CN111866912A - Time-space based traffic volume region classification and analysis method - Google Patents
Time-space based traffic volume region classification and analysis method Download PDFInfo
- Publication number
- CN111866912A CN111866912A CN202010549613.6A CN202010549613A CN111866912A CN 111866912 A CN111866912 A CN 111866912A CN 202010549613 A CN202010549613 A CN 202010549613A CN 111866912 A CN111866912 A CN 111866912A
- Authority
- CN
- China
- Prior art keywords
- tri
- traffic
- data
- triangle
- area
- 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
Images
Classifications
-
- H—ELECTRICITY
- H04—ELECTRIC COMMUNICATION TECHNIQUE
- H04W—WIRELESS COMMUNICATION NETWORKS
- H04W24/00—Supervisory, monitoring or testing arrangements
- H04W24/02—Arrangements for optimising operational condition
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F18/00—Pattern recognition
- G06F18/20—Analysing
- G06F18/23—Clustering techniques
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F18/00—Pattern recognition
- G06F18/20—Analysing
- G06F18/24—Classification techniques
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F18/00—Pattern recognition
- G06F18/20—Analysing
- G06F18/24—Classification techniques
- G06F18/241—Classification techniques relating to the classification model, e.g. parametric or non-parametric approaches
- G06F18/2413—Classification techniques relating to the classification model, e.g. parametric or non-parametric approaches based on distances to training or reference patterns
- G06F18/24133—Distances to prototypes
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F18/00—Pattern recognition
- G06F18/20—Analysing
- G06F18/24—Classification techniques
- G06F18/241—Classification techniques relating to the classification model, e.g. parametric or non-parametric approaches
- G06F18/2413—Classification techniques relating to the classification model, e.g. parametric or non-parametric approaches based on distances to training or reference patterns
- G06F18/24147—Distances to closest patterns, e.g. nearest neighbour classification
-
- H—ELECTRICITY
- H04—ELECTRIC COMMUNICATION TECHNIQUE
- H04W—WIRELESS COMMUNICATION NETWORKS
- H04W24/00—Supervisory, monitoring or testing arrangements
- H04W24/06—Testing, supervising or monitoring using simulated traffic
Abstract
The invention discloses a time-space based traffic volume region classification and analysis method, and belongs to the technical field of prediction and network optimization. Firstly, a plurality of adjacent base stations are selected, a coordinate point set of each base station is constructed, and corresponding historical traffic of each base station is inquired. And then constructing a Delaunay triangulation network, recording the coordinates of three base stations of each triangle, and generating a triangle linked list P. Traversing a triangular linked list P, searching a Weinuo side by using the outer center of the triangle, drawing a Weinuo graph, and automatically dividing all base stations into a plurality of polygonal areas; simultaneously, recording adjacent areas of each area to realize spatial division; and simultaneously, classifying the service volume data of all the areas by adopting a k-NN algorithm to realize time division. And finally, marking adjacent base station areas with the same historical traffic change trend in the Veno diagram as the same color, and testing the causality of the traffic data of each classified area by using the Glange causality relationship. The invention improves the accuracy of network optimization.
Description
Technical Field
The invention belongs to the technical field of prediction and network optimization in a cellular network, and particularly relates to a time-space-based traffic volume region classification and analysis method.
Background
With the rapid development of mobile networks, data traffic in wireless networks is also growing dramatically. According to the statistical data and prediction of Cisco visual network indexes, namely Forecast and Trends, 2017-reservoir 2022,' [ R ] Cisco, Tech.Rep., November,2018, the flow of the Internet will increase by 4.6 times in a busy period and 3.2 times in an idle period from 2016 to 2021, and the flow of the mobile phone end will exceed that of the PC end; the multiplied increase of the traffic brings great challenges to the operation and maintenance of the network, and therefore, the analysis of the wireless network data and the research on the traffic model are particularly critical.
The coming future of 5G communication brings about a surge of traffic for cellular networks, and the traffic in the cellular networks has strong time variability and strong correlation of space and time, which brings great difficulty to network performance analysis.
In order to make reasonable use of communication network resources, the way traffic information in cellular networks is handled becomes especially critical. However, there still exist some problems in the existing wireless network optimization, for example, when the network optimization is performed, workers often divide areas according to experience, such as business areas and residential areas. This results in certain errors in resource allocation and network parameter adjustment.
