CN110175711A - One kind being based on joint LSTM base station cell method for predicting and device - Google Patents
One kind being based on joint LSTM base station cell method for predicting and device Download PDFInfo
- Publication number
- CN110175711A CN110175711A CN201910412460.8A CN201910412460A CN110175711A CN 110175711 A CN110175711 A CN 110175711A CN 201910412460 A CN201910412460 A CN 201910412460A CN 110175711 A CN110175711 A CN 110175711A
- Authority
- CN
- China
- Prior art keywords
- cell
- prediction
- sample data
- lstm
- generates
- 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.)
- Pending
Links
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
- G06N3/00—Computing arrangements based on biological models
- G06N3/02—Neural networks
- G06N3/04—Architecture, e.g. interconnection topology
- G06N3/045—Combinations of networks
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06Q—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES, NOT OTHERWISE PROVIDED FOR
- G06Q10/00—Administration; Management
- G06Q10/04—Forecasting or optimisation specially adapted for administrative or management purposes, e.g. linear programming or "cutting stock problem"
Abstract
The disclosure is directed to one kind to be based on the base station cell joint LSTM method for predicting, device, electronic equipment and computer readable storage medium.Wherein, this method comprises: obtaining the user data of each cell in scene to be predicted respectively, and processing is carried out to the user data and generates sample data；It is based on long Memory Neural Networks (LSTM) in short-term to the sample data of each cell respectively to model, the prediction for carrying out the periodic characteristic of the sample data calculates, and generates periodic characteristic prediction result；Two layers of artificial neural network (ANN) modeling is based on to all sample datas, the prediction for carrying out the linked character of the sample data calculates, and generates linked character prediction result；Merge above-mentioned periodic characteristic prediction result and linked character prediction result, generates each community associated traffic prediction value.The disclosure realizes the Accurate Prediction in advance to multiple cell flow by the method based on joint LSTM.
Description
Technical field
This disclosure relates to which field of computer technology, pre- based on the base station cell joint LSTM flow in particular to one kind
Survey method, apparatus, electronic equipment and computer readable storage medium.
Background technique
The wireless base station of cell is first important bridge that user data moves towards internet, therefore, for wireless network
For the operator of network service, estate performance is optimized and is to maintain one of the important means of network performance.
Existing network optimization method is mainly by traditional network optimized approach based on detection and to be based on common timing
The method of model, but the method based on drive test and fixed point voice quality test, in addition to being had an impact to user's normal use, people
Work testing cost cost is big, needs to consume a large amount of manpower and material resources；Based on the method for current forecasting of time series model base station flow, such as
Auto regressive moving average (ARIMA) model, though having certain advantage, the method first can only be according to historical cell flow
Service condition prediction, it is contemplated that characteristic factor it is less, such as information such as user volume of base station cell access.Secondly, an area
Base station cell is more in domain, can generate certain influence between cell and cell mutually surely, and the method can not consider cell
Between influence factor.
Therefore, it is badly in need of a kind of following a period of time cell flow method of more accurate prediction, it can be as desired and small
Area's characteristic distributions, can be from time, Spatial Dimension, in conjunction with some features needed for other, analysis prediction base station cell flow.
From the above, it can be seen that, it is desirable to provide one or more technical solutions for being at least able to solve the above problem.
It should be noted that information is only used for reinforcing the reason to the background of the disclosure disclosed in above-mentioned background technology part
Solution, therefore may include the information not constituted to the prior art known to persons of ordinary skill in the art.
Summary of the invention
The one kind that is designed to provide of the disclosure is set based on the base station cell joint LSTM method for predicting, device, electronics
Standby and computer readable storage medium, and then overcome the limitation and defect due to the relevant technologies at least to a certain extent and lead
One or more problem caused.
According to one aspect of the disclosure, it provides a kind of based on the base station cell joint LSTM method for predicting, comprising:
Sample data processing step obtains the user data of each cell in scene to be predicted respectively, and to the number of users
Sample data is generated according to processing is carried out；
Periodic characteristic prediction steps are based on long Memory Neural Networks in short-term to the sample data of each cell respectively
(LSTM) it models, the prediction for carrying out the periodic characteristic of the sample data calculates, and generates periodic characteristic prediction result；
Linked character prediction steps are based on two layers of artificial neural network (ANN) to all sample datas and model, described in progress
The prediction of the linked character of sample data calculates, and generates linked character prediction result；
Traffic prediction value generation step merges above-mentioned periodic characteristic prediction result and linked character prediction result, generates each
Community associated traffic prediction value.
In a kind of exemplary embodiment of the disclosure, the sample data processing step further include:
The user data of each cell derives from each base station reported data, and flow, user volume, spare time including each base station are busy
When, festivals or holidays, the time granularity of the user data is hour.
In a kind of exemplary embodiment of the disclosure, the sample data processing step further includes to the user data
It is pre-processed:
Based on nearest polishing principle, polishing processing is carried out to missing data using last moment data；
Determine the user volume, flow of all cell unit time and corresponding not busy busy, festivals or holidays etc. in a certain range
Characteristic information；
Cell flow, the user volume of historical juncture are standardized respectively, and not busy busy, festivals or holidays are quantified
The mute coding of feature.
In a kind of exemplary embodiment of the disclosure, the periodic characteristic prediction steps further include:
Each cell is established respectively based on long Memory Neural Networks (LSTM) model in short-term, with each cell in the more of fixed cycle
Flow, the user volume at a time point are used as input to carry out prediction calculating, generate periodic characteristic prediction result.
In a kind of exemplary embodiment of the disclosure, the linked character prediction steps further include:
Two layers of artificial neural network (ANN) modeling is based on to other correlated characteristics of each cell, it is busy with the spare time in sample data
When, festivals or holidays linked character as input carry out prediction calculating, generate linked character prediction result.
