CN111225382A

CN111225382A - LTE functional characteristic selection method and system

Info

Publication number: CN111225382A
Application number: CN201811425700.XA
Authority: CN
Inventors: 郁文尧
Original assignee: China Mobile Communications Group Co Ltd; China Mobile Group Shanghai Co Ltd
Current assignee: China Mobile Communications Group Co Ltd; China Mobile Group Shanghai Co Ltd
Priority date: 2018-11-27
Filing date: 2018-11-27
Publication date: 2020-06-02
Anticipated expiration: 2038-11-27
Also published as: CN111225382B

Abstract

The embodiment of the invention provides a method and a system for selecting LTE functional characteristics, wherein the method comprises the following steps: acquiring a feature vector of each LTE functional characteristic of a cell to be tested; and inputting the feature vector of each LTE functional characteristic of the cell to be tested into a prediction model to obtain the target LTE functional characteristic of the cell to be tested, wherein the prediction model is obtained by training the feature vector of each LTE functional characteristic of each cell in a training sample set. According to the LTE functional characteristic selection method and system provided by the embodiment of the invention, the prediction model is established by counting the historical data of the cell and the selection result of the historical functional characteristic, the real-time condition of the cell to be tested is predicted by the prediction model, the functional characteristic suitable for the current cell to be tested is selected, the obvious benefit is generated compared with the application strategy based on the cell flow or the cell scene, and the utilization rate of resources is improved.

Description

LTE functional characteristic selection method and system

Technical Field

The embodiment of the invention relates to the technical field of communication, in particular to a method and a system for selecting LTE (Long term evolution) functional characteristics.

Background

Long Term Evolution (LTE) is a Long Term Evolution organized by The 3rd generation Partnership Project (3 GPP) to establish a technical standard, and LTE networks are capable of providing a download rate of 300Mbit/s and an upload rate of 75Mbit/s by means of functional features of LTE.

The LTE terminal has different support performances for various functional characteristics due to different LTE terminal performances.

The commonly used function characteristic application strategy of the existing network is based on cell flow or based on cell scene application. For example, a Multiple Input Multiple Output (MIMO) technology and a Quadrature Amplitude Modulation (QAM) technology are generally applied to a high cell with service traffic, and a carrier aggregation technology is generally applied to a cell with an important scene.

The existing network LTE functional characteristic application is mainly based on cell flow or cell scenes, but the application strategy has great limitation.

The LTE functional characteristics are applied according to the cell traffic or the cell scenario, the application strategy is too extensive, and it is certainly thought that there are more functional characteristic requirements in a high traffic or important scenario without strict demonstration, and the validity of the application strategy in a relevant scenario cannot be guaranteed at all.

On the premise that the application strategy is not effective, a large amount of resources and energy are required to be invested subsequently to realize the adjustment of the LTE functional characteristics.

The method aims to improve the bearing capacity of a cell on the premise of not increasing resources by putting MIMO and QAM characteristics into a high-flow cell, but is limited by the performance of a terminal, and if the number and the proportion of terminals supporting corresponding characteristics cannot be confirmed, the application effect cannot be basically guaranteed.

The carrier aggregation characteristic is input in an important scene, the aim is to improve the peak rate of a single user on the premise of not increasing resources, but the peak rate is limited by the performance of the terminal, and if the number and the proportion of the terminals supporting the corresponding characteristic cannot be confirmed, the application effect cannot be ensured.

In summary, screening dimensions are artificially defined according to cell traffic or cell scenes, and effective utilization of resources cannot be basically realized according to demands.

Disclosure of Invention

The embodiment of the invention provides a method and a system for selecting LTE (Long term evolution) functional characteristics, which are used for solving the problem that the functional characteristics of a cell in the prior art cannot be selected according to actual conditions, so that the resource utilization rate is low.

