CN107426759A - The Forecasting Methodology and system of newly-increased base station data portfolio - Google Patents
The Forecasting Methodology and system of newly-increased base station data portfolio Download PDFInfo
- Publication number
- CN107426759A CN107426759A CN201710676951.4A CN201710676951A CN107426759A CN 107426759 A CN107426759 A CN 107426759A CN 201710676951 A CN201710676951 A CN 201710676951A CN 107426759 A CN107426759 A CN 107426759A
- Authority
- CN
- China
- Prior art keywords
- base station
- data
- newly
- cell
- factor
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Granted
Links
Classifications
-
- H—ELECTRICITY
- H04—ELECTRIC COMMUNICATION TECHNIQUE
- H04W—WIRELESS COMMUNICATION NETWORKS
- H04W24/00—Supervisory, monitoring or testing arrangements
- H04W24/06—Testing, supervising or monitoring using simulated traffic
-
- H—ELECTRICITY
- H04—ELECTRIC COMMUNICATION TECHNIQUE
- H04B—TRANSMISSION
- H04B17/00—Monitoring; Testing
- H04B17/30—Monitoring; Testing of propagation channels
- H04B17/391—Modelling the propagation channel
- H04B17/3913—Predictive models, e.g. based on neural network models
-
- H—ELECTRICITY
- H04—ELECTRIC COMMUNICATION TECHNIQUE
- H04W—WIRELESS COMMUNICATION NETWORKS
- H04W24/00—Supervisory, monitoring or testing arrangements
- H04W24/02—Arrangements for optimising operational condition
-
- H—ELECTRICITY
- H04—ELECTRIC COMMUNICATION TECHNIQUE
- H04W—WIRELESS COMMUNICATION NETWORKS
- H04W28/00—Network traffic management; Network resource management
- H04W28/02—Traffic management, e.g. flow control or congestion control
Abstract
The present invention relates to a kind of Forecasting Methodology and system of newly-increased base station data portfolio, the first object cell related to the newly-increased base station cell is chosen from each adjacent base station cell of the newly-increased base station cell of acquisition, and obtain the KPI Key Performance Indicator data of the first object cell, from the first factor of influence data of the data business volume for increasing base station cell described in the KPI Key Performance Indicator extracting data newly, using the first factor of influence data, pass through the data business volume of the newly-increased base station of preset predictive model prediction.Increase the first factor of influence data of the data business volume of base station cell newly from the KPI Key Performance Indicator extracting data of first object cell, fully take into account has influence factor of the adjacent base station cell to newly-increased base station cell of considerable influence to increasing base station cell newly, utilize the relation between the first factor of influence data and newly-increased base station cell, the collection of historical data in conventional art can be not only reduced, but also the Accurate Prediction of the data business volume to increasing base station newly can be realized.
Description
Technical field
The present invention relates to network communication technology field, more particularly to a kind of Forecasting Methodology of newly-increased base station data portfolio
And system.
Background technology
With the gradual perfection of networking, the planning construction of communication network has been taken leave of traditional layout type network and built
If turn to networking to become more meticulous the stage.The important reference index of one of which is exactly the data business volume of each planning base station, should
The prediction of index plans that the resource utilization of base station and reasonable arrangement construction progress tool are of great significance for weighing.
Existing data business volume prediction has based on the method for moving average, exponential smoothing, trend extrapolation, Rayleigh point at present
The Forecasting Methodologies such as the multifactor method of cloth.
The method of moving average, exponential smoothing and the extrapolation that becomes are required for collecting historical data as detailed as possible, including
Portfolio developing history under the statistics background of communications industry difference developing period.For newly-increased base station, historical data
It is exactly an insurmountable problem to collect and prepare, and with the fast development of the communication technology, the base station industry of different times
Business amount changes greatly, and rolling average and trend extropolation belong to steady state time series model, is only capable of reflecting that fluctuation range is little
Overall data trend, it is more difficult to carry out precisely prediction for newly-increased base station.The multifactor method of rayleigh distributed is a kind of from macroeconomy
The Forecasting Methodology that the angle of environment and various influence factors is set out.But the multifactor method of rayleigh distributed is when carrying out each factors quantization
The subjective judgement of individual is relied primarily on, prediction result carries certain subjective colo(u)r, the quantization field being more suitable on integral macroscopic
Scape, this method are not particularly suited for carrying out the scene of data business volume prediction to increasing base station newly.
When the data business volume prediction of newly-increased base station is therefore carried out in engineering, the practical application accuracy of the above method compared with
It is low.
The content of the invention
Based on this, it is necessary to for traditional newly-increased base station data business volume Forecasting Methodology accuracy it is relatively low the problem of,
A kind of Forecasting Methodology and system of newly-increased base station data portfolio are provided.
A kind of Forecasting Methodology of newly-increased base station data portfolio, comprises the following steps：
Each adjacent base station cell of newly-increased base station cell is obtained, is chosen and newly-increased base station cell from each adjacent base station cell
Related first object cell；
The KPI Key Performance Indicator data of first object cell are obtained, it is small to increase base station newly from KPI Key Performance Indicator extracting data
First factor of influence data of the data business volume in area；
By the first factor of influence data input into preset predictive model, newly-increased base station is obtained by preset predictive model
Data business volume.
A kind of forecasting system of newly-increased base station data portfolio, including：
Object selection unit, for obtaining each adjacent base station cell of newly-increased base station cell, from each adjacent base station cell
Choose the first object cell related to newly-increased base station cell；
Factor extraction unit, for obtaining the KPI Key Performance Indicator data of first object cell, from KPI Key Performance Indicator number
According to the first factor of influence data of the data business volume of the newly-increased base station cell of middle extraction；
Traffic forecast unit, for into preset predictive model, the first factor of influence data input to be passed through into preset predictive
Model obtains the data business volume of newly-increased base station.
According to the Forecasting Methodology and system of the newly-increased base station data portfolio of the invention described above, it is the newly-increased base from acquisition
The first object cell related to newly-increased base station cell is chosen in each adjacent base station cell for cell of standing, and it is small to obtain first object
The KPI Key Performance Indicator data in area, increase the first shadow of the data business volume of base station cell newly from KPI Key Performance Indicator extracting data
Factor data is rung, using the first factor of influence data, passes through the data business volume of the newly-increased base station of preset predictive model prediction.Herein
In scheme, it is determined that the first object cell that newly-increased base station cell is related, from the KPI Key Performance Indicator data of first object cell
First factor of influence data of the data business volume of the newly-increased base station cell of extraction, fully take into account to newly-increased base station cell have it is larger
The adjacent base station cell of influence utilizes the first factor of influence data and newly-increased base station cell to the influence factor of newly-increased base station cell
Between relation, can not only reduce the collection of historical data in conventional art, but also can realize to increase newly base station number
According to the Accurate Prediction of portfolio.
A kind of readable storage medium storing program for executing, is stored thereon with executable program, and the program is realized above-mentioned when being executed by processor
The step of Forecasting Methodology of newly-increased base station data portfolio.
