CN112291706A - Network optimization method based on big data and artificial intelligence - Google Patents
Network optimization method based on big data and artificial intelligence Download PDFInfo
- Publication number
- CN112291706A CN112291706A CN202011162894.6A CN202011162894A CN112291706A CN 112291706 A CN112291706 A CN 112291706A CN 202011162894 A CN202011162894 A CN 202011162894A CN 112291706 A CN112291706 A CN 112291706A
- Authority
- CN
- China
- Prior art keywords
- user
- data
- cell
- network
- learning model
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Pending
Links
- 238000005457 optimization Methods 0.000 title claims abstract description 54
- 238000000034 method Methods 0.000 title claims abstract description 51
- 238000013473 artificial intelligence Methods 0.000 title claims abstract description 26
- 230000011664 signaling Effects 0.000 claims abstract description 36
- 238000004458 analytical method Methods 0.000 claims abstract description 16
- 238000006073 displacement reaction Methods 0.000 claims abstract description 16
- 230000003068 static effect Effects 0.000 claims abstract description 16
- 238000004140 cleaning Methods 0.000 claims abstract description 11
- 238000010801 machine learning Methods 0.000 claims abstract description 9
- 230000000295 complement effect Effects 0.000 claims abstract description 8
- 230000004927 fusion Effects 0.000 claims abstract description 8
- 230000006399 behavior Effects 0.000 claims abstract description 6
- 238000005070 sampling Methods 0.000 claims abstract description 6
- 230000008569 process Effects 0.000 claims description 18
- 230000033001 locomotion Effects 0.000 claims description 15
- 230000002860 competitive effect Effects 0.000 claims description 10
- 238000012549 training Methods 0.000 claims description 10
- 238000005259 measurement Methods 0.000 claims description 9
- 238000009826 distribution Methods 0.000 claims description 8
- 238000012360 testing method Methods 0.000 claims description 7
- 230000008901 benefit Effects 0.000 claims description 4
- 238000004891 communication Methods 0.000 claims description 4
- 238000012545 processing Methods 0.000 claims description 4
- 230000008859 change Effects 0.000 claims description 3
- 238000013480 data collection Methods 0.000 claims description 3
- 238000005516 engineering process Methods 0.000 claims description 3
- 238000011156 evaluation Methods 0.000 claims description 3
- 238000000605 extraction Methods 0.000 claims description 3
- 238000001914 filtration Methods 0.000 claims description 3
- 230000003993 interaction Effects 0.000 claims description 3
- 238000002372 labelling Methods 0.000 claims description 3
- 238000007433 macroscopic evaluation Methods 0.000 claims description 3
- 238000012384 transportation and delivery Methods 0.000 claims description 3
- 238000012795 verification Methods 0.000 claims description 3
- 238000013439 planning Methods 0.000 description 5
- 238000013500 data storage Methods 0.000 description 2
- 230000006872 improvement Effects 0.000 description 2
- 238000012423 maintenance Methods 0.000 description 2
- 238000013433 optimization analysis Methods 0.000 description 2
- 238000011160 research Methods 0.000 description 2
- 230000002159 abnormal effect Effects 0.000 description 1
- 230000009286 beneficial effect Effects 0.000 description 1
- 230000007547 defect Effects 0.000 description 1
- 238000011161 development Methods 0.000 description 1
- 238000010586 diagram Methods 0.000 description 1
- 230000000694 effects Effects 0.000 description 1
- 238000007726 management method Methods 0.000 description 1
- 230000007246 mechanism Effects 0.000 description 1
- 238000012986 modification Methods 0.000 description 1
- 230000004048 modification Effects 0.000 description 1
- 230000008447 perception Effects 0.000 description 1
Images
Classifications
-
- H—ELECTRICITY
- H04—ELECTRIC COMMUNICATION TECHNIQUE
- H04W—WIRELESS COMMUNICATION NETWORKS
- H04W4/00—Services specially adapted for wireless communication networks; Facilities therefor
- H04W4/02—Services making use of location information
- H04W4/025—Services making use of location information using location based information parameters
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F30/00—Computer-aided design [CAD]
- G06F30/20—Design optimisation, verification or simulation
- G06F30/27—Design optimisation, verification or simulation using machine learning, e.g. artificial intelligence, neural networks, support vector machines [SVM] or training a model
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
- G06N20/00—Machine learning
-
- H—ELECTRICITY
- H04—ELECTRIC COMMUNICATION TECHNIQUE
- H04W—WIRELESS COMMUNICATION NETWORKS
- H04W24/00—Supervisory, monitoring or testing arrangements
- H04W24/02—Arrangements for optimising operational condition
-
- H—ELECTRICITY
- H04—ELECTRIC COMMUNICATION TECHNIQUE
- H04W—WIRELESS COMMUNICATION NETWORKS
- H04W24/00—Supervisory, monitoring or testing arrangements
- H04W24/06—Testing, supervising or monitoring using simulated traffic
-
- H—ELECTRICITY
- H04—ELECTRIC COMMUNICATION TECHNIQUE
- H04W—WIRELESS COMMUNICATION NETWORKS
- H04W4/00—Services specially adapted for wireless communication networks; Facilities therefor
- H04W4/02—Services making use of location information
- H04W4/029—Location-based management or tracking services
Abstract
The invention discloses a network optimization method based on big data and artificial intelligence, which relates to the technical field of network optimization and comprises the following steps: collecting multi-dimensional wireless network data and standard + customized signaling XDR data, and sorting and cleaning the data through complementary fusion to realize the backfilling and warehousing of user sampling points through latitude and longitude; obtaining accurate user behaviors through association of a displacement algorithm and a GIS building map layer; the mobile state and the static state of the user level are judged through a displacement algorithm and indoor and outdoor user analysis, meanwhile, the indoor and outdoor conditions of the user are obtained from signaling analysis, and modeling and calibration of high-speed mobile users and static indoor users are realized by combining the two information; based on an artificial intelligence machine learning algorithm, cell users are classified to form different cell sets, the cell sets are divided randomly to train and verify a parameter learning model, and a network optimization scheme can be formulated for the cell sets through the verified parameter learning model. The invention can optimize and widen the network.