In a first aspect, an embodiment of the present invention provides an LTE function characteristic selection method, including:

acquiring a feature vector of each LTE functional characteristic of a cell to be tested, wherein for any LTE functional characteristic, the feature vector of any LTE functional characteristic of the cell to be tested comprises one or more of the following nine parameters: the ratio of the supported number of any LTE functional characteristic in the cell to be tested, the RSRP level of any LTE functional characteristic support terminal, the average RSRQ level of any LTE functional characteristic support terminal, the frequency point of the cell to be tested, the data traffic of the cell to be tested, the PRB utilization rate of the cell to be tested, the MR coverage rate of the cell to be tested, the terminal transmission power allowance of the cell to be tested and the uplink signal-to-noise ratio of the cell to be tested;

and inputting the feature vector of each LTE functional characteristic of the cell to be tested into a prediction model to obtain the target LTE functional characteristic of the cell to be tested, wherein the prediction model is obtained by training the feature vector of each LTE functional characteristic of each cell in a training sample set.

In a second aspect, an embodiment of the present invention provides an LTE function characteristic selection system, including:

the feature module is configured to obtain a feature vector of each LTE functional characteristic of the cell to be tested, where, for any LTE functional characteristic, the feature vector of any LTE functional characteristic of the cell to be tested includes one or more of the following nine parameters: the ratio of the supported number of any LTE functional characteristic in the cell to be tested, the RSRP level of any LTE functional characteristic support terminal, the average RSRQ level of any LTE functional characteristic support terminal, the frequency point of the cell to be tested, the cell data traffic of the cell to be tested, the PRB utilization rate of the cell to be tested, the MR coverage rate of the cell to be tested, the terminal transmission power allowance of the cell to be tested and the uplink signal-to-noise ratio of the cell to be tested;

the prediction module is used for inputting the feature vector of each LTE functional characteristic of the cell to be tested into a prediction model to obtain the target LTE functional characteristic of the cell to be tested, and the prediction model is obtained by training the feature vector of each LTE functional characteristic of each cell in a training sample set.

In a third aspect, an embodiment of the present invention provides an electronic device, including:

at least one processor, at least one memory, a communication interface, and a bus; wherein the content of the first and second substances,

the processor, the memory and the communication interface complete mutual communication through the bus;

the communication interface is used for information transmission between the test equipment and the communication equipment of the display device;

the memory stores program instructions executable by the processor, and the processor calls the program instructions to execute an LTE functional characteristic selection method provided by the first aspect.

In a fourth aspect, an embodiment of the present invention provides a non-transitory computer-readable storage medium, which stores computer instructions, where the computer instructions cause the computer to execute the LTE functional characteristic selection method provided in the first aspect.

According to the LTE functional characteristic selection method and system provided by the embodiment of the invention, the prediction model is established by counting the historical data of the cell and the selection result of the historical functional characteristic, the real-time condition of the cell to be tested is predicted by the prediction model, the functional characteristic suitable for the current cell to be tested is selected, the obvious benefit is generated compared with the application strategy based on the cell flow or the cell scene, and the utilization rate of resources is improved.

Drawings

In order to more clearly illustrate the embodiments of the present invention or the technical solutions in the prior art, the drawings used in the description of the embodiments or the prior art will be briefly described below, and it is obvious that the drawings in the following description are some embodiments of the present invention, and those skilled in the art can also obtain other drawings according to the drawings without creative efforts.

Fig. 1 is a flowchart of an LTE functional characteristic selection method according to an embodiment of the present invention;

fig. 2 is a schematic structural diagram of an LTE functional characteristic selection system according to an embodiment of the present invention;

fig. 3 illustrates a physical structure diagram of an electronic device.

Detailed Description

In order to make the objects, technical solutions and advantages of the embodiments of the present invention clearer, the technical solutions in the embodiments of the present invention will be clearly and completely described below with reference to the drawings in the embodiments of the present invention, and it is obvious that the described embodiments are some, but not all, embodiments of the present invention. All other embodiments, which can be derived by a person skilled in the art from the embodiments given herein without making any creative effort, shall fall within the protection scope of the present invention.