A kind of pre- measurement equipment, including memory, processor and storage on a memory and can run on a processor can
Configuration processor, the step of realizing the Forecasting Methodology of above-mentioned newly-increased base station data portfolio during computing device program.
According to the Forecasting Methodology of the newly-increased base station data portfolio of the invention described above, the present invention also provides a kind of readable storage
Medium and pre- measurement equipment, for realizing the Forecasting Methodology of above-mentioned newly-increased base station data portfolio by program.
Brief description of the drawings
Fig. 1 is the schematic flow sheet of the Forecasting Methodology of the newly-increased base station data portfolio of one of embodiment；
Fig. 2 is the structural representation of the forecasting system of the newly-increased base station data portfolio of one of embodiment；
Fig. 3 is the structural representation of the forecasting system of the newly-increased base station data portfolio of one of embodiment；
Fig. 4 is the structural representation of the forecasting system of the newly-increased base station data portfolio of one of embodiment；
Fig. 5 is the overall flow signal of the Forecasting Methodology of the newly-increased base station data portfolio of one of specific embodiment
Figure；
Fig. 6 is the schematic flow sheet of the factor extraction of the influence cell business volume of one of specific embodiment；
Fig. 7 is the extraction schematic diagram of adjacent base station in conventional art.
Embodiment
For the objects, technical solutions and advantages of the present invention are more clearly understood, below in conjunction with drawings and Examples, to this
Invention is described in further detail.It should be appreciated that embodiment described herein is only to explain the present invention,
Do not limit protection scope of the present invention.
Shown in Figure 1, the flow for the Forecasting Methodology of the newly-increased base station data portfolio of one embodiment of the invention is shown
It is intended to.The Forecasting Methodology of newly-increased base station data portfolio in the embodiment, comprises the following steps：
Step S101：Each adjacent base station cell of newly-increased base station cell is obtained, is chosen from each adjacent base station cell and new
Increase the related first object cell in base station cell；
In this step, increasing newly around base station has adjacent base station, and there is corresponding cell each base station, and it is small to increase base station newly
Area can correspond to multiple adjacent base station cells；
Step S102：The KPI Key Performance Indicator data of first object cell are obtained, from KPI Key Performance Indicator extracting data
First factor of influence data of the data business volume of newly-increased base station cell；
In this step, KPI Key Performance Indicator (KPI, Key Performance Indicator) data are each built
The statistical nature data of cell, the related number of data and the cell adjacent to the cell including corresponding cell itself
According to；
Step S103：By the first factor of influence data input into preset predictive model, obtained by preset predictive model
The data business volume of newly-increased base station.
In this step, preset predictive model is trained forecast model, and the forecast model can be according to input
First factor of influence data prediction goes out the data business volume of newly-increased base station.
In the present embodiment, chosen and newly-increased base station cell from each adjacent base station cell of the newly-increased base station cell of acquisition
Related first object cell, and the KPI Key Performance Indicator data of first object cell are obtained, from KPI Key Performance Indicator data
First factor of influence data of the data business volume of the newly-increased base station cell of extraction, using the first factor of influence data, by preset
The data business volume of the newly-increased base station of forecast model prediction.In this scheme, it is determined that the first object of newly-increased base station cell correlation is small
Area, increased newly from the KPI Key Performance Indicator extracting data of first object cell the first of the data business volume of base station cell influence because
Subdata, fully take into account to increase newly base station cell have the adjacent base station cell of considerable influence to increase newly the influence of base station cell because
Element, using the relation between the first factor of influence data and newly-increased base station cell, it can not only reduce history number in conventional art
According to collection, but also can realize to increase newly base station data business volume Accurate Prediction.
Optionally, first object cell can be each adjacent base station cell for having overlapping region with newly-increased base station cell.
In one of the embodiments, the step of each adjacent base station cell for obtaining newly-increased base station cell, includes following step
Suddenly：
Each adjacent base station of newly-increased base station is obtained, wireless coverage emulation is carried out to newly-increased base station and each adjacent base station, obtained
Newly-increased base station and the signal distributions data of each adjacent base station；
Each adjacent base station cell is determined according to the signal distributions data of newly-increased base station and each adjacent base station.
In the present embodiment, emulated using wireless coverage and the network coverage situation for increasing base station and each adjacent base station newly is carried out
Emulation, the signal distributions data of the overall network of newly-increased base station and each adjacent base station can be obtained, above-mentioned signal distributions data can
Objective data foundation is provided with the determination for adjacent base station cell, it is adjacent with newly-increased base station cell each adjacent so as to filter out
Base station cell.
In one of the embodiments, the first object related to newly-increased base station cell is chosen from each adjacent base station cell
The step of cell, comprises the following steps：
Newly-increased base station cell and each adjacent base are calculated respectively according to the signal distributions data of newly-increased base station and each adjacent base station
The overlapping region area for cell of standing；
Descending sort is carried out to each adjacent base station cell according to the size of overlapping region area, chooses most preceding one that sorts
Or multiple adjacent base station cells are first object cell.
In the present embodiment, using the signal distributions data of newly-increased base station and each adjacent base station can obtain newly-increased base station and
The signal coverage areas of each adjacent base station, and then the overlapping region face of newly-increased base station cell and each adjacent base station cell is calculated
Product；By overlapping region area according to descending sort is carried out to each adjacent base station cell for increasing base station cell newly from big to small, therefrom
The forward adjacent base station cell that sorts is chosen as first object cell, can so improve first object cell and newly-increased base station
The correlation of cell, be advantageous to the acquisition of follow-up first factor of influence data.
Optionally, first object cell can be three adjacent base station cells for sorting most preceding.
In one of the embodiments, the data business volume of base station cell is increased newly from KPI Key Performance Indicator extracting data
The step of first factor of influence data, comprises the following steps：
Increase all factor of influence data of the data business volume of base station cell, meter newly from KPI Key Performance Indicator extracting data
The degree of correlation of all factor of influence data and the data business volume of newly-increased base station cell is calculated, by the degree of correlation higher than relevance threshold
Factor of influence data are as the first factor of influence data.
In the present embodiment, there are a variety of different achievement datas in the KPI Key Performance Indicator data of first object cell, its
In part index number data and newly-increased base station cell it is closely related, i.e., the factor of influence number of the data business volume of newly-increased base station cell
According to, by calculating the degree of correlation of the factor of influence data to data business volume, the first factor of influence data can be therefrom filtered out, the
One factor of influence data are higher to the degree of correlation for increasing the data business volume of base station cell newly, can make the data of newly-increased base station cell
The prediction of portfolio is more accurate.
Optionally, can be by the statistical law of radio service data come to factor of influence data and newly-increased base station cell
The degree of correlation of data business volume carries out analysis calculating, sets relevance threshold to reject the data service with newly-increased base station cell
Measure unrelated or influence small factor of influence data, the first higher factor of influence data of the degree of correlation are filtered out, for increasing base newly
The prediction of the data business volume for cell of standing.
Optionally, increasing the factor of influence data of the data business volume of base station cell newly includes the area of coverage of first object cell
Domain area, first object cell and the overlapping region area, overlapping region area and newly-increased base station cell of newly-increased base station cell
The ratio of overlay area area, the data business volume of first object cell, the high-speed slender body theory of first object cell are used
Amount, the cell average user number of first object cell, the actual use physical resource number of blocks and first of first object cell
Physical Resource Block total amount of Target cell etc..
