CN109586836A - A kind of user's cluster algorithm based on artificial intelligence - Google Patents
A kind of user's cluster algorithm based on artificial intelligence Download PDFInfo
- Publication number
- CN109586836A CN109586836A CN201811401988.7A CN201811401988A CN109586836A CN 109586836 A CN109586836 A CN 109586836A CN 201811401988 A CN201811401988 A CN 201811401988A CN 109586836 A CN109586836 A CN 109586836A
- Authority
- CN
- China
- Prior art keywords
- bosk
- wab
- group
- value
- interference
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Granted
Links
Classifications
-
- H—ELECTRICITY
- H04—ELECTRIC COMMUNICATION TECHNIQUE
- H04J—MULTIPLEX COMMUNICATION
- H04J11/00—Orthogonal multiplex systems, e.g. using WALSH codes
- H04J11/0023—Interference mitigation or co-ordination
- H04J11/005—Interference mitigation or co-ordination of intercell interference
-
- H—ELECTRICITY
- H04—ELECTRIC COMMUNICATION TECHNIQUE
- H04J—MULTIPLEX COMMUNICATION
- H04J11/00—Orthogonal multiplex systems, e.g. using WALSH codes
- H04J11/0023—Interference mitigation or co-ordination
- H04J11/0026—Interference mitigation or co-ordination of multi-user interference
-
- H—ELECTRICITY
- H04—ELECTRIC COMMUNICATION TECHNIQUE
- H04J—MULTIPLEX COMMUNICATION
- H04J11/00—Orthogonal multiplex systems, e.g. using WALSH codes
- H04J11/0023—Interference mitigation or co-ordination
- H04J11/005—Interference mitigation or co-ordination of intercell interference
- H04J11/0053—Interference mitigation or co-ordination of intercell interference using co-ordinated multipoint transmission/reception
-
- H—ELECTRICITY
- H04—ELECTRIC COMMUNICATION TECHNIQUE
- H04W—WIRELESS COMMUNICATION NETWORKS
- H04W16/00—Network planning, e.g. coverage or traffic planning tools; Network deployment, e.g. resource partitioning or cells structures
- H04W16/18—Network planning tools
- H04W16/20—Network planning tools for indoor coverage or short range network deployment
Abstract
The invention discloses a kind of user's cluster algorithm based on artificial intelligence first has to the numerical procedure of design cell weight, a kind of secondary cell cluster algorithm for proposing customer-centric on this basis to improve system velocity as target.Compared with prior art, the present invention improves the handling capacity of edge customer, also has improvement to average system throughput.Method provided by the invention can be used for cell sub-clustering, can be used for user's sub-clustering, has the value of popularization and application.
Description
Technical field
The present invention relates to mobile communication technology field more particularly to a kind of user's cluster algorithms based on artificial intelligence.
Background technique
In order to adapt to the explosive growth of the following magnanimity mobile data, accelerate the exploitation of new business new opplication, in global industry
Under the promotion energetically on boundary, the 5th third-generation mobile communication (5G) network comes into being.Have and researchs and analyses the data industry for showing future communications
Most of business is to occur in outdoor hot spot region and indoor office region, and user distribution and business demand have significantly
Regional feature.In this market context, industry introduces the concept of super-intensive on-premise network, will be a large amount of different types of low
Power minimizes base station deployment under the coverage area of macro base station, so that traditional macrocellular network isomerization.
Low-power miniaturization base-station node have the features such as coverage area is small, transmission power is low, between user at a distance from it is close, this
Sample can be effectively reduced path loss, enhancing useful signal, and minimize advantage it is also possible that the deployment of network more just
It is prompt.
Super-intensive networking is exactly to be shortened distance between sites by the wireless network deployment of more densification, keeps its website close
Degree greatly increases, to improve the network capacity of spectral multiplexing rate, user experience rate and unit area.However, super-intensive group
Net also brings new problem while hoist capacity.It is dry in super-intensive networking scene compared to traditional cellular network
It disturbs situation to become more complicated, not only there is the same layer interference between original macro base station, and it is small with low-power to introduce macro base station
Same layer interference between cross-layer interference between type base station and low-power miniaturization base station.Thus more cell edges can be generated
User, these Cell Edge User incite somebody to action therefore performance degradation.
Summary of the invention
The object of the invention is that providing a kind of user's sub-clustering calculation based on artificial intelligence to solve the above-mentioned problems
Method.
