CN106301984B - A kind of mobile communications network capacity prediction methods and device - Google Patents
A kind of mobile communications network capacity prediction methods and device Download PDFInfo
- Publication number
- CN106301984B CN106301984B CN201510294026.6A CN201510294026A CN106301984B CN 106301984 B CN106301984 B CN 106301984B CN 201510294026 A CN201510294026 A CN 201510294026A CN 106301984 B CN106301984 B CN 106301984B
- Authority
- CN
- China
- Prior art keywords
- subset
- predicted
- cluster
- value
- class
- 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.)
- Active
Links
Landscapes
- Data Exchanges In Wide-Area Networks (AREA)
Abstract
The invention discloses a kind of mobile communications network capacity prediction methods and devices, to improve the accuracy of network capacity prediction result.Mobile communications network capacity prediction methods, comprising: obtain in default measurement period and be located at the data statistics of cell and/or user in region to be predicted;According to the data statistics of acquisition, if using preset clustering algorithm in the region to be predicted cell or user clustered to obtain Ganlei's subset;If the cluster feature value of every a kind of subset is determined according to Ganlei's subset that cluster obtains, wherein the cluster feature value is to obtain according to the data statistics of acquisition;For every class subset that cluster obtains, the network capacity of such subset is predicted according to the cluster feature value of such subset;The network capacity for determining the region to be predicted is the sum of the network capacity of every class subset in the region to be predicted.
Description
Technical field
The present invention relates to mobile communication technology field more particularly to a kind of mobile communications network capacity prediction methods and dress
It sets.
Background technique
The prediction of mobile communications network capacity is significant for network construction, can guarantee that network capacity is meeting not
To reduce network construction cost under the premise of business demand as far as possible, reduces the waste of radio resource investment.Existing network capacity
Prediction technique is to carry out whole volume prediction for region to be predicted, is referred to by every factor of analyzing influence network capacity
Mark, is predicted, and then predict the overall network capacity in the region to be predicted respectively for each factor index.
Every factor index that existing network capacity prediction technique is directed to entire region to be predicted carries out whole prediction, but
It is the cell due to being in diverse geographic location, network capacity situation of change gap is larger, and existing network capacity is caused to be predicted
As a result there is large error.
Summary of the invention
The embodiment of the present invention provides a kind of mobile communications network capacity prediction methods and device, pre- to improve network capacity
Survey the accuracy of result.
The embodiment of the present invention provides a kind of mobile communications network capacity prediction methods, comprising:
It obtains in default measurement period and is located at the data statistics of cell and/or user in region to be predicted;
According to the data statistics of acquisition, using preset clustering algorithm in the region to be predicted cell or
If user is clustered to obtain Ganlei's subset;
If the cluster feature value of every a kind of subset is determined according to Ganlei's subset that cluster obtains, wherein the cluster feature value
For what is obtained according to the data statistics of acquisition;
For every class subset that cluster obtains, predict that the network of such subset holds according to the cluster feature value of such subset
Amount;
The network capacity for determining the region to be predicted is the sum of the network capacity of every class subset in the region to be predicted.
The embodiment of the present invention provides a kind of mobile communications network capacity prediction meanss, comprising:
Acquiring unit is located at cell and/or the data of user system in region to be predicted for obtaining in default measurement period
Count information;
Cluster cell, for the data statistics according to acquisition, using preset clustering algorithm to the area to be predicted
If the cell or user in domain are clustered to obtain Ganlei's subset;
Determination unit, if Ganlei's subset for being obtained according to cluster determines the cluster feature value of every a kind of subset, wherein
The cluster feature value is to be obtained according to the data statistics of acquisition;
Predicting unit, every class subset for obtaining for cluster, predicts such according to the cluster feature value of such subset
The network capacity of subset;And determine that the network capacity in the region to be predicted is the network of every class subset in the region to be predicted
The sum of capacity.
Region to be predicted is included by mobile communications network capacity prediction methods and device provided in an embodiment of the present invention first
Each cell classify, every one kind cell has common characteristic value, predict its network capacity for every a kind of cell respectively,
And then the network total capacity in region to be predicted is obtained, due to no longer carrying out network appearance for entire region to be predicted in the above process
Amount prediction, but predicted respectively according to every class cell that region to be predicted includes, since the network capacity of homogeneous cells becomes
It is relatively small to change difference, so as to reduce the error of network capacity prediction result.
Other features and advantages of the present invention will be illustrated in the following description, also, partly becomes from specification
It obtains it is clear that understand through the implementation of the invention.The objectives and other advantages of the invention can be by written explanation
Specifically noted structure is achieved and obtained in book, claims and attached drawing.
Detailed description of the invention
The drawings described herein are used to provide a further understanding of the present invention, constitutes a part of the invention, this hair
Bright illustrative embodiments and their description are used to explain the present invention, and are not constituted improper limitations of the present invention.In the accompanying drawings:
Fig. 1 is the implementation process diagram of mobile communications network capacity prediction methods in the embodiment of the present invention;
Fig. 2 is the schematic illustration of mobile communications network capacity prediction in the embodiment of the present invention;
Fig. 3 is the structural schematic diagram of mobile communications network capacity prediction meanss in the embodiment of the present invention.
Specific embodiment
In order to reduce the error of mobile communications network capacity prediction, the embodiment of the invention provides a kind of mobile communications networks
Capacity prediction methods and device.