In one of the embodiments, the Forecasting Methodology for increasing base station data portfolio newly is further comprising the steps of：
Each adjacent base station cell that history increases base station cell newly is obtained, each adjacent base station for increasing base station cell newly from history is small
Second Target cell related to the newly-increased base station cell of history is chosen in area；
The KPI Key Performance Indicator data of the second Target cell are obtained, from the KPI Key Performance Indicator data of the second Target cell
Extract the second factor of influence data that history increases the data business volume of base station cell newly；
Obtain the actual data traffic amount that history increases base station cell newly；
Initial predicted model is established, using the second factor of influence data as input, using actual data traffic amount as output, to first
Beginning forecast model is trained, and obtains preset predictive model.
In the present embodiment, the acquisition process of the second factor of influence data and the acquisition process phase of the first factor of influence data
Seemingly, the influence object of the second factor of influence data be history increase newly base station data business volume, by history increase newly base station cell with
Its actual data business volume, the second factor of influence data are associated, and are available for training the training number of initial predicted model
It is input by the second factor of influence data according to collection, using actual data traffic amount as output, initial predicted model is trained,
So as to obtain that the forecast model of newly-increased base station cell data business volume can be predicted.
Further, the different history of multigroup correspondence can be utilized to increase the training data set pair initial predicted of base station cell newly
Model is trained.
In one of the embodiments, base station is increased newly from the KPI Key Performance Indicator extracting data history of the second Target cell
The step of second factor of influence data of the data business volume of cell, comprises the following steps：
Increase the data business volume of base station cell newly from the KPI Key Performance Indicator extracting data history of the second Target cell
All factor of influence data, the degree of correlation that all factor of influence data increase the data business volume of base station cell newly with history is calculated,
Using factor of influence data of the degree of correlation higher than relevance threshold as the second factor of influence data.
In the present embodiment, there are a variety of different achievement datas in the KPI Key Performance Indicator data of the second Target cell, its
In part index number data and history increase newly base station cell it is closely related, i.e., history increase newly base station cell data business volume shadow
Ring factor data, by calculating the degree of correlation of the factor of influence data to data business volume, can therefrom filter out the second influence because
Subdata, the degree of correlation for the data business volume that the second factor of influence data increase base station cell newly to history is higher, can make initial
The training dataset of forecast model is more accurate.
Optionally, can be by the statistical law of radio service data come small to factor of influence data and the newly-increased base station of history
The degree of correlation of the data business volume in area carries out analysis calculating, sets relevance threshold to reject and increases base station cell newly with history
Data business volume is unrelated or influences small factor of influence data, filters out the second higher factor of influence data of the degree of correlation, uses
In the training of initial predicted model.
In one of the embodiments, the overlay area area of all factor of influence data including the second Target cell, the
Two Target cells increase the overlapping region area of base station cell, overlapping region area and history newly with history and increase covering for base station cell newly
The ratio of cover area area, the data business volume of the second Target cell, the high-speed downlink packet access users of the second Target cell
Number, the actual use physical resource number of blocks and the second mesh of the cell average user number of the second Target cell, the second Target cell
Mark the Physical Resource Block total amount of cell.
In the present embodiment, it is small to increase base station newly for the overlay area area of the second Target cell, the second Target cell and history
Overlapping region area, overlapping region area and the history in area increase ratio, the second target of the overlay area area of base station cell newly
The data business volume of cell, the high-speed downlink packet access users number of the second Target cell, the cell of the second Target cell are averaged
Number of users, the second Target cell actual use physical resource number of blocks and the second Target cell Physical Resource Block total amount these
Data are to increase that base station cell is closely related newly with history, and can be from the KPI Key Performance Indicator data of the second Target cell
Extracted, the factor of influence data that history increases base station cell newly are obtained in a manner of relatively simple.
In one of the embodiments, initial predicted model includes multivariate regression models, using the second factor of influence data as
Input, comprise the following steps using actual data traffic amount as the step of exporting, initial predicted model is trained：
Using the second factor of influence data as independent variable, actual data traffic amount is dependent variable, is increased newly for different history
Base station is repeatedly trained to multivariate regression models, obtains the regression coefficient of multivariate regression models；
Preset predictive model includes housebroken multivariate regression models, the regression equation of housebroken multivariate regression models
For：
Yi=β0+β1X1i+β2X2i+…+βkXki+μi(i=1,2 ..., n)
In formula, YiThe prediction data portfolio of i-th of newly-increased base station cell is represented, k represents independent variable number, βj(j=1,
2 ..., k) it is regression coefficient, XkiRepresent the first factor of influence data of the data business volume of i-th of newly-increased base station cell, μiTable
Show the stochastic variable of i-th of newly-increased base station cell.
In the present embodiment, initial predicted model can include multivariate regression models, be increased newly using corresponding different history
The the second factor of influence data and actual data traffic amount of base station are repeatedly trained to multivariate regression models, can be obtained polynary
The regression coefficient of the regression equation of regression model, and then the regression equation of housebroken multivariate regression models is obtained, it belongs to pre-
Forecast model is put, by the regression equation of housebroken multivariate regression models, with reference to the data business volume of newly-increased base station cell
First factor of influence data, each side influence factor of the data business volume of newly-increased base station cell is taken into full account, it is right in all directions
The data business volume of newly-increased base station cell carries out Accurate Prediction.
In one of the embodiments, by the step in the first factor of influence data input to preset predictive model include with
Lower step：
History is obtained according to the regression equation of the second factor of influence data and housebroken multivariate regression models and increases base station newly
The regression fit data business volume of cell；
The recurrence of housebroken multivariate regression models is calculated according to regression fit data business volume and actual data traffic amount
The degree of fitting and relative error of equation；
When degree of fitting is higher than degree of fitting threshold value and relative error is less than error threshold, by the first factor of influence data input
To housebroken multivariate regression models.
In the present embodiment, the pre- of the data service of newly-increased base station cell is being carried out using housebroken multivariate regression models
Before survey, accuracy evaluation, including the degree of fitting of regression equation and relative mistake can be carried out to housebroken multivariate regression models
Difference is assessed, and when degree of fitting and relative error meet threshold condition, represents that currently housebroken multivariate regression models is effective,
Now the data service for increasing base station cell newly is predicted, it is ensured that the accuracy of prediction.
Optionally, the equation of degree of fitting can beIn formula, R2Degree of fitting is represented, y is represented
History increases the observation of the actual data traffic amount of base station cell, y newlycRepresent that housebroken multivariate regression models increases newly to history
The predicted value of the data business volume of base station cell,Represent that multiple history increase the observation of the actual data traffic amount of base station cell newly
Average value, summation are that the related data that multiple different history are increased newly with base station cell calculates.
Optionally, the equation of relative error can beIn formula, δ represents relative error, and y is represented
History increases the observation of the actual data traffic amount of base station cell, y newlycRepresent that housebroken multivariate regression models increases newly to history
The predicted value of the data business volume of base station cell.