The present invention through the following technical solutions to achieve the above objectives:
The present invention the following steps are included:
(1) regard each node as independent tree, and all nodes in network in any one node set are asked
Take the interference weight Wab between node two-by-two；
(2) by Wab by being ranked up to obtain Wab_order from small to large；
(3) first value in Wab_order, i.e., the minimum value in all Wab are obtained, corresponding node serial number is
(i, j) unites i and j for first bosk；
(4) since second of Wab_order value, if the corresponding node serial number of second value be (i, m) or (m,
J), i.e. then one group of m self-contained；If the corresponding node serial number of second value is (n, m), n and m small gloomy as second
Woods；
(5) since the third of Wab_order value, if two of Wab_order connection set with constitute before it is small gloomy
The tree of woods does not repeat, then it is individually for new bosk；If there is one tree repeats, then remaining another one tree Cheng little Sen alone
Woods；If all repeated, next weight is jumped to；Until having considered all weight Wab_order；Thus by all use
Family constitutes bosk, and has interference value as small as possible inside bosk；
(6) first round merges bosk: the basic thought that bosk merges is the interference minimum newly increased: in (5) step
Orderly bosk, the basis that first group is merged as first time, remaining group is interfered with first group respectively
And calculating, the method for calculating are as follows: assuming that the group merges with first group, interfere summation two-by-two in group after record merges.Consider all
The remaining group of interference merged with first group and.Interference and the smallest merging mode are retained, complete to merge for the first time,
In the remaining bosk not merged, sequentially successively merged by the method that first time merges, until all bosks are all complete
At the merging of this wheel.
(7) n-th (n >=2) wheel merging bosk: it is successively carried out on the basis of n-1 takes turns and merges according to the method for step 5
Merge, bosk scale is increasing, and number is being reduced；
(8) bosk is successively merged, the number of the cluster until reaching needs then stops.
The beneficial effects of the present invention are:
The present invention is a kind of user's cluster algorithm based on artificial intelligence, and compared with prior art, the present invention improves side
The handling capacity of edge user also has improvement to average system throughput.Method provided by the invention can be used for cell sub-clustering, can also be with
For user's sub-clustering, have the value of popularization and application.
Detailed description of the invention
Fig. 1 is MATLAB simulation result curve graph of the invention.
Specific embodiment
The present invention will be further explained below with reference to the attached drawings:
Coordinated multipoint transmission reception technique is transmission skill together using the multiple transmission point collaboration separated on geographical location
Art.By reducing inter-cell interference the progress Combined Treatment of inter-cell interference suffered by Cell Edge User or cooperative scheduling,
Enhance the communication performance of edge customer, improves data rate covering and cell edge throughput.
Coordinated multipoint transmission reception technique is applied to super-intensive networking scene and is improving network throughput and Protect edge information
Customer-side shows great potential, therefore is considered as a kind of solution inter-cell interference, protective plot edge customer communication quality
One of most efficient method.
Coordinated multipoint transmission reception technique can effectively inhibit inter-cell interference, improve edge user throughput.Cooperative cluster
It is selected as the pith of coordinated multipoint transmission technology, mainly passes through classifying rationally cooperative cluster set, is mentioned for edge customer
For more preferable and more reliable federated service collection of base stations, suitable cooperative cluster can reduce the interference of minizone, will not be ultra dense
The computation complexity and energy consumption of more signal processings are brought in collection network, but is avoided that between serious user and does
It disturbs, achievees the purpose that lifting system capacity, therefore cluster algorithm becomes particularly significant.
The present invention the following steps are included:
(1) regard each node as independent tree, and all nodes in network in any one node set are asked
Take the interference weight Wab between node two-by-two；
(2) by Wab by being ranked up to obtain Wab_order from small to large；
(3) first value in Wab_order, i.e., the minimum value in all Wab are obtained, corresponding node serial number is
(i, j) unites i and j for first bosk；
(4) since second of Wab_order value, if the corresponding node serial number of second value be (i, m) or (m,
J), i.e. then one group of m self-contained；If the corresponding node serial number of second value is (n, m), n and m small gloomy as second
Woods；
(5) since the third of Wab_order value, if two of Wab_order connection set with constitute before it is small gloomy
The tree of woods does not repeat, then it is individually for new bosk；If there is one tree repeats, then remaining another one tree Cheng little Sen alone
Woods；If all repeated, next weight is jumped to；Until having considered all weight Wab_order；Thus by all use
Family constitutes bosk, and has interference value as small as possible inside bosk；
(6) merge bosk: the basic thought that bosk merges is the interference minimum newly increased: in step 5 orderly
Bosk, such as (3,5), 4, (6,7), (1,2), 8；Based on first group (3,5), by remaining 4, (6,7), (1,2),
8 respectively with first group in node 3 and node 5, find corresponding Wab value, choose increased interference it is the smallest be incorporated to cluster (3,
5) (such as compare Wab (3,4)+Wab (5,4) and Wab (3,6)+Wab (3,7)+Wab (5,6)+Wab (5,7) and Wab (3,1)
The value of+Wab (3,2)+Wab (5,1)+Wab (5,1) and Wab (3,8)+Wab (5,8), these values the inside is the smallest, is incorporated to
First cluster；Assuming that the smallest is 4, so become (3,4,5) after the end of the step, (6,7), (1,2), 8.Thus become 4 small
Forest；
(7) it according to step (6), to next bosk merged not yet, finds and increases interference minimum newly for which
, the number of bosk is being reduced；
(8) bosk is successively merged, the number of the cluster until reaching needs then stops.