Below in conjunction with Figure of description, preferred embodiment of the present invention will be described, it should be understood that described herein
Preferred embodiment only for the purpose of illustrating and explaining the present invention and is not intended to limit the present invention, and in the absence of conflict, this hair
The feature in embodiment and embodiment in bright can be combined with each other.
In the embodiment of the present invention, if any cell that user includes with region to be predicted in default measurement period is established
RRC connection is at least once, then it is assumed that the user is located in region to be predicted;If the user establishes RRC with any cell and connect
Afterwards, data transmission service occurs and regards the user then as the number biography user in region to be predicted.
As shown in Figure 1, it shows for the implementing procedure of mobile communications network capacity prediction methods provided in an embodiment of the present invention
It is intended to:
S11, the data statistics of cell and/or user in region to be predicted are obtained in default measurement period.
In order to carry out the prediction of network capacity, it is necessary first to carry out necessary data acquisition.It, can be in the embodiment of the present invention
Obtain the data in certain measurement period, wherein measurement period duration can be configured according to actual needs, such as measurement period
It can be, but not limited to be set as half a year or three months or one month.
The data for needing to acquire can be, but not limited to include cell-level data or user-level data.Wherein, cell parameter
Information refers to the relevant data of cell, as cell number passes flow information, cell radio resource utilization rate information, user parameter information
Refer to the relevant parameter information of user, such as cell average user quantity information, community user ARPU (Average Revenue
Per User, every user's average income) value;The data of acquisition can also include regional planning number of users information to be predicted etc..
When it is implemented, above-mentioned data can also be the average data in a period of time.As shown in table 1, believe for the data statistics of acquisition
Cease a kind of possible format, the data obtained in table 1 include wireless resource utility efficiency, average user quantity and number is spreaded per capita
Amount.
Table 1
Cell name | Wireless resource utility efficiency (%) | Average user quantity (a) | The number amount of spreading (MB/ people) per capita |
Cell 1 | 29.56 | 31.9 | 12.75 |
Cell 2 | 9.86 | 5.18 | 12.74 |
Cell 3 | 8.96 | 3.13 | 46.19 |
Cell 4 | 16.32 | 22.7 | 28.43 |
Cell 5 | 20.94 | 21.33 | 18.9 |
Cell 6 | 61.3 | 70.03 | 14.77 |
Cell 7 | 51.62 | 58.86 | 10.59 |
Cell 8 | 8.48 | 4.25 | 35.96 |
Cell 9 | 76.7 | 99.38 | 4.59 |
Cell 10 | 84.64 | 137.52 | 9.33 |
S12, the data statistics according to acquisition, using preset clustering algorithm treat cell in estimation range or
If user is clustered to obtain Ganlei's subset.
Treating the purpose that each cell that estimation range includes is classified is to carry out the cell of different flow model of growth
It distinguishes and classifies, to carry out network capacity prediction respectively, reduce prediction error.
The method of cluster has very much, for example, division methods, hierarchical method, the method based on density, the method based on grid,
Method etc. based on model, the cluster result that different clustering methods obtains are also different.In view of needing to carry out to numerous cells
Classification, the index of classification is multiple dimensions again, it is therefore advantageous to, it can choose K mean cluster algorithm in the embodiment of the present invention
(K-means algorithm).Certainly, when it is implemented, can choose other clustering methods, the embodiment of the present invention does not do this any yet
It limits.
K mean cluster algorithm is according to initially specified mass center, by the mistake of the Euclidean distance of data point to nearest mass center
Poor quadratic sum judgement, continuous iteration updates mass center, until mass center is stablized.When it is implemented, can will be located at according to the value of K
Cell or user in region to be predicted are divided into K class.After the completion of cluster, it can use Cluster Assessment function and cluster tied
Fruit is evaluated, it is assumed that is optimal when K=6.
After number of clusters has been determined, it is also necessary to cluster foundation, i.e. cluster feature value are determined, when it is implemented, can basis
Wireless resource utility efficiency, counts the amount of spreading, cell resident user ARPU (every user's average income) per capita at average user quantity per capita
If at least one parameter in the parameters such as value classifies cell or user to obtain Ganlei's subset, if including in each subset
Dry cell or several users.For ease of description, it is described for being clustered to cell in the embodiment of the present invention.
Wherein, cell resident user refers to: any user is directed to, if the user in the second measurement period, produces in any cell
When the ratio for the total flow that raw total flow and the user generates in the second measurement period is more than or equal to preset value, the use is determined
Family is the resident user of the cell.Here preset value can be arbitrarily arranged, for example, can be set to 30%.Such as Mr. Yu is used
For family, in one month measurement period, the total flow generated is 300M, is 150M, In in the flow that cell A is generated
The flow that cell B is generated is 100M, is 20M in the flow that cell C is generated, is 30M in the flow that cell D is generated, if in advance
If value is set as 30%, then the user is the resident user of cell A and cell B.
When it is implemented, certain user may be the resident user of multiple cells simultaneously as above described in example, calculated specifically
Journey summarizes, and ARPU value of the user in default measurement period will be embodied directly in respectively in its corresponding all resident cell,
To characterize the user to the flow demand of these cells.
If S13, the cluster feature value for determining every a kind of subset according to Ganlei's subset that cluster obtains.
In step s 12, it is further mentioned according to cluster result according to certain cluster according to after being clustered to cell
Take the cluster feature value of homogeneous cells.Wherein, cluster feature value is to be obtained according to the data statistics of acquisition, for example, cluster
Characteristic value can be the data statistics of acquisition itself, or certain fortune is carried out according to the data statistics of acquisition
It obtains.
