, it is difficult to accurately evaluate a single functional characteristic in practical applications. This proposal achieves the evaluation of a single functional characteristic in the initial step, and realizes the simulation evaluation of functional characteristics through grid segmentation. It achieves accurate evaluation of the effects in scenarios where multiple functional characteristics are used in combination without a large investment of production resources.

[0085] Specifically, in practice, to obtain the KPI indicator matrix for the first time period before the application of a single functional feature and the KPI indicator matrix for the second time period after the application of the single functional feature in the target pilot application area, the following steps can be taken:

[0086] Step 210: Based on daily network optimization needs and equipment support capabilities, determine the list of 4G or 5G functional features to be evaluated for application assessment;

[0087] Specifically, the first step is to select 4 / 5G functional features. Based on daily network optimization needs and equipment support capabilities, a list of 4 / 5G functional features to be evaluated for application assessment is selected.

[0088] Step 220: Determine the call traffic statistics indicators, which include, but are not limited to: access capacity, hold capacity, call setup capacity, and carrying capacity;

[0089] Secondly, the pilot application area and call volume statistics indicators should be selected. The goal of the small-scale pilot application is to assess the actual application effect of the corresponding functional features. It is recommended that the pilot application scale be controlled to around 30 sites (around 100 cells), and the pilot comparison period should be at least one week, with statistics collected one week before and after the application for comparison. Call volume statistics indicators should include the main daily call volume statistics KPIs, including but not limited to access capacity, hold capacity, call setup capacity, and bearer capacity. It is recommended to include 20 4G indicators, 13 5G indicators, and 5 voice indicators.

[0090] Step 230: Obtain the first time-segment KPI indicators for the target pilot application area before the application of a single functional feature. The sum of the single functional feature indicators yields p rows of data. After removing blank items, sort the statistical indicators in ascending order and divide the statistical area into equal parts to form a grid, thus obtaining the first time-segment KPI indicator matrix M. abc ;

[0091] Step 240: Obtain the second time-segment KPI indicators for the target pilot application area after the application of a single functional feature. The sum of the single functional feature indicators yields p rows of data. After removing blank items, sort the statistical indicators in ascending order and divide the statistical area into equal parts to form a grid, thus obtaining the second time-segment KPI indicator matrix N. abc ;

[0092] Where M represents all time-segmented KPI information before the application of all features to be evaluated, and N represents all time-segmented KPI information after the application of all features to be evaluated; a, b, and c represent two matrix dimensions, where a represents the number of all features, b represents the number of all KPI statistical indicator items, and c represents the number of indicator segment grids.

[0093] Specifically, after completing a small-scale pilot test of a single functional feature according to the above steps, extract the hourly statistical indicators at the community level before and after the application. The total number of single indicators for a single functional feature is p rows of data. After removing blank items, arrange each statistical indicator in ascending order and divide the statistical area into grids (it is recommended to divide it into 100 grids).

[0094] After the above steps, two three-dimensional matrices M are obtained. abc and N abcWhere M represents all time-segmented KPI information for all features to be evaluated before application, and N represents all time-segmented KPI information for all features to be evaluated after application. abc represent two matrix dimensions: a represents the number of all features, b represents the number of all KPI statistical indicators, and c represents the number of indicator segment grids.

[0095] Furthermore, the determination of whether the indicators in the difference matrix have changed significantly based on the preset confidence level and the difference matrix between the first time-segmented KPI indicator matrix and the second time-segmented KPI indicator matrix includes:

[0096] Step 310: For a sample set M before adjustment, and N after adjustment, containing a × b × c sample groups, where M... abc and N abc These represent individual raster sample groups before and after adjustment, respectively. Each sample group contains d data points, and the number d is obtained using the following formula:

[0097] d = Ceiling(p / c), where Ceiling represents rounding up;

[0098] Step 320: Each raster data sample group contains d data points, M abci N represents the individual unadjusted data in each sample group. abci Represents a single adjusted data point in each sample group, where i∈(0,d]; N abci -M abci This represents the fluctuation of the indicator before and after the adjustment; X represents N. abci -M abci The difference, Let X represent the mean, and S represent the standard deviation of the sample X. After obtaining the above data, construct a single statistic t according to the central limit theorem, such that t:

[0099]

[0100] Step 330: Using 95% confidence level, determine whether the p-value corresponding to t is significantly different from 0 based on hypothesis testing, and obtain the judgment result of whether the indicators in the difference matrix have changed significantly.

[0101] Specifically, in order to achieve the comparison of corresponding KPI indicators at each grid level based on hypothesis testing, we can statistically analyze two sets of comparative data before and after the application of functional characteristics within each grid, and use comparative hypothesis testing to confirm significance.