According to the Forecasting Methodology of above-mentioned newly-increased base station data portfolio, the present invention also provides a kind of newly-increased base station data business
The forecasting system of amount, just the embodiment of the forecasting system of the newly-increased base station data portfolio of the present invention is described in detail below.
Shown in Figure 2, the structure for the forecasting system of the newly-increased base station data portfolio of one embodiment of the invention is shown
It is intended to.The forecasting system of newly-increased base station data portfolio in the embodiment includes：
Object selection unit 210, for obtaining each adjacent base station cell of newly-increased base station cell, from each adjacent base station cell
It is middle to choose the first object cell related to newly-increased base station cell；
Factor extraction unit 220, for obtaining the KPI Key Performance Indicator data of first object cell, from KPI Key Performance Indicator
Extracting data increases the first factor of influence data of the data business volume of base station cell newly；
Traffic forecast unit 230, for by the first factor of influence data input into preset predictive model, by preset pre-
Survey the data business volume that model obtains newly-increased base station.
In one of the embodiments, Object selection unit 210 obtains each adjacent base station of newly-increased base station, to increasing base station newly
Wireless coverage emulation is carried out with each adjacent base station, obtains the signal distributions data of newly-increased base station and each adjacent base station；According to newly-increased
The signal distributions data of base station and each adjacent base station determine each adjacent base station cell.
In one of the embodiments, Object selection unit 210 is according to newly-increased base station and the signal distributions of each adjacent base station
Data calculate the overlapping region area of newly-increased base station cell and each adjacent base station cell respectively；According to the size of overlapping region area
Descending sort is carried out to each adjacent base station cell, it is small for first object to choose the most preceding one or more adjacent base station cells that sort
Area.
In one of the embodiments, factor extraction unit 220 is small from the newly-increased base station of KPI Key Performance Indicator extracting data
All factor of influence data of the data business volume in area, calculate the data service of all factor of influence data and newly-increased base station cell
The degree of correlation of amount, using factor of influence data of the degree of correlation higher than relevance threshold as the first factor of influence data.
In one of the embodiments, as shown in figure 3, the forecasting system of newly-increased base station data portfolio also includes model structure
Unit 240 is built, each adjacent base station cell of base station cell is increased newly for obtaining history, each adjacent of base station cell is increased newly from history
Second Target cell related to the newly-increased base station cell of history is chosen in base station cell；Obtain the Key Performance of the second Target cell
Achievement data, increase the data business volume of base station cell newly from the KPI Key Performance Indicator extracting data history of the second Target cell
Second factor of influence data；Obtain the actual data traffic amount that history increases base station cell newly；Initial predicted model is established, with second
Factor of influence data are input, and using actual data traffic amount as output, initial predicted model is trained, obtains preset predictive
Model.
In one of the embodiments, model construction unit 240 is from the KPI Key Performance Indicator data of the second Target cell
All factor of influence data that history increases the data business volume of base station cell newly are extracted, calculate all factor of influence data and history
The degree of correlation of the data business volume of newly-increased base station cell, using factor of influence data of the degree of correlation higher than relevance threshold as second
Factor of influence data.
In one of the embodiments, the overlay area area of all factor of influence data including the second Target cell, the
Two Target cells increase the overlapping region area of base station cell, overlapping region area and history newly with history and increase covering for base station cell newly
The ratio of cover area area, the data business volume of the second Target cell, the high-speed downlink packet access users of the second Target cell
Number, the actual use physical resource number of blocks and the second mesh of the cell average user number of the second Target cell, the second Target cell
Mark the Physical Resource Block total amount of cell.
In one of the embodiments, initial predicted model includes multivariate regression models, and model construction unit 240 is with second
Factor of influence data are independent variable, and actual data traffic amount is dependent variable, and base station is increased newly to multiple regression for different history
Model is repeatedly trained, and obtains the regression coefficient of multivariate regression models；Preset predictive model includes housebroken multiple regression
Model, the regression equation of housebroken multivariate regression models are：
Yi=β0+β1X1i+β2X2i+…+βkXki+μi(i=1,2 ..., n)
In formula, YiThe prediction data portfolio of i-th of newly-increased base station cell is represented, k represents independent variable number, βj(j=1,
2 ..., k) it is regression coefficient, XkiRepresent the first factor of influence data of the data business volume of i-th of newly-increased base station cell, μiTable
Show the stochastic variable of i-th of newly-increased base station cell.
In one of the embodiments, as shown in figure 4, the forecasting system of newly-increased base station data portfolio is also commented including model
Estimate unit 250, it is new for obtaining history according to the regression equation of the second factor of influence data and housebroken multivariate regression models
Increase the regression fit data business volume of base station cell；Calculated according to regression fit data business volume and actual data traffic amount through instruction
The degree of fitting and relative error of the regression equation of experienced multivariate regression models；
Traffic forecast unit 230 is when degree of fitting is higher than degree of fitting threshold value and relative error is less than error threshold, by first
Factor of influence data input is to housebroken multivariate regression models.
The ordinal number such as above-mentioned " first ", " second " is intended merely to distinguish described object, is not to description object sheet
The restriction of body.
The forecasting system of the newly-increased base station data portfolio of the present invention is pre- with the newly-increased base station data portfolio of the present invention
Survey method correspond, above-mentioned newly-increased base station data portfolio Forecasting Methodology embodiment illustrate technical characteristic and its have
Beneficial effect is suitable for the embodiment of forecasting system for increasing base station data portfolio newly.
According to the Forecasting Methodology of above-mentioned newly-increased base station data portfolio, the embodiment of the present invention also provides a kind of readable storage medium
Matter and a kind of pre- measurement equipment.Executable program is stored with readable storage medium storing program for executing, the program is realized above-mentioned when being executed by processor
The step of Forecasting Methodology of newly-increased base station data portfolio；Pre- measurement equipment includes memory, processor and storage on a memory
And the executable program that can be run on a processor, the pre- of above-mentioned newly-increased base station data portfolio is realized during computing device program
The step of survey method.
In a specific embodiment, the solution of the present invention can be applied in the scene of base station construction.
From the target for improving base station construction benefit, the present invention have studied a kind of newly-increased based on adjacent base station correlation
Base station data Traffic prediction method.By with the phase of adjacent base station cell KPI indexs and newly-increased base station cell data business volume
Close property be point of penetration, establish increase newly base station cell DBMS Traffic prediction model, and by degree of fitting and relative error come pair
Forecast model is assessed, and when assessment result meets condition, the data business volume for increasing base station newly is carried out using forecast model
Prediction.
Technical scheme overall flow figure as figure 5 illustrates, carries out cell-level wireless coverage emulation, is extracted according to wireless simulation result
History increases the adjacent base station cell of base station cell newly, calculates adjacent base station cell and history increases the overlapping coverage areas of base station cell newly
Domain area, weighing factor is determined according to the area, first three adjacent base station cell is taken as the second Target cell by weight descending,
Increase the data business volume of the second Target cell when base station is not opened newly to history, according to wireless traffic rule extraction second influence because
Subdata, and then the second factor of influence matrix is generated, it is increased newly to the data business volume composition training after opening base station with history
Data set, the initial regression model of structure is trained, obtains preset predictive model, when not opened to existing newly-increased base station
The data business volume of first object cell is extracted, and is obtained the first factor of influence data, is input to preset predictive model,
Obtain the predicted value of the data business volume of existing newly-increased base station.
