Poor quality cell identification method, device, equipment and readable storage medium

Abstract

The present invention discloses a method, device, equipment and readable storage medium for identifying poor quality cells. The method comprises: obtaining target download rates of at least two cells, wherein the target download rates include at least one of the download rates corresponding to the service types of the cells and the comprehensive download rates of the service types of the cells; obtaining the number of cells corresponding to each target download rate; determining the mathematical expectation of the target download rate and the standard deviation of the target download rate according to the target download rate and the number of cells; determining the normal distribution function of the target download rate according to the mathematical expectation and the standard deviation; determining the poor quality threshold according to the inflection point of the normal distribution function and the mathematical expectation; and treating the cell corresponding to the target download rate less than the poor quality threshold as a poor quality cell. The present invention identifies the poor quality cell by determining the poor quality threshold.

Description

Technical Field

[0001] The present invention relates to the field of mobile communication technology, and in particular to a method, device, equipment and readable storage medium for identifying a poor quality cell. Background Art

[0002] With the development of mobile communication technology, the continuous construction, transformation and upgrading of mobile communication networks have led to the rapid development of mobile data services based on mobile communication networks. With this development, users' demand for mobile Internet services is getting higher and higher, which also puts higher requirements on mobile communication networks.

[0003] With the development of data services in mobile communication networks, the size of service packets has an increasingly significant impact on the evaluation of mobile communication networks. Judging poor-quality cells based on user behavior characteristics is easily interfered by data from individual service servers, resulting in low accuracy in judging poor-quality cells. Summary of the invention

[0004] The main purpose of the present invention is to provide a method, device, equipment and readable storage medium for identifying poor quality cells, aiming to solve the problem of how to improve the accuracy of poor quality cell judgment.

[0005] To achieve the above object, the present invention provides a method for identifying a poor quality cell, the method comprising the following steps:

[0006] Obtaining target download rates of at least two cells, where the target download rates include at least one of download rates corresponding to respective service types of the cells and comprehensive download rates of respective service types of the cells;

[0007] Get the number of cells corresponding to each target download rate;

[0008] Determine a mathematical expectation of the target download rate and a standard deviation of the target download rate according to the target download rate and the number of cells;

[0009] Determine a normal distribution function of the target download rate according to the mathematical expectation and the standard deviation;

[0010] Determining a quality difference threshold according to the inflection point of the normal distribution function and the mathematical expectation;

[0011] The cell corresponding to the target download rate that is less than the quality difference threshold is regarded as a poor quality cell.

[0012] In one embodiment, the step of obtaining target download rates of at least two cells includes:

[0013] Acquire extended XDR data of call data records of at least two cells, wherein the XDR data includes at least one XDR record;

[0014] A target download rate for each cell is determined according to the XDR data.

[0015] In one embodiment, the step of determining the target download rate of each cell according to the XDR data includes:

[0016] Determine the downlink traffic and downlink duration of each cell within a preset time period according to the XDR data;

[0017] Determine the sum of the downlink traffic and the sum of the downlink duration;

[0018] The comprehensive download rate of each cell is determined according to the sum of the downlink traffic and the sum of the downlink duration, and the target download rate includes the comprehensive download rate.

[0019] In one embodiment, the step of determining the target download rate of each cell according to the XDR data includes:

[0020] Determine the file type corresponding to each cell according to the XDR data;

[0021] Determine the service type of each cell according to the file type;

[0022] Determine the downlink traffic and downlink duration of each service type in each cell respectively;

[0023] The download rates corresponding to the various service types of each cell are determined according to the downlink traffic and downlink duration of each service type, and the target download rate includes the download rates corresponding to the various service types.

[0024] In one embodiment, the step of determining the quality difference threshold according to the inflection point of the normal distribution function and the mathematical expectation includes:

[0025] Determine a comprehensive threshold according to the inflection point and mathematical expectation of the normal distribution function corresponding to the comprehensive download rate, wherein the quality difference threshold includes the comprehensive threshold;

[0026] and / or,

[0027] The service threshold is determined according to the inflection point and mathematical expectation of the normal distribution function corresponding to the download rate corresponding to each service type, and the quality difference threshold includes the service threshold of each service type.

[0028] In one embodiment, before the step of determining the downlink traffic and downlink duration of each cell within a preset time period according to the XDR data, the step further includes:

[0029] Determine a packet threshold of a session packet corresponding to the XDR record according to the XDR data;

[0030] The XDR record in the XDR data in which the session packet is smaller than the small packet threshold is deleted.

[0031] In one embodiment, the step of determining the packet threshold of the session packet corresponding to the XDR record according to the XDR data includes:

[0032] Determine the data volume and download rate of the session packet corresponding to the XDR record;

[0033] Dividing the data volume into a plurality of intervals according to a preset step size, and sorting the intervals according to the size of the data volume;

[0034] Determine a download rate of the XDR record corresponding to each interval;

[0035] Determine the growth rate of the download rate according to the download rates of the adjacent intervals;

[0036] determining a growth threshold according to a maximum value among the growth rates;

[0037] When the number of growth rates less than the growth threshold is greater than a preset threshold, and the number of growth rates greater than the preset threshold is continuous, a reference growth rate is determined from the preset continuous number of growth rates;

[0038] The lower limit value of the data volume in the interval corresponding to the reference growth rate is used as the packet threshold of the session packet corresponding to the XDR record.