[0102] Given a sample set of M before adjustment and N after adjustment, containing a × b × c sample groups, where M... abc and N abcThese represent individual raster sample groups before and after adjustment, respectively. Each sample group contains d data points, and the number d is obtained using the following formula:

[0103] d = Ceiling(p / c), where Ceiling represents rounding up.

[0104] Each raster contains d data samples, assuming M... abci N represents the individual unadjusted data in each sample group. abci Let represent a single adjusted data point in each sample group, where i ∈ (0, d). And N abci -M abci This represents the fluctuation of the indicator before and after the adjustment. X represents N. abci -M abci The difference, Let X represent the mean, and S represent the standard deviation of the sample X. After obtaining the above data, construct a single statistic t according to the central limit theorem, such that t:

[0105]

[0106] Using a 95% confidence level, we determine whether the p-value corresponding to the t-test is significantly different from 0 based on hypothesis testing. That is, when the p-value corresponding to the t-test is significantly different from 0, it means that the relevant indicators within the grid before and after adjustment have significantly improved or deteriorated, which has statistical significance.

[0107] Further, the step of performing sparsification and / or medianization on the difference matrix based on the judgment result to obtain a sparse median matrix includes:

[0108] Step 410: Retain the grid indicators that have reached the preset information level;

[0109] Step 420: Sparse all invalid indicator data that do not reach the preset confidence level;

[0110] Step 430: Sort the grid indicators that have reached the preset confidence level in ascending order, select the median value and fill it into the corresponding grid to obtain a sparse median matrix, which is used to characterize the improvement of indicators in the corresponding grid segment after the application of the corresponding functional characteristics.

[0111] Specifically, based on the above output results, the effectiveness of the relevant functional characteristics in improving the indicators within each raster segment interval can be determined. To reduce subsequent computational load, the data is sparsified; raster indicators that reach the 95% confidence level threshold are retained, while invalid indicator data that do not reach the confidence level threshold are all sparsified.

[0112] N abc –M abc =0 (if the t-test statistic in the corresponding raster is significant to 0)

[0113] Invalid raster indicator data sparsification has been implemented. For valid raster data, median filling is used. Before and after the application of a functional feature, the fluctuation ranges of all before-and-after samples in the valid raster are subtracted and sorted in ascending order. The median value is then selected and filled into the corresponding raster to represent the indicator improvement in the corresponding raster segment after the application of the functional feature. That is, for...

[0114] Med = N abc –M abc =Median(N) abci； –M abci ), where i∈(0,d) (if

[0115] (The t-test statistic in the corresponding raster is significantly different from 0)

[0116] Furthermore, the functional evaluation results of applying single functional characteristics to the region to be applied by the sparse median matrix include:

[0117] Step 510: Determine the application area and the expected thresholds for each indicator;

[0118] Step 520: Extract user simulation evaluation data of community-level KPI indicators within the designated implementation area of ​​the region to be applied;

[0119] Step 530: Based on the sparse median matrix, use a weighted method to determine the regional or cell-level KPI improvement data for the area to be applied;

[0120] Step 540: Based on the KPI improvement data, compare it with the expected threshold of each indicator, and output the corresponding preset functional feature application suggestion decision.

[0121] Specifically, to achieve simulated application effect evaluation of corresponding functional characteristics and output application decision suggestions to assist in the application of functional characteristics, the application area and KPI threshold settings are first determined. The area to be applied is selected, and the expected thresholds for each KPI indicator are set. The area to be applied is used for simulation calculation, and the expected thresholds for KPI indicators are used to help engineers make functional characteristic decisions. Furthermore, user simulation evaluations of cell-level KPI indicators are extracted within the planned implementation area. Using the grid-level improvement effect data of each indicator obtained in the above steps, a weighted method is used to calculate the area-level or cell-level KPI improvement data. The grid-level KPI indicator improvement magnitude is shown in the following formula, where n represents the number of samples in the grid, M represents the indicator data after application, and N represents the indicator data before application. That is:

[0122] Increase magnitude = ∑(Mi-Ni) / n, where i∈(0,n);

[0123] Based on the improvement of KPI indicators obtained from the above steps, compare them with the input threshold in step C1, and give suggestions for the application of functional features.

[0124] Compared with the prior art, the embodiments of the present invention have the following obvious advantages and effects:

[0125] Based on the results of a small-scale pilot project, this proposal achieves accurate simulation and evaluation of 4G / 5G functional characteristics. Compared with existing technologies, its advantages are mainly concentrated in the following three aspects.