Description
Technical Field
The invention relates to the technical field of network optimization, in particular to a network optimization method based on big data and artificial intelligence.
Background
The traditional optimization analysis means based on network management data and field test data and a single, universal and solidified optimization scheme can not meet the requirements of complex customer perception improvement and network quality optimization in the 4G era. Aiming at some regions with extensive and rare land, the problems of complex network structure (large difficulty in network planning and optimization), late 4G commercial start (gap between 4G planning and optimization experience and advanced province) and insufficient optimization personnel exist, and in addition, the original centralized optimization electronic process lacks an effective automatic and intelligent support means, so that a large amount of self-owned and manufacturer technicians are required to carry out planning and analysis work, the optimization experiences in different regions are different, and the optimization experiences cannot be shared in time, thereby causing high dependence on manufacturers.
Aiming at the problems, a network optimization method based on big data and artificial intelligence is designed and researched based on big data and artificial intelligence, the on-line flow of the electronic is optimized in a centralized manner, the network optimization rate is maximized, and maintenance staff are supported to carry out optimization processing work.
Disclosure of Invention
Aiming at the requirements and the defects of the prior art development, the invention provides a network optimization method based on big data and artificial intelligence, so as to improve the working efficiency of network planning, optimization and maintenance, widen and deepen the optimization work.
The invention relates to a network optimization method based on big data and artificial intelligence, which adopts the following technical scheme for solving the technical problems:
a network optimization method based on big data and artificial intelligence comprises the following steps:
collecting multi-dimensional wireless network data mainly comprising MR and standard + customized signaling XDR data, and sorting and cleaning the data in a complementary fusion mode to realize the backfilling and warehousing of user sampling points through latitude and longitude;
obtaining accurate user behaviors through association of a displacement algorithm and a GIS building map layer;
the method comprises the steps of judging the moving state and the static state of a user level through a displacement algorithm and indoor and outdoor user analysis, obtaining indoor and outdoor conditions of the user from signaling analysis, and realizing modeling and calibration of high-speed moving users and static indoor users through combination of the two key information;
the method comprises the steps of classifying cell users based on an artificial intelligence machine learning algorithm to form different cell sets, training and verifying a parameter learning model by randomly dividing the cell sets, and formulating different network optimization schemes based on the different cell sets through the verified parameter learning model.
Optionally, multi-dimensional wireless network data mainly including MR and standard + customized signaling XDR data are acquired, in this process, based on an APP used by the smartphone and capable of acquiring location information, user data acquisition is performed through a Gb \ IuPS \ S1 interface, longitude and latitude information reported by a user in the data signaling is analyzed in combination with a cell occupied by the user, so that a fixed point location of the user can be obtained, and finally, data acquisition is acquired.
Further optionally, when the user uses the APP capable of acquiring the location information through the smart phone, the user needs to call a data interface for auxiliary positioning, the information interaction process contains potential location information, relevant signaling and protocols are analyzed, the user location information is subjected to signaling extraction, the accuracy of the user location information is verified, and the accurate available fields are used for realizing data storage.
Further optionally, after data sorting and cleaning are performed in a complementary fusion mode, an MRO database table and a signaling XDR database table are obtained;
in the process, modeling is carried out through historical wireless measurement information and known position information contained in an MRO, relevance of an MRO database table and a signaling XDR database table is constructed based on a Hadoop big data processing layer, and then the position of a user only containing wireless measurement information is obtained through fingerprint database matching reverse positioning, so that full-quantity position information of a full-quantity user is realized, and the requirement of position continuity is met.