Fig. 1 is a flowchart of an LTE function characteristic selection method according to an embodiment of the present invention, where as shown in fig. 1, the method includes:

s1, acquiring a feature vector of each LTE functional characteristic of the cell to be tested, where, for any LTE functional characteristic, the feature vector of any LTE functional characteristic of the cell to be tested includes one or more of the following nine parameters: the ratio of the supported number of any LTE functional characteristic in the cell to be tested, the RSRP level of any LTE functional characteristic support terminal, the average RSRQ level of any LTE functional characteristic support terminal, the frequency point of the cell to be tested, the data traffic of the cell to be tested, the PRB utilization rate of the cell to be tested, the MR coverage rate of the cell to be tested, the terminal transmission power allowance of the cell to be tested and the uplink signal-to-noise ratio of the cell to be tested;

s2, inputting the feature vector of each LTE functional characteristic of the cell to be tested into a prediction model to obtain the target LTE functional characteristic of the cell to be tested, wherein the prediction model is obtained by training the feature vector of each LTE functional characteristic of each cell in a training sample set.

The LTE functional characteristics comprise an MIMO technology, a QAM technology, a carrier aggregation technology and the like.

Firstly, a feature vector of each LTE functional characteristic in a cell to be tested is obtained, and for any LTE functional characteristic feature vector, parameters of the feature vector include one or more of the following nine parameters, which are described as an example in the embodiment of the present invention:

the support quantity proportion of the LTE functional characteristics, the average RSRP level of the LTE functional characteristic support terminal, the average RSRQ level of the LTE functional characteristic support terminal and the traffic extraction statistical index of the cell to be tested, wherein the traffic extraction statistical index of the cell to be tested comprises the following steps: the method comprises the steps of frequency points of a cell to be tested, data traffic of the cell to be tested, PRB utilization rate of the cell to be tested, MR coverage rate of the cell to be tested, terminal transmission power allowance of the cell to be tested and uplink signal-to-noise ratio of the cell to be tested.

It should be noted that RSRP (Reference Signal Receiving Power) is one of the key parameters that can represent the wireless Signal strength in the LTE network and the physical layer measurement requirement, and is the average value of the received Signal Power on all REs (resource elements) that carry Reference signals within a certain symbol.

The RSRQ (Reference Signal Receiving Quality) represents the LTE Reference Signal reception Quality, and this metric mainly ranks different LTE candidate cells according to Signal Quality. This measurement is used as input for handover and cell reselection decisions.

PRB (Physical Resource Block) denotes a Physical Resource Block.

Secondly, inputting the obtained feature vector of each LTE functional characteristic of the cell to be tested into a prediction model, wherein the prediction model is obtained by training the feature vector of each LTE functional characteristic of each cell in a training sample set, and the prediction model is obtained by training the feature vector of each LTE functional characteristic in the training sample set. The content in the training sample set is data information of the cell historical functional characteristics.

It should be further noted that the average RSRP level of the LTE functional characteristic supporting terminal, the average RSRQ level of the LTE functional characteristic supporting terminal, and the extracted traffic statistic index of the cell to be measured may be obtained by:

step A1: and mapping the terminal IMEITAC and the terminal name.

The IMEITAC can uniquely identify the terminal name or the terminal model, the terminal IMEITAC and the terminal name mapping information are stored in a file, and software is imported to realize mapping.

Step A2: and mapping the terminal name and the terminal chip.

The terminal name and the terminal chip can be obtained by extracting main IT website information, and mapping of the LTE terminal and the chip model thereof is realized by using a web crawler through software coding. And accessing an IT website page in a circulation mode, and acquiring the terminal chip capacity information in the HTML webpage through a regular expression.

Step A3: mapping the chip and the terminal Category capability.

The capability of the chip and the terminal Category can be obtained by crawling GSA information, mapping information of the chip and the terminal Category is downloaded, and software is introduced to realize mapping.

Step A4: terminal Category capability is mapped with support function characteristics.

The terminal Category capability and the support function characteristic can be acquired by inquiring the 3GPP protocol, mapping information is acquired by inquiring, and software is imported to realize mapping.

Step A5: and mapping the terminal IMEITAC with the support function characteristics.

And based on the obtained data of A1 (mapping the terminal IMEITAC and the terminal name), A2 (mapping the terminal name and the terminal chip), A3 (mapping the chip and the terminal Category capability) and A4 (mapping the terminal Category capability and the support function characteristic), keyword screening is carried out, and the software automatically completes the mapping of the terminal IMEITAC and the support function characteristic.

