CN117241374A - Method, device and medium for establishing mobile network fingerprint positioning model - Google Patents
Method, device and medium for establishing mobile network fingerprint positioning model Download PDFInfo
- Publication number
- CN117241374A CN117241374A CN202311140325.5A CN202311140325A CN117241374A CN 117241374 A CN117241374 A CN 117241374A CN 202311140325 A CN202311140325 A CN 202311140325A CN 117241374 A CN117241374 A CN 117241374A
- Authority
- CN
- China
- Prior art keywords
- grid
- data
- fingerprint positioning
- geographic
- topography
- 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
- 238000000034 method Methods 0.000 title claims abstract description 60
- 230000001788 irregular Effects 0.000 claims abstract description 74
- 238000012876 topography Methods 0.000 claims abstract description 57
- 238000012549 training Methods 0.000 claims abstract description 33
- 238000005070 sampling Methods 0.000 claims description 29
- 230000015654 memory Effects 0.000 claims description 17
- 238000010801 machine learning Methods 0.000 claims description 14
- 238000004590 computer program Methods 0.000 claims description 11
- 238000005259 measurement Methods 0.000 claims description 6
- 238000005516 engineering process Methods 0.000 description 12
- 238000010586 diagram Methods 0.000 description 9
- 238000011166 aliquoting Methods 0.000 description 5
- 238000004422 calculation algorithm Methods 0.000 description 3
- 230000006870 function Effects 0.000 description 3
- 241001148624 Areae Species 0.000 description 2
- 238000004364 calculation method Methods 0.000 description 2
- 238000012545 processing Methods 0.000 description 2
- 238000012827 research and development Methods 0.000 description 2
- 230000011218 segmentation Effects 0.000 description 2
- 238000012360 testing method Methods 0.000 description 2
- 239000003643 water by type Substances 0.000 description 2
- 230000007547 defect Effects 0.000 description 1
- 238000000605 extraction Methods 0.000 description 1
- 238000012986 modification Methods 0.000 description 1
- 230000004048 modification Effects 0.000 description 1
- 230000003287 optical effect Effects 0.000 description 1
- 230000008569 process Effects 0.000 description 1
- 239000000126 substance Substances 0.000 description 1
Landscapes
- Image Analysis (AREA)
Abstract
The application provides a method, a device and a medium for establishing a mobile network fingerprint positioning model, wherein the method comprises the following steps: collecting MDT data of a mobile network; acquiring Geographic Information System (GIS) topography/landform data of a target area, wherein the GIS topography/landform data comprises a topography/landform boundary list extracted based on different topography/landforms; performing geographic rasterization on the target area to obtain a grid boundary list; performing intersection operation on the topography/landform boundary list and the grid boundary list to generate an irregular grid region; and obtaining a fingerprint positioning model according to the MDT data and the irregular grid area. The method, the device and the medium can solve the problems of low positioning accuracy and high model training cost of the fingerprint positioning model caused by excessive equally divided grids and no obvious data characteristic at the grid boundary in the existing fingerprint positioning method based on rasterization.
Description
Technical Field
The present application relates to the field of network technologies, and in particular, to a method, an apparatus, and a medium for establishing a mobile network fingerprint positioning model.
Background
The mobile network terminal equipment and the position information of the user have high application value, so the industry develops a plurality of positioning technologies. Among these positioning technologies, the fingerprint positioning technology has the advantages of low cost, wide positioning range, no newly added hardware resources and the like, and has been widely applied to mobile networks such as 4G, 5G and the like in recent years.
However, the existing fingerprint positioning method based on rasterization has the following disadvantages:
1) Multiple geographic attributes may exist in a single grid region, and different wireless environment characteristics are caused, so that the training difficulty of a machine learning model is high, and the model positioning accuracy is reduced;
2) The equally divided geographic grids have no more information value, but increase the operation complexity of the fingerprint positioning model and reduce the accuracy of the model;
3) The equal grid joint has no data difference of any dimension, and time-varying wireless propagation environment and instability of wireless channels are superimposed, so that the generalization difficulty of the model is increased, the accuracy of model positioning is reduced, and the research and development difficulty and model training cost are increased.
Disclosure of Invention
The application aims to solve the technical problems of the prior art, and provides a method, a device and a medium for establishing a mobile network fingerprint positioning model, so as to at least solve the problems of low positioning accuracy and high model training cost of the fingerprint positioning model caused by excessive equally dividing grids in the existing fingerprint positioning method based on rasterization and no obvious data characteristic at the grid boundary.