Disclosure of Invention
In order to optimize the manual network resource allocation and improve the rationality of network resource allocation, the invention provides a time-space-based traffic volume region classification and analysis method, which classifies historical traffic volume data by using a space-time model, applies causal analysis to communication data analysis on the basis of the classification, and is an effective region classification and data analysis method for the accurate network optimization of 4G network and future 5G network data.
The time-space based traffic volume region classification and analysis method comprises the following specific steps:
step one, selecting a plurality of adjacent base stations, constructing a set B of coordinate points of each base station, and inquiring corresponding historical traffic of each base station to form a matrix T.
Step two, constructing a Delaunay triangulation network by the coordinate position of each base station through a Delaunay algorithm, recording the coordinates of three base station points forming each triangle in the Delaunay triangulation network, and generating a triangular linked list P;
P＝[Tri1,Tri2......,Trit.....,Trin]
wherein, TritIs the t-th triangle after triangulation; n is the number of triangles after all base stations are triangulated.
Traversing the triangular linked list P, searching a Weinuo edge by using the outer center of the triangle, and storing the Weinuo edge into the linked list;
The specific process is as follows:
step 301, aiming at the current triangle TritTraversing the triangle chain table P, finding the triangle TritThree adjacent triangles Tri sharing one sideA，TriBAnd TriC；
TriA∈P，TriB∈P，TriC∈P。
Step 302, respectively calculating the current triangle TritAnd three adjacent triangles TriA，TriBAnd TriCThe center of the circumscribed circle of (1) is totally four outsoles;
step 303, the current triangle TritThe outer centers of the three triangles are respectively connected with the outer centers of the three triangles which are adjacent to each other and are stored in the Voronoi edge chain table;
three connecting line segments between four outsides are stored in the Voronoi chain table; if the triangle with common edges is not enough, the current triangle Tri is solvedtThe outermost perpendicular bisector is stored in the voronoi linked list.
And step 304, repeating the steps until the traversal of the triangular linked list P is finished, finding all the voronoi edges, and storing the voronoi edges into the linked list.
Drawing a voronoi diagram according to voronoi edges, and automatically dividing all base stations into a plurality of polygonal areas; simultaneously, recording adjacent regions of each region to obtain an adjacent region list of each region, and realizing spatial division;
classifying the service volume data of all the areas by adopting a k-NN algorithm, dividing the areas where the base stations with the same historical service volume change trend along with time into a class, and finally obtaining clusters of all base station coverage areas to realize time division;
The classification criterion of the k-NN algorithm adopts Euclidean distance, and the calculation formula is as follows:
wherein [ x ]1,x2,...,xt]And [ y1,y2,...,yt]Respectively historical traffic data sets for the two base stations.
And step six, according to the classification result of the k-NN algorithm and the adjacent region list, marking the polygonal regions in the Voronoi diagram, marking the adjacent base station regions with the same historical traffic change trend and the same visual color, and finishing accurate scene division based on time-space.
And step seven, aiming at each accurate classification area, utilizing the Glange causal relationship to test the causality of the traffic data of each classification area, finding the relationship between the traffic of each area and the adjacent area from the causal perspective, and applying the relationship to a communication data analysis optimization network.
The glange causal relationship test refers to: the method comprises the steps of analyzing historical data sequences distributed in the space of each area, carrying out causal relationship inspection among a plurality of time sequences to obtain the causal relationship between the traffic of adjacent areas and the traffic change of a central area, obtaining the strength of the causal relationship among each group of data according to the result of the causal relationship of the data of each area, further selecting area data with strong causality with the traffic of a target (central) area in an urban communication network, and carrying out multivariate prediction on the data of the area through combined processing.
The invention has the advantages that:
according to the time-space based traffic volume region classification and analysis method, the obtained analysis result is beneficial to improving the accuracy of network optimization, finding out the base station which is easily influenced by surrounding data, accurately optimizing the network and analyzing the relationship between causal density and geographical distribution.