Preferably, in the embodiment of the present invention by taking cluster feature value is average user quantity and counts the amount of spreading per capita as an example.Its
In, it, can be direct if in the data statistics obtained including average user quantity according to the data statistics of acquisition
Determine that it is cluster feature value;Or if if obtain data statistics in comprising per capita number the amounts of spreading, can be straight
It connects and determines that it is cluster feature value;It is also possible to number of users is passed comprising the number amount of spreading sum number in the data statistics obtained,
Then it is determined as the ratio that the number amount of spreading and number pass number of users to count the amount of spreading per capita.
S14, the every class subset obtained for cluster, the network of such subset is predicted according to the cluster feature value of such subset
Capacity.
S15, the network capacity for determining region to be predicted are the sum of network capacity of every class subset in region to be predicted.
Specifically, the network capacity in region to be predicted can directly sum for the network capacity of every class subset, alternatively, can also
With the accounting according to the network capacity of class subset every in current statistical data in region to be predicted, sum in proportion.Preferably,
It is obtained when it is implemented, the scale factor of every class subset can use historical data training.
When it is implemented, network capacity can be number of users, network flow and demand number of carrier wave etc..It ties below
Specific embodiment is closed to be illustrated the implementation process of step S14.
The first embodiment, network capacity are network flow
When predicting network capacity, the data statistics for needing to obtain can pass flow information sum number for number and pass number of users
Information directly acquires the number amount of spreading per capita;Cluster feature value can be people's mean amount of spreading;If the data statistics obtained
Information is that number passes flow information sum number biography number of users information, then according to the information of acquisition, cluster feature value can be determined for number
The amount of spreading and number pass the ratio of number of users.
Based on this, can implement according to following below scheme in step S14:
Step 1: passing flow information for every class subset that cluster obtains according to the number per capita of such subset and determining such
The number per capita of subset passes traffic prediction value.
Wherein, the amount of spreading of number per capita of every class subset is the mean value of the amount of spreading of number per capita for the cell that such subset includes.
Specifically, for every class subset that cluster obtains, the different time sections for including in the default measurement period according to such subset
Number per capita pass flow informations, be fitted the number amount of the spreading prediction curves per capita of such subset, and obtain such subset per capita
Number passes traffic prediction value.As shown in table 2, it is that every class that cluster obtains is small that first time period-third time hop counts, which pass average flow rate,
Area is fitted according to the data of table 2 using Multiple Non Linear Regression function in the actual average number amount of spreading in different time periods
Prediction obtains the number amount of the spreading prediction curve per capita of every class cell, and the number per capita for obtaining such subset passes traffic prediction value,
If the average transmission flow of the 4th period in table 2 is to pass average flow rate prediction according to first time period-third time hop counts
It obtains.
Table 2
It should be noted that being fitted for the 4th period according to first time period-third period average amount of spreading
When number passes flow curve, other fitting functions can also be used, if exponential function is fitted, linear function fit etc., the present invention
Embodiment does not limit this.
Step 2: determining the number of users predicted value and the number amount of spreading per capita that the network flow of such subset is such subset
The product of predicted value.
Wherein, the every class subset obtained for cluster, can determine number of users predicted value according to following either type:
Mode one is predicted based on user's development
Specifically, may comprise steps of:
Step 1: determining average user quantity and region to be predicted in such subset for every class subset that cluster obtains
The ratio of interior average user quantity.
Wherein, the average user quantity of every class subset is the sum of the cell average user quantity that such subset includes.
Step 2: determining the prediction number of users of every class subset: W*L*I according to following formula, in which: W is such subset
The ratio of average user quantity in middle average user quantity and the region to be predicted, i.e. the number of users of certain class cell account for
The ratio of the number of users of estimation range;L is the planning number of users in preset region to be predicted;I is preset activity factor.
As shown in table 3, the number of users prediction result to treat the every class cell obtained after estimation range is clustered:
Table 3
Mode two is predicted based on historical data
Specifically, the every class subset of every class is obtained for cluster, the difference for including in default measurement period according to such subset
The average user quantity of period is fitted the average user quantity prediction curve of such subset.When it is implemented, can use more
First nonlinear solshing is fitted.It should be noted that when it is implemented, other fitting functions, such as finger can also be used
Number Function Fitting, linear function fit etc., it is not limited in the embodiment of the present invention.
As shown in table 4, first time period-third period average user quantity is each cell in reality in different time periods
Average user quantity, the 4th period average user quantity (mode one) are using aforesaid way one according to first time period-the
Three period average user quantity are the 4th period that each cell is predicted in actual average number of users in different time periods
Average user quantity, the 4th period average user quantity (mode two) are using aforesaid way two according to first time period-the
Three period average user quantity are the 4th period that each cell is predicted in actual average number of users in different time periods
Average user quantity:
Table 4
According to the data of table 4, prediction is fitted using Multiple Non Linear Regression function, obtains the average use of every class cell
Amount amount prediction curve obtains the number of users predicted value of such subset.
Second of embodiment, network capacity are number of users
When predicting number of users, the data statistics for needing to obtain can pass flow information sum number for number and pass number of users
Information directly acquires the number amount of spreading per capita;Cluster feature value can be people's mean amount of spreading;If the data statistics obtained
Information is that number passes flow information sum number biography number of users information, then according to the information of acquisition, cluster feature value can be determined for number
The amount of spreading and number pass the ratio of number of users.