[0039] In one embodiment, the step of taking a cell corresponding to the target download rate less than the quality difference threshold as a poor quality cell includes:

[0040] The cell corresponding to the target download rate which is less than the quality difference threshold and whose duration of being less than the quality difference threshold is greater than a preset time is regarded as a poor quality cell.

[0041] In one embodiment, the step of determining the mathematical expectation of the target download rate and the standard deviation of the target download rate according to the target download rate and the number of cells includes:

[0042] Determine the mathematical expectation according to the target download rate, the number of cells and the total number of cells;

[0043] The standard deviation is determined according to the target download rate corresponding to each cell and the mathematical expectation.

[0044] In one embodiment, the step of determining the quality difference threshold according to the inflection point of the normal distribution function and the mathematical expectation includes:

[0045] Determine an inflection point of the normal distribution function at which the target download rate is less than the mathematical expectation;

[0046] Determine a target download rate corresponding to the inflection point;

[0047] The target download rate corresponding to the inflection point is used as the quality difference threshold.

[0048] To achieve the above object, the present invention further provides a poor quality cell identification device, the poor quality cell identification device comprising:

[0049] A rate acquisition module, used to acquire target download rates of at least two cells, wherein the target download rates include at least one of download rates corresponding to respective service types of the cells and comprehensive download rates of respective service types of the cells;

[0050] A quantity acquisition module is used to obtain the number of cells corresponding to each target download rate;

[0051] A calculation module, used to determine the mathematical expectation of the target download rate and the standard deviation of the target download rate according to the target download rate and the number of cells;

[0052] A function determination module, used to determine a normal distribution function of the target download rate according to the mathematical expectation and the standard deviation;

[0053] A threshold determination module, used to determine a quality difference threshold according to the inflection point of the normal distribution function and the mathematical expectation;

[0054] The identification module is used to regard the cell corresponding to the target download rate less than the quality difference threshold as a poor quality cell.

[0055] To achieve the above-mentioned purpose, the present invention also provides a device for identifying poor quality cells, which includes a memory, a processor, and a poor quality cell identification program stored in the memory and executable on the processor. When the poor quality cell identification program is executed by the processor, it implements the various steps of the poor quality cell identification method described above.

[0056] To achieve the above objectives, the present invention also provides a readable storage medium, which stores a poor quality cell identification program, and when the poor quality cell identification program is executed by a processor, the various steps of the poor quality cell identification method described above are implemented.

[0057] The present invention provides a method, device, equipment and readable storage medium for identifying a poor quality cell. By calculating the mathematical expectation and standard deviation corresponding to the target download rate, a normal distribution function of the target download rate of the cell is obtained. The poor quality threshold is determined according to the inflection point of the normal distribution function and the mathematical expectation of the target download rate, and the cell corresponding to the target download rate less than the poor quality threshold is regarded as a poor quality cell. The poor quality threshold is determined by the target download rate of the cell, and the poor quality cell is determined according to the poor quality threshold, thereby improving the accuracy of poor quality cell determination. BRIEF DESCRIPTION OF THE DRAWINGS

[0058] Figure 1 A schematic diagram of the hardware structure of a poor quality cell identification device according to an embodiment of the present invention;

[0059] Figure 2 It is a flowchart of a first embodiment of a method for identifying a poor quality cell according to the present invention;

[0060] Figure 3 This is a detailed flowchart of step S10 of the second embodiment of the method for identifying a poor quality cell of the present invention;

[0061] Figure 4 This is a detailed flow chart of step S12 of the third embodiment of the method for identifying a poor quality cell of the present invention;

[0062] Figure 5 This is a detailed flow chart of step S12 of the fourth embodiment of the method for identifying a poor quality cell of the present invention;

[0063] Figure 6 Detailed flowchart of step S12 of the fifth embodiment of the method for identifying a poor quality cell of the present invention;

[0064] Figure 7 The figure is a schematic diagram of the logical structure of the apparatus for identifying poor quality cells involved in an embodiment of the present invention.

[0065] The realization of the purpose, functional features and advantages of the present invention will be further explained in conjunction with embodiments and with reference to the accompanying drawings. DETAILED DESCRIPTION

[0066] It should be understood that the specific embodiments described herein are only used to explain the present invention, and are not used to limit the present invention.

[0067] The main solution of the embodiment of the present invention is: by calculating the mathematical expectation and standard deviation corresponding to the target download rate, the normal distribution function of the target download rate of the cell is obtained, and the quality difference threshold is determined according to the inflection point of the normal distribution function and the mathematical expectation of the target download rate, and the cell corresponding to the target download rate less than the quality difference threshold is regarded as a poor quality cell.

[0068] The quality difference threshold is determined by the target download rate of the cell, and the poor quality cell is determined according to the quality difference threshold, thereby improving the accuracy of poor quality cell determination.

[0069] As an implementation solution, the identification device of the poor quality cell can be as follows Figure 1 shown.

[0070] The embodiment of the present invention relates to a poor quality cell identification device, which includes: a processor 101, such as a CPU, a memory 102, and a communication bus 103. The communication bus 103 is used to realize connection and communication between these components.