The flow of the factor of influence data extraction of cell business volume is being influenceed as shown in fig. 6, according to adjacent base station cell
Some KPI index sets and data business volume, the statistical indicators such as the degree of correlation are calculated, irrelevant factor is rejected according to index, obtains shadow
Ring factor data.
Specifically, the step of newly-increased BTS service amount Forecasting Methodology based on adjacent base station cell correlation, is as follows：
(1) determine to influence to increase newly the adjacent base station of the data business volume of base station；
In existing correlative study, the adjacent base station not clearly definition for influenceing newly-increased base station is general to use
Centered on the newly-increased base station of extraction, the website of a circle is as adjacent base station in the range of some radius R, as shown in Figure 7.But this sentence
The method for determining adjacent base station is too rough, does not account for the actual conditions of wireless coverage, because actual wireless coverage has greatly
There is small rule to differ, can cause to omit some websites for influenceing newly-increased base station according to uniform radial delineation, or redundancy is some right
Newly-increased base station has no the website of influence.In consideration of it, the present invention fully takes into account the reality of wireless coverage when extracting adjacent base station
Situation, mainly including following two steps：
(a) cell-level high precision wireless emulation of coverage capability
Because each cell configuration of base station is not quite similar, coverage also can be variant, is carried out in units of cell wireless
Emulation of coverage capability can more accurately reflect the actual coverage condition of website.By simulation result can obtain newly-increased base station with it is adjacent
The signal distributions data of base station.
(b) degree of correlation of adjacent base station cell is determined
According to step (a), cell-level wireless coverage distributed data is obtained, then calculates each adjacent base station cell and newly-increased base
Stand the overlapping region area of MPS process, determine that weighing factor (accounts for the newly-increased base station cell gross area using overlapping region area
Ratio is as weighing factor), weighing factor is then as the index for judging adjacent base station cell and newly-increased base station cell degree of correlation.
Preceding 3 cells are taken as related cell according to the arrangement of weighing factor descending.
(2) extraction influences the factor of influence data of the data business volume of newly-increased base station cell；
According to the business experience of expert, KPI (Key Performance Indicator, Key Performance from related cell
Index) extracting data may be related to the data business volume of newly-increased base station factor of influence data, mainly include：Related cell
Area coverage, related cell and newly-increased base station cell covering overlapping area, the weighing factor to increasing base station cell newly, mutually turn down
The data business volume in area, HSDPA (High Speed Downlink Packet Access, the high speed downlink packet of related cell
Access) number of users, the cell average user number of related cell, actual use PRB (the physical resource of related cell
Block, Physical Resource Block) quantity, the quantity of PRB altogether of related cell etc..
Further according to the statistical law of wireless traffic the factor and the data industry of newly-increased base station cell are influenceed to what is extracted
The degree of correlation between business amount is calculated and analyzed, and rejects the factor unrelated with the data business volume of newly-increased base station cell, screening
Go out the high factor of the degree of correlation, the training characteristics collection established as following model.
(3) model training collection is established according to historical data；
After step (2), the related cell required for model and factor of influence data are obtained establishing, then will
Historical traffic data, the factor of influence data of newly-increased base station cell and related cell are associated, and form a wide table, this
Wide table includes factor of influence matrix, can be as the training dataset of modeling.
(4) regression model is generated；
By step (3), the training dataset that modeling needs is formed, the target to be realized is that the newly-increased base station of prediction is small
The data business volume in area, data business volume belong to prediction numeric type data, select homing method to establish model.It is newly-increased due to influenceing
The factor of the data business volume of base station cell may include multiple, therefore use multivariate regression models.It is new for different history
Increase base station to carry out after repeatedly training multivariate regression models, finally build shaped like Yi=β0+β1X1i+β2X2i+…+βkXki+μi(i=
1,2 ..., n) regression equation.Wherein YiThe prediction data portfolio of i-th of newly-increased base station cell is represented, k represents independent variable number
Mesh, βj(j=1,2 ..., k) is regression coefficient, XkiRepresent the first of the data business volume of i-th of newly-increased base station cell influence because
Subdata, μiRepresent the stochastic variable of i-th of newly-increased base station cell.
(5) assessment of forecast model；
Digital simulation degree R2 and relative error δ.
Degree of fitting：
Wherein, y-ycThe difference of actual observed value and regression fit value is represented, represents the deviation do not explained by regression equation,The difference of actual observed value and the average value of observation is represented, represents the part that can be explained by regression equation；
Relative error：In formula, δ represents relative error, and y represents that history increases base station cell newly
The observation of actual data traffic amount, ycRepresent that housebroken multivariate regression models increases the data service of base station cell newly to history
The predicted value of amount；
Degree of fitting and relative error threshold value thresholding are set according to expertise, when model-fitting degree is higher and relative error
Relatively low, when reaching the threshold value of expert's setting, it is exactly effective to assess the model, can be used for engineer applied.
(6) model application；
According to preceding 5 steps, the data business volume of newly-increased base station cell just can be calculated and predicted.A collection upper cycle and rule
The adjacent cell indices value in base station cell is drawn, the first factor of influence data are extracted according to wireless traffic rule, by built
Good forecast model, it is possible to calculate the data business volume that accordingly planning base station future is likely to be breached.With network topology
Constantly change, can gather new data to update and expand historic training data, and model is carried out more according to preceding 5 steps
Newly, just more accurately the data business volume for increasing base station cell newly can be predicted.So can be becoming more meticulous for operator
Network planning construction or precision marketing etc. provide relatively accurately foundation, can also greatly reduce corresponding operation cost.
The solution of the present invention, by the analysis in terms of the statistics degree of correlation, focuses on phase using demand in engineering as main target
The high index of closing property, can play reduction data dimension, the portfolio for increasing base station newly be carried out by less KPI indexs accurate
Prediction；Objective science definition is carried out to newly-increased base station ambient stations, by the emulation that becomes more meticulous of cell-level, and according to covering weight
Folded region area screening adjacent base station cell, compared to centered on newly-increased base station, some radius draws a circle to approve the side of adjacent base station
Method, more meet actual scene, also with more Impersonal authenticity, more accurate data supporting is provided for the foundation of forecast model；
Compared to conventional arts such as the methods of moving average, the solution of the present invention can solve the data business volume in the single planning base station short cycle
Forecasting problem, and the traffic forecast practical application for carrying out cell-level has higher accurate fixed and correlation.
All it is mainly to use to draw a circle to approve scope by geographical position to the choosing method of planning surrounding sites base station in conventional art,
Simple but inadequate science.The present invention is not only from geographical position, the actual conditions also according to wireless network covering, and use is more objective
The means of sight science, are extracted to adjacent base station.First according to the work parameter evidence of existing website, become more meticulous using wireless coverage imitative
True software emulates to network coverage situation, obtains overall network coverage situation, includes the actual covering of each base station cell
Scope, and the information such as cell conditions of covering adjacent thereto, these information provide objective section for the extraction of adjacent base station cell
Data foundation.