Based on this, can implement according to following below scheme in step S14:
Step 1: passing flow information for every class subset that cluster obtains according to the number per capita of such subset and determining such
The number per capita of subset passes traffic prediction value.
Wherein, the amount of spreading of number per capita of every class subset is the mean value of the amount of spreading of number per capita for the cell that such subset includes.
It in the default measurement period include not according to such subset specifically, for the obtained every class subset of cluster
Number per capita with the period passes flow information, and the amount of the spreading prediction curve of number per capita for being fitted such subset obtains such subset
Number passes traffic prediction value per capita.
The embodiment that may refer to that number biography traffic prediction value per capita is determined in the first embodiment is embodied in it, here
It repeats no more.
Step 2: determining the predicting network flow value and the number amount of spreading per capita that the number of users of such subset is such subset
The ratio of predicted value.
Wherein, the every class subset obtained for cluster, its predicting network flow value can be determined according to following either type:
Mode one,
Step 1: determining the ratio of the network flow in such subset network flow and region to be predicted.
When it is implemented, the sum of web-based history flow of each cell that can include according to such subset determines such son
Collect network flow, and the sum of network flow of all kinds of cells for including according to region to be predicted determines the network flow in region to be predicted
Amount.
Step 2: determining the predicting network flow value of every class subset according to following formula:
Wherein: α is the ratio of the network flow in such subset network flow and the region to be predicted;For it is preset,
The planning network flow in the region to be predicted.
Mode two, for the obtained every class subset of cluster, in the default measurement period include not according to such subset
With the network traffic information of period, it is fitted the predicting network flow curve of such subset, obtains the network flow of such subset
Predicted value.
The third embodiment, network capacity are number of carrier wave predicted value
Based on the first embodiment and second of embodiment, in the number of users or net for predicting every a kind of subset
After network flow, (see that the number of users of carrying or single carrier see the network flow of carrying including single carrier according to single-carrier capacity
Amount), determine the number of carrier wave that such subset needs.Specifically, determining such subset for every a kind of subset that cluster obtains
Number of carrier wave predicted value is the ratio that number of users predicted value and single carrier can carry number of users;Or determine such subset
Number of carrier wave predicted value be predicting network flow value and single carrier can bearer network flow ratio.
It should be noted that after determining the number of carrier wave predicted value of every class subset in the manner described above, specific implementation
When, the number of carrier wave of every class subset is directly summed or number of carrier wave that available region to be predicted of summing in proportion is total needs
Seek predicted value.
When it is implemented, when determining the total number of carrier wave requirement forecasting value in region to be predicted first can also be being utilized
Kind embodiment is predicted to obtain the network flow in region to be predicted or is predicted to obtain region to be predicted using second embodiment
Number of users on the basis of, determine the number of users in region to be predicted and single carrier can carry number of users ratio be to pre-
Survey the total number of carrier wave requirement forecasting value in region;Or determine that the network flow in region to be predicted and single carrier can bearer network streams
The ratio of amount is the total number of carrier wave requirement forecasting value in region to be predicted.
As shown in Fig. 2, being treated pre- for the schematic illustration of mobile communications network capacity provided in an embodiment of the present invention prediction
It surveys each cell that region includes to classify, for every a kind of cell that classification obtains, predicts such cell respectively per capita
The number amount of spreading, and the accounting prediction for combining the average user quantity of every class cell to pass in total number of users amount in number of regions to be predicted is every
The average user quantity of class cell, and then determine the network flow of every class cell and the network total flow in region to be predicted.
Mobile communications network capacity prediction methods provided in an embodiment of the present invention, first by region to be predicted include it is each small
Area classifies, and every one kind cell has common characteristic value, and predicts its network capacity for every a kind of cell respectively, in turn
The network total capacity in region to be predicted is obtained, it is pre- due to no longer carrying out network capacity for entire region to be predicted in the above process
It surveys, but is predicted respectively according to every class cell that region to be predicted includes, due to the network capacity difference in change of homogeneous cells
It is different relatively small, so as to reduce the error of network capacity prediction result.
Based on the same inventive concept, a kind of mobile communications network capacity prediction meanss are additionally provided in the embodiment of the present invention,
Since the principle that above-mentioned apparatus solves the problems, such as is similar to mobile communications network capacity prediction methods, the implementation of above-mentioned apparatus can
With referring to the implementation of method, overlaps will not be repeated.
As shown in figure 3, being mobile communications network capacity prediction meanss structural schematic diagram provided in an embodiment of the present invention, packet
It includes:
Acquiring unit 31 is located at cell and/or the data of user in region to be predicted for obtaining in default measurement period
Statistical information;
Cluster cell 32, for the data statistics according to acquisition, using preset clustering algorithm to described to be predicted
If the cell or user in region are clustered to obtain Ganlei's subset;
Determination unit 33, if Ganlei's subset for being obtained according to cluster determines the cluster feature value of every a kind of subset,
Described in cluster feature value be to be obtained according to the data statistics of acquisition;
Predicting unit 34, every class subset for obtaining for cluster should according to the prediction of the cluster feature value of such subset
The network capacity of class subset;And determine that the network capacity in the region to be predicted is the net of every class subset in the region to be predicted
The sum of network capacity.
When it is implemented, network capacity includes network flow;The data statistics include that number passes flow information sum number
It passes number of users information or the data statistics includes counting the amount of spreading per capita;And the cluster feature value is people's mean biography
Flow.