In a kind of exemplary embodiment of the disclosure, the traffic prediction value generation step further include:
Calculating is iterated to the model based on LSTM and ANN, default cut-off condition is introduced, generates each community associated
Traffic prediction value.
In a kind of exemplary embodiment of the disclosure, the traffic prediction value generation step further include:
Calculating is iterated to the model based on LSTM and ANN, model is allowed with early method (early stopping) of stopping
Convergence after then presetting the number of iterations, generates model optimal solution, and then obtain each community associated traffic prediction value.
In one aspect of the present disclosure, it provides a kind of based on joint LSTM base station cell volume forecasting device, comprising:
Sample data processing module, for obtaining the user data of each cell in scene to be predicted respectively, and to the use
User data carries out processing and generates sample data；
Periodic characteristic prediction module is based on long Memory Neural Networks in short-term for the sample data respectively to each cell
(LSTM) it models, the prediction for carrying out the periodic characteristic of the sample data calculates, and generates periodic characteristic prediction result；
Linked character prediction module is carried out for being based on two layers of artificial neural network (ANN) modeling to all sample datas
The prediction of the linked character of the sample data calculates, and generates linked character prediction result；
Traffic prediction value generation module, it is raw for merging above-mentioned periodic characteristic prediction result and linked character prediction result
At each community associated traffic prediction value.
In one aspect of the present disclosure, a kind of electronic equipment is provided, comprising:
Processor；And
Memory is stored with computer-readable instruction on the memory, and the computer-readable instruction is by the processing
The method according to above-mentioned any one is realized when device executes.
In one aspect of the present disclosure, a kind of computer readable storage medium is provided, computer program is stored thereon with, institute
State realization method according to above-mentioned any one when computer program is executed by processor.
In the exemplary embodiment of the disclosure based on joint the base station cell LSTM method for predicting, obtained respectively to pre-
The user data of each cell in scene is surveyed, and processing is carried out to the user data and generates sample data；Respectively to described each small
The sample data in area is based on long Memory Neural Networks (LSTM) in short-term and models, and carries out the prediction of the periodic characteristic of the sample data
It calculates, generates periodic characteristic prediction result；Two layers of artificial neural network (ANN) modeling is based on to all sample datas, carries out institute
The prediction for stating the linked character of sample data calculates, and generates linked character prediction result；Merge above-mentioned periodic characteristic prediction result
And linked character prediction result, generate each community associated traffic prediction value.On the one hand, the disclosure can be according to based on joint
The method of LSTM modeling generates the volume forecasting of multiple cells, discloses each minizone flow interaction prediction result；On the other hand, by
In based on ANN modeling introduce it is other influence volume forecasting results factors, can respectively to the progress such as not busy busy, festivals or holidays more
Add accurately volume forecasting.
It should be understood that above general description and following detailed description be only it is exemplary and explanatory, not
The disclosure can be limited.
Detailed description of the invention
Its example embodiment is described in detail by referring to accompanying drawing, the above and other feature and advantage of the disclosure will become
It is more obvious.
Fig. 1 is shown according to one exemplary embodiment of the disclosure based on the joint base station cell LSTM method for predicting
Flow chart；
Fig. 2 shows according to one exemplary embodiment of the disclosure based on the joint base station cell LSTM method for predicting
Model structure；
Fig. 3 is shown according to one exemplary embodiment of the disclosure based on the joint base station cell LSTM method for predicting
LSTM model unit composite structural diagram；
Fig. 4 is shown according to one exemplary embodiment of the disclosure based on the joint base station cell LSTM method for predicting
Real traffic and predicted flow rate comparison diagram；
Fig. 5 is shown according to one exemplary embodiment of the disclosure based on the joint base station cell LSTM volume forecasting device
Schematic block diagram；
Fig. 6 diagrammatically illustrates the block diagram of the electronic equipment according to one exemplary embodiment of the disclosure；And
Fig. 7 diagrammatically illustrates the schematic diagram of the computer readable storage medium according to one exemplary embodiment of the disclosure.
Specific embodiment
Example embodiment is described more fully with reference to the drawings.However, example embodiment can be real in a variety of forms
It applies, and is not understood as limited to embodiment set forth herein；On the contrary, thesing embodiments are provided so that the disclosure will be comprehensively and complete
It is whole, and the design of example embodiment is comprehensively communicated to those skilled in the art.Identical appended drawing reference indicates in figure
Same or similar part, thus repetition thereof will be omitted.
In addition, described feature, structure or characteristic can be incorporated in one or more implementations in any suitable manner
In example.In the following description, many details are provided to provide and fully understand to embodiment of the disclosure.However,
It will be appreciated by persons skilled in the art that can be with technical solution of the disclosure without one in the specific detail or more
It is more, or can be using other methods, constituent element, material, device, step etc..In other cases, it is not shown in detail or describes
Known features, method, apparatus, realization, material or operation are to avoid fuzzy all aspects of this disclosure.
Block diagram shown in the drawings is only functional entity, not necessarily must be corresponding with physically separate entity.
I.e., it is possible to realize these functional entitys using software form, or these are realized in the module of one or more softwares hardening
A part of functional entity or functional entity, or realized in heterogeneous networks and/or processor device and/or microcontroller device
These functional entitys.
In this exemplary embodiment, it provides firstly a kind of based on the base station cell joint LSTM method for predicting；With reference to
Shown in Fig. 1, it should be may comprise steps of based on the base station cell joint LSTM method for predicting:
Sample data processing step S110 obtains the user data of each cell in scene to be predicted respectively, and to the use
User data carries out processing and generates sample data；
Periodic characteristic prediction steps S120 is based on long Memory Neural Networks in short-term to the sample data of each cell respectively
(LSTM) it models, the prediction for carrying out the periodic characteristic of the sample data calculates, and generates periodic characteristic prediction result；
Linked character prediction steps S130 is based on two layers of artificial neural network (ANN) modeling to all sample datas, carries out
The prediction of the linked character of the sample data calculates, and generates linked character prediction result；
Traffic prediction value generation step S140 merges above-mentioned periodic characteristic prediction result and linked character prediction result, raw
At each community associated traffic prediction value.