Drawings
FIG. 1 is a flow chart of a time-space based traffic region classification and analysis method of the present invention;
FIG. 2 is a schematic diagram of a selected time-space based traffic volume region classification and analysis method in an embodiment of the present invention;
FIG. 3 is a detailed diagram of a causal verification and multivariate timing prediction method in an embodiment of the invention;
FIG. 4 is a delaunay diagram generated from geographical location information of a base station in an embodiment of the present invention;
FIG. 5 is a Voronoi diagram resulting from triangulating the delaunay diagram according to an embodiment of the present invention;
FIG. 6 is a Voronoi diagram obtained by the k-NN algorithm according to the embodiment of the invention and combined with the historical data of each region.
Detailed Description
The invention is further described with reference to the following detailed description of embodiments in conjunction with the accompanying drawings.
The invention discloses a traffic area classification and prediction method aiming at the time-space characteristics of a cellular network. In the first part, the time-space characteristics of wireless network data are considered for the measured data of an actual network, and a base station-level wireless network data classification scheme is designed according to the ideas of delaunay triangulation, voronoi diagram and K-NN. The traffic data classification is expanded from two dimensions of time and space, clustering is carried out on historical traffic time sequences of each base station from time, and the traffic data classification is used for clustering the traffic data with the same variation trend into one class. Meanwhile, spatially, the information of the adjacent regions is considered; therefore, base stations which are adjacent in space and have the same historical data change trend are classified into one type. And the second part researches the mutual influence relationship between the classified service volume data of each region and the adjacent regions and between the base stations in each region by taking the regions generated by classification as the basis and adopting the concept of the Glange causal check. The final analysis results are helpful for improving the accuracy of network optimization, finding base stations susceptible to surrounding data, precise network optimization, and analyzing the relationship between causal density and geographical distribution.
As shown in fig. 1, the specific steps are as follows:
step one, selecting a plurality of adjacent base stations, constructing a set B of coordinate points of each base station, and inquiring corresponding historical traffic of each base station to form a matrix T.
Secondly, constructing a neighbor network by the coordinate positions of all the base stations through a Delaunay algorithm, namely constructing a Delaunay triangulation network; simultaneously recording the coordinates of three base station points forming each triangle in the Delaunay triangulation network, and generating a triangle linked list P;
P＝[Tri1,Tri2......,Trit.....,Trin]
wherein, TritIs the t-th triangle after triangulation; n is the number of triangles after all base stations are triangulated.
Step three, traversing a triangular linked list P, searching a voronoi edge by utilizing the outer center of the triangle, and storing the voronoi edge into the linked list;
the specific process is as follows:
step 301, aiming at the current triangle TritTraversing the triangle chain table P, finding the triangle TritThree adjacent triangles Tri sharing one sideA，TriBAnd TriC；
TriA∈P，TriB∈P，TriC∈P。
Step 302, respectively calculating the current triangle TritAnd three adjacent triangles TriA，TriBAnd TriCThe center of the circumscribed circle of (1) is totally four outsoles;
step 303, the current triangle TritThe outer centers of the three triangles are respectively connected with the outer centers of the three triangles which are adjacent to each other and are stored in the Voronoi edge chain table;
three connecting line segments between four outsides are stored in the Voronoi chain table; if the triangle with common edges is not enough, the current triangle Tri is solved tThe outermost perpendicular bisector is stored in the voronoi linked list.
And step 304, repeating the steps until the traversal of the triangular linked list P is finished, finding all the voronoi edges, and storing the voronoi edges into the linked list.
Drawing a voronoi diagram according to voronoi edges, and automatically dividing all base stations into a plurality of polygonal areas; simultaneously, recording adjacent regions of each region to obtain an adjacent region list of each region, and realizing spatial division;
classifying the service volume data of all areas by adopting a k-NN (k-nearest neighbor) algorithm, dividing the areas where the base stations with the same historical service volume change trend along with time into a class, and finally obtaining clusters of all base station coverage areas to realize time division;
the classification criterion of the k-NN algorithm adopts Euclidean distance, and the calculation formula is as follows:
wherein, [ x ]1,x2,...,xt]And [ y1,y2,...,yt]Respectively historical traffic data sets for the two base stations.
And step six, according to the classification result of the k-NN algorithm and the adjacent region list, marking the polygonal regions in the Voronoi diagram, marking the adjacent base station regions with the same historical traffic change trend and the same visual color, and finishing accurate scene division based on time-space.
And step seven, aiming at each accurate classification area, utilizing the Glange causal relationship to test the causality of the traffic data of each classification area, finding the relationship between the traffic of each area and the adjacent area from the causal perspective, and applying the relationship to a communication data analysis optimization network.