Step a shows a mapping relationship between the terminal and the functional characteristics supported by the terminal, and after the terminal and the functional characteristics supported by the terminal are obtained, the traffic statistic data of the cell to be measured also needs to be obtained, which is specifically obtained in step B.

And B: MRO and OTT data extraction and summarization.

And C, completing statistics of terminal model feature support conditions based on the energy mapping in the step A, and acquiring terminal feature support capability information under the cell to be tested.

And step B, extracting and summarizing MRO and OTT data, counting model number information of the cell level terminal to be detected, and outputting a multi-dimensional characteristic vector.

The specific implementation process is as follows:

step B1: MRO and OTT data extraction.

MRO refers to a measurement report reported by a mobile phone, and the measurement period is reported once every 5 seconds;

OTT is an abbreviation for "Over The Top," and means providing various application services to users through The internet. And backfilling the MRO by using the time stamp, the network measurement information such as MME UE S1AP ID associated user position information and MRO RSRP and the like, and the information such as longitude and latitude and the like, so as to realize the output of the processed MRO data.

Step B2: MRO and OTT data processing is realized based on the functional characteristics.

And B, obtaining the functional characteristics based on the query in the step A, and counting the ratio of the number of each characteristic support in the cell level range to be tested, the average RSRP level of the functional characteristic support terminal and the average RSRQ level of the functional characteristic support terminal by taking the terminal IMEITAC as an index.

Step B3: and extracting cell traffic statistic data.

And extracting telephone traffic statistical indexes including frequency points of the cell to be detected, data traffic of the cell to be detected, PRB utilization rate of the cell to be detected, MR coverage rate of the cell to be detected, terminal transmission power allowance of the cell to be detected and uplink signal-to-noise ratio of the cell to be detected according to the cell level range to be detected.

Step B4: and realizing data summarization based on the cell dimension.

And summarizing 9 data dimensions in a cell-level range to be detected, wherein the data dimensions comprise the support quantity proportion of each functional characteristic, the average RSRP level of each functional characteristic support terminal, the average RSRQ level of each functional characteristic support terminal and traffic statistic data extraction of the cell to be detected.

Each functional characteristic is output by the 9 feature vectors.

According to the LTE functional characteristic selection method provided by the embodiment of the invention, the prediction model is established by counting the historical data of the cell and the selection result of the historical functional characteristic, the real-time condition of the cell to be tested is predicted through the prediction model, the functional characteristic of the current cell to be tested is selected, the method has obvious benefits compared with the application strategy based on the cell flow or the cell scene, and the utilization rate of resources is improved.

On the basis of the above embodiment, preferably, the prediction model is obtained by:

acquiring each LTE functional characteristic of each cell in the training sample set;

acquiring a feature vector of each LTE functional characteristic of each cell in the training sample set;

and obtaining the prediction model according to the feature vector of each LTE functional characteristic of each cell in the training sample set and a gradient lifting algorithm.

Specifically, the prediction model is obtained by training a feature vector of each LTE functional characteristic of each cell in a training sample set by using a gradient boosting algorithm.

The so-called gradient boosting algorithm, which is a machine learning technique that solves the regression and classification problem, generates a prediction model by integrating weak prediction models (such as decision trees). It builds models in a step-wise fashion like other lifting methods and generalizes them by allowing the use of arbitrarily differentiable loss functions.

Gradient lifting squareThe method also combines weak learners by an iterative method to form a strong learner. The principle of the algorithm is easily explained in least squares regression, which aims to "teach" the model F by subtracting the square error

Minimize (average training set) to predict the form

The value of (c).

And (4) obtaining the feature vector of the LTE functional characteristic of each cell in the training sample set through the step A and the step B, and iterating to perform gradient enhancement calculation, so that the residual error is reduced, and the application accuracy is improved.

Specifically, after the feature vector of the LTE functional characteristic of the cell to be tested is obtained, the feature vector needs to be preprocessed, and the feature vector of each LTE functional characteristic of each cell in the test sample set also needs to be preprocessed.