In a first aspect, the present application provides a method for establishing a fingerprint positioning model of a mobile network, where the method includes:
collecting MDT data of a mobile network;
acquiring Geographic Information System (GIS) topography/landform data of a target area, wherein the GIS topography/landform data comprises a topography/landform boundary list extracted based on different topography/landforms;
performing geographic rasterization on the target area to obtain a grid boundary list;
performing intersection operation on the topography/landform boundary list and the grid boundary list to generate an irregular grid region;
and obtaining a fingerprint positioning model according to the MDT data and the irregular grid area.
Further, the obtaining geographic information system GIS terrain/topography data of the target area specifically includes:
acquiring GIS map information of a target area;
extracting topographic/geomorphic boundary information of different topographic/geomorphic features in the GIS map information based on an image recognition model;
and storing the extracted topographic/geomorphic boundary information according to a preset format to obtain the topographic/geomorphic boundary list.
Further, the performing geographic rasterization on the target area to obtain a grid boundary list specifically includes:
and carrying out geographic rasterization on the target area by adopting a preset rectangle or hexagon to obtain a grid boundary list.
Further, the obtaining a fingerprint positioning model according to the MDT data and the irregular grid area specifically includes:
adding a corresponding region identifier for each sampling point according to an irregular grid region in which each sampling point in the MDT data falls;
and inputting the MDT data added with the area identifier as a data set into a preset machine learning model for training to obtain the fingerprint positioning model.
Further, the adding a corresponding region identifier for each sampling point according to the irregular grid region in which each sampling point in the MDT data falls, specifically includes:
coding all the generated irregular grid areas to obtain area identifiers corresponding to each irregular grid area;
adding a column of irregular grid information column into the MDT data;
and adding an area identifier corresponding to each sampling point in the irregular grid information column according to the irregular grid area in which each sampling point in the MDT data falls.
Further, the machine learning model is an extreme gradient lift tree XGBoost model.
Further, after the obtaining the fingerprint positioning model according to the MDT data and the irregular grid region, the method further includes:
and inputting the acquired measurement report MR data without longitude and latitude information into the fingerprint positioning model to realize fingerprint positioning.
In a second aspect, the present application provides a device for establishing a fingerprint positioning model of a mobile network, including:
the data acquisition module is used for acquiring MDT data of the mobile network;
the geographic/geomorphic acquisition module is connected with the data acquisition module and is used for acquiring Geographic Information System (GIS) geographic/geomorphic data of a target area, wherein the GIS geographic/geomorphic data comprises a geographic/geomorphic boundary list extracted based on different geographic/geomorphic;
the geographic rasterization module is connected with the topography/landform acquisition module and is used for carrying out geographic rasterization on the target area to obtain a grid boundary list;
the irregular grid generation module is connected with the geographic rasterization module and is used for carrying out intersection operation on the topographic/geomorphic boundary list and the grid boundary list to generate an irregular grid region;
and the fingerprint positioning model building module is connected with the irregular grid generating module and is used for obtaining a fingerprint positioning model according to the MDT data and the irregular grid area.
In a third aspect, the present application provides a device for building a mobile network fingerprint positioning model, including a memory and a processor, where the memory stores a computer program, and the processor is configured to run the computer program to implement the method for building a mobile network fingerprint positioning model according to the first aspect.
In a fourth aspect, the present application provides a computer readable storage medium, where a computer program is stored, where the computer program when executed by a processor implements the method for building a mobile network fingerprint positioning model according to the first aspect.
The application provides a method, a device and a medium for establishing a mobile network fingerprint positioning model, which are used for firstly collecting MDT data of a mobile network; then, geographic information system GIS topography/landform data of the target area is obtained, wherein the GIS topography/landform data comprises a topography/landform boundary list extracted based on different topography/landforms; performing geographic rasterization on the target area to obtain a grid boundary list; finally, carrying out intersection operation on the topography/landform boundary list and the grid boundary list to generate an irregular grid region; and obtaining a fingerprint positioning model according to the MDT data and the irregular grid area. According to the application, an irregular grid area is automatically established through the landform/landform boundary and the grid boundary to replace the original halving grid method, and the method is applied to fingerprint positioning model training and application, so that the positioning accuracy of the fingerprint positioning model can be improved, the operation amount of model training is reduced, and the problems of low positioning accuracy and high model training cost of the fingerprint positioning model caused by excessive halving of the scenes in the grid and no obvious data characteristic of the grid boundary in the existing fingerprint positioning method based on the grid are solved.