The glange causal relationship test refers to: the method comprises the steps of analyzing historical data sequences distributed in the space of each area, carrying out causal relationship inspection among a plurality of time sequences to obtain the causal relationship between the traffic of adjacent areas and the traffic change of a central area, obtaining the strength of the causal relationship among each group of data according to the result of the causal relationship of the data of each area, further selecting area data with strong causality with the traffic of a target (central) area in an urban communication network, and carrying out multivariate prediction on the data of the area through combined processing.
Equivalently, finding out the 'reason' of the traffic change of the prediction area, and combining the traffic data of the prediction area with the causal data to finish the accurate analysis of the wireless network. The specific process is as follows:
the first step is as follows: verifying the original hypothesis: x is not the glandor cause of y.
First, the following two regression models are estimated:
Wherein alpha is0Representing a constant term, p and q are the maximum hysteresis amounts of y and x, respectively,tis white noise and t is the time series length.
The residual sum of squares of the two regression models is then calculated to construct the F statistic.
Wherein m is the number of samples, and r represents a constrained model, namely formula (2); u denotes an unconstrained model, i.e., equation (1). The original hypothesis can be verified by equation (3). If f ≧ f _ (q, m-p-q-1), then significantly not 0, rejecting the assumption that x is not the Glanberg cause of y; instead, this assumption cannot be rejected. Wherein f is RSS in the above formularAnd RSSuThe ratio of the two variances.
The second step is that: exchanging the positions of y and x, and checking the original hypothesis, as in the first step: "y is not the granger cause of x change".
The third step: to conclude that "x is the grand cause of y" the original hypothesis that "x is not the grand cause of y" must be negated, and the original hypothesis that "y is not the grand cause of x" is accepted, thereby obtaining the final result.
Example (b):
the present embodiment is a traffic volume region classification and prediction method based on cellular network time-space characteristics, and the overall flow is shown in fig. 2: firstly, constructing a base station position coordinate point set B and a corresponding historical traffic matrix T; constructing an adjacent station network according to the base station position and the Delaunay algorithm, and simultaneously recording which three base stations each triangle consists of to generate a triangle linked list: p ═ Tri 1,Tri2......,Trin]In this embodiment, the number n of triangles after triangulation of the base station is 53.
Then, traversing the triangle linked list, and drawing a Voronoi diagram according to the Voronoi edge connected with the outer center of the triangle. And simultaneously recording the adjacent area of each area to obtain an adjacent area list of each area.
Meanwhile, clustering is carried out by adopting a K-NN algorithm based on historical traffic data of the base station.
And screening and marking polygons in the Voronoi diagram according to the clustering result of the K-NN and the adjacent area list, and marking the adjacent areas with consistent traffic trends by adopting a visual coloring method.
Finally, for causal verification and data prediction, as shown in fig. 3, data processing (DataDetrending), time-space modeling (Means remove), and causal verification (Granger Causality Check) are mainly included. The traffic data in this embodiment is downlink traffic data of an LTE network.
Data processing: in this embodiment, the data set is provided by a main urban area in china. The selected data is the geographical position information of all base stations and the downlink flow data of each base station. As shown in table 1, the granularity of data sampling time is 1 hour, in this case, data of 9 base stations in a data set is selected for testing, and in the table, downlink traffic data of each base station in 24 hours of a working day is shown, where a statistical unit of the downlink data is GB.
TABLE 1
Time-space modeling: the method mainly combines the geographical position of a base station and historical telephone traffic data to carry out accurate regional division on the urban wireless network. As shown in fig. 4, the Delaunay diagram is generated according to geographical location information, which is an engineering parameter of the base station, and the horizontal and vertical coordinates are the longitude and latitude of the base station, respectively. And performing k-NN clustering on the historical data on the basis, classifying the base stations with the consistent historical data change trend into one class, and screening according to the adjacent region information of each base station on the basis of clustering. Since the characterization of the regions after classification is not obvious, a voronoi (voronoi) map is computed on the basis of the triangulation delaunay map, as shown in fig. 5. And (3) calculating to obtain a voronoi diagram after combining a k-NN algorithm of historical data of each region, and finishing accurate scene division based on time and space by obtaining a final result as shown in FIG. 6.