[0126] 1) When evaluating the optimization effect of a single functional feature after a small-scale pilot test, the use of raster matrix data sparsification and medianization in the calculation process improves computational efficiency, significantly reduces computational overhead, and increases the computational speed for application effect evaluation. Taking 100 raster elements, 17 functional features, and 25 KPI indicators as an example, the estimated number of raster elements is 1.275 million (17 functional features × 100 raster elements × 15 hours × 7 days × 2 weeks × 25 KPI indicators). Through sparsification, it is expected that the effective number of raster elements can be reduced to less than 50,000, improving computational efficiency by 96.8%.

[0127] 2) Since multiple functional features are often used in the existing network area to achieve network optimization, it is difficult to accurately evaluate a single functional feature in practical applications. This proposal achieves the evaluation of a single functional feature in the initial step and realizes the simulation evaluation of functional features through grid segmentation. It achieves accurate evaluation of the effect in scenarios where multiple functional features are used in combination without a large investment of production resources.

[0128] 3) The solution provided in this embodiment of the invention is simple to implement, requires minimal engineer skills, and reduces human resource investment. Furthermore, the method does not require changes to the network structure, is easy to operate, and poses no network risks. Therefore, the solution provided in this embodiment of the invention can simply and effectively improve the evaluation efficiency and application benefits of 4 / 5G functional characteristics.

[0129] This invention provides a method for simulating and evaluating the functional characteristics of mobile networks. Based on gridded hypothesis testing and data sparsification, it simulates the application of 4 / 5G cell functional characteristics. After small-scale application of a single functional characteristic, it analyzes the results of call statistics data to simulate and evaluate the application effect of a large-scale mixed application of functional characteristics. The method determines the effectiveness of functional characteristics through hypothesis testing, and uses data sparsification and medianization for data processing to obtain an application effect evaluation. This guides the application and simulation evaluation of corresponding functional characteristics in other areas. The entire method effectively improves the utilization efficiency of functional characteristics, enabling effective simulation and accurate evaluation of 4 / 5G cell functional characteristics, and guiding the reasonable and effective application of functional characteristics.

[0130] like Figure 6As shown, examples are provided to illustrate this technical solution more clearly.

[0131] In another embodiment of the present invention, the functional characteristics of a mobile network can be simulated and evaluated according to the following steps:

[0132] Step A1: Statistical calculations are performed using the downlink 256QAM and uplink 64QAM features of the LTE network as examples.

[0133] Step A2: Select two independent areas of 100 cells each in Pudong as pilot areas. Observe and statistical indicators including 20 4G-related indicators and 5 voice-related indicators, involving 25 KPI indicators extracted on an hourly basis, extracting 15 busy hours per day, with an extraction cycle of one week before and after the adjustment. A total of 525,000 statistical indicators were extracted for each functional feature (100 cells × 15 hours × 7 days × 2 weeks × 25 KPI indicators). The downlink 256QAM and uplink 64QAM functional features totaled 1,050,000 statistical indicators.

[0134] Step A3: Taking call connection rate and uplink user speed as examples of KPI statistics, the downlink 256QAM and uplink 64QAM metrics are used as the reference group for 10,500 statistical indicators (100 cells × 15 hours × first 7 days) in the first week of the verification area, and as the comparison group for 10,500 statistical indicators (100 cells × 15 hours × last 7 days) in the week after the pilot is completed. Blank values ​​are removed from both KPI statistics, and the data is sorted in ascending order to form raster data.

[0135] The above steps yield two three-dimensional matrices M. abc and N abc The two matrices are 2 (two functional characteristics of downlink 256QAM and uplink 64QAM) × 2 (two KPI indicators of connection rate and uplink user rate) × 100 (100 grids).

[0136] Step A4: Matrix M abc and N abc Each sample contains 400 sample groups, with approximately 100 data points in each group. An example of the uplink 64QAM feature data is as follows:

[0137] Uplink 64QAM call completion rate statistics:

[0138]

[0139] Uplink user rate statistics for 64QAM:

[0140]

[0141] Similarly, the downlink 256QAM functional characteristic data sample was used. Taking the before-and-after data of the uplink 64QAM functional characteristic call completion rate statistics group 1 as an example, the significance of the call completion rate improvement was verified through hypothesis testing. This sample group contains 100 data points, and the call completion rate before adjustment ranged from 15.68% to 27.35%, represented as M. abci [15.68%, 16.23%...27.23%, 27.33%] A total of 100 data points; these 100 call connection rates correspond to N, the adjusted call connection rate indicators for the same cell and time period. abci This refers to a total of 100 statistical data items, namely [14.98%, 15.39%...30.23%, 50.36%]. Subtracting the corresponding pre-adjustment connection rate data from the adjusted connection rate data yields:

[0142] N abci -M abci =[-0.7%, -0.84%, ..., 3%, 23.03%], a total of 100 data points. Based on the Central Limit Theorem, the formula is used... Calculate the corresponding t-value (where S is the sample standard deviation and n is the number of samples), and determine N. abci -M abci Is there a significant difference from 0? The calculated t=1.72, corresponding to a p-value probability of 0.26, is significantly higher than 0.05, making it impossible to reject the null hypothesis. This indicates that the uplink 64QAM group 1 did not significantly improve the connection rate before and after the adjustment. Using the same method, the improvement effect on the connection rate of the remaining 99 sample groups of uplink 64QAM after adjustment was calculated, confirming that there was no significant difference in the connection rate of all sample groups before and after the adjustment. This indicates that the application of the uplink 64QAM functional features has no optimization effect on the connection rate.

[0143] The same method was used to calculate the optimization effect of uplink 64QAM on uplink user rate. There was no significant difference in uplink user rate before and after the adjustment for groups 1-27. However, the p-values ​​for groups 28-100 were all less than 0.05, which indicates that there was a significant difference in uplink user rate before and after the adjustment of uplink 64QAM. This confirms that the uplink user rate in some grid groups was significantly improved after the application of the uplink 64QAM feature.

[0144] Similarly, after applying the downlink 256QAM functional characteristics, the P-values ​​of the corresponding grids for the connection rate and uplink speed are greater than 0.05, indicating that there were no significant fluctuations in the connection rate and uplink speed before and after the application of downlink 256QAM.

[0145] Step B1: For grids with a P-value greater than 0.05 (indicating that the corresponding KPI indicators do not fluctuate significantly after the application of the corresponding functional characteristics), namely the uplink 64QAM connection rate indicator, the downlink 256QAM connection rate indicator, the downlink 256QAM uplink rate indicator, and the uplink 64QAM uplink rate 1-27 grid indicators, the corresponding grid (N) abci -M abci All values ​​are assigned to 0.

[0146] Step B2: For graticles with a P-value less than 0.05 (indicating a significant difference in the corresponding KPI after the application of the relevant functional characteristics), i.e., the uplink 64QAM uplink rate 28-100 graticle index, the corresponding graticle (N) abci -M abci The values ​​are assigned using the sample median. For example, in sample group 100, the corresponding raster sample group data before and after the 64QAM adjustment are [31.91, 31.98...40.29, 40.82] and [33.23, 38.39...31.87, 39.59], respectively. Will After sorting in ascending order, the 50th data point out of 100 samples is taken as the representative value: 6.23. The same process is applied to the raster index processing for groups 28-99.

[0147] Step C1 selects 1000 cells in the surrounding area where no relevant features are applied as simulation verification calculations, and inputs the KPI threshold as the uplink user rate improvement (i.e., the uplink user rate improvement is greater than 0 Mbps).

[0148] Step C2 extracts a total of 2000 raw KPI data points, including connection rate and uplink speed, from 1000 cells before application.

[0149] Step C3, based on the raw KPI call statistics data extracted in C2, and according to the cell uplink rate data, divides the cells into corresponding MABC grids according to the uplink rate range. As shown below:

[0150]

[0151] The average improvement expected for 1000 cells was calculated using the formula ∑(Mi-Ni) / n. Simulation results showed that after the application of the uplink 64QAM function, it is expected that 653 cells in the entire area will be improved, and the uplink user speed of the affected cells can be improved by an average of 3.84Mbps.

[0152] Since the uplink rate improvement test value after applying downlink 256QAM in step B is not significant, downlink 256QAM has no effect on improving the uplink rate.

[0153] Step C4: Since it is calculated that after applying uplink 64QAM to 1000 cells, it is expected that 653 cells in the overall area can be improved, and the overall uplink user rate can be improved by 3.84Mbps, which is higher than the input threshold of step C1, it is recommended to apply uplink 64QAM to 653 cells to improve the uplink user rate.

[0154] The functional characteristic simulation evaluation method for mobile networks provided in this invention provides a method and apparatus for comparing the effects of functional characteristics before and after application. Compared with traditional manual observation methods or average value methods, it is more reasonable and accurate. During the calculation of functional characteristic effects, the corresponding modules perform sparsification and medianization processing on the functional characteristic evaluation data matrix, reducing the computational overhead by more than 95%. Based on raster hypothesis testing and sparsification processing, functional characteristic simulation evaluation is achieved, guiding the effective application in the current network.