Optionally, the determination of the moving state and the static state at the user level is realized through a displacement algorithm and indoor and outdoor user analysis, and in the process:
firstly, labeling a user motion state through a displacement algorithm to distinguish road user data and fixed-point user data;
then, the user is distinguished which users are high-speed moving users by setting a time window,
(A) and (3) judging the high-speed mobile user: in the time window, the cell replacement times are more than or equal to the specified times, and the cell replacement distance is more than the threshold distance,
(B) and judging the low-speed mobile user: in the time window, the cell replacement times are less than the specified times or the distance between the first cell and the last cell is less than the threshold distance,
(C) and (3) judging by the user in the static state: within the time window, the cell change number is equal to 0.
Further optionally, based on an artificial intelligence machine learning algorithm, classifying cell users to form different cell sets, specifically:
setting a threshold value based on the ISODATA algorithm, classifying according to the basic information of the cell configuration and the map to obtain the geographic environment characteristics of the cell,
setting a threshold value based on an ISODATA algorithm, classifying according to the acquired data to obtain the user service distribution characteristics,
optimizing based on the collected data after sorting to obtain the optimized configuration characteristics related to the network,
based on the collected data after cleaning, combining with network indexes and parameters, filtering to obtain index characteristics,
the cell users are divided into different cell sets through the four aspects of cell geographic environment characteristics, user service distribution characteristics, optimization configuration characteristics and index characteristics.
Further optionally, the parameter learning model is trained and verified by randomly dividing the cell sets, and different network optimization schemes can be formulated based on different cell sets through the verified parameter learning model, where the specific operations include:
selecting any cell set, and randomly dividing data contained in the set into a training set and a test set according to the ratio of 7: 3;
a GBDT algorithm is applied to learn a training set to obtain a parameter learning model;
and verifying the effectiveness of the parameter learning model by using the test set, and in the verification process, when the parameter learning model outputs a parameter configuration scheme which is most similar to the service requirement of the cell and has excellent index characteristics, considering that the parameter learning model is effective, and at the moment, outputting a network optimization scheme of the cell by the parameter learning model, wherein the network optimization scheme has the best parameter configuration.
Optionally, based on MR-based multidimensional wireless network data and standard + customized signaling XDR data, network problems can be accurately located, and point-line-surface network coverage of buildings, roads, and grids can be realized, wherein,
(a) the method has the advantages that the buildings are used as points, the whole coverage assessment of the whole network of buildings is realized, on one hand, floors are layered in combination with building height information, the layered coverage assessment of the buildings is realized in combination with MR data reported by indoor resident users, on the other hand, the movement and competitive coverage and quality conditions of the area can be comprehensively presented, and the floor can be accurate to the indoor 10 meters;
(b) the road is taken as a line, macroscopic evaluation on road indexes can be realized according to positioning data at different time, and cells moving along the line and competing for the road indexes are analyzed;
(c) taking the outdoor part as a surface, and according to the positioning result of the user MR data in the motion state, covering and filling the outdoor area according to 40-by-40 grids, so as to realize the whole-network outdoor coverage evaluation and guide the site resource delivery;
based on the (a), (b) and (c), the MR data acquired after the competitive pair frequency point measurement is started can be combined with the MR indoor and outdoor positioning technology to realize competitive comparison of coverage conditions, realize coverage comparison results of single building movement and telecommunication and movement and communication, and realize comprehensive three-dimensional presentation.
Compared with the prior art, the network optimization method based on big data and artificial intelligence has the following beneficial effects:
1) the method converts the traditional single network problem research into the inter-related network problem research set, integrates and processes the network problems with high relevance, and makes a unique and targeted network optimization scheme aiming at problem cells of different types, so that on one hand, the original optimization work efficiency can be improved, on the other hand, the range, means and content of the optimization work can be widened and deepened, and the economic benefit can be improved;
2) the invention can find the abnormal problems of network structures such as latitude, direction angle and the like in the network without manual work, and provides a solution;
3) the invention provides a means for accurately positioning network problems for network optimization by using multi-dimensional wireless network data mainly based on MR and standard + customized signaling XDR data, realizes comprehensive presentation of point-line-surface network coverage, quality and competitive pairing of buildings, roads and grids and a series of optimization analysis applications, and supports network planning optimization work of various provinces.
Drawings
Fig. 1 is a detailed flow diagram of the first implementation (4) of the present invention.
Detailed Description
In order to make the technical scheme, the technical problems to be solved and the technical effects of the present invention more clearly apparent, the following technical scheme of the present invention is clearly and completely described with reference to the specific embodiments.
The first embodiment is as follows:
the embodiment provides a network optimization method based on big data and artificial intelligence, and the implementation content of the method comprises the following steps:
(1) collecting multi-dimensional wireless network data mainly comprising MR and standard + customized signaling XDR data, and sorting and cleaning the data in a complementary fusion mode to realize the backfilling and warehousing of user sampling points through latitude and longitude.