In current existing correlative study, some mainly judge newly-increased base according to the data business volume of adjacent base station cell
The data business volume for cell of standing, seldom consider other indexs of correlation；Some can be according to network planning network optimization expertise to indices
The judgement in business experience is carried out, selects some indexs that may have influence on cell business volume, but these methods are with stronger
Subjectivity.The solution of the present invention is assessed existing KPI index sets, used from the objective angle of data one by one
Statistical relatedness computation is screened to factor of influence data, and the higher factor of influence data of the degree of correlation are extracted
As the factor of influence data acquisition system for influenceing cell data portfolio.
Traditional engineer applied is that the data business volume of the adjacent circle in base station is carried out averagely or other are simple to increase newly mostly
The data business volume as newly-increased base station is calculated, though method is simple, result and actual volume error are big, and this method is effective
Property is poor.The solution of the present invention combines the multiple factors for influenceing cell data portfolio, and prediction mould is established using multiple regression procedure
Type, the factor of each side is fully taken into account, data business volume precisely can be predicted in all directions.
The solution of the present invention closing to reality engineer applied, with more scientific effective method find adjacent base station cell with treat it is pre-
The relation surveyed between base station cell.It is main to solve conventional art in the judgement to adjacent base station cell still without good
The problem of selection rule, by the simulation means that become more meticulous, filter out to objective science the adjacent base related to newly-increased base station cell
Stand set of cells, more authentic and valid data are provided for follow-up Traffic prediction.
Further, with data driven technique, newly-increased base station cell is realized by means such as statistical analysis and data minings
The accurate prediction of data business volume.During forecast model is established, do not use what is generally used on current engineer applied
Simple average method, but consider analysis various factors and index, and be modeled using the method for data mining, method is more
Science.
In addition, the solution of the present invention patch have can continuity, and reduce manual intervention work, can progressively tend to automate
Realize.From the emulation that becomes more meticulous, related cell is extracted, extracts factor of influence data, Traffic prediction modeling to the end can
Realized using program meanses, reach the target of automation modeling.As long as being updated to data, corresponding modeling process is just
The model after renewal can be exported, to adapt to the data for constantly changing and updating.
Each technical characteristic of embodiment described above can be combined arbitrarily, to make description succinct, not to above-mentioned reality
Apply all possible combination of each technical characteristic in example to be all described, as long as however, the combination of these technical characteristics is not deposited
In contradiction, the scope that this specification is recorded all is considered to be.
Can be with one of ordinary skill in the art will appreciate that realizing that all or part of step in above-described embodiment method is
The hardware of correlation is instructed to complete by program.Described program can be stored in read/write memory medium.The program exists
During execution, including the step described in the above method.Described storage medium, including：ROM/RAM, magnetic disc, CD etc..
Embodiment described above only expresses the several embodiments of the present invention, and its description is more specific and detailed, but simultaneously
Can not therefore it be construed as limiting the scope of the patent.It should be pointed out that come for one of ordinary skill in the art
Say, without departing from the inventive concept of the premise, various modifications and improvements can be made, these belong to the protection of the present invention
Scope.Therefore, the protection domain of patent of the present invention should be determined by the appended claims.
Claims (10)
1. a kind of Forecasting Methodology of newly-increased base station data portfolio, it is characterised in that comprise the following steps：
Each adjacent base station cell of newly-increased base station cell is obtained, is chosen and the newly-increased base station from each adjacent base station cell
The related first object cell of cell；
The KPI Key Performance Indicator data of the first object cell are obtained, from new described in the KPI Key Performance Indicator extracting data
Increase the first factor of influence data of the data business volume of base station cell；
By the first factor of influence data input into preset predictive model, newly-increased base is obtained by the preset predictive model
The data business volume stood.
2. the Forecasting Methodology of newly-increased base station data portfolio according to claim 1, it is characterised in that described obtain increases newly
The step of each adjacent base station cell of base station cell, comprises the following steps：
Each adjacent base station of newly-increased base station is obtained, wireless coverage emulation is carried out to the newly-increased base station and each adjacent base station,
Obtain the signal distributions data of the newly-increased base station and each adjacent base station；
Each adjacent base station cell is determined according to the signal distributions data of the newly-increased base station and each adjacent base station.
3. the Forecasting Methodology of newly-increased base station data portfolio according to claim 2, it is characterised in that described from each described
The step of first object cell related to the newly-increased base station cell is chosen in adjacent base station cell comprises the following steps：
According to the signal distributions data of the newly-increased base station and each adjacent base station calculate respectively the newly-increased base station cell with
The overlapping region area of each adjacent base station cell；
Descending sort is carried out to each adjacent base station cell according to the size of the overlapping region area, it is most preceding to choose sequence
One or more adjacent base station cells are the first object cell.
4. the Forecasting Methodology of newly-increased base station data portfolio according to claim 3, it is characterised in that described from the pass
The step of the first factor of influence data for increasing the data business volume of base station cell described in key performance indicators extracting data newly, includes
Following steps：
From all factor of influence numbers for the data business volume for increasing base station cell described in the KPI Key Performance Indicator extracting data newly
According to the degree of correlation of calculating all factor of influence data and the data business volume of the newly-increased base station cell, by the correlation
Degree is higher than the factor of influence data of relevance threshold as the first factor of influence data.
5. the Forecasting Methodology of newly-increased base station data portfolio according to claim 1, it is characterised in that also including following step
Suddenly：
Each adjacent base station cell that history increases base station cell newly is obtained, each adjacent base station for increasing base station cell newly from the history is small
Second Target cell related to the newly-increased base station cell of the history is chosen in area；
The KPI Key Performance Indicator data of second Target cell are obtained, from the KPI Key Performance Indicator number of second Target cell
Increase the second factor of influence data of the data business volume of base station cell newly according to the middle extraction history；
Obtain the actual data traffic amount that the history increases base station cell newly；
Initial predicted model is established, using the second factor of influence data as input, using the actual data traffic amount as output,
The initial predicted model is trained, obtains the preset predictive model.
6. the Forecasting Methodology of newly-increased base station data portfolio according to claim 5, it is characterised in that described from described
History described in the KPI Key Performance Indicator extracting data of two Target cells increases the second influence of the data business volume of base station cell newly
The step of factor data, comprises the following steps：
History described in KPI Key Performance Indicator extracting data from second Target cell increases the data service of base station cell newly
All factor of influence data of amount, calculate the data service that all factor of influence data increase base station cell newly with the history
The degree of correlation of amount, using factor of influence data of the degree of correlation higher than relevance threshold as the second factor of influence data.
7. the Forecasting Methodology of newly-increased base station data portfolio according to claim 6, it is characterised in that described to be had an impact
Factor data includes the overlay area area of second Target cell, second Target cell and the history and increases base station newly
The overlapping region area of cell, the overlapping region area increase the ratio of the overlay area area of base station cell newly with the history
The high-speed downlink packet access users number, described of example, the data business volume of second Target cell, second Target cell
The cell average user number of second Target cell, the actual use physical resource number of blocks of second Target cell and described
The Physical Resource Block total amount of two Target cells.