Drawings
FIG. 1 is a schematic diagram of a prior art aliquoting grid;
fig. 2 is a flowchart of a method for establishing a mobile network fingerprint positioning model according to embodiment 1 of the present application;
FIG. 3 is a schematic diagram of a GIS map according to an embodiment of the present application;
FIG. 4 is a schematic illustration of a terrain/topography boundary according to an embodiment of the present application;
FIG. 5 is a schematic view of a grid boundary according to an embodiment of the present application;
FIG. 6 is a schematic view of an irregular grid according to an embodiment of the present application;
fig. 7 is a schematic structural diagram of a device for establishing a fingerprint positioning model of a mobile network according to embodiment 2 of the present application;
fig. 8 is a schematic structural diagram of a device for establishing a fingerprint positioning model of a mobile network according to embodiment 3 of the present application.
Detailed Description
In order to make the technical scheme of the present application better understood by those skilled in the art, the following detailed description of the embodiments of the present application will be given with reference to the accompanying drawings.
It is to be understood that the specific embodiments and figures described herein are merely illustrative of the application, and are not limiting of the application.
It is to be understood that the various embodiments of the application and the features of the embodiments may be combined with each other without conflict.
It is to be understood that only the portions relevant to the present application are shown in the drawings for convenience of description, and the portions irrelevant to the present application are not shown in the drawings.
It should be understood that each unit and module in the embodiments of the present application may correspond to only one physical structure, may be formed by a plurality of physical structures, or may be integrated into one physical structure.
It will be appreciated that the terms "first," "second," and the like in embodiments of the present application are used to distinguish between different objects or to distinguish between different processes on the same object, and are not used to describe a particular order of objects.
It will be appreciated that, without conflict, the functions and steps noted in the flowcharts and block diagrams of the present application may occur out of the order noted in the figures.
It is to be understood that the flowcharts and block diagrams of the present application illustrate the architecture, functionality, and operation of possible implementations of systems, apparatuses, devices, methods according to various embodiments of the present application. Where each block in the flowchart or block diagrams may represent a unit, module, segment, code, or the like, which comprises executable instructions for implementing the specified functions. Moreover, each block or combination of blocks in the block diagrams and flowchart illustrations can be implemented by hardware-based systems that perform the specified functions, or by combinations of hardware and computer instructions.
It should be understood that the units and modules related in the embodiments of the present application may be implemented by software, or may be implemented by hardware, for example, the units and modules may be located in a processor.
In order to enable those skilled in the art to better understand the technical solutions in the embodiments of the present application, some technical terms related to the embodiments of the present application are briefly described below.
1) Model positioning accuracy: refers to the proximity of position information (typically coordinates) to its actual position.
2) Model positioning accuracy: refers to whether the grid in which the model is positioned is the correct grid.
Summary of the application
The mobile network terminal equipment and the position information of the user have high application value, so the industry develops a plurality of positioning technologies. Among these positioning technologies, the fingerprint positioning technology has the advantages of low cost, wide positioning range, no newly added hardware resources and the like, and has been widely applied to mobile networks such as 4G, 5G and the like in recent years.
In the existing mobile network fingerprint positioning technology, data information such as coverage, interference and the like of a wireless network in a specific geographic position is generally collected, fingerprint library modeling is performed through algorithms such as machine learning and the like, finally, the similarity between MR (Measurement Report ) data newly reported by a terminal/user and the fingerprint library is compared, and the nearest fingerprint position is selected as the terminal position.
The mobile network fingerprint positioning technology can be divided into two types, namely indoor positioning and outdoor positioning. Because of the wide outdoor positioning space area, the existing outdoor fingerprint positioning technology generally performs grid processing on a GIS (Geographic Information System ) map, and divides the map into equal grids according to longitude and latitude, taking account of sparsity of terminal data, uncertainty of wireless signal propagation and cost of hardware computing capability. The grid size is customized according to fingerprint modeling algorithm capability, model training data precision and other factors. After geographic rasterization, fingerprint models are built for each grid through model training, and finally a complete fingerprint library is formed. The fingerprint positioning method based on rasterization has the positioning precision of about 20 meters, and the side length of a common rectangular grid is 30 meters to 100 meters. The rasterization method is a method of balancing positioning accuracy with system operation costs, and it positions a terminal or user to a geographic grid. This can meet the positioning requirements of some specific locations of interest to the user.