(1.1) collecting multi-dimensional wireless network data mainly comprising MR and signaling XDR data mainly comprising standard + customized signaling, in the process, based on APP which is used by the smart phone and can obtain position information, user data collection is carried out through Gb \ IuPS \ S1 interfaces, longitude and latitude information reported by a user in the data signaling is analyzed to be combined with the signaling occupied cell, then the fixed point position of the user can be obtained, and finally the data collection is obtained.
When a user uses an APP capable of acquiring position information through a smart phone, a data interface for auxiliary positioning needs to be called, potential position information is contained in the information interaction process, relevant signaling and protocols are analyzed, the user position information is subjected to signaling extraction, the accuracy of the user position information is verified, and data storage is achieved through accurate available fields.
And (1.2) collecting multi-dimensional wireless network data mainly comprising MR and standard + customized signaling XDR data, and performing data sorting and cleaning in a complementary fusion mode to obtain an MRO database table and a signaling XDR database table.
And (1.3) backfilling and warehousing of the longitude and latitude of a sampling point of a user are realized, in the process, modeling is carried out through historical wireless measurement information and known position information contained in an MRO, the relevance of an MRO database table and a signaling XDR database table is established based on a Hadoop big data processing layer, and then the position of the user only containing the wireless measurement information is obtained through fingerprint database matching reverse positioning, so that the full-amount position information of the full-amount user is realized, and the requirement of position continuity is met.
(2) And obtaining accurate user behaviors through association of a displacement algorithm and the GIS building map layer.
(3) The method realizes the judgment of the moving state and the static state of the user level through a displacement algorithm and indoor and outdoor user analysis, simultaneously obtains the indoor and outdoor conditions of the user from signaling analysis, and realizes the modeling and calibration of high-speed moving users and static indoor users through the combination of the two key information.
(3.1) judging the moving state and the static state of the user level through a displacement algorithm and indoor and outdoor user analysis, wherein in the process:
(3.1.1) firstly, labeling the motion state of a user by a displacement algorithm to distinguish road user data from fixed-point user data;
(3.1.2) subsequently, by setting a time window, it is discriminated which users are high-speed moving users,
(A) and (3) judging the high-speed mobile user: in the time window, the cell replacement times are more than or equal to the specified times, and the cell replacement distance is more than the threshold distance,
(B) and judging the low-speed mobile user: in the time window, the cell replacement times are less than the specified times or the distance between the first cell and the last cell is less than the threshold distance,
(C) and (3) judging by the user in the static state: within the time window, the cell change number is equal to 0.
(4) The method comprises the steps of classifying cell users based on an artificial intelligence machine learning algorithm to form different cell sets, training and verifying a parameter learning model by randomly dividing the cell sets, and formulating different network optimization schemes based on the different cell sets through the verified parameter learning model.
The artificial intelligence machine learning algorithm can be an ISODATA algorithm, namely iterative self-organizing analysis, by setting initial parameters and using a mechanism of merging and splitting, when the center distance of certain two classes is smaller than a certain threshold value, the two classes are merged into one class, and when the standard deviation of the certain class is larger than the certain threshold value or the number of samples exceeds the certain threshold value, the two classes are separated. When the number of samples of a certain type is less than a certain threshold, it is cancelled. Therefore, a relatively ideal classification result is finally obtained according to the initial clustering center and the set parameters such as the number of classes and the like.
(4.1) classifying users in the cells based on an artificial intelligence machine learning algorithm to form different cell sets, wherein the specific operation is as follows:
(4.1.1) setting a threshold value based on the ISODATA algorithm, classifying according to the basic information of the cell configuration and the map to obtain the geographic environment characteristics of the cell,
(4.1.2) setting a threshold value based on the ISODATA algorithm, classifying according to the acquired data to obtain the user service distribution characteristics,
(4.1.3) optimizing based on the collected data after arrangement to obtain the optimized configuration characteristics related to the network,
(4.1.4) based on the collected data after cleaning, combining network indexes and parameters, filtering to obtain index characteristics,
and (4.1.5) dividing the cell users into different cell sets through the four aspects of cell geographic environment characteristics, user service distribution characteristics, optimized configuration characteristics and index characteristics.
(4.2) training and verifying the parameter learning model by randomly dividing the cell sets, and formulating different network optimization schemes based on different cell sets through the verified parameter learning model, wherein the specific operation comprises the following steps:
(4.2.1) selecting any cell set, and randomly dividing the set containing data into a training set and a test set according to the ratio of 7: 3;
(4.2.2) learning the training set by using a GBDT algorithm to obtain a parameter learning model;
(4.2.3) verifying the effectiveness of the parameter learning model by using the test set, and in the verification process, when the parameter learning model outputs a parameter configuration scheme which is most similar to the service requirement of the cell and has excellent index characteristics, considering that the parameter learning model is effective, and at the moment, outputting a network optimization scheme of the cell by the parameter learning model, wherein the network optimization scheme has the optimal parameter configuration.