Based on this, predicting unit is passed specifically for the every class subset obtained for cluster according to the number per capita of such subset
Flow information determines that the number per capita of such subset passes traffic prediction value;The network flow for determining such subset is the use of such subset
Amount amount predicted value passes the product of traffic prediction value with number per capita.
Wherein, predicting unit, specifically for the every class subset obtained for cluster, according to such subset in the default system
The number in different time periods that the meter period includes passes number of users, and the number for being fitted such subset passes number of users prediction curve, obtains
The number of users predicted value of such subset.Alternatively, predicting unit, specifically for the every class subset obtained for cluster, determining should
Number passes the ratio that the number in number of users and the region to be predicted passes number of users in class subset;It is determined according to following formula every
The prediction number of users of class subset: W*L*I, in which: W is that number passes in number of users and the region to be predicted in such subset
Number passes the ratio of number of users;L is the planning number of users in preset, the described region to be predicted;I is preset activity factor.
When it is implemented, predicting unit, for every class subset that cluster obtains, according to such subset in the default statistics
The number per capita in different time periods that period includes passes flow information, is fitted the number amount of the spreading prediction curve per capita of such subset, and
The number per capita for obtaining such subset passes traffic prediction value.
Preferably, when it is implemented, network capacity can also include number of users;The data statistics include that number passes
Flow information sum number passes number of users information or the data statistics include counting the amount of spreading per capita;And the cluster feature
Value is people's mean amount of spreading.
Based on this, predicting unit is passed specifically for the every class subset obtained for cluster according to the number per capita of such subset
Flow information determines that the number per capita of such subset passes traffic prediction value;The number of users for determining such subset is the net of such subset
Network traffic prediction value passes the ratio of traffic prediction value with number per capita.
Wherein, predicting unit, specifically for the every class subset obtained for cluster, according to such subset in the default system
The network traffic information in different time periods that the meter period includes, is fitted the predicting network flow curve of such subset, obtains such
The predicting network flow value of subset.Or predicting unit determines such subset specifically for the every class subset obtained for cluster
The ratio of network flow and the network flow in the region to be predicted;Determine that the network flow of every class subset is pre- according to following formula
Measured value:Wherein: α is the ratio of the network flow in such subset network flow and the region to be predicted;It is default
, the planning network flow in the region to be predicted.
When it is implemented, predicting unit, specifically for the every class subset obtained for cluster, according to such subset described
The number per capita in different time periods that default measurement period includes passes flow information, and the number per capita for being fitted such subset passes volume forecasting
Curve, and the number per capita for obtaining such subset passes traffic prediction value.
When it is implemented, network capacity further includes number of carrier wave predicted value: and predicting unit, it is specifically used for for cluster
Obtained every class subset determines that the number of carrier wave predicted value of such subset is the number of users predicted value and single carrier of such subset
The ratio of number of users can be carried;Or determine that the number of carrier wave predicted value of such subset is the predicting network flow of such subset
Value and single carrier can bearer network flow ratio.
For convenience of description, above each section is divided by function describes respectively for each module (or unit).Certainly, In
Implement to realize the function of each module (or unit) in same or multiple softwares or hardware when the present invention.
It should be understood by those skilled in the art that, the embodiment of the present invention can provide as method, system or computer program
Product.Therefore, complete hardware embodiment, complete software embodiment or reality combining software and hardware aspects can be used in the present invention
Apply the form of example.Moreover, it wherein includes the computer of computer usable program code that the present invention, which can be used in one or more,
The computer program implemented in usable storage medium (including but not limited to magnetic disk storage, CD-ROM, optical memory etc.) produces
The form of product.
The present invention be referring to according to the method for the embodiment of the present invention, the process of equipment (system) and computer program product
Figure and/or block diagram describe.It should be understood that every one stream in flowchart and/or the block diagram can be realized by computer program instructions
The combination of process and/or box in journey and/or box and flowchart and/or the block diagram.It can provide these computer programs
Instruct the processor of general purpose computer, special purpose computer, Embedded Processor or other programmable data processing devices to produce
A raw machine, so that being generated by the instruction that computer or the processor of other programmable data processing devices execute for real
The device for the function of being specified in present one or more flows of the flowchart and/or one or more blocks of the block diagram.
These computer program instructions, which may also be stored in, is able to guide computer or other programmable data processing devices with spy
Determine in the computer-readable memory that mode works, so that it includes referring to that instruction stored in the computer readable memory, which generates,
Enable the manufacture of device, the command device realize in one box of one or more flows of the flowchart and/or block diagram or
The function of being specified in multiple boxes.
These computer program instructions also can be loaded onto a computer or other programmable data processing device, so that counting
Series of operation steps are executed on calculation machine or other programmable devices to generate computer implemented processing, thus in computer or
The instruction executed on other programmable devices is provided for realizing in one or more flows of the flowchart and/or block diagram one
The step of function of being specified in a box or multiple boxes.
Although preferred embodiments of the present invention have been described, it is created once a person skilled in the art knows basic
Property concept, then additional changes and modifications can be made to these embodiments.So it includes excellent that the following claims are intended to be interpreted as
It selects embodiment and falls into all change and modification of the scope of the invention.
Obviously, various changes and modifications can be made to the invention without departing from essence of the invention by those skilled in the art
Mind and range.In this way, if these modifications and changes of the present invention belongs to the range of the claims in the present invention and its equivalent technologies
Within, then the present invention is also intended to include these modifications and variations.