According in this example embodiment based on the joint base station cell LSTM method for predicting, on the one hand, the disclosure can
To generate the volume forecasting of multiple cells according to the method based on joint LSTM modeling, each minizone flow interaction prediction knot is disclosed
Fruit；On the other hand, due to based on ANN modeling introduce it is other influence volume forecasting results factors, can respectively to not busy busy,
Festivals or holidays etc. carry out more accurate volume forecasting.
In the following, by being carried out further in this example embodiment based on the joint base station cell LSTM method for predicting
Explanation.
In sample data processing step S110, the user data of each cell in scene to be predicted can be obtained respectively, and
Processing is carried out to the user data and generates sample data.
In this exemplary embodiment, the sample data processing step further include:
The user data of each cell derives from each base station reported data, and flow, user volume, spare time including each base station are busy
When, festivals or holidays, the time granularity of the user data is hour.
In this exemplary embodiment, the sample data processing step further includes being located in advance to the user data
Reason:
Based on nearest polishing principle, polishing processing is carried out to missing data using last moment data；
Determine the user volume, flow of all cell unit time and corresponding not busy busy, festivals or holidays etc. in a certain range
Characteristic information；
Cell flow, the user volume of historical juncture are standardized respectively, and not busy busy, festivals or holidays are quantified
The mute coding of feature.
In this exemplary embodiment, data set derives from flow, the user volume that base station cell reports, and time granularity is small
When, i.e., the time interval before sample is 1 hour.Select the data of three cells of continuous three months same base stations as experiment
Data set, totally 6480 data.
Data set includes flow, user volume, time, cell i d, latitude and longitude information and not busy busy, festivals or holidays data.Specifically
Feature Selection and coding are as follows:
| day_key | 000 | Time |
| cell_id | 001 | Cell i d |
| latitude | 002 | Latitude |
| longitude | 003 | Longitude |
| user_flow | 004 | Cell flow |
| user_num | 005 | Community user number |
| Busytime | 006 | Not busy busy |
| Holidaytime | 007 | Festivals or holidays |
Table 1: Feature Selection and coding schedule
It with cell is user volume, a flow number that group counts each chronomere by data, missing data is former with nearest polishing
Then, using the Data-parallel language of last moment.
Determine the user volume, flow of all cell unit time and corresponding not busy busy, festivals or holidays etc. in a certain range
Characteristic information.
Data normalization: being standardized cell flow, the user volume of historical juncture respectively, and by not busy busy,
Festivals or holidays do the mute coding of quantitative characteristic.
Determine period of time T, using the data in T time section as sample characteristics X, T+1 time data is as sample data
Y, sample data should the cell flow comprising the T moment of each cell and user volumes, T moment corresponding not busy busy, section vacation
Day etc., totally 2136 samples.
The sample data according to needed for model stores data with dictionary format, and storage organization is as follows:
Table 2: sample data storage organization table
In periodic characteristic prediction steps S120, the sample data to each cell it can be based on long short-term memory respectively
Neural network (LSTM) modeling, the prediction for carrying out the periodic characteristic of the sample data calculate, and generate periodic characteristic prediction result.
In this exemplary embodiment, the periodic characteristic prediction steps further include:
Each cell is established respectively based on long Memory Neural Networks (LSTM) model in short-term, with each cell in the more of fixed cycle
Flow, the user volume at a time point are used as input to carry out prediction calculating, generate periodic characteristic prediction result.
In linked character prediction steps S130, two layers of artificial neural network (ANN) can be based on to all sample datas
Modeling, the prediction for carrying out the linked character of the sample data calculate, and generate linked character prediction result.
In this exemplary embodiment, the linked character prediction steps further include:
Two layers of artificial neural network (ANN) modeling is based on to other correlated characteristics of each cell, it is busy with the spare time in sample data
When, festivals or holidays linked character as input carry out prediction calculating, generate linked character prediction result.
In this exemplary embodiment, the thought modeling based on deep learning, each cell is utilized respectively long short-term memory
Neural network (LSTM) modeling, and the model for merging each cell exports the incidence relation between feature learning cell, further accordance with
The features such as festivals or holidays, not busy busy learn assemblage characteristic using artificial neural network (ANN), finally by cell characteristic and festivals or holidays,
Not busy busy feature merges jointly, establishes deep learning model, to predict flow, the user volume of future time instance cell.
Model algorithm: LSTM+ANN, using the user volume of lstm model learning history, the temporal aspect of flow, and to phase
Close other characteristic use artificial neural networks building model.
As shown in Figure 2 illustrating realizes process based on joint LSTM base station cell discharge model.In Fig. 2, model
Output be T+1 time point each cell flow (cell A, cell B), input data include each cell (cell A, cell B)
In data such as the corresponding not busy busy of n time point (cycle T) cell flow and user volume and T+1 time point, festivals or holidays.Model
Respectively to cell A, cell B) it is modeled with long Memory Neural Networks (LSTM) in short-term, excavate the periodic feature of cell flow；Its
He is modeled correlated characteristic (data such as not busy busy, festivals or holidays) using two layers of artificial neural network (ANN).Then will obtain cell A,
Cell B carries out feature merging in the output for other correlated characteristics that T+1 time point exports and learns, then passes through the complete of neural network
Connectionist learning further learns linked character；Finally by the flow of two hiding neural unit output cell A and cell B.