[0071] The memory 102 may be a high-speed RAM memory or a stable memory (non-volatile memory), such as a disk memory. Figure 1 As shown, the memory 102 as a readable storage medium may include a poor quality cell identification program; and the processor 101 may be used to call the poor quality cell identification program stored in the memory 102, and perform the following operations:

[0072] Obtaining target download rates of at least two cells, where the target download rates include at least one of download rates corresponding to respective service types of the cells and comprehensive download rates of respective service types of the cells;

[0073] Get the number of cells corresponding to each target download rate;

[0074] Determine a mathematical expectation of the target download rate and a standard deviation of the target download rate according to the target download rate and the number of cells;

[0075] Determine a normal distribution function of the target download rate according to the mathematical expectation and the standard deviation;

[0076] Determining a quality difference threshold according to the inflection point of the normal distribution function and the mathematical expectation;

[0077] The cell corresponding to the target download rate that is less than the quality difference threshold is regarded as a poor quality cell.

[0078] In one embodiment, the processor 101 may be used to call a poor quality cell identification program stored in the memory 102, and perform the following operations:

[0079] Acquire extended XDR data of call data records of at least two cells, wherein the XDR data includes at least one XDR record;

[0080] A target download rate for each cell is determined according to the XDR data.

[0081] In one embodiment, the processor 101 may be used to call a poor quality cell identification program stored in the memory 102, and perform the following operations:

[0082] Determine the downlink traffic and downlink duration of each cell within a preset time period according to the XDR data;

[0083] Determine the sum of the downlink traffic and the sum of the downlink duration;

[0084] The comprehensive download rate of each cell is determined according to the sum of the downlink traffic and the sum of the downlink duration, and the target download rate includes the comprehensive download rate.

[0085] In one embodiment, the processor 101 may be used to call a poor quality cell identification program stored in the memory 102, and perform the following operations:

[0086] Determine the file type corresponding to each cell according to the XDR data;

[0087] Determine the service type of each cell according to the file type;

[0088] Determine the downlink traffic and downlink duration of each service type in each cell respectively;

[0089] The download rates corresponding to the various service types of each cell are determined according to the downlink traffic and downlink duration of each service type, and the target download rate includes the download rates corresponding to the various service types.

[0090] In one embodiment, the processor 101 may be used to call a poor quality cell identification program stored in the memory 102, and perform the following operations:

[0091] Determine a comprehensive threshold according to the inflection point and mathematical expectation of the normal distribution function corresponding to the comprehensive download rate, wherein the quality difference threshold includes the comprehensive threshold;

[0092] and / or,

[0093] The service threshold is determined according to the inflection point and mathematical expectation of the normal distribution function corresponding to the download rate corresponding to each service type, and the quality difference threshold includes the service threshold of each service type.

[0094] In one embodiment, the processor 101 may be used to call a poor quality cell identification program stored in the memory 102, and perform the following operations:

[0095] Determine a packet threshold of a session packet corresponding to the XDR record according to the XDR data;

[0096] The XDR record in the XDR data in which the session packet is smaller than the small packet threshold is deleted.

[0097] In one embodiment, the processor 101 may be used to call a poor quality cell identification program stored in the memory 102, and perform the following operations:

[0098] Determine the data volume and download rate of the session packet corresponding to the XDR record;

[0099] Dividing the data volume into a plurality of intervals according to a preset step size, and sorting the intervals according to the size of the data volume;

[0100] Determine a download rate of the XDR record corresponding to each interval;

[0101] Determine the growth rate of the download rate according to the download rates of the adjacent intervals;

[0102] determining a growth threshold according to a maximum value among the growth rates;

[0103] When the number of growth rates less than the growth threshold is greater than a preset threshold, and the number of growth rates greater than the preset threshold is continuous, a reference growth rate is determined from the preset continuous number of growth rates;

[0104] The lower limit value of the data volume in the interval corresponding to the reference growth rate is used as the packet threshold of the session packet corresponding to the XDR record.

[0105] In one embodiment, the processor 101 may be used to call a poor quality cell identification program stored in the memory 102, and perform the following operations:

[0106] The cell corresponding to the target download rate which is less than the quality difference threshold and whose duration of being less than the quality difference threshold is greater than a preset time is regarded as a poor quality cell.

[0107] In one embodiment, the processor 101 may be used to call a poor quality cell identification program stored in the memory 102, and perform the following operations:

[0108] Determine the mathematical expectation according to the target download rate, the number of cells and the total number of cells;

[0109] The standard deviation is determined according to the target download rate corresponding to each cell and the mathematical expectation.

[0110] In one embodiment, the processor 101 may be used to call a poor quality cell identification program stored in the memory 102, and perform the following operations:

[0111] Determine an inflection point of the normal distribution function at which the target download rate is less than the mathematical expectation;

[0112] Determine a target download rate corresponding to the inflection point;

[0113] The target download rate corresponding to the inflection point is used as the quality difference threshold.

[0114] Based on the hardware architecture of the above-mentioned poor quality cell identification device, an embodiment of the poor quality cell identification method of the present invention is proposed.