For the present embodiment, what needs to be supplemented is: based on multi-dimensional wireless network data mainly based on MR and standard + customized signaling XDR data, network problems can be accurately positioned, and point-line-surface network coverage of buildings, roads and grids can be realized, wherein,
(a) the method has the advantages that the buildings are used as points, the whole coverage assessment of the whole network of buildings is realized, on one hand, floors are layered in combination with building height information, the layered coverage assessment of the buildings is realized in combination with MR data reported by indoor resident users, on the other hand, the movement and competitive coverage and quality conditions of the area can be comprehensively presented, and the floor can be accurate to the indoor 10 meters;
(b) the road is taken as a line, macroscopic evaluation on road indexes can be realized according to positioning data at different time, and cells moving along the line and competing for the road indexes are analyzed;
(c) taking the outdoor part as a surface, and according to the positioning result of the user MR data in the motion state, covering and filling the outdoor area according to 40-by-40 grids, so as to realize the whole-network outdoor coverage evaluation and guide the site resource delivery;
based on the (a), (b) and (c), the MR data acquired after the competitive pair frequency point measurement is started can be combined with the MR indoor and outdoor positioning technology to realize competitive comparison of coverage conditions, realize coverage comparison results of single building movement and telecommunication and movement and communication, and realize comprehensive three-dimensional presentation.
Taking the analysis and optimization of the problem of the Xinjiang mobile network-the Quinun network as an example, the network optimization method based on big data and artificial intelligence provided by the embodiment is explained in detail. At this time:
and (2) completing acquisition of multi-dimensional wireless network data mainly comprising MR and standard + customized signaling XDR data according to the step (1), and sorting and cleaning the data in a complementary fusion mode to realize the backfilling and warehousing of the longitude and the latitude of a user sampling point.
And (3) associating the user behavior with the GIS building map layer through a displacement algorithm according to the step (2) to obtain the accurate user behavior.
And (3) judging the moving state and the static state of the user level through a displacement algorithm and indoor and outdoor user analysis, simultaneously obtaining the indoor and outdoor conditions of the user from signaling analysis, and realizing modeling and calibration of high-speed moving users and static indoor users through the combination of the two key information.
Referring to the attached drawings, according to (4) an artificial intelligence machine learning algorithm, four aspects of cell geographic environment characteristics, user service distribution characteristics, optimization configuration characteristics and index characteristics are considered, cell users are classified to form different cell sets, a parameter learning model is trained and verified by randomly dividing the cell sets, and different network optimization schemes can be formulated based on the different cell sets through the verified parameter learning model.
What needs to be supplemented is:
in the specific implementation process of the step (4), the user service requirements can be analyzed according to experience, and are basically 5 categories of downlink video traffic, uplink video traffic, WeChat traffic, QQ communication traffic, and web browsing traffic, so that the initialization clustering center can be set to 2^5 ═ 32 categories (downlink video traffic (large, small), uplink video traffic (large, small) …), and based on the basic requirements of service users, the cell user classification is completed by considering the cell geographic environment characteristics, the user service distribution characteristics, the optimization configuration characteristics, and the index characteristics.
In the specific implementation process of (4), based on different network optimization schemes formulated by the parameter learning model for different cell sets, variance demonstration can be specifically performed on the average downlink QCI packet loss rate, the average uplink and downlink QCI packet loss rate, the wireless call-through rate and the switching success rate before and after a certain cell set, and whether the network optimization scheme provided by the parameter learning model improves network indexes is verified.
The principles and embodiments of the present invention have been described in detail using specific examples, which are provided only to aid in understanding the core technical content of the present invention. Based on the above embodiments of the present invention, those skilled in the art should make any improvements and modifications to the present invention without departing from the principle of the present invention, and therefore, the present invention should fall into the protection scope of the present invention.
Claims (8)
1. A network optimization method based on big data and artificial intelligence is characterized in that the implementation content comprises the following steps:
collecting multi-dimensional wireless network data mainly comprising MR and standard + customized signaling XDR data, and sorting and cleaning the data in a complementary fusion mode to realize the backfilling and warehousing of user sampling points through latitude and longitude;
obtaining accurate user behaviors through association of a displacement algorithm and a GIS building map layer;
the method comprises the steps of judging the moving state and the static state of a user level through a displacement algorithm and indoor and outdoor user analysis, obtaining indoor and outdoor conditions of the user from signaling analysis, and realizing modeling and calibration of high-speed moving users and static indoor users through combination of the two key information;
the method comprises the steps of classifying cell users based on an artificial intelligence machine learning algorithm to form different cell sets, training and verifying a parameter learning model by randomly dividing the cell sets, and formulating different network optimization schemes based on the different cell sets through the verified parameter learning model.