If Fig. 3 is length used Memory Neural Networks neural unit structure chart in short-term.Cell A and the definition of cell B input data
For XA、XB, XAFor n-dimensional vector: XA=(x1x2…xn), wherein xiFor bivector: xi=(f, u), f, u indicate the stream at i time point
Amount and user volume.
In traffic prediction value generation step S140, above-mentioned periodic characteristic prediction result and linked character prediction can be merged
As a result, generating each community associated traffic prediction value.
In this exemplary embodiment, the traffic prediction value generation step further include:
Calculating is iterated to the model based on LSTM and ANN, default cut-off condition is introduced, generates each community associated
Traffic prediction value.
In this exemplary embodiment, the traffic prediction value generation step further include:
Calculating is iterated to the model based on LSTM and ANN, with the early mould that allows for stopping method (early stopping)
Type convergence after then presetting the number of iterations, generates model optimal solution, and then obtain each community associated traffic prediction value.
In this exemplary embodiment, the model training of the disclosure be in a manner of two kinds of iteration cut-off condition, first with
The mode for early stopping method (early stopping) allows model to restrain as early as possible, then defines the mode of the number of iterations, model is allowed to find
It is optimal.Model exports the predicted value for the future traffic that H is all cells in base station.As Fig. 4 shows certain cell in test data set
True and predicted flow rate comparison diagram, wherein solid line indicates that cell true flow rate value, dotted line indicate cell predicted flow rate value.
In this exemplary embodiment, by introducing the information characteristics such as not busy busy, festivals or holidays, dug using artificial neural network
Pick, and be fused in joint LSTM model, jointly constructs deep neural network model.The user volume and flow that the disclosure proposes close
The scheme for joining analysis, increases the dynamic factor in volume forecasting, improves the precision of volume forecasting.
It should be noted that although describing each step of method in the disclosure in the accompanying drawings with particular order,
This does not require that or implies must execute these steps in this particular order, or have to carry out step shown in whole
Just it is able to achieve desired result.Additional or alternative, it is convenient to omit multiple steps are merged into a step and held by certain steps
Row, and/or a step is decomposed into execution of multiple steps etc..
In addition, in this exemplary embodiment, additionally providing a kind of based on joint LSTM base station cell volume forecasting device.Ginseng
According to shown in Fig. 5, should may include: based on joint LSTM base station cell volume forecasting device 200 sample data processing module 210,
Periodic characteristic prediction module 220, linked character prediction module 230 and traffic prediction value generation module 240.Wherein:
Sample data processing module 210, for obtaining the user data of each cell in scene to be predicted respectively, and to described
User data carries out processing and generates sample data；
Periodic characteristic prediction module 220 is based on long short-term memory nerve for the sample data respectively to each cell
Network (LSTM) modeling, the prediction for carrying out the periodic characteristic of the sample data calculate, and generate periodic characteristic prediction result；
Linked character prediction module 230, for being based on two layers of artificial neural network (ANN) modeling to all sample datas,
The prediction for carrying out the linked character of the sample data calculates, and generates linked character prediction result；
Traffic prediction value generation module 240, for merging above-mentioned periodic characteristic prediction result and linked character prediction result,
Generate each community associated traffic prediction value.
The detail respectively based on the joint base station cell LSTM volume forecasting apparatus module is in corresponding sound among the above
Frequency range falls in recognition methods and is described in detail, therefore details are not described herein again.
It should be noted that although being referred in the above detailed description based on joint LSTM base station cell volume forecasting device
200 several modules or unit, but this division is not enforceable.In fact, according to embodiment of the present disclosure, on
Two or more modules of text description or the feature and function of unit can embody in a module or unit.Instead
It, the feature and function of an above-described module or unit can be with further division by multiple modules or unit Lai
It embodies.
In addition, in an exemplary embodiment of the disclosure, additionally providing a kind of electronic equipment that can be realized the above method.
Person of ordinary skill in the field it is understood that various aspects of the invention can be implemented as system, method or
Program product.Therefore, various aspects of the invention can be embodied in the following forms, it may be assumed that complete hardware embodiment, completely
Software implementation (including firmware, microcode etc.) or hardware and software in terms of combine embodiment, may be collectively referred to as here
Circuit, " module " or " system ".
The electronic equipment 300 of this embodiment according to the present invention is described referring to Fig. 6.The electronics that Fig. 6 is shown is set
Standby 300 be only an example, should not function to the embodiment of the present invention and use scope bring any restrictions.
As shown in figure 3, electronic equipment 300 is showed in the form of universal computing device.The component of electronic equipment 300 can wrap
It includes but is not limited to: at least one above-mentioned processing unit 310, at least one above-mentioned storage unit 320, the different system components of connection
The bus 330 of (including storage unit 320 and processing unit 310), display unit 340.
Wherein, the storage unit is stored with program code, and said program code can be held by the processing unit 310
Row, so that various according to the present invention described in the execution of the processing unit 310 above-mentioned " illustrative methods " part of this specification
The step of exemplary embodiment.For example, the processing unit 310 can execute step S110 as shown in fig. 1 to step
S140。
Storage unit 320 may include the readable medium of volatile memory cell form, such as Random Access Storage Unit
(RAM) 3201 and/or cache memory unit 3202, it can further include read-only memory unit (ROM) 3203.
Storage unit 320 can also include program/utility with one group of (at least one) program module 3205
3204, such program module 3205 includes but is not limited to: operating system, one or more application program, other program moulds
It may include the realization of network environment in block and program data, each of these examples or certain combination.
Bus 330 can be to indicate one of a few class bus structures or a variety of, including storage unit bus or storage
Cell controller, peripheral bus, graphics acceleration port, processing unit use any bus structures in a variety of bus structures
Local bus.