[0115] Reference Figure 2 , Figure 2 This is a first embodiment of a method for identifying a poor quality cell of the present invention, and the method for identifying a poor quality cell comprises the following steps:

[0116] Step S10, obtaining target download rates of at least two cells, wherein the target download rates include at least one of download rates corresponding to respective service types of the cells and a comprehensive download rate of respective service types of the cells.

[0117] Specifically, the target download rate is the download rate of the session packet of the cell, and the target download rate includes at least one of the download rates corresponding to each service type of the cell and the comprehensive download rate of each service type of the cell. The service type may be a video download, an application download, a browsing download, and other service types.

[0118] Obtaining the target download rates of at least two cells may be to determine the target download rates of each cell according to the downlink traffic and downlink duration of the cell. The download rate corresponding to the service type is the ratio of the downlink traffic corresponding to the service type of the cell to the downlink duration, and the comprehensive download rate is the ratio of the total downlink traffic of each service type of the cell to the total downlink duration.

[0119] Step S20, obtaining the number of cells corresponding to each target download rate.

[0120] Specifically, each cell corresponds to a comprehensive download rate, and / or a download rate corresponding to at least one service type. The number of cells corresponding to each target download rate is obtained. For example, if the target download rate is the comprehensive download rate, and the comprehensive download rate is 40 KB / s, the number of cells with a comprehensive download rate of 40 KB / s is determined.

[0121] Step S30: determining the mathematical expectation of the target download rate and the standard deviation of the target download rate according to the target download rate and the number of cells.

[0122] Specifically, the distribution of the target download rate v and the number of cells conforms to the normal distribution v~N(μ,σ 2 ), the mathematical expectation is determined according to the target download rate, the number of cells and the total number of cells, as shown in the following formula:

[0123]

[0124] Where μ is the mathematical expectation of the target download rate; k is the number of target download rate values ​​after deduplication; v i is the target download rate after deduplication of the ith item, n i is the number of cells corresponding to the i-th target download rate value; m is the sum of the number of cells corresponding to the target download rate.

[0125] The standard deviation is determined based on the target download rate and mathematical expectation of each cell, as shown in the following formula:

[0126]

[0127] Where σ is the standard deviation of the target download rate; v j is the target download rate of the jth cell; m is the sum of the number of cells corresponding to the target download rate; μ is the mathematical expectation of the target download rate.

[0128] Step S40: determining a normal distribution function of the target download rate according to the mathematical expectation and the standard deviation.

[0129] Specifically, the normal distribution function of the target download rate is determined by substituting the mathematical expectation and the standard deviation into a preset normal function distribution formula, as shown in the following formula:

[0130] The target download rate conforms to the normal distribution function:

[0131]

[0132] Where f(v) is the normal distribution function, v is the target download rate, μ is the mathematical expectation of the target download rate, and σ is the standard deviation of the target download rate.

[0133] Step S50, determining a quality difference threshold according to the inflection point of the normal distribution function and the mathematical expectation;

[0134] Specifically, the normal distribution function has two inflection points, the inflection point of the normal distribution function at which the target download rate is less than the mathematical expectation is determined, the target download rate corresponding to the inflection point is determined, and the target download rate corresponding to the inflection point is used as the quality difference threshold.

[0135] It can be seen from the normal distribution function corresponding to the target download rate that when ν<μ, the number of cells corresponding to the target download rate gradually increases with the increase of the target download rate, but there is an inflection point in the growth rate, that is, the change point where the f(ν) function changes from a concave function to a convex function, and the change point can be obtained by taking the second-order derivative of the f(ν) function. Let f"(ν)=0 and satisfy ν<μ, and solve ν=μ-σ according to the normal distribution characteristics, which is the quality difference threshold of the comprehensive download rate of the cell.

[0136] Since the target download rate may include at least one of the comprehensive download rate and the download rate corresponding to each service type. The quality difference threshold is determined according to the inflection point and mathematical expectation of the normal distribution function corresponding to the target download rate. The comprehensive threshold can be determined according to the inflection point and mathematical expectation of the normal distribution function corresponding to the comprehensive download rate, and the quality difference threshold includes the comprehensive threshold; and / or, the service threshold is determined according to the inflection point and mathematical expectation of the normal distribution function corresponding to the download rate corresponding to each service type, and the quality difference threshold includes the service threshold of each service type. If the target download rate includes the comprehensive download rate and the download rate corresponding to each service type, the quality difference threshold can be determined according to the comprehensive threshold and the service threshold. Exemplarily, the minimum value of the comprehensive threshold and each service threshold is used as the quality difference threshold.

[0137] Step S60: The cell corresponding to the target download rate that is less than the quality difference threshold is regarded as a poor quality cell.

[0138] Specifically, the target download rate of the cell is compared with the quality difference threshold, and the cell corresponding to the target download rate less than the quality difference threshold is regarded as a poor quality cell. Alternatively, the cell corresponding to the target download rate less than the quality difference threshold and the duration less than the quality difference threshold is greater than a preset time is regarded as a poor quality cell.

[0139] In the technical solution of this embodiment, the normal distribution function of the target download rate of the cell is obtained by calculating the mathematical expectation and standard deviation corresponding to the target download rate, and the quality difference threshold is determined according to the inflection point of the normal distribution function and the mathematical expectation of the target download rate, and the cell corresponding to the target download rate less than the quality difference threshold is regarded as a poor quality cell. The quality difference threshold is determined by the target download rate of the cell, and the poor quality cell is determined according to the quality difference threshold, thereby improving the accuracy of the poor quality cell determination.