8. the Forecasting Methodology of newly-increased base station data portfolio according to claim 5, it is characterised in that the initial predicted
Model includes multivariate regression models, described using the second factor of influence data as input, using the actual data traffic amount as
The step of exporting, being trained to the initial predicted model comprises the following steps：
Using the second factor of influence data as independent variable, the actual data traffic amount is dependent variable, for different history
Newly-increased base station is repeatedly trained to the multivariate regression models, obtains the regression coefficient of the multivariate regression models；
The preset predictive model includes housebroken multivariate regression models, the regression equation of housebroken multivariate regression models
For：
Yi=β0+β1X1i+β2X2i+…+βkXki+μi(i=1,2 ..., n)
In formula, YiThe prediction data portfolio of i-th of newly-increased base station cell is represented, k represents independent variable number, βj(j=1,2 ...,
K) it is regression coefficient, XkiRepresent the first factor of influence data of the data business volume of i-th of newly-increased base station cell, μiRepresent i-th
The stochastic variable of individual newly-increased base station cell.
9. the Forecasting Methodology of newly-increased base station data portfolio according to claim 8, it is characterised in that described by described
Step in one factor of influence data input to preset predictive model comprises the following steps：
The history is obtained according to the regression equation of the second factor of influence data and the housebroken multivariate regression models
The regression fit data business volume of newly-increased base station cell；
The housebroken multiple regression mould is calculated according to the regression fit data business volume and the actual data traffic amount
The degree of fitting and relative error of the regression equation of type；
When the degree of fitting is higher than degree of fitting threshold value and the relative error is less than error threshold, by first factor of influence
Data input is to the housebroken multivariate regression models.
A kind of 10. forecasting system of newly-increased base station data portfolio, it is characterised in that including：
Object selection unit, for obtaining each adjacent base station cell of newly-increased base station cell, from each adjacent base station cell
Choose the first object cell related to the newly-increased base station cell；
Factor extraction unit, for obtaining the KPI Key Performance Indicator data of the first object cell, refer to from the Key Performance
Mark the first factor of influence data of the data business volume for increasing base station cell described in extracting data newly；
Traffic forecast unit, for by the first factor of influence data input into preset predictive model, by described preset
Forecast model obtains the data business volume of newly-increased base station.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201710676951.4A CN107426759B (en) | 2017-08-09 | 2017-08-09 | Method and system for predicting data traffic of newly added base station |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201710676951.4A CN107426759B (en) | 2017-08-09 | 2017-08-09 | Method and system for predicting data traffic of newly added base station |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN107426759A true CN107426759A (en) | 2017-12-01 |
| CN107426759B CN107426759B (en) | 2020-06-19 |
Family
ID=60437782
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201710676951.4A Active CN107426759B (en) | 2017-08-09 | 2017-08-09 | Method and system for predicting data traffic of newly added base station |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN107426759B (en) |
Cited By (12)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN109039691A (en) * | 2018-06-01 | 2018-12-18 | 平安科技（深圳）有限公司 | The method and storage medium of server, forecasting system calling amount |
| CN109391950A (en) * | 2018-11-30 | 2019-02-26 | 中国移动通信集团福建有限公司 | Prediction technique, device, equipment and the medium of terminal distribution |
| CN111163482A (en) * | 2019-12-25 | 2020-05-15 | 中国联合网络通信集团有限公司 | Data processing method, device and storage medium |
| CN111225382A (en) * | 2018-11-27 | 2020-06-02 | 中国移动通信集团上海有限公司 | LTE functional characteristic selection method and system |
| CN111278013A (en) * | 2018-12-04 | 2020-06-12 | 中国移动通信集团山东有限公司 | Automatic pseudo base station identification method and device |
| CN112054943A (en) * | 2020-09-17 | 2020-12-08 | 中国电信股份有限公司河北分公司 | Traffic prediction method for mobile network base station |
| CN112492637A (en) * | 2020-12-28 | 2021-03-12 | 中国联合网络通信集团有限公司 | Method and device for predicting cell traffic |
| CN113052582A (en) * | 2019-12-27 | 2021-06-29 | 中移动信息技术有限公司 | Bill checking method, device and equipment and computer storage medium |
| CN113766523A (en) * | 2020-06-02 | 2021-12-07 | 中国移动通信集团河南有限公司 | Network resource utilization rate prediction method and device for serving cell and electronic equipment |
| CN113938893A (en) * | 2020-06-29 | 2022-01-14 | 中国移动通信集团广西有限公司 | Network planning assisting method and device and electronic equipment |
| CN114071525A (en) * | 2021-11-10 | 2022-02-18 | 中国联合网络通信集团有限公司 | Method, device and storage medium for determining optimization order of base station |
| CN114338416A (en) * | 2020-09-29 | 2022-04-12 | 中国移动通信有限公司研究院 | Space-time multi-index prediction method and device and storage medium |
Citations (9)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN101150836A (en) * | 2007-11-06 | 2008-03-26 | 中国移动通信集团江苏有限公司 | Radio precise planning analog and testing system and method |
| CN101964981A (en) * | 2009-07-22 | 2011-02-02 | 中国移动通信集团北京有限公司 | Method and device for pre-evaluating telephone traffic and configuring carrier frequency of new base station |
| CN102111284A (en) * | 2009-12-28 | 2011-06-29 | 北京亿阳信通软件研究院有限公司 | Method and device for predicting telecom traffic |
| CN102595416A (en) * | 2012-02-16 | 2012-07-18 | 北京邮电大学 | PCI (Physical Cell Identity) self-adaptive distribution method in cellular communication system |
| CN103354644A (en) * | 2013-07-31 | 2013-10-16 | 广东省电信规划设计院有限公司 | Method and system for site selection of base station |
| CN103686759A (en) * | 2012-09-26 | 2014-03-26 | 中国移动通信集团广东有限公司 | TD-LTE system base station locating method and TD-LTE system base station locating device |
| CN103931225A (en) * | 2012-11-09 | 2014-07-16 | 华为技术服务有限公司 | Method and device for planning new base station |
| CN104125580A (en) * | 2013-04-23 | 2014-10-29 | 中国移动通信集团浙江有限公司 | Network planning method and apparatus |
| US20160029253A1 (en) * | 2014-07-28 | 2016-01-28 | Telefonaktiebolaget L M Ericsson (Publ) | System and method of automatic neighbor relation (anr) intelligence enhancement for boomer neighbor in lte |
-
2017
- 2017-08-09 CN CN201710676951.4A patent/CN107426759B/en active Active