Electronic equipment 300 can also be with one or more external equipments 370 (such as keyboard, sensing equipment, bluetooth equipment
Deng) communication, can also be enabled a user to one or more equipment interact with the electronic equipment 300 communicate, and/or with make
Any equipment (such as the router, modulation /demodulation that the electronic equipment 300 can be communicated with one or more of the other calculating equipment
Device etc.) communication.This communication can be carried out by input/output (I/O) interface 350.Also, electronic equipment 300 can be with
By network adapter 360 and one or more network (such as local area network (LAN), wide area network (WAN) and/or public network,
Such as internet) communication.As shown, network adapter 360 is communicated by bus 330 with other modules of electronic equipment 300.
It should be understood that although not shown in the drawings, other hardware and/or software module can not used in conjunction with electronic equipment 300, including but not
Be limited to: microcode, device driver, redundant processing unit, external disk drive array, RAID system, tape drive and
Data backup storage system etc..
By the description of above embodiment, those skilled in the art is it can be readily appreciated that example embodiment described herein
It can also be realized in such a way that software is in conjunction with necessary hardware by software realization.Therefore, implemented according to the disclosure
The technical solution of example can be embodied in the form of software products, which can store in a non-volatile memories
In medium (can be CD-ROM, USB flash disk, mobile hard disk etc.) or on network, including some instructions are so that a calculating equipment (can
To be personal computer, server, terminal installation or network equipment etc.) it executes according to the method for the embodiment of the present disclosure.
In an exemplary embodiment of the disclosure, a kind of computer readable storage medium is additionally provided, energy is stored thereon with
Enough realize the program product of this specification above method.In some possible embodiments, various aspects of the invention can be with
It is embodied as a kind of form of program product comprising program code, it is described when described program product is run on the terminal device
Program code is for executing the terminal device described in above-mentioned " illustrative methods " part of this specification according to the present invention
The step of various exemplary embodiments.
Refering to what is shown in Fig. 7, the program product 400 for realizing the above method of embodiment according to the present invention is described,
It can using portable compact disc read only memory (CD-ROM) and including program code, and can in terminal device, such as
It is run on PC.However, program product of the invention is without being limited thereto, in this document, readable storage medium storing program for executing, which can be, appoints
What include or the tangible medium of storage program that the program can be commanded execution system, device or device use or and its
It is used in combination.
Described program product can be using any combination of one or more readable mediums.Readable medium can be readable letter
Number medium or readable storage medium storing program for executing.Readable storage medium storing program for executing for example can be but be not limited to electricity, magnetic, optical, electromagnetic, infrared ray or
System, device or the device of semiconductor, or any above combination.The more specific example of readable storage medium storing program for executing is (non exhaustive
List) include: electrical connection with one or more conducting wires, portable disc, hard disk, random access memory (RAM), read-only
Memory (ROM), erasable programmable read only memory (EPROM or flash memory), optical fiber, portable compact disc read only memory
(CD-ROM), light storage device, magnetic memory device or above-mentioned any appropriate combination.
Computer-readable signal media may include in a base band or as carrier wave a part propagate data-signal,
In carry readable program code.The data-signal of this propagation can take various forms, including but not limited to electromagnetic signal,
Optical signal or above-mentioned any appropriate combination.Readable signal medium can also be any readable Jie other than readable storage medium storing program for executing
Matter, the readable medium can send, propagate or transmit for by instruction execution system, device or device use or and its
The program of combined use.
The program code for including on readable medium can transmit with any suitable medium, including but not limited to wirelessly, have
Line, optical cable, RF etc. or above-mentioned any appropriate combination.
The program for executing operation of the present invention can be write with any combination of one or more programming languages
Code, described program design language include object oriented program language-Java, C++ etc., further include conventional
Procedural programming language-such as " C " language or similar programming language.Program code can be fully in user
It calculates and executes in equipment, partly executes on a user device, being executed as an independent software package, partially in user's calculating
Upper side point is executed on a remote computing or is executed in remote computing device or server completely.It is being related to far
Journey calculates in the situation of equipment, and remote computing device can pass through the network of any kind, including local area network (LAN) or wide area network
(WAN), it is connected to user calculating equipment, or, it may be connected to external computing device (such as utilize ISP
To be connected by internet).
In addition, above-mentioned attached drawing is only the schematic theory of processing included by method according to an exemplary embodiment of the present invention
It is bright, rather than limit purpose.It can be readily appreciated that the time that above-mentioned processing shown in the drawings did not indicated or limited these processing is suitable
Sequence.In addition, be also easy to understand, these processing, which can be, for example either synchronously or asynchronously to be executed in multiple modules.
Those skilled in the art after considering the specification and implementing the invention disclosed here, will readily occur to its of the disclosure
His embodiment.This application is intended to cover any variations, uses, or adaptations of the disclosure, these modifications, purposes or
Adaptive change follow the general principles of this disclosure and including the undocumented common knowledge in the art of the disclosure or
Conventional techniques.The description and examples are only to be considered as illustrative, and the true scope and spirit of the disclosure are by claim
It points out.
It should be understood that the present disclosure is not limited to the precise structures that have been described above and shown in the drawings, and
And various modifications and changes may be made without departing from the scope thereof.The scope of the present disclosure is only limited by the attached claims.
Claims (10)
1. one kind is based on the base station cell joint LSTM method for predicting, which is characterized in that the described method includes:
Sample data processing step, obtains the user data of each cell in scene to be predicted respectively, and to the user data into
Row processing generates sample data；
Periodic characteristic prediction steps are based on long Memory Neural Networks (LSTM) in short-term to the sample data of each cell respectively and build
Mould, the prediction for carrying out the periodic characteristic of the sample data calculate, and generate periodic characteristic prediction result；
Linked character prediction steps are based on two layers of artificial neural network (ANN) modeling to all sample datas, carry out the sample
The prediction of the linked character of data calculates, and generates linked character prediction result；
Traffic prediction value generation step merges above-mentioned periodic characteristic prediction result and linked character prediction result, generates each cell
United traffic prediction value.