[0155] The functional characteristic simulation and evaluation apparatus for mobile networks provided by the present invention is described below. The functional characteristic simulation and evaluation apparatus for mobile networks described below can be referred to in correspondence with the functional characteristic simulation and evaluation method for mobile networks described above.

[0156] Please refer to Figure 7 , Figure 7 This is a schematic diagram of the composition structure of a functional characteristic simulation and evaluation device for mobile networks provided in an embodiment of the present invention.

[0157] In one specific embodiment of the present invention, an embodiment of the present invention provides a functional characteristic simulation and evaluation device 700 for mobile networks, comprising:

[0158] The matrix acquisition module 710 is used to acquire the first time period KPI indicator matrix of the target pilot application area before the application of the single functional feature and the second time period KPI indicator matrix after the application of the single functional feature.

[0159] The significant judgment module 720 is used to determine whether there is a significant change in the indicators in the difference matrix based on the preset confidence level and the difference matrix between the first time-segmented KPI indicator matrix and the second time-segmented KPI indicator matrix.

[0160] The sparse median module 730 is used to perform sparsification and / or medianization on the difference matrix based on the judgment result to obtain a sparse median matrix.

[0161] The evaluation result module 740 is used to evaluate the functional characteristics of the application area based on the sparse median matrix.

[0162] Furthermore, the matrix acquisition module includes:

[0163] The first unit is used to determine the list of 4G or 5G functional features to be evaluated for application assessment based on daily network optimization needs and equipment support capabilities.

[0164] The second unit is used to determine call traffic statistics indicators, which include, but are not limited to, access capacity, hold capacity, connection capacity, and carrying capacity.

[0165] The third unit is used to obtain the first time-segment KPI indicators of the target pilot application area before the application of a single functional feature. The sum of the single functional feature indicators yields p rows of data. After removing blank items, the statistical indicators are sorted in ascending order, and the statistical area is divided into equal parts to form a grid, resulting in the first time-segment KPI indicator matrix M. abc ;

[0166] The fourth unit is used to obtain the second-period KPI indicators of the target pilot application area after the application of a single functional feature. The sum of the individual indicators of the single functional feature yields p rows of data. After removing blank items, the statistical indicators are sorted in ascending order, and the statistical area is divided into equal parts to form a grid, resulting in the second-period KPI indicator matrix N. abc ;

[0167] Where M represents all time-segmented KPI information before the application of all features to be evaluated, and N represents all time-segmented KPI information after the application of all features to be evaluated; a, b, and c represent two matrix dimensions, where a represents the number of all features, b represents the number of all KPI statistical indicator items, and c represents the number of indicator segment grids.

[0168] Furthermore, the significant determination module includes:

[0169] The fifth unit is used for a sample set M before adjustment, and N after adjustment, containing a × b × c sample groups, where M... abc and N abc These represent individual raster sample groups before and after adjustment, respectively. Each sample group contains d data points, and the number d is obtained using the following formula:

[0170] d = Ceiling(p / c), where Ceiling represents rounding up;

[0171] The sixth unit is used when each raster contains d data points, M. abci N represents the individual unadjusted data in each sample group. abci Represents a single adjusted data point in each sample group, where i∈(0,d]; N abci -M abci This represents the fluctuation of the indicator before and after the adjustment; X represents N. abci -M abci The difference, Let X represent the mean, and S represent the standard deviation of the sample X. After obtaining the above data, construct a single statistic t according to the central limit theorem, such that t:

[0172]

[0173] Unit 7 is used to determine whether the p-value corresponding to t is significantly different from 0 based on hypothesis testing at a 95% confidence level, and to obtain the judgment result of whether the indicators in the difference matrix have changed significantly.

[0174] Furthermore, the sparse median module includes:

[0175] The first sparse unit is used to retain the raster index that has reached the preset information level;

[0176] The second sparse unit is used to sparse out all invalid indicator data that do not reach the preset confidence level.

[0177] The first median unit is used to sort the grid indicators that have reached the preset confidence level in ascending order, select the median and fill it into the corresponding grid to obtain a sparse median matrix, which is used to characterize the improvement of indicators in the corresponding grid segment after the application of the corresponding functional characteristics.

[0178] Furthermore, the evaluation result module includes:

[0179] The threshold determination unit is used to determine the expected thresholds for the application area and various indicators.

[0180] The data extraction unit is used to extract user simulation evaluation data of community-level KPI indicators within the designated implementation area of ​​the application region;

[0181] The data determination unit is used to determine the regional or cell-level KPI improvement data for the application area based on the sparse median matrix and using a weighted method.