[0140] Reference Figure 3 , Figure 3 This is a second embodiment of the method for identifying a poor quality cell of the present invention, based on the first embodiment, step S10 includes:

[0141] Step S11, acquiring extended XDR data of call data records of at least two cells, wherein the XDR data includes at least one XDR record;

[0142] Step S12: determining a target download rate for each cell according to the XDR data.

[0143] Specifically, XDR (X Data Recording, extension of call data recording) data is data obtained based on DPI (Deep Packet Inspection) technology. The XDR data of each cell may include at least one XDR record, and the XDR record includes data records of dimensions such as date, downlink time, cell information, file type, downlink traffic, number of downlink session packets, or downlink duration.

[0144] The target download rate of each cell is determined based on the XDR data. The service type of the cell can be determined by the file type of the XDR data. The download rate of the service type can be determined by the downlink traffic and downlink duration corresponding to the service type of the cell in the XDR data, and the comprehensive download rate can be determined by the total downlink traffic and total downlink duration of each service type of the cell in the XDR data.

[0145] In the technical solution of this embodiment, XDR data of each cell is obtained, and the target download rate of each cell is determined based on the XDR data. Since the type of the obtained XDR data is comprehensive, the calculation of the target download rate of the cell is more accurate.

[0146] Reference Figure 4 , Figure 4 This is a third embodiment of the method for identifying a poor quality cell of the present invention, based on the second embodiment, step S12 includes:

[0147] Step S121, determining the downlink traffic and downlink duration of each cell within a preset time period according to the XDR data;

[0148] Step S122, determining the sum of the downlink traffic and the sum of the downlink duration;

[0149] Step S123, determining the comprehensive download rate of each cell according to the sum of the downlink traffic and the sum of the downlink duration, wherein the target download rate includes the comprehensive download rate.

[0150] Specifically, the target download rate of each cell is determined according to the XDR data, wherein the target download rate may be a comprehensive download rate. The comprehensive download rate of the cell is determined as shown in the following formula:

[0151]

[0152] Among them, v z is the comprehensive download rate of the cell, in KB / S; n1 is the total number of XDR records in the same cell; Lm is the downstream traffic corresponding to the mth XDR record, in KB; T m The downlink duration corresponding to the mth XDR record, in seconds.

[0153] In the technical solution of this embodiment, the sum of the downlink traffic and the sum of the downlink duration of each cell within a preset time period is determined based on the XDR data, and the comprehensive download rate of the cell is determined based on the sum of the downlink traffic and the sum of the downlink duration, thereby measuring the download rates corresponding to all service types of the cell.

[0154] Reference Figure 5 , Figure 5 This is a fourth embodiment of the method for identifying a poor quality cell of the present invention, based on the first embodiment, step S12 includes:

[0155] Step S124, determining the file type corresponding to each cell according to the XDR data;

[0156] Step S125, determining the service type of each cell according to the file type;

[0157] Step S126, respectively determining the downlink traffic and downlink duration of each service type of each cell;

[0158] Step S127, determining the download rates corresponding to the various service types of each cell according to the downlink traffic and downlink duration of each service type, wherein the target download rate includes the download rates corresponding to the various service types.

[0159] Specifically, the target download rate of each cell is determined according to the XDR data, wherein the target download rate may be the download rate of each service type.

[0160] The file type of each cell is determined based on the XDR data, where the file type can be video, application software, browsing history, etc. The service type of the cell can be determined based on the file type, and the service type can include services with large download data volume such as video download, application download, and browsing download. The download rate corresponding to each service type is determined based on the downlink traffic corresponding to each service type and the downlink duration. For example, when the service type is video download, the following formula is shown:

[0161]

[0162] Among them, v s is the download rate of video downloads in the cell, in KB / S; n2 is the total number of XDR records with video download service type in the same cell; L nis the downlink traffic corresponding to the nth XDR record with the service type of video download, in KB; T n The downlink duration corresponding to the nth XDR record whose service type is video download, in seconds.

[0163] In the technical solution of this embodiment, the downlink traffic and downlink duration of each service type in each cell are determined based on XDR data, and the download rate corresponding to each service type in each cell is determined based on the downlink traffic and downlink duration of each service type, thereby measuring the download rate corresponding to each service type in the cell.

[0164] Reference Figure 6 , Figure 6 This is a fifth embodiment of the method for identifying a poor quality cell of the present invention. Based on the third embodiment, before step S121, the method further includes:

[0165] Step S128, determining a small packet threshold of a session packet corresponding to the XDR record according to the XDR data;

[0166] Step S129: deleting the XDR record in the XDR data where the session packet is smaller than the small packet threshold.

[0167] Specifically, since there are situations in the XDR data where the data volume of the session packets corresponding to the XDR records is very small and the downlink duration is short, the target download rate is calculated too large. However, such XDR records cannot measure the quality of the cell service. Before the target download rate is reached, the XDR records corresponding to the session packets with smaller data volumes need to be deleted. Therefore, after determining the small packet threshold of the session packets corresponding to the XDR records based on the XDR data, delete the XDR records in the XDR data where the session packets are smaller than the small packet threshold. Among them, the small packet threshold is the threshold for measuring whether the session packet corresponding to the XDR record is a small packet.