Claims (20)
1. a kind of mobile communications network capacity prediction methods characterized by comprising
It obtains in default measurement period and is located at the data statistics of cell and/or user in region to be predicted;
According to the data statistics of acquisition, using preset clustering algorithm in the region to be predicted cell or user
If being clustered to obtain Ganlei's subset;
If the cluster feature value of every a kind of subset is determined according to Ganlei's subset that cluster obtains, wherein the cluster feature value is root
It is obtained according to the data statistics of acquisition;
For every class subset that cluster obtains, the network capacity of such subset, institute are predicted according to the cluster feature value of such subset
Stating network capacity includes network flow or number of users;The data statistics include that number passes flow information sum number biography user
Quantity information or the data statistics include counting the amount of spreading per capita;And the cluster feature value is people's mean amount of spreading;
The network capacity for determining the region to be predicted is the sum of the network capacity of every class subset in the region to be predicted, specifically
It include: that the network capacity in the region to be predicted is directly summed for the network capacity of every class subset, alternatively, according to every class subset
Accounting of the network capacity in the region to be predicted, sums in proportion.
2. the method as described in claim 1, which is characterized in that for every class subset that cluster obtains, according to such subset
Cluster feature value predicts the network capacity of such subset, specifically includes:
For every class subset that cluster obtains, flow information is passed according to the number per capita of such subset and determines counting per capita for such subset
Pass traffic prediction value;
The network flow for determining such subset is that the number of users predicted value of such subset and number per capita pass multiplying for traffic prediction value
Product.
3. method according to claim 2, which is characterized in that for every class subset that cluster obtains, in accordance with the following methods really
The number of users predicted value of such fixed subset:
For every class subset that cluster obtains, the number in different time periods for including in the default measurement period according to such subset
Number of users is passed, the number for being fitted such subset passes number of users prediction curve, obtains the number of users predicted value of such subset.
4. method according to claim 2, which is characterized in that for every class subset that cluster obtains, in accordance with the following methods really
The number of users predicted value of such fixed subset:
Determine that number passes the ratio that number of users passes number of users with the number in the region to be predicted in such subset;
The prediction number of users of every class subset: W*L*I is determined according to following formula, in which:
W is the ratio that the number that number passes in number of users and the region to be predicted in such subset passes number of users;
L is the planning number of users in preset, the described region to be predicted;
I is preset activity factor.
5. method according to claim 2, which is characterized in that for every class subset that cluster obtains, according to such subset
Number passes flow information and determines that the number per capita of such subset passes traffic prediction value per capita, specifically includes:
For every class subset that cluster obtains, the people in different time periods for including in the default measurement period according to such subset
Mean passes flow information, is fitted the number amount of the spreading prediction curve per capita of such subset, and the number per capita for obtaining such subset passes
Traffic prediction value.
6. the method as described in claim 1, which is characterized in that for every class subset that cluster obtains, according to such subset
Cluster feature value predicts the network capacity of such subset, specifically includes:
For every class subset that cluster obtains, flow information is passed according to the number per capita of such subset and determines counting per capita for such subset
Pass traffic prediction value;
The number of users for determining such subset is the predicting network flow value of such subset and number passes the ratio of traffic prediction value per capita
Value.
7. method as claimed in claim 6, which is characterized in that for every class subset that cluster obtains, in accordance with the following methods really
The predicting network flow value of such fixed subset:
For every class subset that cluster obtains, the net in different time periods for including in the default measurement period according to such subset
Network flow information is fitted the predicting network flow curve of such subset, obtains the predicting network flow value of such subset.
8. method as claimed in claim 6, which is characterized in that for every class subset that cluster obtains, in accordance with the following methods really
The predicting network flow value of such fixed subset:
Determine the ratio of the network flow in such subset network flow and the region to be predicted;
The predicting network flow value of every class subset is determined according to following formula:Wherein:
α is the ratio of the network flow in such subset network flow and the region to be predicted;
For the planning network flow in preset, the described region to be predicted.
9. method as claimed in claim 6, which is characterized in that for every class subset that cluster obtains, according to such subset
Number passes flow information and determines that the number per capita of such subset passes traffic prediction value per capita, specifically includes:
For every class subset that cluster obtains, the people in different time periods for including in the default measurement period according to such subset
Mean passes flow information, is fitted the number amount of the spreading prediction curve per capita of such subset, and the number per capita for obtaining such subset passes
Traffic prediction value.
10. the method as described in claim 2 or 6, which is characterized in that the network capacity further includes number of carrier wave predicted value:
And
For every class subset that cluster obtains, the number of carrier wave predicted value of such subset is determined in accordance with the following methods:
The number of carrier wave predicted value for determining such subset is that the number of users predicted value of such subset and single carrier can carry user
The ratio of quantity;Or
The number of carrier wave predicted value for determining such subset is that the predicting network flow value of such subset and single carrier can bearer networks
The ratio of flow.