2. The method as claimed in claim 1, wherein the method comprises collecting MR-based multidimensional wireless network data and standard + customized signaling XDR data, during which user data collection is performed through Gb \ IuPS \ S1 interface based on the APP used by the smart phone to obtain location information, analyzing longitude and latitude information reported by the user in the data signaling in combination with its signaling occupying cell, and obtaining the fixed-point location of the user, and finally, collecting and obtaining the data.
3. The method of claim 2, wherein when a user uses an APP that can obtain location information through a smart phone, the user needs to invoke a data interface for assisting in positioning, the information interaction process includes potential location information, related signaling and protocols are analyzed, the user location information is subjected to signaling extraction, the accuracy of the user location information is verified, and the accurate available fields are used to store data.
4. The network optimization method based on big data and artificial intelligence as claimed in claim 3, wherein MRO database table and signaling XDR database table are obtained after data sorting and cleaning are performed in a complementary fusion mode;
in the process, modeling is carried out through historical wireless measurement information and known position information contained in an MRO, relevance of an MRO database table and a signaling XDR database table is constructed based on a Hadoop big data processing layer, and then the position of a user only containing wireless measurement information is obtained through fingerprint database matching reverse positioning, so that full-quantity position information of a full-quantity user is realized, and the requirement of position continuity is met.
5. The method for optimizing the network based on big data and artificial intelligence as claimed in claim 1, wherein the determination of the moving state and the static state at the user level is realized by a displacement algorithm and indoor and outdoor user analysis, in the process:
firstly, labeling a user motion state through a displacement algorithm to distinguish road user data and fixed-point user data;
then, the user is distinguished which users are high-speed moving users by setting a time window,
(A) and (3) judging the high-speed mobile user: in the time window, the cell replacement times are more than or equal to the specified times, and the cell replacement distance is more than the threshold distance,
(B) and judging the low-speed mobile user: in the time window, the cell replacement times are less than the specified times or the distance between the first cell and the last cell is less than the threshold distance,
(C) and (3) judging by the user in the static state: within the time window, the cell change number is equal to 0.
6. The method according to claim 1, wherein the method for optimizing the network based on big data and artificial intelligence is characterized in that cell users are classified based on an artificial intelligence machine learning algorithm to form different cell sets, and the specific operation is as follows:
setting a threshold value based on the ISODATA algorithm, classifying according to the basic information of the cell configuration and the map to obtain the geographic environment characteristics of the cell,
setting a threshold value based on an ISODATA algorithm, classifying according to the acquired data to obtain the user service distribution characteristics,
optimizing based on the collected data after sorting to obtain the optimized configuration characteristics related to the network,
based on the collected data after cleaning, combining with network indexes and parameters, filtering to obtain index characteristics,
the cell users are divided into different cell sets through the four aspects of cell geographic environment characteristics, user service distribution characteristics, optimization configuration characteristics and index characteristics.
7. The method according to claim 6, wherein the parameter learning model is trained and verified by randomly dividing the cell sets, and different network optimization schemes can be formulated based on different cell sets through the verified parameter learning model, and the specific operations include:
selecting any cell set, and randomly dividing data contained in the set into a training set and a test set according to the ratio of 7: 3;
a GBDT algorithm is applied to learn a training set to obtain a parameter learning model;
and verifying the effectiveness of the parameter learning model by using the test set, and in the verification process, when the parameter learning model outputs a parameter configuration scheme which is most similar to the service requirement of the cell and has excellent index characteristics, considering that the parameter learning model is effective, and at the moment, outputting a network optimization scheme of the cell by the parameter learning model, wherein the network optimization scheme has the best parameter configuration.
8. The method as claimed in claim 1, wherein the network optimization method based on big data and artificial intelligence is based on MR-based multidimensional wireless network data and standard + customized signaling XDR data, which can precisely locate network problems and realize point-line-plane network coverage of buildings, roads and grids, wherein,
(a) the method has the advantages that the buildings are used as points, the whole coverage assessment of the whole network of buildings is realized, on one hand, floors are layered in combination with building height information, the layered coverage assessment of the buildings is realized in combination with MR data reported by indoor resident users, on the other hand, the movement and competitive coverage and quality conditions of the area can be comprehensively presented, and the floor can be accurate to the indoor 10 meters;
(b) the road is taken as a line, macroscopic evaluation on road indexes can be realized according to positioning data at different time, and cells moving along the line and competing for the road indexes are analyzed;
(c) taking the outdoor part as a surface, and according to the positioning result of the user MR data in the motion state, covering and filling the outdoor area according to 40-by-40 grids, so as to realize the whole-network outdoor coverage evaluation and guide the site resource delivery;
based on the (a), (b) and (c), the MR data acquired after the competitive pair frequency point measurement is started can be combined with the MR indoor and outdoor positioning technology to realize competitive comparison of coverage conditions, realize coverage comparison results of single building movement and telecommunication and movement and communication, and realize comprehensive three-dimensional presentation.