[0168] Determine the packet threshold of the session packet corresponding to the XDR record according to the XDR data, first determine the data volume and download rate of the session packet corresponding to the XDR record; count the average size and download rate of the downlink session packet recorded by the XDR according to the cell XDR dimension, and retain the data correspondence,

[0169]

[0170]

[0171] Among them, S is the average size of the downlink session packet, in KB; L is the downlink traffic corresponding to the XDR record, in KB; Z is the number of downlink session packets corresponding to the XDR record; v is the download rate, in KB / S; T is the downlink duration, in seconds.

[0172] The data volume is divided into multiple intervals according to a preset step size, and the intervals are sorted according to the size of the data volume; the download rate of the XDR record corresponding to each interval is determined; and the average download rate within the statistical interval is used as the download rate corresponding to the interval, as shown in the following formula:

[0173] {(0,w],v1}, {(w,2w],v2}, {(2w,3w],v1}......{((i-1)×w,i×w],v i},

[0174]

[0175] Wherein, w is the step size of the downlink session packet interval division, which can be 10KB; i is the total number of interval divisions, where i×w≥max(S); r represents the total number of XDR records in each interval, j represents the jth XDR record; v is the download rate of each XDR record in the interval.

[0176] The growth rate of the download rate is determined based on the download rates of adjacent intervals, as shown in the following formula:

[0177]

[0178] Where n is an integer in the interval (1, i-1); v i is the average download rate of each interval.

[0179] Determine the growth threshold according to the maximum value of the growth rate, and count the download rate growth rate λ in order of intervals. n , the growth threshold G can be expressed as follows:

[0180]

[0181] Among them, max(λ) is the maximum value of the growth rate of the download rate; g is a constant and can be taken as 10.

[0182] If the number of growth rates less than the growth threshold is greater than the preset threshold, and the number of growth rates greater than the preset threshold is continuous, indicating that the download rate tends to be stable, and the session packet size has little effect on the download rate, then a reference growth rate is determined from the preset continuous number of growth rates; the first growth rate less than the growth threshold in the continuous growth rates can be used as the reference growth rate. The lower limit value of the data volume in the interval corresponding to the reference growth rate is used as the packet size threshold of the session packet corresponding to the XDR record.

[0183] In the technical solution of this embodiment, the packet threshold of the session packet corresponding to the XDR record is determined according to the XDR data, and the XDR record whose session packet in the XDR data is smaller than the packet threshold is deleted. Before calculating the target download rate, the XDR record that cannot measure the quality of the cell service is deleted, thereby improving the accuracy of the target download rate.

[0184] Reference Figure 7 The present invention also provides a poor quality cell identification device, characterized in that the poor quality cell identification device comprises:

[0185] The rate acquisition module 100 is used to acquire target download rates of at least two cells, where the target download rates include at least one of the download rates corresponding to the service types of the cells and the comprehensive download rates of the service types of the cells;

[0186] The quantity acquisition module 200 is used to obtain the number of cells corresponding to each target download rate;

[0187] A calculation module 300, configured to determine a mathematical expectation of the target download rate and a standard deviation of the target download rate according to the target download rate and the number of cells;

[0188] A function determination module 400, configured to determine a normal distribution function of the target download rate according to the mathematical expectation and the standard deviation;

[0189] A threshold determination module 500, configured to determine a quality difference threshold according to the inflection point of the normal distribution function and the mathematical expectation;

[0190] The identification module 600 is used to identify the cell corresponding to the target download rate less than the quality difference threshold as a poor quality cell.

[0191] In one embodiment, in terms of acquiring the target download rates of at least two cells, the rate acquisition module 100 is specifically used to:

[0192] Acquire extended XDR data of call data records of at least two cells, wherein the XDR data includes at least one XDR record;

[0193] A target download rate for each cell is determined according to the XDR data.

[0194] In one embodiment, in determining the target download rate of each cell according to the XDR data, the rate acquisition module 100 is specifically used to:

[0195] Determine the downlink traffic and downlink duration of each cell within a preset time period according to the XDR data;

[0196] Determine the sum of the downlink traffic and the sum of the downlink duration;

[0197] The comprehensive download rate of each cell is determined according to the sum of the downlink traffic and the sum of the downlink duration, and the target download rate includes the comprehensive download rate.

[0198] In one embodiment, in determining the target download rate of each cell according to the XDR data, the rate acquisition module 100 is specifically used to:

[0199] Determine the file type corresponding to each cell according to the XDR data;

[0200] Determine the service type of each cell according to the file type;

[0201] Determine the downlink traffic and downlink duration of each service type in each cell respectively;

[0202] The download rates corresponding to the various service types of each cell are determined according to the downlink traffic and downlink duration of each service type, and the target download rate includes the download rates corresponding to the various service types.

[0203] In one embodiment, in determining the quality difference threshold according to the inflection point of the normal distribution function and the mathematical expectation, the threshold determination module 500 is specifically used to:

[0204] Determining a comprehensive threshold according to the inflection point and mathematical expectation of the normal distribution function corresponding to the comprehensive download rate, wherein the quality difference threshold includes the comprehensive threshold;

[0205] and / or,

[0206] The service threshold is determined according to the inflection point and mathematical expectation of the normal distribution function corresponding to the download rate corresponding to each service type, and the quality difference threshold includes the service threshold of each service type.