Similarly, the feature vector of each LTE functional characteristic of each cell in the training sample set is preprocessed, and the specific preprocessing method includes:

step C1: and screening 9 feature vectors of each functional characteristic by taking a cell level as a unit, realizing descending sorting by taking the flow generated by the functional characteristics under the cell as a basis, and marking the first 15 percent as effective and marking the last 15 percent as ineffective. The processed 9 feature vectors of the 30% cells and their labels are input to step C2.

Step C2: and (4) finishing the construction and optimization of the gradient enhancement method model based on the characteristic vectors of the functional characteristics (the input data of the step C1). And obtaining the application effect and the parameter importance of the model.

The gradient enhancement method learns from the residual error of the previous training result, utilizes the loss function negative gradient value in the current model, and realizes residual error optimization through continuous iterative fitting.

The specific implementation method comprises the following steps:

f₀(x)＝0，

f_m(x)＝f_m-1(x)+T(x；Θ_m),m＝1,2,…,M，

when the mth step is realized, the current model f is given_m-1(x) Solving the following formula to obtain

We solve for

I.e. the best fit of the model obtained at this time.

When the sum of squares is used as the loss function, the calculation formula can be simplified as follows: l (y, f (x)) ═ y-f (x))²，

The corresponding solution to the minimum of the loss function is solved at this point.

L(y,f_m-1(x)+T(x；Θ_m))

＝(y-f_m-1(x)-T(x；Θ_m))²，

＝(r-T(x；Θ_m))²

At the beginning of model building, each sample is given equal weight, i.e. the importance is consistent at the beginning. According to the result obtained in each training, the data points are different, so that the weighted values of the feature dimensions are distributed again after each step: and increasing the weight of the misclassified points and reducing the weight of the correct classified points.

After multiple iterations are performed on the feature dimensions, a plurality of base classifiers are obtained, and a final model is obtained by combining the base classifiers and voting.

However, when the gradient enhancement method is adopted, no baseline value is set for part of parameters, and iterative optimization needs to be continuously performed for various problems and methods. Therefore, in the process of constructing the model, iterative optimization is continuously carried out aiming at the maximum depth, the minimum sample number of the root node and the number of the nodes.

And continuously iterating to obtain a parameter optimal solution based on a gradient lifting method, so that a residual value is minimized, and the application accuracy is greatly improved.

And when the accuracy of the application model reaches the standard, completing the optimization process of the model, and realizing the output model. If the accuracy target cannot be reached, the target value is readjusted to obtain the optimal parameters of the prediction model.

On the basis of the above embodiments, preferably, the parameters of the prediction model are optimized by using the feature vectors of each LTE functional characteristic of each cell in the test sample set.

After the prediction model is obtained, the prediction model is tested by using the test sample data in the test sample set, and if the test result does not reach the standard, the prediction model can be optimized according to the test sample data so as to improve the accuracy of the prediction model.

In order to better explain the technical scheme of the embodiment of the invention, a specific embodiment is taken as an example for explanation:

step 1: selecting a downlink 256QAM functional characteristic as a sample, firstly realizing the correlation of the IMEITAC and a terminal Category, wherein the current downlink 256QAM support includes Category6, Category7, Category9, Category12, Category16 and Category18, and a software system carries out screening processing on IMEITAC sampling points of related categories and obtains the mapping relation between the terminal and the support function.

Step 2: the data extraction processing module extracts OTT and MRO data, and outputs 9 dimensional feature vectors (downlink 256QAM characteristic support proportion, average downlink 256QAM characteristic support terminal RSRP level, average downlink 256QAM characteristic support terminal RSRQ level, cell frequency point, cell data traffic, cell PRB utilization rate, cell MR coverage rate, cell terminal transmission power allowance and cell uplink signal-to-noise ratio) by taking a cell as a unit based on the mapping relation between a terminal and a support function.

And step 3: establishing a prediction model, wherein the construction method of the prediction model comprises the following steps: for the downlink 256QAM, the characteristic is labeled by applying dependent variable. For the cell with the applied downlink 256QAM characteristics, the downlink 256QAM flow generated under the cell is sorted in a descending order according to the mark of 'effective' for the first 15% and 'ineffective' for the last 15%. 9 eigenvectors and labels are input for 30% of the cells (700 cells are involved) after processing.