Patent Citations (9)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN101150836A (en) * | 2007-11-06 | 2008-03-26 | 中国移动通信集团江苏有限公司 | Radio precise planning analog and testing system and method |
| CN101964981A (en) * | 2009-07-22 | 2011-02-02 | 中国移动通信集团北京有限公司 | Method and device for pre-evaluating telephone traffic and configuring carrier frequency of new base station |
| CN102111284A (en) * | 2009-12-28 | 2011-06-29 | 北京亿阳信通软件研究院有限公司 | Method and device for predicting telecom traffic |
| CN102595416A (en) * | 2012-02-16 | 2012-07-18 | 北京邮电大学 | PCI (Physical Cell Identity) self-adaptive distribution method in cellular communication system |
| CN103686759A (en) * | 2012-09-26 | 2014-03-26 | 中国移动通信集团广东有限公司 | TD-LTE system base station locating method and TD-LTE system base station locating device |
| CN103931225A (en) * | 2012-11-09 | 2014-07-16 | 华为技术服务有限公司 | Method and device for planning new base station |
| CN104125580A (en) * | 2013-04-23 | 2014-10-29 | 中国移动通信集团浙江有限公司 | Network planning method and apparatus |
| CN103354644A (en) * | 2013-07-31 | 2013-10-16 | 广东省电信规划设计院有限公司 | Method and system for site selection of base station |
| US20160029253A1 (en) * | 2014-07-28 | 2016-01-28 | Telefonaktiebolaget L M Ericsson (Publ) | System and method of automatic neighbor relation (anr) intelligence enhancement for boomer neighbor in lte |
Non-Patent Citations (2)
| Title |
|---|
| 朱李光: ""LTE网络基于MR的小区相关性分析在邻区优化中的运用"", 《移动通信》 * |
| 朱李光: ""CDMA网络中PN规划方法"", 《移动通信》 * |
Cited By (20)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN109039691A (en) * | 2018-06-01 | 2018-12-18 | 平安科技（深圳）有限公司 | The method and storage medium of server, forecasting system calling amount |
| CN111225382A (en) * | 2018-11-27 | 2020-06-02 | 中国移动通信集团上海有限公司 | LTE functional characteristic selection method and system |
| CN111225382B (en) * | 2018-11-27 | 2023-04-25 | 中国移动通信集团上海有限公司 | LTE functional characteristic selection method and system |
| CN109391950A (en) * | 2018-11-30 | 2019-02-26 | 中国移动通信集团福建有限公司 | Prediction technique, device, equipment and the medium of terminal distribution |
| CN109391950B (en) * | 2018-11-30 | 2021-10-29 | 中国移动通信集团福建有限公司 | Method, device, equipment and medium for predicting terminal distribution |
| CN111278013A (en) * | 2018-12-04 | 2020-06-12 | 中国移动通信集团山东有限公司 | Automatic pseudo base station identification method and device |
| CN111163482B (en) * | 2019-12-25 | 2022-10-14 | 中国联合网络通信集团有限公司 | Data processing method, device and storage medium |
| CN111163482A (en) * | 2019-12-25 | 2020-05-15 | 中国联合网络通信集团有限公司 | Data processing method, device and storage medium |
| CN113052582B (en) * | 2019-12-27 | 2024-03-22 | 中移动信息技术有限公司 | Method, device, equipment and computer storage medium for checking bill |
| CN113052582A (en) * | 2019-12-27 | 2021-06-29 | 中移动信息技术有限公司 | Bill checking method, device and equipment and computer storage medium |
| CN113766523B (en) * | 2020-06-02 | 2023-08-01 | 中国移动通信集团河南有限公司 | Method and device for predicting network resource utilization rate of serving cell and electronic equipment |
| CN113766523A (en) * | 2020-06-02 | 2021-12-07 | 中国移动通信集团河南有限公司 | Network resource utilization rate prediction method and device for serving cell and electronic equipment |
| CN113938893A (en) * | 2020-06-29 | 2022-01-14 | 中国移动通信集团广西有限公司 | Network planning assisting method and device and electronic equipment |
| CN112054943A (en) * | 2020-09-17 | 2020-12-08 | 中国电信股份有限公司河北分公司 | Traffic prediction method for mobile network base station |
| CN114338416A (en) * | 2020-09-29 | 2022-04-12 | 中国移动通信有限公司研究院 | Space-time multi-index prediction method and device and storage medium |
| CN114338416B (en) * | 2020-09-29 | 2023-04-07 | 中国移动通信有限公司研究院 | Space-time multi-index prediction method and device and storage medium |
| CN112492637B (en) * | 2020-12-28 | 2023-06-27 | 中国联合网络通信集团有限公司 | Method and device for predicting cell traffic |
| CN112492637A (en) * | 2020-12-28 | 2021-03-12 | 中国联合网络通信集团有限公司 | Method and device for predicting cell traffic |
| CN114071525A (en) * | 2021-11-10 | 2022-02-18 | 中国联合网络通信集团有限公司 | Method, device and storage medium for determining optimization order of base station |
| CN114071525B (en) * | 2021-11-10 | 2023-07-18 | 中国联合网络通信集团有限公司 | Base station optimization order determining method, device and storage medium |
Also Published As
| Publication number | Publication date |
|---|---|
| CN107426759B (en) | 2020-06-19 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN107426759A (en) | The Forecasting Methodology and system of newly-increased base station data portfolio | |
| CN102567391B (en) | Method and device for building classification forecasting mixed model | |
| CN104572449A (en) | Automatic test method based on case library | |
| CN107463730B (en) | A kind of streamflow change attribution recognition methods considering Spatio-temporal Evolution of Land Use | |
| CN106095942B (en) | Strong variable extracting method and device | |
| CN110210686A (en) | A kind of electricity charge risk model construction method of electric power big data | |
| CN112801342A (en) | Adaptive runoff forecasting method based on rainfall runoff similarity | |
| CN106096810B (en) | Method and system for planning based on power distribution network operation data Yu geographical topology information | |
| CN107133436A (en) | A kind of multiple sample model training method and device | |
| CN103106535B (en) | Method for solving collaborative filtering recommendation data sparsity based on neural network | |
| CN106934720A (en) | Equipment insurance intelligent pricing method and system based on Internet of Things | |
| CN107517481A (en) | A kind of load of base station balanced management method and system | |
| Chu et al. | Improving multi-objective reservoir operation optimization with sensitivity-informed dimension reduction | |
| CN107679660A (en) | Based on SVMs by when building energy consumption Forecasting Methodology | |
| CN104252652A (en) | Space load predicting method in electricity system | |
| CN109903182A (en) | Power customer arrears risk analysis method and device based on random forests algorithm | |
| CN116090839B (en) | Multiple risk analysis and evaluation method and system for water resource coupling system | |
| CN110097220A (en) | A kind of monthly power predicating method of wind-power electricity generation | |
| CN113256022A (en) | Method and system for predicting electric load of transformer area | |
| CN108197848A (en) | A kind of energy quality comprehensive assessment method and device based on intuitionistic fuzzy theory | |
| CN109741177A (en) | Appraisal procedure, device and the intelligent terminal of user credit | |
| CN111008870A (en) | Regional logistics demand prediction method based on PCA-BP neural network model | |
| CN113033081A (en) | Runoff simulation method and system based on SOM-BPNN model | |
| CN108462965A (en) | Share planing method in a kind of steel tower demand site | |
| Bo et al. | Establishment and application of fuzzy comprehensive evaluation of green building design based on data mining |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| GR01 | Patent grant | ||
| GR01 | Patent grant |