[0207] In one embodiment, in determining the downlink traffic and downlink duration of each cell within a preset time period according to the XDR data, the rate acquisition module 100 is specifically used to:

[0208] Determine a packet threshold of a session packet corresponding to the XDR record according to the XDR data;

[0209] The XDR record in the XDR data in which the session packet is smaller than the small packet threshold is deleted.

[0210] In one embodiment, in determining the packet threshold of the session packet corresponding to the XDR record according to the XDR data, the rate acquisition module 100 is specifically configured to:

[0211] Determine the data volume and download rate of the session packet corresponding to the XDR record;

[0212] Dividing the data volume into a plurality of intervals according to a preset step size, and sorting the intervals according to the size of the data volume;

[0213] Determine the download rate of the XDR record corresponding to each interval; determine the growth rate of the download rate according to the download rates of adjacent intervals;

[0214] determining a growth threshold according to a maximum value among the growth rates;

[0215] When the number of growth rates less than the growth threshold is greater than a preset threshold, and the number of growth rates greater than the preset threshold is continuous, a reference growth rate is determined from the preset continuous number of growth rates;

[0216] The lower limit value of the data volume in the interval corresponding to the reference growth rate is used as the packet threshold of the session packet corresponding to the XDR record.

[0217] In one embodiment, in terms of taking a cell corresponding to the target download rate less than the quality difference threshold as a poor quality cell, the identification module 600 is specifically used to:

[0218] The cell corresponding to the target download rate which is less than the quality difference threshold and whose duration of being less than the quality difference threshold is greater than a preset time is regarded as a poor quality cell.

[0219] In one embodiment, in determining the mathematical expectation of the target download rate and the standard deviation of the target download rate according to the target download rate and the number of cells, the calculation module 300 is specifically used to:

[0220] Determine the mathematical expectation according to the target download rate, the number of cells and the total number of cells;

[0221] The standard deviation is determined according to the target download rate corresponding to each cell and the mathematical expectation.

[0222] In one embodiment, in determining the quality difference threshold according to the inflection point of the normal distribution function and the mathematical expectation, the threshold determination module 500 is specifically used to:

[0223] Determine an inflection point of the normal distribution function at which the target download rate is less than the mathematical expectation;

[0224] Determine a target download rate corresponding to the inflection point;

[0225] The target download rate corresponding to the inflection point is used as the quality difference threshold.

[0226] The present invention also provides a device for identifying cells of poor quality, which includes a memory, a processor, and a program for identifying cells of poor quality stored in the memory and executable on the processor. When the program for identifying cells of poor quality is executed by the processor, the various steps of the method for identifying cells of poor quality described in the above embodiment are implemented.

[0227] The present invention also provides a readable storage medium, which stores a poor quality cell identification program. When the poor quality cell identification program is executed by a processor, the various steps of the poor quality cell identification method described in the above embodiment are implemented.

[0228] The serial numbers of the above embodiments of the present invention are only for description and do not represent the advantages or disadvantages of the embodiments.

[0229] It should be noted that, in this article, the terms "include", "comprises" or any other variations thereof are intended to cover non-exclusive inclusion, so that a process, system, article or device including a series of elements includes not only those elements, but also other elements not explicitly listed, or also includes elements inherent to such process, system, article or device. In the absence of further restrictions, an element defined by the sentence "comprises a ..." does not exclude the existence of other identical elements in the process, system, article or device including the element.

[0230] Through the description of the above implementation methods, those skilled in the art can clearly understand that the above-mentioned embodiment system can be implemented by means of software plus a necessary general hardware platform, and of course it can also be implemented by hardware, but in many cases the former is a better implementation method. Based on such an understanding, the technical solution of the present invention is essentially or the part that contributes to the prior art can be embodied in the form of a software product, which is stored in a computer-readable storage medium (such as ROM / RAM, magnetic disk, optical disk) as described above, and includes a number of instructions for a terminal device (which can be a mobile phone, a computer, a parking management device, an air conditioner, or a network device, etc.) to execute the system described in each embodiment of the present invention.

[0231] The above are only preferred embodiments of the present invention, and are not intended to limit the patent scope of the present invention. Any equivalent structure or equivalent process transformation made using the contents of the present invention specification and drawings, or directly or indirectly applied in other related technical fields, are also included in the patent protection scope of the present invention.

Claims

1. A method for identifying a poor quality cell, characterized in that: The method for identifying a poor quality cell comprises: Obtaining target download rates of at least two cells, where the target download rates include at least one of download rates corresponding to respective service types of the cells and comprehensive download rates of respective service types of the cells; Get the number of cells corresponding to each target download rate; Determine a mathematical expectation of the target download rate and a standard deviation of the target download rate according to the target download rate and the number of cells; Determine a normal distribution function of the target download rate according to the mathematical expectation and the standard deviation; Determining a quality difference threshold according to the inflection point of the normal distribution function and the mathematical expectation; The cell corresponding to the target download rate that is less than the quality difference threshold is regarded as a poor quality cell.