2. the method as described in claim 1, which is characterized in that the sample data processing step further include:
The user data of each cell derives from each base station reported data, flow, user volume including each base station, not busy busy,
Festivals or holidays, the time granularity of the user data are hour.
3. the method as described in claim 1, which is characterized in that the sample data processing step further includes to the number of users
According to being pre-processed:
Based on nearest polishing principle, polishing processing is carried out to missing data using last moment data；
Determine the features such as the user volume, flow of all cell unit time and corresponding not busy busy, festivals or holidays in a certain range
Information；
Cell flow, the user volume of historical juncture are standardized respectively, and not busy busy, festivals or holidays are done into quantitative characteristic
Mute coding.
4. the method as described in claim 1, which is characterized in that the periodic characteristic prediction steps further include:
Each cell is established respectively based on long Memory Neural Networks (LSTM) model in short-term, with each cell fixed cycle it is multiple when
Between flow, the user volume put as input carry out prediction calculating, generate periodic characteristic prediction result.
5. the method as described in claim 1, which is characterized in that the linked character prediction steps further include:
Two layers of artificial neural network (ANN) modeling is based on to other correlated characteristics of each cell, with the not busy busy in sample data, section
Holiday linked character carries out prediction calculating as input, generates linked character prediction result.
6. the method as described in claim 1, which is characterized in that the traffic prediction value generation step further include:
Calculating is iterated to the model based on LSTM and ANN, default cut-off condition is introduced, generates each community associated stream
Measure predicted value.
7. method as claimed in claim 6, which is characterized in that the traffic prediction value generation step further include:
Calculating is iterated to the model based on LSTM and ANN, with it is early stop method (early stopping) allow model to receive
It holds back, after then presetting the number of iterations, generates model optimal solution, and then obtain each community associated traffic prediction value.
8. one kind is based on joint LSTM base station cell volume forecasting device, which is characterized in that described device includes:
Sample data processing module, for obtaining the user data of each cell in scene to be predicted respectively, and to the number of users
Sample data is generated according to processing is carried out；
Periodic characteristic prediction module is based on long Memory Neural Networks in short-term for the sample data respectively to each cell
(LSTM) it models, the prediction for carrying out the periodic characteristic of the sample data calculates, and generates periodic characteristic prediction result；
Linked character prediction module is modeled for being based on two layers of artificial neural network (ANN) to all sample datas, described in progress
The prediction of the linked character of sample data calculates, and generates linked character prediction result；
Traffic prediction value generation module generates each for merging above-mentioned periodic characteristic prediction result and linked character prediction result
Community associated traffic prediction value.
9. a kind of electronic equipment, which is characterized in that including
Processor；And
Memory is stored with computer-readable instruction on the memory, and the computer-readable instruction is held by the processor
Method according to any one of claim 1 to 7 is realized when row.
10. a kind of computer readable storage medium, is stored thereon with computer program, the computer program is executed by processor
Shi Shixian is according to claim 1 to any one of 7 the methods.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201910412460.8A CN110175711A (en) | 2019-05-17 | 2019-05-17 | One kind being based on joint LSTM base station cell method for predicting and device |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201910412460.8A CN110175711A (en) | 2019-05-17 | 2019-05-17 | One kind being based on joint LSTM base station cell method for predicting and device |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN110175711A true CN110175711A (en) | 2019-08-27 |
Family
ID=67691456
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201910412460.8A Pending CN110175711A (en) | 2019-05-17 | 2019-05-17 | One kind being based on joint LSTM base station cell method for predicting and device |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN110175711A (en) |
Cited By (8)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN111246552A (en) * | 2020-01-14 | 2020-06-05 | 南京工业大学 | Base station dormancy method based on mobile network flow prediction |
| CN111369042A (en) * | 2020-02-27 | 2020-07-03 | 山东大学 | Wireless service flow prediction method based on weighted federal learning |
| CN111585783A (en) * | 2020-03-18 | 2020-08-25 | 宁波送变电建设有限公司永耀科技分公司 | Flow dynamic scheduling method based on space-time flow matrix prediction |
| CN112468312A (en) * | 2019-09-09 | 2021-03-09 | 中兴通讯股份有限公司 | Network flow prediction method, communication equipment and storage medium |
| CN112686483A (en) * | 2019-10-17 | 2021-04-20 | 中国移动通信集团陕西有限公司 | Early warning area identification method and device, computing equipment and computer storage medium |