Priority Applications (2)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202011162894.6A CN112291706A (en) | 2020-10-27 | 2020-10-27 | Network optimization method based on big data and artificial intelligence |
| PCT/CN2021/117472 WO2022089031A1 (en) | 2020-10-27 | 2021-09-09 | Network optimization method based on big data and artificial intelligence |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202011162894.6A CN112291706A (en) | 2020-10-27 | 2020-10-27 | Network optimization method based on big data and artificial intelligence |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN112291706A true CN112291706A (en) | 2021-01-29 |
Family
ID=74373127
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202011162894.6A Pending CN112291706A (en) | 2020-10-27 | 2020-10-27 | Network optimization method based on big data and artificial intelligence |
Country Status (2)
| Country | Link |
|---|---|
| CN (1) | CN112291706A (en) |
| WO (1) | WO2022089031A1 (en) |
Cited By (8)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN114245392A (en) * | 2021-12-20 | 2022-03-25 | 哈尔滨入云科技有限公司 | 5G network optimization method and system |
| WO2022089031A1 (en) * | 2020-10-27 | 2022-05-05 | 浪潮天元通信信息系统有限公司 | Network optimization method based on big data and artificial intelligence |
| CN114826941A (en) * | 2022-04-27 | 2022-07-29 | 中国电子科技集团公司第五十四研究所 | AI model configuration method for wireless communication network |
| CN115243300A (en) * | 2021-04-23 | 2022-10-25 | 中国移动通信集团四川有限公司 | Network feature portrait generation method and device and electronic equipment |
| WO2022242447A1 (en) * | 2021-05-19 | 2022-11-24 | 中兴通讯股份有限公司 | Road user recognition method and device, storage medium, and electronic device |
| WO2023020086A1 (en) * | 2021-08-20 | 2023-02-23 | 中兴通讯股份有限公司 | Network optimization methods and apparatus, electronic device and storage medium |
| CN116578924A (en) * | 2023-07-12 | 2023-08-11 | 太极计算机股份有限公司 | Network task optimization method and system for machine learning classification |
| CN115243300B (en) * | 2021-04-23 | 2024-04-19 | 中国移动通信集团四川有限公司 | Network feature image generation method and device and electronic equipment |
Families Citing this family (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN115038103B (en) * | 2022-08-10 | 2022-11-22 | 中通服建设有限公司 | Intelligent tuning system and method for 5G wireless network parameters |
| CN116828529B (en) * | 2022-11-14 | 2024-04-16 | 新疆八达科技发展有限公司 | Intelligent community communication signal safety monitoring and early warning platform based on Internet of things |
| CN115883366B (en) * | 2023-01-29 | 2023-05-12 | 河北六联通信科技有限公司 | Data security supervision system and method based on communication network optimization |
| CN116193574B (en) * | 2023-02-22 | 2023-10-13 | 中电建建筑集团有限公司 | 5g network-based observation information fusion positioning key technical method and system |
Citations (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN108243495A (en) * | 2016-12-23 | 2018-07-03 | 亿阳信通股份有限公司 | A kind of location fingerprint database building method, device and method of locating terminal |
| CN108683527A (en) * | 2018-04-25 | 2018-10-19 | 武汉虹信技术服务有限责任公司 | A kind of user's perceived depth detection method based on MR and XDR |
| CN109769201A (en) * | 2018-12-28 | 2019-05-17 | 科大国创软件股份有限公司 | A kind of smart city management platform for realizing user's precise positioning |
Family Cites Families (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN107396312A (en) * | 2017-07-18 | 2017-11-24 | 浪潮天元通信信息系统有限公司 | The accurate recognition methods of customer location based on neutral net |
| CN109996186B (en) * | 2017-12-29 | 2021-02-26 | 中国移动通信集团陕西有限公司 | Network coverage problem identification method and device and readable storage medium |
| US10631263B2 (en) * | 2018-09-14 | 2020-04-21 | Viavi Solutions Inc. | Geolocating a user equipment |
| CN112291706A (en) * | 2020-10-27 | 2021-01-29 | 浪潮天元通信信息系统有限公司 | Network optimization method based on big data and artificial intelligence |
-
2020
- 2020-10-27 CN CN202011162894.6A patent/CN112291706A/en active Pending
-
2021
- 2021-09-09 WO PCT/CN2021/117472 patent/WO2022089031A1/en active Application Filing
Patent Citations (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN108243495A (en) * | 2016-12-23 | 2018-07-03 | 亿阳信通股份有限公司 | A kind of location fingerprint database building method, device and method of locating terminal |
| CN108683527A (en) * | 2018-04-25 | 2018-10-19 | 武汉虹信技术服务有限责任公司 | A kind of user's perceived depth detection method based on MR and XDR |
| CN109769201A (en) * | 2018-12-28 | 2019-05-17 | 科大国创软件股份有限公司 | A kind of smart city management platform for realizing user's precise positioning |