[0182] The suggested output unit is used to compare the KPI improvement data with the expected thresholds of each indicator and output the corresponding preset functional feature application suggestion decision.

[0183] This invention provides a functional characteristic simulation and evaluation device for mobile networks. Based on gridded hypothesis testing and data sparsification methods, it simulates the application of 4 / 5G cell functional characteristics. After small-scale application of a single functional characteristic, it analyzes the results of call statistics data to simulate and evaluate the application effect of a large-scale mixed application of functional characteristics. Hypothesis testing is used to determine the effectiveness of functional characteristics, and data sparsification and medianization are employed for data processing to obtain an application effect evaluation. This guides the application and simulation evaluation of corresponding functional characteristics in other areas. The entire method effectively improves the utilization efficiency of functional characteristics, enabling effective simulation and accurate evaluation of 4 / 5G cell functional characteristics, and guiding the reasonable and effective application of functional characteristics.

[0184] Figure 8 An example is a schematic diagram of the physical structure of an electronic device, such as... Figure 8 As shown, the electronic device may include: a processor 810, a communication interface 820, a memory 830, and a communication bus 840, wherein the processor 810, the communication interface 820, and the memory 830 communicate with each other via the communication bus 840. The processor 810 can call a computer program in the memory 830 to execute steps of a functional characteristic simulation evaluation method for mobile networks, such as including:

[0185] Obtain the KPI indicator matrix for the first time period before the application of the single functional feature in the target pilot application area and the KPI indicator matrix for the second time period after the application of the single functional feature.

[0186] The judgment result of whether there is a significant change in the indicators in the difference matrix is ​​determined based on the preset credit level and the difference matrix between the first time period KPI indicator matrix and the second time period KPI indicator matrix;

[0187] Based on the judgment result, the difference matrix is ​​subjected to sparsification and / or medianization to obtain a sparse median matrix.

[0188] The functional evaluation results of applying individual functional characteristics to the region to be applied are based on the sparse median matrix.

[0189] Furthermore, the logical instructions in the aforementioned memory 830 can be implemented as software functional units and, when sold or used as independent products, can be stored in a computer-readable storage medium. Based on this understanding, the technical solution of the present invention, or the part that contributes to the prior art, or a part of the technical solution, can be embodied in the form of a software product. This computer software product is stored in a storage medium and includes several instructions to cause a computer device (which may be a personal computer, server, or network device, etc.) to execute all or part of the steps of the methods described in the various embodiments of the present invention. The aforementioned storage medium includes various media capable of storing program code, such as USB flash drives, portable hard drives, read-only memory (ROM), random access memory (RAM), magnetic disks, or optical disks.

[0190] On the other hand, the present invention also provides a computer program product, the computer program product comprising a computer program stored on a non-transitory computer-readable storage medium, the computer program comprising program instructions, wherein when the program instructions are executed by a computer, the computer is able to execute the functional characteristic simulation and evaluation method for mobile networks provided by the above methods, the method comprising:

[0191] Obtain the KPI indicator matrix for the first time period before the application of the single functional feature in the target pilot application area and the KPI indicator matrix for the second time period after the application of the single functional feature.

[0192] The judgment result of whether there is a significant change in the indicators in the difference matrix is ​​determined based on the preset credit level and the difference matrix between the first time period KPI indicator matrix and the second time period KPI indicator matrix;

[0193] Based on the judgment result, the difference matrix is ​​subjected to sparsification and / or medianization to obtain a sparse median matrix.

[0194] The functional evaluation results of applying individual functional characteristics to the region to be applied are based on the sparse median matrix.

[0195] On the other hand, embodiments of this application also provide a processor-readable storage medium storing a computer program for causing the processor to execute the functional characteristic simulation and evaluation method for mobile networks provided in the above embodiments, for example including:

[0196] Obtain the KPI indicator matrix for the first time period before the application of the single functional feature in the target pilot application area and the KPI indicator matrix for the second time period after the application of the single functional feature.

[0197] The judgment result of whether there is a significant change in the indicators in the difference matrix is ​​determined based on the preset credit level and the difference matrix between the first time period KPI indicator matrix and the second time period KPI indicator matrix;

[0198] Based on the judgment result, the difference matrix is ​​subjected to sparsification and / or medianization to obtain a sparse median matrix.

[0199] The functional evaluation results of applying individual functional characteristics to the region to be applied are based on the sparse median matrix.