The model randomly selects 70% of cells as a training set (490 cells) and 30% of cells as a test set (210 cells) in 700 sampling samples, performs model optimization by using training set data, and evaluates the application effectiveness of the test set cells by using an output model. Iterative optimization is respectively performed on the maximum depth, the minimum sample number and the number of iterative predictors, the threshold of model accuracy is set to be 96%, table 1 is an iterative process of a gradient enhancement algorithm, and as shown in table 1:

TABLE 1

And 4, step 4: and inputting the feature vector of each LTE functional characteristic in the test set into a prediction model, judging whether the prediction accuracy rate reaches the standard, if not, optimizing calculation, and if so, performing data processing application in a cell in the whole network range to obtain the target LTE functional characteristic.

The commonly used function characteristic application strategy of the existing network is based on cell flow or based on cell scene application. For example, MIMO technology and QAM technology are generally applied to a traffic high cell, and carrier aggregation technology is generally applied to a cell in an important scenario. The use of related methods does not effectively and accurately assess user needs. The embodiment of the invention realizes reasonable application of functional characteristics based on the enhancement method, and greatly improves the characteristic application efficiency.

The method has the following advantages:

by the gradient enhancement method, the functional characteristic application is started, the characteristic application efficiency is improved, the resource input cost of a company is saved, and the use experience of a user is not influenced basically.

The method does not need to change the network structure, is simple to operate and has no network risk. Analysis is performed based on the derived data and application results are given after calculation.

Fig. 2 is a schematic structural diagram of an LTE functional characteristic selection system according to an embodiment of the present invention, and as shown in fig. 2, the system includes a feature module 201 and a prediction module 202, where:

the feature module 201 is configured to obtain a feature vector of each LTE functional characteristic of the cell to be tested, where, for any LTE functional characteristic, the feature vector of any LTE functional characteristic of the cell to be tested includes one or more of the following nine parameters: the ratio of the supported number of any LTE functional characteristic in the cell to be tested, the RSRP level of any LTE functional characteristic support terminal, the average RSRQ level of any LTE functional characteristic support terminal, the frequency point of the cell to be tested, the cell data traffic of the cell to be tested, the PRB utilization rate of the cell to be tested, the MR coverage rate of the cell to be tested, the terminal transmission power allowance of the cell to be tested and the uplink signal-to-noise ratio of the cell to be tested;

the prediction module 202 is configured to input the feature vector of each LTE functional characteristic of the cell to be measured into a prediction model, and obtain a target LTE functional characteristic of the cell to be measured, where the prediction model is obtained by training the feature vector of each LTE functional characteristic of each cell in a training sample set.

Firstly, the feature module 201 is configured to obtain a feature vector of each LTE functional characteristic of the cell to be tested, and for any LTE functional characteristic, the feature vector of the LTE functional characteristic of the cell to be tested includes: the supporting quantity proportion of the LTE functional characteristics, the RSRP level of the LTE functional characteristic supporting terminal, the average RSRQ level of the LTE functional characteristic supporting terminal, the frequency point of the cell to be tested, the data traffic of the cell to be tested, the PRG utilization rate of the cell to be tested, the MR coverage rate of the cell to be tested, the terminal transmitting power allowance of the cell to be tested and the uplink signal-to-noise ratio of the cell to be tested.

The prediction module 202 inputs the feature vector of each LTE functional characteristic acquired by the feature module 201 into the prediction model to obtain the target LTE functional characteristic of the cell to be measured.

The specific implementation process of the embodiment of the system is the same as that of the embodiment of the method, and please refer to the embodiment of the method for details, which is not described herein again.

According to the LTE functional characteristic selection system provided by the embodiment of the invention, the prediction model is established by counting the historical data of the cell and the selection result of the historical functional characteristic, the real-time condition of the cell to be tested is predicted through the prediction model, the functional characteristic of the current cell to be tested is selected, the obvious benefit is generated compared with the application strategy based on the cell flow or the cell scene, and the utilization rate of resources is improved.