11. a kind of mobile communications network capacity prediction meanss characterized by comprising
Acquiring unit, for obtaining the data statistics letter for being located at cell and/or user in region to be predicted in default measurement period
Breath;
Cluster cell, for the data statistics according to acquisition, using preset clustering algorithm in the region to be predicted
If cell or user clustered to obtain Ganlei's subset;
Determination unit, if Ganlei's subset for being obtained according to cluster determines the cluster feature value of every a kind of subset, wherein described
Cluster feature value is to be obtained according to the data statistics of acquisition;
Predicting unit, every class subset for obtaining for cluster, predicts such subset according to the cluster feature value of such subset
Network capacity, the network capacity includes network flow or number of users;The data statistics include the number amount of spreading
Information sum number passes number of users information or the data statistics include counting the amount of spreading per capita;And the cluster feature value is
The number amount of spreading per capita;And determine that the network capacity in the region to be predicted is the network appearance of every class subset in the region to be predicted
The sum of amount, wherein hold in the network that the network capacity for determining the region to be predicted is every class subset in the region to be predicted
When the sum of amount, predicting unit is specifically used for: the network capacity in the region to be predicted is that the network capacity of every class subset is directly asked
With alternatively, the accounting according to the network capacity of every class subset in the region to be predicted, sums in proportion.
12. device as claimed in claim 11, which is characterized in that
The predicting unit, specifically for the every class subset obtained for cluster, according to the amount of the spreading letter of number per capita of such subset
It ceases and determines that the number per capita of such subset passes traffic prediction value;The network flow for determining such subset is the number of users of such subset
Predicted value passes the product of traffic prediction value with number per capita.
13. device as claimed in claim 12, which is characterized in that
The predicting unit, specifically for the every class subset obtained for cluster, according to such subset in the default statistics week
The number in different time periods that phase includes passes number of users, and the number for being fitted such subset passes number of users prediction curve, obtains such
The number of users predicted value of subset.
14. device as claimed in claim 12, which is characterized in that
The predicting unit, specifically for for the obtained every class subset of cluster, determine in such subset number pass number of users with
Number in the region to be predicted passes the ratio of number of users;The prediction number of users of every class subset is determined according to following formula:
W*L*I, in which: W is the ratio that the number that number passes in number of users and the region to be predicted in such subset passes number of users;L
For the planning number of users in preset, the described region to be predicted;I is preset activity factor.
15. device as claimed in claim 12, which is characterized in that
The predicting unit includes in the default measurement period according to such subset for the obtained every class subset of cluster
Number per capita in different time periods passes flow information, is fitted the number amount of the spreading prediction curve per capita of such subset, and obtains such son
The number per capita of collection passes traffic prediction value.
16. device as claimed in claim 11, which is characterized in that
The predicting unit, specifically for the every class subset obtained for cluster, according to the amount of the spreading letter of number per capita of such subset
It ceases and determines that the number per capita of such subset passes traffic prediction value;The number of users for determining such subset is the network flow of such subset
Predicted value passes the ratio of traffic prediction value with number per capita.
17. device as claimed in claim 16, which is characterized in that
The predicting unit, specifically for the every class subset obtained for cluster, according to such subset in the default statistics week
The network traffic information in different time periods that phase includes is fitted the predicting network flow curve of such subset, obtains such subset
Predicting network flow value.
18. device as claimed in claim 16, which is characterized in that
The predicting unit, specifically for for the obtained every class subset of cluster, determine such subset network flow and it is described to
The ratio of the network flow of estimation range;The predicting network flow value of every class subset is determined according to following formula:Wherein:
α is the ratio of the network flow in such subset network flow and the region to be predicted;For preset, the described region to be predicted
Planning network flow.
19. device as claimed in claim 16, which is characterized in that
The predicting unit, specifically for the every class subset obtained for cluster, according to such subset in the default statistics week
The number per capita in different time periods that phase includes passes flow information, is fitted the number amount of the spreading prediction curve per capita of such subset, and obtains
The number per capita for obtaining such subset passes traffic prediction value.
20. the device as described in claim 12 or 16, which is characterized in that the network capacity further includes number of carrier wave prediction
Value: and
The predicting unit determines the number of carrier wave predicted value of such subset specifically for the every class subset obtained for cluster
The ratio of number of users can be carried for the number of users predicted value and single carrier of such subset;Or determine the carrier wave of such subset
Quantitative forecast value be such subset predicting network flow value and single carrier can bearer network flow ratio.