[0200] The processor-readable storage medium can be any available medium or data storage device that the processor can access, including but not limited to magnetic memory (e.g., floppy disk, hard disk, magnetic tape, magneto-optical disk (MO)), optical memory (e.g., CD, DVD, BD, HVD), and semiconductor memory (e.g., ROM, EPROM, EEPROM, non-volatile memory (NAND FLASH), solid-state drive (SSD)).

[0201] The device embodiments described above are merely illustrative. The units described as separate components may or may not be physically separate. The components shown as units may or may not be physical units; that is, they may be located in one place or distributed across multiple network units. Some or all of the modules can be selected to achieve the purpose of this embodiment according to actual needs. Those skilled in the art can understand and implement this without any creative effort.

[0202] Through the above description of the embodiments, those skilled in the art can clearly understand that each embodiment can be implemented by means of software plus necessary general-purpose hardware platforms, and of course, it can also be implemented by hardware. Based on this understanding, the above technical solutions, in essence or the part that contributes to the prior art, can be embodied in the form of a software product. This computer software product can be stored in a computer-readable storage medium, such as ROM / RAM, magnetic disk, optical disk, etc., and includes several instructions to cause a computer device (which may be a personal computer, server, or network device, etc.) to execute the methods described in the various embodiments or some parts of the embodiments.

[0203] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention, and not to limit them; although the present invention has been described in detail with reference to the foregoing embodiments, those skilled in the art should understand that modifications can still be made to the technical solutions described in the foregoing embodiments, or equivalent substitutions can be made to some of the technical features; and these modifications or substitutions do not cause the essence of the corresponding technical solutions to deviate from the spirit and scope of the technical solutions of the embodiments of the present invention.

Claims

1. A method for simulating and evaluating the functional characteristics of mobile networks, characterized in that, include: Obtain the KPI indicator matrix for the first time period before the application of the single functional feature in the target pilot application area and the KPI indicator matrix for the second time period after the application of the single functional feature. The judgment result of whether there is a significant change in the indicators in the difference matrix is ​​determined based on the preset credit level and the difference matrix between the first time period KPI indicator matrix and the second time period KPI indicator matrix; Based on the judgment result, the difference matrix is ​​subjected to sparsification and / or medianization to obtain a sparse median matrix. Based on the sparse median matrix, the functional evaluation results of the application of individual functional characteristics in the application area are performed; The determination of whether there are significant changes in the indicators in the difference matrix based on the preset confidence level and the difference matrix between the first time-segmented KPI indicator matrix and the second time-segmented KPI indicator matrix includes: For a sample set M before adjustment, and N after adjustment, the total number of samples is a. b c sample groups of data, where M abc and N abc These represent individual raster sample groups before and after adjustment, respectively. Each sample group contains d data points, and the number d is obtained using the following formula: d = Ceiling(p / c), where Ceiling represents rounding up; Each raster contains d data samples, M. abci N represents the individual unadjusted data in each sample group. abci Represents a single adjusted data point in each sample group, where i (0, d];N abci - M abci This represents the fluctuation of the indicator before and after the adjustment; X represents N. abci - M abci The difference, Let X represent the mean, and S represent the standard deviation of the sample X. After obtaining the above data, construct a single statistic t according to the central limit theorem, such that t: t = / ； Using a 95% confidence level, we determine whether the p-value corresponding to t is significantly different from 0 based on hypothesis testing, and obtain the judgment result of whether the indicators in the difference matrix have changed significantly. The acquisition of the target pilot application area's KPI indicator matrix for the first time period before the application of a single functional feature and the KPI indicator matrix for the second time period after the application of the single functional feature includes: Based on daily network optimization needs and equipment support capabilities, a list of 4G or 5G functional features to be evaluated for application assessment is determined. Determine the call traffic statistics indicators, which include, but are not limited to: access capacity, hold capacity, call setup capacity, and carrying capacity; Obtain the first time-segment KPI indicators for the target pilot application area before the application of a single functional feature. The sum of the individual indicators for each functional feature yields p rows of data. After removing blank items, sort the statistical indicators in ascending order and divide the statistical area into equal grids to obtain the first time-segment KPI indicator matrix M. abc ; Obtain the second-period KPI indicators for the target pilot application area after the application of a single functional feature. The sum of the individual indicators for each functional feature yields p rows of data. After removing blank items, sort the statistical indicators in ascending order and divide the statistical area into equal grids to obtain the second-period KPI indicator matrix N. abc ; Where M represents all time-segmented KPI information before the application of all features to be evaluated, and N represents all time-segmented KPI information after the application of all features to be evaluated; a, b, and c represent two matrix dimensions, where a represents the number of all features, b represents the number of all KPI statistical indicator items, and c represents the number of indicator segment grids.