Non-Patent Citations (3)
| Title |
|---|
| 伍进: ""基于移动互联网的多运营商网络覆盖评估系统设计与实现"", 《中国优秀博硕士学位论文全文数据库(硕士)信息科技辑》 * |
| 张进等: "" 基于神经网络的MR精准定位"", 《邮电设计技术》 * |
| 王浩: ""基于用户感知的LTE业务质量评价与提升研究"", 《中国优秀博硕士学位论文全文数据库(硕士)经济与管理科学辑》 * |
Cited By (9)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| WO2022089031A1 (en) * | 2020-10-27 | 2022-05-05 | 浪潮天元通信信息系统有限公司 | Network optimization method based on big data and artificial intelligence |
| CN115243300A (en) * | 2021-04-23 | 2022-10-25 | 中国移动通信集团四川有限公司 | Network feature portrait generation method and device and electronic equipment |
| CN115243300B (en) * | 2021-04-23 | 2024-04-19 | 中国移动通信集团四川有限公司 | Network feature image generation method and device and electronic equipment |
| WO2022242447A1 (en) * | 2021-05-19 | 2022-11-24 | 中兴通讯股份有限公司 | Road user recognition method and device, storage medium, and electronic device |
| WO2023020086A1 (en) * | 2021-08-20 | 2023-02-23 | 中兴通讯股份有限公司 | Network optimization methods and apparatus, electronic device and storage medium |
| CN114245392A (en) * | 2021-12-20 | 2022-03-25 | 哈尔滨入云科技有限公司 | 5G network optimization method and system |
| CN114245392B (en) * | 2021-12-20 | 2022-07-01 | 哈尔滨入云科技有限公司 | 5G network optimization method and system |
| CN114826941A (en) * | 2022-04-27 | 2022-07-29 | 中国电子科技集团公司第五十四研究所 | AI model configuration method for wireless communication network |
| CN116578924A (en) * | 2023-07-12 | 2023-08-11 | 太极计算机股份有限公司 | Network task optimization method and system for machine learning classification |
Also Published As
| Publication number | Publication date |
|---|---|
| WO2022089031A1 (en) | 2022-05-05 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN112291706A (en) | Network optimization method based on big data and artificial intelligence | |
| CN110677859B (en) | Method and device for determining weak coverage area and computer readable storage medium | |
| CN104484993B (en) | Processing method of cell phone signaling information for dividing traffic zones | |
| CN109548041B (en) | Wireless coverage analysis method and system | |
| CN105282758A (en) | Self-adaptive dynamic construction method of WIFI indoor positioning system fingerprint database | |
| CN114173356B (en) | Network quality detection method, device, equipment and storage medium | |
| CN102332210A (en) | Method for extracting real-time urban road traffic flow data based on mobile phone positioning data | |
| CN108243435B (en) | Parameter optimization method and device in LTE cell scene division | |
| CN105813092B (en) | A kind of building network coverage optimization method and device based on ticket | |
| CN103530347A (en) | Internet resource quality assessment method and system based on big data mining | |
| CN113570867B (en) | Urban traffic state prediction method, device, equipment and readable storage medium | |
| CN110856186A (en) | Method and system for constructing wireless network knowledge graph | |
| CN107392252A (en) | Computer deep learning characteristics of image and the method for quantifying perceptibility | |
| CN111294730B (en) | Method and device for processing network problem complaint information | |
| CN111465025B (en) | Tourism city 5G network networking method based on novel capacity prediction model | |
| CN101835173B (en) | System and analysis method of TD-SCDMA (Time Division-Synchronization Code Division Multiple Access) wireless network based on geographical position coverage effect | |
| CN110493720A (en) | Localization method, device and the storage medium of terminal | |
| CN103440328B (en) | A kind of user classification method based on mouse behavior | |
| CN111163482B (en) | Data processing method, device and storage medium | |
| CN114184881B (en) | Fault event positioning method based on topology model tracking analysis | |
| CN101741992A (en) | GIS-based traffic distribution display method and GIS-based traffic distribution display system | |
| CN104684004A (en) | Complex wireless communication network operation quality evaluation method based on fuzzy analysis | |
| CN109995549B (en) | Method and device for evaluating flow value | |
| CN112422650B (en) | Building positioning method, building positioning device, building positioning equipment and computer readable storage medium | |
| CN112990605B (en) | Method and device for planning and compiling on whole flow line |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| RJ01 | Rejection of invention patent application after publication | ||
| RJ01 | Rejection of invention patent application after publication |
Application publication date: 20210129 |