Use MATLAB for emulation tool by computer, the weight design scheme and user's cluster algorithm to proposition carry out
Simulation verifying, and be compared using K-mean cluster algorithm as with reference to algorithm.Herein in system integration project,
Cell is formed by corresponding cell Clustering Algorithm first, then carries out cell selection and multi-cell scheduling, finally to scheduling user
Carry out beam forming and power distribution.Simulation parameter specifically see the table below.
In order to ensure the fairness and authenticity of various algorithm simulating results in simulation process, in addition to point of two kinds of algorithms
Cluster algorithm is different outer, has been all made of same algorithm herein with power control algorithm in system parameter setting, power segmentation.Emulation knot
Fruit sees Fig. 1.It can be seen from the figure that on 5% position of cumulative probability Density Distribution, the handling capacity of cluster algorithm of the invention
It can be significantly better than and effectively improve the handling capacity of edge customer with reference to algorithm；Meanwhile for cumulative probability Density Distribution
On 85% position, this algorithm also has improvement, and the handling capacity of central user is also suitable with comparison algorithm.
Basic principles and main features and advantages of the present invention of the invention have been shown and described above.The technology of the industry
Personnel are it should be appreciated that the present invention is not limited to the above embodiments, and the above embodiments and description only describe this
The principle of invention, without departing from the spirit and scope of the present invention, various changes and improvements may be made to the invention, these changes
Change and improvement all fall within the protetion scope of the claimed invention.The claimed scope of the invention by appended claims and its
Equivalent thereof.
Claims (1)
1. a kind of user's cluster algorithm based on artificial intelligence, which comprises the following steps:
(1) regard each node as independent tree, and two are sought to all nodes in network in any one node set
Interference weight Wab between two nodes；
(2) by Wab by being ranked up to obtain Wab_order from small to large；
(3) obtain first value in Wab_order, i.e., the minimum value in all Wab, corresponding node serial number be (i,
J), i and j are united as first bosk；
(4) since second value of Wab_order, if node serial number corresponding to second value is (i, m) or (m, j),
That is then one group of m self-contained；If the corresponding node serial number of second value is (n, m), n and m become second bosk；
(5) since the third of Wab_order value, if two trees of Wab_order connection and composition bosk before
Tree does not repeat, then it is individually for new bosk；If there is one tree repeats, then remaining another one tree is alone at bosk；
If all repeated, next weight is jumped to；Until having considered all weight Wab_order；Thus by all users
Bosk is constituted, and has interference value as small as possible inside bosk；
(6) first round merges bosk: the basic thought that bosk merges is the interference minimum newly increased: in (5) step
Orderly bosk, the basis that first group is merged as first time, remaining group is interfered and is counted with first group respectively
It calculates, the method for calculating are as follows: assuming that the group merges with first group, interfere summation two-by-two in group after record merges.Consider all remaining
Group and first group of interference merged and.Interference and the smallest merging mode are retained, complete to merge for the first time, remaining
In the bosk not merged, sequentially successively merged by the method that first time merges, until all bosks all complete this
The merging of one wheel.