2. The method for identifying a poor quality cell according to claim 1, characterized in that: The step of obtaining target download rates of at least two cells comprises: Acquire extended XDR data of call data records of at least two cells, wherein the XDR data includes at least one XDR record; A target download rate for each cell is determined according to the XDR data.

3. The method for identifying poor quality cells according to claim 2, characterized in that: The step of determining the target download rate of each cell according to the XDR data comprises: Determine the downlink traffic and downlink duration of each cell within a preset time period according to the XDR data; Determine the sum of the downlink traffic and the sum of the downlink duration; The comprehensive download rate of each cell is determined according to the sum of the downlink traffic and the sum of the downlink duration, and the target download rate includes the comprehensive download rate.

4. The method for identifying poor quality cells according to claim 2, characterized in that: The step of determining the target download rate of each cell according to the XDR data comprises: Determine the file type corresponding to each cell according to the XDR data; Determine the service type of each cell according to the file type; Determine the downlink traffic and downlink duration of each service type in each cell respectively; The download rates corresponding to the various service types of each cell are determined according to the downlink traffic and downlink duration of each service type, and the target download rate includes the download rates corresponding to the various service types.

5. The method for identifying poor quality cells according to claim 1, characterized in that: The step of determining the quality difference threshold according to the inflection point of the normal distribution function and the mathematical expectation comprises: Determining a comprehensive threshold according to an inflection point and a mathematical expectation of a normal distribution function corresponding to the comprehensive download rate, wherein the quality difference threshold includes the comprehensive threshold; and / or, The service threshold is determined according to the inflection point and mathematical expectation of the normal distribution function corresponding to the download rate corresponding to each service type, and the quality difference threshold includes the service threshold of each service type.

6. The method for identifying poor quality cells as claimed in claim 3, characterized in that: Before the step of determining the downlink traffic and downlink duration of each cell within a preset time period according to the XDR data, the method further includes: Determine a packet threshold of a session packet corresponding to the XDR record according to the XDR data; The XDR record in the XDR data in which the session packet is smaller than the small packet threshold is deleted.

7. The method for identifying poor quality cells according to claim 6, characterized in that: The step of determining the packet threshold of the session packet corresponding to the XDR record according to the XDR data comprises: Determine the data volume and download rate of the session packet corresponding to the XDR record; Dividing the data volume into a plurality of intervals according to a preset step size, and sorting the intervals according to the size of the data volume; Determine a download rate of the XDR record corresponding to each interval; Determine the growth rate of the download rate according to the download rates of the adjacent intervals; determining a growth threshold according to a maximum value among the growth rates; When the number of growth rates less than the growth threshold is greater than a preset threshold, and the number of growth rates greater than the preset threshold is continuous, a reference growth rate is determined from the preset continuous number of growth rates; The lower limit value of the data volume in the interval corresponding to the reference growth rate is used as the packet threshold of the session packet corresponding to the XDR record.

8. The method for identifying poor quality cells according to claim 1, characterized in that: The step of taking the cell corresponding to the target download rate less than the quality difference threshold as a poor quality cell comprises: The cell corresponding to the target download rate which is less than the quality difference threshold and whose duration of being less than the quality difference threshold is greater than a preset time is regarded as a poor quality cell.

9. The method for identifying poor quality cells according to claim 1, characterized in that: The step of determining the mathematical expectation of the target download rate and the standard deviation of the target download rate according to the target download rate and the number of cells comprises: Determine the mathematical expectation according to the target download rate, the number of cells and the total number of cells; The standard deviation is determined according to the target download rate corresponding to each cell and the mathematical expectation.

10. The method for identifying poor quality cells according to claim 1, characterized in that: The step of determining the quality difference threshold according to the inflection point of the normal distribution function and the mathematical expectation comprises: Determine an inflection point of the normal distribution function at which the target download rate is less than the mathematical expectation; Determine a target download rate corresponding to the inflection point; The target download rate corresponding to the inflection point is used as the quality difference threshold.

11. A device for identifying a poor quality cell, characterized in that: The poor quality cell identification device comprises: A rate acquisition module, used to acquire target download rates of at least two cells, wherein the target download rates include at least one of download rates corresponding to respective service types of the cells and comprehensive download rates of respective service types of the cells; A quantity acquisition module is used to obtain the number of cells corresponding to each target download rate; A calculation module, used to determine the mathematical expectation of the target download rate and the standard deviation of the target download rate according to the target download rate and the number of cells; A function determination module, used to determine a normal distribution function of the target download rate according to the mathematical expectation and the standard deviation; A threshold determination module, used to determine a quality difference threshold according to the inflection point of the normal distribution function and the mathematical expectation; The identification module is used to regard the cell corresponding to the target download rate less than the quality difference threshold as a poor quality cell.

12. A device for identifying a poor quality cell, characterized in that: The poor quality cell identification device includes a memory, a processor, and a poor quality cell identification program stored in the memory and executable on the processor. When the poor quality cell identification program is executed by the processor, it implements the various steps of the poor quality cell identification method as described in any one of claims 1-10.

13. A readable storage medium, characterized in that: The readable storage medium stores a poor quality cell identification program, and when the poor quality cell identification program is executed by a processor, each step of the poor quality cell identification method as described in any one of claims 1-10 is implemented.

Images