| CN113743684A (en) * | 2021-11-05 | 2021-12-03 | 通号通信信息集团有限公司 | People flow prediction method, electronic device and computer readable storage medium |
| CN115396328A (en) * | 2021-05-08 | 2022-11-25 | 中国移动通信有限公司研究院 | Network index prediction method and device and electronic equipment |
| US11863397B2 (en) | 2019-12-25 | 2024-01-02 | Zte Corporation | Traffic prediction method, device, and storage medium |
Citations (19)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US20140258198A1 (en) * | 2013-02-22 | 2014-09-11 | Bottlenose, Inc. | System and method for revealing correlations between data streams |
| CN104394538A (en) * | 2014-11-28 | 2015-03-04 | 重庆大学 | Mobile network data flow analysis and prediction method |
| CN105376097A (en) * | 2015-11-30 | 2016-03-02 | 沈阳工业大学 | Hybrid prediction method for network traffic |
| TW201628384A (en) * | 2015-01-28 | 2016-08-01 | Chunghwa Telecom Co Ltd | Traffic prediction and resource scheduling automatic analyzing system of mobile network base station and method thereof |
| US20170109756A1 (en) * | 2014-07-30 | 2017-04-20 | Huawei Technologies Co., Ltd. | User Unsubscription Prediction Method and Apparatus |
| CN107171848A (en) * | 2017-05-27 | 2017-09-15 | 华为技术有限公司 | A kind of method for predicting and device |
| US20180137412A1 (en) * | 2016-11-16 | 2018-05-17 | Cisco Technology, Inc. | Network traffic prediction using long short term memory neural networks |
| CN108446794A (en) * | 2018-02-25 | 2018-08-24 | 西安电子科技大学 | One kind being based on multiple convolutional neural networks combination framework deep learning prediction techniques |
| CN108733792A (en) * | 2018-05-14 | 2018-11-02 | 北京大学深圳研究生院 | A kind of entity relation extraction method |
| CN108764568A (en) * | 2018-05-28 | 2018-11-06 | 哈尔滨工业大学 | A kind of data prediction model tuning method and device based on LSTM networks |
| CN108900346A (en) * | 2018-07-06 | 2018-11-27 | 西安电子科技大学 | Wireless network traffic prediction technique based on LSTM network |
| CN109194498A (en) * | 2018-07-27 | 2019-01-11 | 南京理工大学 | A kind of network flow prediction method based on LSTM |
| CN109257760A (en) * | 2018-09-28 | 2019-01-22 | 西安交通大学深圳研究院 | Customer flow forecasting system in wireless network |
| CN109495318A (en) * | 2018-12-17 | 2019-03-19 | 广东宜通世纪科技股份有限公司 | A kind of mobile communications network method for predicting, device and readable storage medium storing program for executing |
| CN109640351A (en) * | 2019-01-25 | 2019-04-16 | 南京邮电大学 | A kind of unified prediction of base station flow |
| CN109670569A (en) * | 2017-10-16 | 2019-04-23 | 优酷网络技术（北京）有限公司 | Neural net prediction method and device |
| CN109697522A (en) * | 2017-10-20 | 2019-04-30 | 北京京东尚科信息技术有限公司 | A kind of method and apparatus of data prediction |
| CN109711865A (en) * | 2018-12-07 | 2019-05-03 | 恒安嘉新（北京）科技股份公司 | A method of prediction is refined based on the mobile radio communication flow that user behavior excavates |
| CN109754126A (en) * | 2019-01-30 | 2019-05-14 | 银江股份有限公司 | Short-time Traffic Flow Forecasting Methods based on temporal correlation and convolutional neural networks |
-
2019
- 2019-05-17 CN CN201910412460.8A patent/CN110175711A/en active Pending
Patent Citations (19)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US20140258198A1 (en) * | 2013-02-22 | 2014-09-11 | Bottlenose, Inc. | System and method for revealing correlations between data streams |
| US20170109756A1 (en) * | 2014-07-30 | 2017-04-20 | Huawei Technologies Co., Ltd. | User Unsubscription Prediction Method and Apparatus |
| CN104394538A (en) * | 2014-11-28 | 2015-03-04 | 重庆大学 | Mobile network data flow analysis and prediction method |
| TW201628384A (en) * | 2015-01-28 | 2016-08-01 | Chunghwa Telecom Co Ltd | Traffic prediction and resource scheduling automatic analyzing system of mobile network base station and method thereof |
| CN105376097A (en) * | 2015-11-30 | 2016-03-02 | 沈阳工业大学 | Hybrid prediction method for network traffic |
| US20180137412A1 (en) * | 2016-11-16 | 2018-05-17 | Cisco Technology, Inc. | Network traffic prediction using long short term memory neural networks |
| CN107171848A (en) * | 2017-05-27 | 2017-09-15 | 华为技术有限公司 | A kind of method for predicting and device |
| CN109670569A (en) * | 2017-10-16 | 2019-04-23 | 优酷网络技术（北京）有限公司 | Neural net prediction method and device |
| CN109697522A (en) * | 2017-10-20 | 2019-04-30 | 北京京东尚科信息技术有限公司 | A kind of method and apparatus of data prediction |
| CN108446794A (en) * | 2018-02-25 | 2018-08-24 | 西安电子科技大学 | One kind being based on multiple convolutional neural networks combination framework deep learning prediction techniques |
| CN108733792A (en) * | 2018-05-14 | 2018-11-02 | 北京大学深圳研究生院 | A kind of entity relation extraction method |
| CN108764568A (en) * | 2018-05-28 | 2018-11-06 | 哈尔滨工业大学 | A kind of data prediction model tuning method and device based on LSTM networks |
| CN108900346A (en) * | 2018-07-06 | 2018-11-27 | 西安电子科技大学 | Wireless network traffic prediction technique based on LSTM network |
| CN109194498A (en) * | 2018-07-27 | 2019-01-11 | 南京理工大学 | A kind of network flow prediction method based on LSTM |
| CN109257760A (en) * | 2018-09-28 | 2019-01-22 | 西安交通大学深圳研究院 | Customer flow forecasting system in wireless network |
| CN109711865A (en) * | 2018-12-07 | 2019-05-03 | 恒安嘉新（北京）科技股份公司 | A method of prediction is refined based on the mobile radio communication flow that user behavior excavates |
| CN109495318A (en) * | 2018-12-17 | 2019-03-19 | 广东宜通世纪科技股份有限公司 | A kind of mobile communications network method for predicting, device and readable storage medium storing program for executing |
| CN109640351A (en) * | 2019-01-25 | 2019-04-16 | 南京邮电大学 | A kind of unified prediction of base station flow |
| CN109754126A (en) * | 2019-01-30 | 2019-05-14 | 银江股份有限公司 | Short-time Traffic Flow Forecasting Methods based on temporal correlation and convolutional neural networks |