Priority Applications (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201510294026.6A CN106301984B (en) | 2015-06-01 | 2015-06-01 | A kind of mobile communications network capacity prediction methods and device |
Applications Claiming Priority (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201510294026.6A CN106301984B (en) | 2015-06-01 | 2015-06-01 | A kind of mobile communications network capacity prediction methods and device |
Publications (2)
Publication Number | Publication Date |
---|---|
CN106301984A CN106301984A (en) | 2017-01-04 |
CN106301984B true CN106301984B (en) | 2019-11-15 |
Family
ID=57656054
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
CN201510294026.6A Active CN106301984B (en) | 2015-06-01 | 2015-06-01 | A kind of mobile communications network capacity prediction methods and device |
Country Status (1)
Country | Link |
---|---|
CN (1) | CN106301984B (en) |
Families Citing this family (14)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN107566207A (en) * | 2017-08-07 | 2018-01-09 | 北京天元创新科技有限公司 | The appraisal procedure and device of a kind of network resource consumption |
CN109981392B (en) * | 2017-12-28 | 2020-12-22 | ***通信集团公司 | Method and device for predicting flow speed, electronic equipment and storage medium |
CN110493796B (en) * | 2018-05-15 | 2022-12-02 | ***通信集团湖北有限公司 | LTE network capacity estimation method, device, equipment and medium |
CN111132208A (en) * | 2018-10-30 | 2020-05-08 | 上海华为技术有限公司 | Capacity prediction method and capacity prediction device |
CN111225392B (en) * | 2018-11-27 | 2023-12-15 | ***通信集团辽宁有限公司 | Cell load balancing method, device, equipment and computer storage medium |
CN111294812B (en) * | 2018-12-10 | 2022-12-13 | 中兴通讯股份有限公司 | Resource capacity-expansion planning method and system |
CN109495318B (en) * | 2018-12-17 | 2021-11-16 | 宜通世纪科技股份有限公司 | Mobile communication network flow prediction method, device and readable storage medium |
CN110213714B (en) * | 2019-05-10 | 2020-08-14 | 中国联合网络通信集团有限公司 | Terminal positioning method and device |
CN112202585B (en) * | 2019-07-08 | 2022-07-01 | ***通信集团浙江有限公司 | Wireless network flow prediction method and device and electronic equipment |
CN110572699A (en) * | 2019-10-09 | 2019-12-13 | 北京邮电大学 | Network-assisted video streaming media transmission optimization method based on multi-cell cluster |
CN112654063B (en) * | 2020-12-15 | 2022-09-02 | 中国联合网络通信集团有限公司 | Uplink capacity assessment method and device |
CN112688821B (en) * | 2021-01-28 | 2023-04-18 | 中国铁塔股份有限公司 | Network capacity prediction method and device and electronic equipment |
CN112929907B (en) * | 2021-02-04 | 2023-11-07 | 中国联合网络通信集团有限公司 | Method and device for determining antenna parameters |
CN115269670A (en) * | 2021-04-30 | 2022-11-01 | 华为技术有限公司 | Flow prediction method, device and storage medium |
Citations (4)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN101203018A (en) * | 2007-11-30 | 2008-06-18 | ***通信集团重庆有限公司 | Method for distributing business density time and business in a mobile communication |
CN103024762A (en) * | 2012-12-26 | 2013-04-03 | 北京邮电大学 | Service feature based communication service forecasting method |
CN103313307A (en) * | 2013-05-31 | 2013-09-18 | 中国联合网络通信集团有限公司 | Network resource allocation method and device |
CN103796224A (en) * | 2014-02-20 | 2014-05-14 | 大唐移动通信设备有限公司 | Base station device carrier self-adaptive deploying method and base station |
Family Cites Families (1)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
US10097329B2 (en) * | 2013-11-08 | 2018-10-09 | Spidercloud Wireless, Inc. | Fractional frequency reuse schemes assigned to radio nodes in an LTE network |
-
2015
- 2015-06-01 CN CN201510294026.6A patent/CN106301984B/en active Active
Patent Citations (4)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN101203018A (en) * | 2007-11-30 | 2008-06-18 | ***通信集团重庆有限公司 | Method for distributing business density time and business in a mobile communication |
CN103024762A (en) * | 2012-12-26 | 2013-04-03 | 北京邮电大学 | Service feature based communication service forecasting method |
CN103313307A (en) * | 2013-05-31 | 2013-09-18 | 中国联合网络通信集团有限公司 | Network resource allocation method and device |
CN103796224A (en) * | 2014-02-20 | 2014-05-14 | 大唐移动通信设备有限公司 | Base station device carrier self-adaptive deploying method and base station |
Also Published As
Publication number | Publication date |
---|---|
CN106301984A (en) | 2017-01-04 |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
CN106301984B (en) | A kind of mobile communications network capacity prediction methods and device | |
CN105637925B (en) | Dynamic frequency and power resource allocation with refinement tactical management | |
CN105682105B (en) | A kind of method and apparatus for planning website evaluation | |
CN102098684B (en) | System and method for allocating cross-layer resources in cognitive radio network | |
CN107171848A (en) | A kind of method for predicting and device | |
CN106454857B (en) | A kind of appraisal procedure and device of the network planning | |
CN103634807B (en) | WIFI data hotspot cell data monitoring method and WLAN deployment ordering method and device | |
CN104980950A (en) | Network optimization server, mobile device of realizing network optimization and system of realizing network optimization | |
Cano et al. | On optimal infrastructure sharing strategies in mobile radio networks | |
CN108243435B (en) | Parameter optimization method and device in LTE cell scene division | |
CN104901827B (en) | A kind of network resource evaluation method and device based on customer service structure | |
Yu et al. | Spectrum investment under uncertainty: A behavioral economics perspective | |
CN105873087B (en) | A kind of network index prediction technique, device and electronic equipment | |
Bakri et al. | Data-driven RAN slicing mechanisms for 5G and beyond | |
CN106934514A (en) | The generation method and device of a kind of regression model | |
CN108322915B (en) | Information processing method and device for base station planning | |
CN107864478A (en) | The release plan optimization method of data-driven | |
CN107147514A (en) | A kind of powerline network is optimized allocation of resources method and system | |
Gemici et al. | Wealth inequality and the price of anarchy | |
CN114066184A (en) | Area coverage assessment method and device and electronic equipment | |
Chang et al. | Edge server placement for vehicular ad hoc networks in metropolitans | |
Shao et al. | A learning based framework for MEC server planning with uncertain BSs demands | |
CN103796218A (en) | Wireless access point site selection method and apparatus | |
CN106714223A (en) | Method and device for building base station energy consumption model | |
CN108271181A (en) | A kind of PCI optimization method and devices of wireless network |
Legal Events
Date | Code | Title | Description |
---|---|---|---|
C06 | Publication | ||
PB01 | Publication | ||
C10 | Entry into substantive examination | ||
SE01 | Entry into force of request for substantive examination | ||
GR01 | Patent grant | ||
GR01 | Patent grant |