On the basis of the above embodiment, preferably, the prediction model in the prediction module is obtained by:

Fig. 3 illustrates a physical structure diagram of an electronic device, and as shown in fig. 3, the server may include: a processor (processor)310, a communication Interface (communication Interface)320, a memory (memory)330 and a bus 340, wherein the processor 310, the communication Interface 320 and the memory 330 complete communication with each other through the bus 340. The communication interface 340 may be used for information transmission between the server and the smart tv. The processor 310 may call logic instructions in the memory 330 to perform the following method:

In addition, the logic instructions in the memory 330 may be implemented in the form of software functional units and stored in a computer readable storage medium when the software functional units are sold or used as independent products. Based on such understanding, the technical solution of the present invention may be embodied in the form of a software product, which is stored in a storage medium and includes instructions for causing a computer device (which may be a personal computer, a server, or a network device) to execute all or part of the steps of the method according to the embodiments of the present invention. And the aforementioned storage medium includes: a U-disk, a removable hard disk, a Read-Only Memory (ROM), a Random Access Memory (RAM), a magnetic disk or an optical disk, and other various media capable of storing program codes.

The present embodiments provide a non-transitory computer-readable storage medium storing computer instructions that cause the computer to perform the methods provided by the above method embodiments, for example, including:

Those of ordinary skill in the art will understand that: all or part of the steps for implementing the method embodiments may be implemented by hardware related to program instructions, and the program may be stored in a computer readable storage medium, and when executed, the program performs the steps including the method embodiments; and the aforementioned storage medium includes: various media that can store program codes, such as ROM, RAM, magnetic or optical disks.

Through the above description of the embodiments, those skilled in the art will clearly understand that each embodiment can be implemented by software plus a necessary general hardware platform, and certainly can also be implemented by hardware. With this understanding in mind, the above-described technical solutions may be embodied in the form of a software product, which can be stored in a computer-readable storage medium such as ROM/RAM, magnetic disk, optical disk, etc., and includes instructions for causing a computer device (which may be a personal computer, a server, or a network device, etc.) to execute the methods described in the embodiments or some parts of the embodiments.

Finally, it should be noted that: the above examples are only intended to illustrate the technical solution of the present invention, but not to limit it; although the present invention has been described in detail with reference to the foregoing embodiments, it will be understood by those of ordinary skill in the art that: the technical solutions described in the foregoing embodiments may still be modified, or some technical features may be equivalently replaced; and such modifications or substitutions do not depart from the spirit and scope of the corresponding technical solutions of the embodiments of the present invention.

Claims

1. An LTE functional characteristic selection method, comprising:

2. The method of claim 1, wherein the predictive model is obtained by:

3. The method of claim 1, wherein the parameters of the predictive model are optimized using feature vectors for each LTE functional characteristic of each cell in the test sample set.

4. The method of claim 2, wherein the inputting the feature vector of each LTE functional characteristic of the cell to be tested into a prediction model to obtain the target LTE functional characteristic of the cell to be tested, further comprises:

and preprocessing the characteristic vector of each LTE functional characteristic of the cell to be detected.

5. The method of claim 2, wherein the obtaining the prediction model according to the target LTE functional characteristics of each cell, the feature vector of each cell, and a gradient boosting algorithm further comprises:

pre-processing a feature vector for each LTE functional characteristic for each cell in the training sample set.

6. An LTE functional characteristic selection system, comprising:

the feature module is configured to obtain a feature vector of each LTE functional characteristic of the cell to be tested, where, for any LTE functional characteristic, the feature vector of any LTE functional characteristic of the cell to be tested includes one or more of the following nine parameters: the ratio of the supported number of any LTE functional characteristic in the cell to be tested, the RSRP level of any LTE functional characteristic support terminal, the average RSRQ level of any LTE functional characteristic support terminal, the frequency point of the cell to be tested, the data traffic of the cell to be tested, the PRB utilization rate of the cell to be tested, the MR coverage rate of the cell to be tested, the terminal transmission power allowance of the cell to be tested and the uplink signal-to-noise ratio of the cell to be tested;

7. The system of claim 6, wherein the prediction model in the prediction module is obtained by:

8. An electronic device, comprising:

the memory stores program instructions executable by the processor, the processor invoking the program instructions to perform the method of any one of claims 1-5.

9. A non-transitory computer-readable storage medium storing computer instructions that cause a computer to perform the method of any one of claims 1 to 5.