However, the existing fingerprint positioning method based on rasterization has the following disadvantages:
1) Multiple geographic attributes may exist in a single grid region, and different wireless environment characteristics are caused, so that the training difficulty of a machine learning model is high, and the model positioning accuracy is reduced;
2) The equally divided geographic grids have no more information value, but rather increase the operation complexity of the fingerprint positioning model and reduce the accuracy of the model.
3) The equal grid joint has no data difference of any dimension, and time-varying wireless propagation environment and instability of wireless channels are superimposed, so that the generalization difficulty of the model is increased, the accuracy of model positioning is reduced, and the research and development difficulty and model training cost are increased. Fig. 1 shows a schematic diagram of a conventional aliquoting grid, and as can be seen from fig. 1, the learning of a fingerprint positioning model is disturbed based on the aliquoting grid boundary due to the instability of a wireless channel.
Aiming at the technical problems, the application provides a method, a device and a medium for establishing a mobile network fingerprint positioning model, which automatically establish an irregular grid area through a topography/landform boundary and a grid boundary to replace the original aliquoting grid method, are applied to fingerprint positioning model training and application, can improve the positioning accuracy of the fingerprint positioning model, reduce the operation of model training, and solve the problems of low positioning accuracy and high model training cost of the fingerprint positioning model caused by excessive aliquoting of the grid internal scene and no obvious data characteristic of the grid boundary in the traditional fingerprint positioning method based on the grid.
Having described the basic principles of the present application, various non-limiting embodiments of the present application will now be described in detail with reference to the accompanying drawings.
Example 1:
the embodiment provides a method for establishing a fingerprint positioning model of a mobile network, as shown in fig. 2, the method includes:
step S101: MDT (Minimization Drive Test, minimization of drive tests) data for the mobile network is collected.
In this embodiment, the acquired MDT data of the mobile network at least includes the serving cell RSRP (Reference Signal Receiving Power, reference signal received power), all neighbor cell RSRP and location information reported by the terminal, and may further include information such as serving cell RSQR (Reference Signal Received Quality ), TA (Tracking Area), PCI (Physical Cell Identities physical cell ID), ul SINR (UpLink SINR), and the like. Server cells PCI, RSRP, RSRQ, TA and ul sinr for each sample point in the MDT data and PCI, RSRP information for up to 7 neighbors may be stored in one csv file.
It should be noted that, since the MDT data is a data acquisition mode of measurement information such as network coverage of a user terminal supported by 3Gpp protocol, the method has the advantages of wide acquisition range, large data volume and no additional hardware input cost, so that machine learning based on the MDT data is a technical method for training and constructing a fingerprint library with the lowest cost and the widest application of operators.
Step S102: and acquiring Geographic Information System (GIS) topography/landform data of the target area, wherein the GIS topography/landform data comprises a topography/landform boundary list extracted based on different topography/landforms.
In this embodiment, the target area is an area where a fingerprint positioning model needs to be established. Because of the obvious difference of the influence of different terrains/landforms on signal propagation loss, the following terrains/landforms data can be acquired firstly: open spaces, buildings, sparse tree areas, dense tree areas, viaducts, waters, roads, etc.
Optionally, the obtaining geographic information system GIS terrain/topography data of the target area specifically includes:
acquiring GIS map information of a target area;
extracting topographic/geomorphic boundary information of different topographic/geomorphic features in the GIS map information based on an image recognition model;
and storing the extracted topographic/geomorphic boundary information according to a preset format to obtain the topographic/geomorphic boundary list.
In this embodiment, a common electronic map (OpenStreetMap) is used, and geographic boundary information of different terrains/landforms is extracted based on an image recognition model (preferably, a semantic segmentation model is used). Taking the GIS map shown in fig. 3 as an example, four types of topography/landform scenes of a building, a forest, a road and an open area can be extracted to obtain a topography/landform boundary shown in fig. 4, and the extracted topography/landform boundary information is stored in a GeoJSON format and expressed as area i Each border encloses a topographical/geomorphic region.
In this embodiment, the GIS topography/landform data may also be directly collected based on a GIS information database with high precision (less than 5 meters precision), which has the disadvantage of higher cost of such map.
Step S103: and carrying out geographic rasterization on the target area to obtain a grid boundary list.
Specifically, the target area may be subjected to geographic rasterization by adopting a preset rectangle or hexagon to obtain a grid boundary list.
In the present embodiment, assuming that a rectangular geographic grid is used, the radius length of the grid is 25 meters (i.e., the side length is 50 meters), the resulting grid boundary is shown in fig. 5, and the grid boundary list may be expressed as area b i 。
Step S104: and carrying out intersection operation on the topographic/geomorphic boundary list and the grid boundary list to generate an irregular grid region.