The classification in the invention is combined with time and space, and the k-NN algorithm is adopted to cluster the base stations with the same trend of the change of the traffic along with the time in time; in space, a Delaunay algorithm and a Voronoi diagram are adopted, and the area of the base station is classified by combining the result of k neighbor.
And (3) causal verification: after accurate region classification, the regions where the base stations with the same traffic variation trend are located are divided together, so that the network optimization accuracy is ensured, and the complexity of network optimization is reduced. The causality verification part mainly performs causality-based analysis on the traffic data of each region obtained by classification.
By means of the Glange causal verification method, causal relationships among the regions can be obtained, and the relationship between the traffic of each region and the traffic of the adjacent regions can be found from the causal perspective. Therefore, the strength of causal relationship among the areas is found, data of the adjacent areas with strong causality are selected to conduct multivariate prediction, clear guidance is provided for network optimization, and great help is provided for network load estimation.
After the causal relationship among the regions is obtained, historical data of the regions and the causal relationship among the region traffic can be combined to predict future traffic change more accurately. For example, geographically adjacent and causal areas are selected, and the data for these areas are jointly processed for prediction.
Claims (4)
1. The time-space based traffic volume region classification and analysis method is characterized by comprising the following specific steps of:
selecting a plurality of adjacent base stations, constructing a set B of coordinate points of each base station, and inquiring corresponding historical traffic of each base station to form a matrix T;
step two, constructing a Delaunay triangulation network by the coordinate position of each base station through a Delaunay algorithm, recording the coordinates of three base station points forming each triangle in the Delaunay triangulation network, and generating a triangular linked list P;
P＝[Tri1,Tri2......,Trit.....,Trin]
Wherein, TritIs the t-th triangle after triangulation; n is the number of triangles after all base stations are triangulated;
traversing the triangular linked list P, searching a Weinuo edge by using the outer center of the triangle, and storing the Weinuo edge into the linked list;
drawing a voronoi diagram according to voronoi edges, and automatically dividing all base stations into a plurality of polygonal areas; simultaneously, recording adjacent regions of each region to obtain an adjacent region list of each region, and realizing spatial division;
classifying the service volume data of all the areas by adopting a k-NN algorithm, dividing the areas where the base stations with the same historical service volume change trend along with time into a class, and finally obtaining clusters of all base station coverage areas to realize time division;
marking polygonal areas in the Voronoi diagram according to the classification result of the k-NN algorithm and an adjacent area list, marking adjacent base station areas with the same historical traffic volume change trend as well as the same visual color, and finishing accurate scene division based on time-space;
and step seven, aiming at each accurate classification area, utilizing the Glange causal relationship to test the causality of the traffic data of each classification area, finding the relationship between the traffic of each area and the adjacent area from the causal perspective, and applying the relationship to a communication data analysis optimization network.
2. The method for classifying and analyzing time-space based traffic volume regions according to claim 1, wherein the step three comprises the following specific processes:
step 301, aiming at the current triangle TritTraversing the triangle chain table P, finding the triangle TritThree adjacent triangles Tri sharing one sideA，TriBAnd TriC；
TriA∈P，TriB∈P，TriC∈P；
Step 302, respectively calculating the current triangle TritAnd three adjacent triangles TriA，TriBAnd TriCThe center of the circumscribed circle of (1) is totally four outsoles;
step 303, the current triangle TritThe outer centers of the three triangles are respectively connected with the outer centers of the three triangles which are adjacent to each other and are stored in the Voronoi edge chain table;
three connecting line segments between four outsides are stored in the Voronoi chain table; if the triangle with common edges is not enough, the current triangle Tri is solvedtThe outmost perpendicular bisector is stored in the Voronoi linked list;
and step 304, repeating the steps until the traversal of the triangular linked list P is finished, finding all the voronoi edges, and storing the voronoi edges into the linked list.
3. The method for time-space based traffic region classification and analysis according to claim 1, wherein the classification criterion of the k-NN algorithm is euclidean distance, and the formula is as follows:
wherein [ x ]1,x2,...,xt]And [ y1,y2,...,yt]History service for two base stations respectively The data set is measured.