Non-Patent Citations (8)
| Title |
|---|
| ATA AKBARI ASANJAN 等: "Short-term precipitation forecast based on the PERSIANN system and LSTM recurrent neural networks", 《JGR ATMOSPHERES》 * |
| RUI FU 等: "Using LSTM and GRU neural network methods for traffic flow prediction", 《2016 31ST YOUTH ACADEMIC ANNUAL CONFERENCE OF CHINESE ASSOCIATION OF AUTOMATION》 * |
| ZONGMING YIN 等: "Forecast customer flow using long short-term memory networks", 《2017 INTERNATIONAL CONFERENCE ON SECURITY, PATTERN ANALYSIS, AND CYBERNETICS》 * |
| 党小超等: "季节周期性Elman网络的网络流量分析与应用", 《计算机工程与应用》 * |
| 周鑫 等: "GPRS小区流量预测中时序模型的比较研究", 《计算机应用》 * |
| 彭铎 等: "基于空间合作关系的基站流量预测模型", 《计算机应用》 * |
| 李校林 等: "基于PF-LSTM网络的高效网络流量预测方法", 《计算机应用研究》 * |
| 谭晖 等: "一种基于LSTM神经网络的流量预测模型", 《信息通信》 * |
Cited By (14)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN112468312A (en) * | 2019-09-09 | 2021-03-09 | 中兴通讯股份有限公司 | Network flow prediction method, communication equipment and storage medium |
| WO2021047270A1 (en) * | 2019-09-09 | 2021-03-18 | 中兴通讯股份有限公司 | Network traffic prediction method, communication device and storage medium |
| CN112686483A (en) * | 2019-10-17 | 2021-04-20 | 中国移动通信集团陕西有限公司 | Early warning area identification method and device, computing equipment and computer storage medium |
| US11863397B2 (en) | 2019-12-25 | 2024-01-02 | Zte Corporation | Traffic prediction method, device, and storage medium |
| CN111246552A (en) * | 2020-01-14 | 2020-06-05 | 南京工业大学 | Base station dormancy method based on mobile network flow prediction |
| CN111246552B (en) * | 2020-01-14 | 2021-11-26 | 南京工业大学 | Base station dormancy method based on mobile network flow prediction |
| WO2021169577A1 (en) * | 2020-02-27 | 2021-09-02 | 山东大学 | Wireless service traffic prediction method based on weighted federated learning |
| CN111369042B (en) * | 2020-02-27 | 2021-09-24 | 山东大学 | Wireless service flow prediction method based on weighted federal learning |
| CN111369042A (en) * | 2020-02-27 | 2020-07-03 | 山东大学 | Wireless service flow prediction method based on weighted federal learning |
| CN111585783A (en) * | 2020-03-18 | 2020-08-25 | 宁波送变电建设有限公司永耀科技分公司 | Flow dynamic scheduling method based on space-time flow matrix prediction |
| CN111585783B (en) * | 2020-03-18 | 2022-07-26 | 宁波送变电建设有限公司永耀科技分公司 | Flow dynamic scheduling method based on space-time flow matrix prediction |
| CN115396328A (en) * | 2021-05-08 | 2022-11-25 | 中国移动通信有限公司研究院 | Network index prediction method and device and electronic equipment |
| CN113743684A (en) * | 2021-11-05 | 2021-12-03 | 通号通信信息集团有限公司 | People flow prediction method, electronic device and computer readable storage medium |
| CN113743684B (en) * | 2021-11-05 | 2022-03-15 | 通号通信信息集团有限公司 | People flow prediction method, electronic device and computer readable storage medium |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN110175711A (en) | One kind being based on joint LSTM base station cell method for predicting and device | |
| JP7157154B2 (en) | Neural Architecture Search Using Performance Prediction Neural Networks | |
| KR102302609B1 (en) | Neural Network Architecture Optimization | |
| US10929614B2 (en) | Automated contextual dialog generation for cognitive conversation | |
| JP6828001B2 (en) | Voice wakeup method and equipment | |
| US20220121906A1 (en) | Task-aware neural network architecture search | |
| CN114362367B (en) | Cloud-edge-cooperation-oriented power transmission line monitoring system and method, and cloud-edge-cooperation-oriented power transmission line identification system and method | |
| Xie et al. | A survey on machine learning-based mobile big data analysis: Challenges and applications | |
| JP2020512639A (en) | Device placement optimization using reinforcement learning | |
| JP6516660B2 (en) | Data prediction apparatus, information terminal, program and method for learning based on data of different periodic hierarchy | |
| US20200227143A1 (en) | Machine learning framework for finding materials with desired properties | |
| CN113505537A (en) | Building energy consumption detection method and device, computer equipment and storage medium | |
| KR20180103671A (en) | Electronic apparatus for compressing language model, electronic apparatus for providing recommendation word and operation methods thereof | |
| CN111950295A (en) | Method and system for training natural language processing model | |
| CN113204577A (en) | Information pushing method and device, electronic equipment and computer readable medium | |
| CN111950791A (en) | Flight delay prediction method, device, server and storage medium | |
| CN116703131B (en) | Power resource allocation method, device, electronic equipment and computer readable medium | |
| KR102382211B1 (en) | Citizen satisfaction prediction system and operation method for smart city construction | |
| CN111401980B (en) | Method and device for improving sample sequencing diversity | |
| CN113011195A (en) | Recommendation system effect enhancement method and device based on pre-training language model | |
| Sonawane et al. | ChatBot for college website | |
| CN115269786B (en) | Interpretable false text detection method and device, storage medium and terminal | |
| US10824777B2 (en) | Employing natural language processing to facilitate geospatial analysis | |
| CN116467629A (en) | Training method of user identification model, user identification method and system | |
| US20230168411A1 (en) | Using machine learning for modeling climate data |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| RJ01 | Rejection of invention patent application after publication |
Application publication date: 20190827 |
|
| RJ01 | Rejection of invention patent application after publication |