2. The method for simulating and evaluating the functional characteristics of mobile networks according to claim 1, characterized in that, The step of performing sparsification and / or medianization on the difference matrix based on the judgment result to obtain a sparse median matrix includes: Retain the grid indicators that have reached the preset information level; All invalid indicator data that do not reach the preset confidence level will be sparsified. The grid indicators that have reached the preset confidence level are sorted in ascending order, and the median value is selected and filled into the corresponding grid to obtain a sparse median matrix, which is used to characterize the improvement of indicators in the corresponding grid segment after the application of the corresponding functional characteristics.

3. The method for simulating and evaluating the functional characteristics of mobile networks according to claim 1, characterized in that, The functional evaluation results of applying a single functional characteristic to the region to be applied using the sparse median matrix include: Determine the application area and the expected thresholds for each indicator; Extract user simulation evaluation data of community-level KPI indicators within the designated implementation area of ​​the application region; Based on the sparse median matrix, a weighted method is used to determine the regional or cell-level KPI improvement data for the area to be applied. Based on the comparison between the KPI improvement data and the expected thresholds of each indicator, corresponding preset functional feature application suggestions are output.

4. A device for simulating and evaluating the functional characteristics of mobile networks, characterized in that, include: The matrix acquisition module is used to acquire the first time-segment KPI indicator matrix of the target pilot application area before the application of a single functional feature and the second time-segment KPI indicator matrix after the application of the single functional feature. The significant judgment module is used to determine whether there is a significant change in the indicators in the difference matrix based on the preset confidence level and the difference matrix between the first time period KPI indicator matrix and the second time period KPI indicator matrix; The sparse median module is used to perform sparsification and / or medianization on the difference matrix based on the judgment result to obtain a sparse median matrix. The evaluation result module is used to evaluate the functional characteristics of the application area based on the sparse median matrix. The significance determination module includes: The fifth unit is used for a sample set M before adjustment, and N after adjustment, containing a total of a b c sample groups of data, where M abc and N abc These represent individual raster sample groups before and after adjustment, respectively. Each sample group contains d data points, and the number d is obtained using the following formula: d = Ceiling(p / c), where Ceiling represents rounding up; The sixth unit is used when each raster contains d data points in its data sample group, M. abci N represents the individual unadjusted data in each sample group. abci Represents a single adjusted data point in each sample group, where i (0, d];N abci - M abci This represents the fluctuation of the indicator before and after the adjustment; X represents N. abci - M abci The difference, Let X represent the mean, and S represent the standard deviation of the sample X. After obtaining the above data, construct a single statistic t according to the central limit theorem, such that t: t = / ； Unit 7 is used to determine whether the p-value corresponding to t is significantly different from 0 based on hypothesis testing at a 95% confidence level, and to obtain the judgment result of whether the indicators in the difference matrix have changed significantly. The matrix acquisition module includes: The first unit is used to determine the list of 4G or 5G functional features to be evaluated for application assessment based on daily network optimization needs and equipment support capabilities. The second unit is used to determine call traffic statistics indicators, which include, but are not limited to, access capacity, hold capacity, connection capacity, and carrying capacity. The third unit is used to obtain the first time-segment KPI indicators of the target pilot application area before the application of a single functional feature. The sum of the single functional feature indicators yields p rows of data. After removing blank items, the statistical indicators are sorted in ascending order, and the statistical area is divided into equal parts to form a grid, resulting in the first time-segment KPI indicator matrix M. abc ; The fourth unit is used to obtain the second-period KPI indicators of the target pilot application area after the application of a single functional feature. The sum of the individual indicators of the single functional feature yields p rows of data. After removing blank items, the statistical indicators are sorted in ascending order, and the statistical area is divided into equal parts to form a grid, resulting in the second-period KPI indicator matrix N. abc ; Where M represents all time-segmented KPI information before the application of all features to be evaluated, and N represents all time-segmented KPI information after the application of all features to be evaluated; a, b, and c represent two matrix dimensions, where a represents the number of all features, b represents the number of all KPI statistical indicator items, and c represents the number of indicator segment grids.

5. An electronic device comprising a processor and a memory storing a computer program, characterized in that, When the processor executes the computer program, it implements the steps of the functional characteristic simulation evaluation method for mobile networks according to any one of claims 1 to 3.

6. A processor-readable storage medium, characterized in that, The processor-readable storage medium stores a computer program for causing the processor to perform the steps of the functional characteristic simulation evaluation method for mobile networks as described in any one of claims 1 to 3.

Images