4. The time-space based traffic zone classification and analysis method of claim 1, wherein the granger causal relationship test is: the method comprises the steps of analyzing historical data sequences distributed in the space of each area, carrying out causal relationship inspection among a plurality of time sequences to obtain the causal relationship between the traffic of adjacent areas and the traffic change of a central area, obtaining the strength of the causal relationship among each group of data according to the result of the causal relationship of the data of each area, further selecting area data with strong causality with the traffic of a target (central) area in an urban communication network, and carrying out multivariate prediction on the data of the area through combined processing.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202010549613.6A CN111866912B (en) | 2020-06-16 | 2020-06-16 | Time-space based traffic volume region classification and analysis method |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202010549613.6A CN111866912B (en) | 2020-06-16 | 2020-06-16 | Time-space based traffic volume region classification and analysis method |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN111866912A true CN111866912A (en) | 2020-10-30 |
| CN111866912B CN111866912B (en) | 2021-07-20 |
Family
ID=72987224
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202010549613.6A Active CN111866912B (en) | 2020-06-16 | 2020-06-16 | Time-space based traffic volume region classification and analysis method |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN111866912B (en) |
Cited By (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN113191416A (en) * | 2021-04-26 | 2021-07-30 | 杭州电子科技大学 | Large-scale geographic point data-oriented spatial attribute associated Voronoi diagram generation method |
| CN113449111A (en) * | 2021-08-31 | 2021-09-28 | 苏州工业园区测绘地理信息有限公司 | Social governance hot topic automatic identification method based on time-space semantic knowledge migration |
| CN114980136A (en) * | 2022-05-20 | 2022-08-30 | 西安电子科技大学 | Method for covering low-altitude three-dimensional signal by high-energy-efficiency ground base station |
Citations (10)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US20080072115A1 (en) * | 2006-09-20 | 2008-03-20 | Samsung Electronics Co., Ltd. | Apparatus and method for processing received data in a broadband wireless communication system |
| CN102625322A (en) * | 2012-02-27 | 2012-08-01 | 北京邮电大学 | Multi-mode intelligent configurable method for implementing optimization of wireless network |
| CN103796227A (en) * | 2012-10-31 | 2014-05-14 | 中国电信股份有限公司 | Method and system for performing wireless signal coverage analysis |
| CN106304105A (en) * | 2015-05-18 | 2017-01-04 | 中国移动通信集团湖北有限公司 | A kind of lane place optimization method and device |
| CN108112018A (en) * | 2017-12-21 | 2018-06-01 | 北京科技大学 | A kind of definite method and device of mobile communications network zone boundary to be optimized |
| CN109213169A (en) * | 2018-09-20 | 2019-01-15 | 湖南万为智能机器人技术有限公司 | The paths planning method of mobile robot |
| CN109862580A (en) * | 2017-11-30 | 2019-06-07 | 华为技术有限公司 | A kind of network optimized approach and device |
| US20190215700A1 (en) * | 2018-01-09 | 2019-07-11 | P. I. Works TR Bilisim Hizm. San. ve Tic A.S. | Method for Capacity and Coverage Optimization of a Multi-RAT Network |
| CN110120883A (en) * | 2018-02-07 | 2019-08-13 | 中国移动通信有限公司研究院 | A kind of method, apparatus and computer readable storage medium for assessing network performance |
| CN110972149A (en) * | 2019-11-27 | 2020-04-07 | 河海大学 | Node optimization deployment method of circular ring type wireless sensor network |
-
2020
- 2020-06-16 CN CN202010549613.6A patent/CN111866912B/en active Active
Patent Citations (10)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US20080072115A1 (en) * | 2006-09-20 | 2008-03-20 | Samsung Electronics Co., Ltd. | Apparatus and method for processing received data in a broadband wireless communication system |
| CN102625322A (en) * | 2012-02-27 | 2012-08-01 | 北京邮电大学 | Multi-mode intelligent configurable method for implementing optimization of wireless network |
| CN103796227A (en) * | 2012-10-31 | 2014-05-14 | 中国电信股份有限公司 | Method and system for performing wireless signal coverage analysis |
| CN106304105A (en) * | 2015-05-18 | 2017-01-04 | 中国移动通信集团湖北有限公司 | A kind of lane place optimization method and device |