(7) n-th (n >=2) wheel merging bosk: it is successively merged on the basis of n-1 takes turns and merges according to the method for step 5,
Bosk scale is increasing, and number is being reduced；
(8) bosk is successively merged, the number of the cluster until reaching needs then stops.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201811401988.7A CN109586836B (en) | 2018-11-22 | 2018-11-22 | User clustering algorithm based on artificial intelligence |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201811401988.7A CN109586836B (en) | 2018-11-22 | 2018-11-22 | User clustering algorithm based on artificial intelligence |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN109586836A true CN109586836A (en) | 2019-04-05 |
| CN109586836B CN109586836B (en) | 2023-09-15 |
Family
ID=65923802
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201811401988.7A Active CN109586836B (en) | 2018-11-22 | 2018-11-22 | User clustering algorithm based on artificial intelligence |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN109586836B (en) |
Citations (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US20150131537A1 (en) * | 2013-11-08 | 2015-05-14 | Spidercloud Wireless, Inc. | Fractional frequency reuse schemes assigned to radio nodes in an lte network |
| CN108738028A (en) * | 2018-05-25 | 2018-11-02 | 西安邮电大学 | A kind of cluster-dividing method that super-intensive group is off the net |
| CN108809470A (en) * | 2018-07-04 | 2018-11-13 | 西安邮电大学 | A kind of cluster algorithm in super-intensive cellular network |
-
2018
- 2018-11-22 CN CN201811401988.7A patent/CN109586836B/en active Active
Patent Citations (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US20150131537A1 (en) * | 2013-11-08 | 2015-05-14 | Spidercloud Wireless, Inc. | Fractional frequency reuse schemes assigned to radio nodes in an lte network |
| CN108738028A (en) * | 2018-05-25 | 2018-11-02 | 西安邮电大学 | A kind of cluster-dividing method that super-intensive group is off the net |
| CN108809470A (en) * | 2018-07-04 | 2018-11-13 | 西安邮电大学 | A kind of cluster algorithm in super-intensive cellular network |
Non-Patent Citations (1)
| Title |
|---|
| 黄俊伟;周朋光;张仁迟;滕得阳;徐浩;: "超密集网络中小小区分簇和子载波分配算法", 电子技术应用, no. 07 * |
Also Published As
| Publication number | Publication date |
|---|---|
| CN109586836B (en) | 2023-09-15 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN104955077B (en) | A kind of heterogeneous network cell cluster-dividing method and device based on user experience speed | |
| Dong et al. | Energy-efficient hierarchical resource allocation in uplink-downlink decoupled NOMA HetNets | |
| CN109862573B (en) | LTE hybrid networking self-planning method based on multi-target particle swarm | |
| CN107172682A (en) | Super-intensive network radio resources distribution method based on dynamic clustering | |
| CN107613556B (en) | Full-duplex D2D interference management method based on power control | |
| CN105916198B (en) | Resource allocation and Poewr control method based on efficiency justice in a kind of heterogeneous network | |
| CN104394535A (en) | Base station planning method for cooperative communication | |
| CN102355718B (en) | Method for coordinating uplink interference by combining high interference indication and overload indication | |
| CN112637907B (en) | Combined optimization method for user multi-association and downlink power distribution in millimeter wave network | |
| CN106341826A (en) | Resource optimized distribution method for wireless power private network based on virtualization technology | |
| CN105554771B (en) | Femto subscriber resource distribution method based on dense distribution | |
| CN106792995A (en) | The user access method of content low time delay transmission is ensured in a kind of following 5G networks | |
| CN108366426A (en) | A kind of virtual map method based on simulated annealing under super-intensive environment | |
| CN106209336B (en) | A kind of resource allocation methods in software definition mobile network | |
| CN107454601A (en) | The wireless dummy mapping method of inter-cell interference is considered under a kind of super-intensive environment | |
| CN108809470B (en) | Clustering algorithm in ultra-dense cellular network | |
| CN108282888A (en) | A kind of D2D resource allocation methods based on improvement fuzzy clustering | |
| CN109586836A (en) | A kind of user's cluster algorithm based on artificial intelligence | |
| Cuadra et al. | A Lamarckian Hybrid Grouping Genetic Algorithm with repair heuristics for resource assignment in WCDMA networks | |
| CN108738028B (en) | Clustering method under ultra-dense networking | |
| CN107613565A (en) | A kind of radio resource management method in full duplex super-intensive network | |
| Alaghehband et al. | Efficient fuzzy based uav positioning in iot environment data collection | |
| CN106817691A (en) | A kind of PCI distribution method and device | |
| CN105979591B (en) | Based on minimum effectively SINR optimal user scheduling method under the big connection scene of low-power consumption | |
| CN108566645A (en) | Network user's cut-in method based on Neyman-Scott cluster processes |
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 | ||
| GR01 | Patent grant | ||
| GR01 | Patent grant |