| CN109862580A (en) * | 2017-11-30 | 2019-06-07 | 华为技术有限公司 | A kind of network optimized approach and device |
| CN108112018A (en) * | 2017-12-21 | 2018-06-01 | 北京科技大学 | A kind of definite method and device of mobile communications network zone boundary to be optimized |
| US20190215700A1 (en) * | 2018-01-09 | 2019-07-11 | P. I. Works TR Bilisim Hizm. San. ve Tic A.S. | Method for Capacity and Coverage Optimization of a Multi-RAT Network |
| CN110120883A (en) * | 2018-02-07 | 2019-08-13 | 中国移动通信有限公司研究院 | A kind of method, apparatus and computer readable storage medium for assessing network performance |
| CN109213169A (en) * | 2018-09-20 | 2019-01-15 | 湖南万为智能机器人技术有限公司 | The paths planning method of mobile robot |
| CN110972149A (en) * | 2019-11-27 | 2020-04-07 | 河海大学 | Node optimization deployment method of circular ring type wireless sensor network |
Non-Patent Citations (2)
| Title |
|---|
| KAISA ZHANG 等: "Traffic-Aware and Energy-Efficiency Network Oriented Spatio-Temporal Analysis and Traffic Prediction", 《 2019 IEEE GLOBECOM WORKSHOPS》 * |
| 张佑生: "一种新颖的网格综合优化算法", 《系统仿真学报》 * |
Cited By (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN113191416A (en) * | 2021-04-26 | 2021-07-30 | 杭州电子科技大学 | Large-scale geographic point data-oriented spatial attribute associated Voronoi diagram generation method |
| CN113191416B (en) * | 2021-04-26 | 2024-02-09 | 杭州电子科技大学 | Large-scale geographic point data-oriented space attribute correlation voronoi diagram generation method |
| CN113449111A (en) * | 2021-08-31 | 2021-09-28 | 苏州工业园区测绘地理信息有限公司 | Social governance hot topic automatic identification method based on time-space semantic knowledge migration |
| CN114980136A (en) * | 2022-05-20 | 2022-08-30 | 西安电子科技大学 | Method for covering low-altitude three-dimensional signal by high-energy-efficiency ground base station |
| CN114980136B (en) * | 2022-05-20 | 2023-06-30 | 西安电子科技大学 | High-energy-efficiency ground base station low-altitude three-dimensional signal coverage method |
Also Published As
| Publication number | Publication date |
|---|---|
| CN111866912B (en) | 2021-07-20 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN111866912B (en) | Time-space based traffic volume region classification and analysis method | |
| CN106792514B (en) | User position analysis method based on signaling data | |
| EP2620024B1 (en) | Generation and use of coverage area models | |
| CN111867049B (en) | Positioning method, positioning device and storage medium | |
| JP2002057614A (en) | Method and device for evaluating rf propagation in radio communication system | |
| CN103068035A (en) | Wireless network location method, device and system | |
| CN108243435B (en) | Parameter optimization method and device in LTE cell scene division | |
| CN108462965B (en) | Iron tower demand site sharing planning method | |
| CN110298687B (en) | Regional attraction assessment method and device | |
| CN111062466A (en) | Method for predicting field intensity distribution of cell after antenna adjustment based on parameters and neural network | |
| WO2020024597A1 (en) | Indoor positioning method and apparatus | |
| CN103491544A (en) | Method and device for selecting shared station | |
| US10880754B1 (en) | Network planning tool for retention analysis in telecommunications networks | |
| Liang et al. | Mobile traffic prediction based on densely connected CNN for cellular networks in highway scenarios | |
| CN112399458A (en) | Big data analysis method for mobile communication network flow | |
| CN112801411A (en) | Network flow prediction method based on generation countermeasure network | |
| CN108207005B (en) | LTE wireless network evaluation method and server | |
| CN114828026A (en) | Base station planning method, device, equipment, storage medium and program product | |
| CN106922017B (en) | Positioning method and terminal | |
| CN108271181A (en) | A kind of PCI optimization method and devices of wireless network | |
| Wang et al. | Terrain-based coverage manifold estimation: Machine learning, stochastic geometry, or simulation? | |
| CN115099385A (en) | Spectrum map construction method based on sensor layout optimization and adaptive Kriging model | |
| Lan et al. | Intelligent GSM Cell Coverage Analysis System Based on GIS. | |
| Li et al. | Partitioning the wireless environment for determining radio coverage and traffic distribution with user feedback | |
| CN109286900B (en) | Wi-Fi sample data optimization method |
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 |