In the present embodiment, the topography/relief boundary list area i With grid boundary list AreaB i Performing intersection operation to generate new irregular grid AreaX i ：
AreaX i ＝AreaE i ∩AreaB i
Wherein within each grid, the differently colored regions become new irregular grids, the resulting irregular grids may be as shown in fig. 6.
Step S105: and obtaining a fingerprint positioning model according to the MDT data and the irregular grid area.
In this embodiment, the fingerprint positioning model is trained through the irregular grid area corresponding to each sampling point in the MDT data, so that a trained fingerprint positioning model can be obtained.
Optionally, the obtaining a fingerprint positioning model according to the MDT data and the irregular grid area specifically includes:
adding a corresponding region identifier for each sampling point according to an irregular grid region in which each sampling point in the MDT data falls;
and inputting the MDT data added with the area identifier as a data set into a preset machine learning model for training to obtain the fingerprint positioning model.
In this embodiment, the machine learning model preferably employs an extreme gradient lifted tree XGBoost model. The XGBoost model can automatically use multithreading of the CPU to perform parallel calculation, has higher efficiency, and simultaneously uses a plurality of strategies (such as regularization terms) to prevent overfitting, and is superior to other algorithms in prediction accuracy.
Optionally, the adding a corresponding area identifier for each sampling point according to the irregular grid area in which each sampling point in the MDT data falls specifically includes:
coding all the generated irregular grid areas to obtain area identifiers corresponding to each irregular grid area;
adding a column of irregular grid information column into the MDT data;
and adding an area identifier corresponding to each sampling point in the irregular grid information column according to the irregular grid area in which each sampling point in the MDT data falls.
In this embodiment, by comparing the latitude and longitude information of each sampling point in the MDT data to which AreaX the latitude and longitude information falls i Within the region, the boundary information AreaX of the irregular grid is added for each sampling information i Preferably, areaX is added i Id (area identification) to replace AreaX i 。
Specifically, a column of irregular raster information column "polygon_Id" is added at the end of MDT data, and each sampling point (i.e. each row of data) and list AreaX in MDT data are searched in a traversing manner i If there is geographic overlap, then the region Id is added to the polygon_Id column.
And inputting MDT data added with the region identifier into an Xgboost model as a training data set, and training a fingerprint positioning model:
model＝xgb.XGBClassifier(max_depth＝4,min_child_weight＝1,gamma＝0.5,subsample＝0.8,colsample_bytree＝0.3,learning_rate＝0.1,n_estimators＝100)
model.fit(X_train,y_train)
wherein the newly added polygon_id is the target data y_train, and the other data is X_train.
Optionally, after the obtaining the fingerprint positioning model according to the MDT data and the irregular grid region, the method further includes:
and inputting the acquired measurement report MR data without longitude and latitude information into the fingerprint positioning model to realize fingerprint positioning.
In this embodiment, MR data without longitude and latitude in the region is acquired, and the data is input into a trained fingerprint positioning model, so that polyon_id information can be added to each MR sampling point data, thereby realizing fingerprint position positioning.
The method is applied to MR positioning of the mobile network, solves the problems of excessive scenes in the equally divided grids and no obvious data characteristics at the boundaries of the grids, and can improve the positioning accuracy of the fingerprint positioning model and reduce the calculation amount of model training.
In a specific embodiment, the method for establishing the mobile network fingerprint positioning model may include the following steps:
the first step: MDT data acquisition
Collecting MDT data of a mobile network: at least comprises a serving cell RSRP, all neighbor cell RSRP and position information reported by a terminal. Preferably, information such as serving cell RSQR, TA and the like can be further included.
And a second step of: acquiring GIS topographic/geomorphic data
A general mobile network fingerprint positioning model mainly performs learning modeling through a plurality of cell signals covering the same location, and therefore, a main factor affecting the modeling efficiency and generalization performance of the fingerprint positioning model is the influence of the location on signal propagation.
Among them, the influence of different topography/topography on signal propagation loss is obvious, and general we need to acquire the following topography/topography data: open space, buildings, sparse tree areas, dense tree areas, viaducts, and waters.
Other terrain divisions that have a greater impact on signal propagation may be added as a preference.
Specifically, the geographic/geomorphic data of the GIS may be obtained by any one of the following methods:
1) Based on the direct collection of a GIS information database with high precision (below 5 meters precision), the map has the defect of higher cost;
2) Based on a common electronic map, the image is extracted by an image recognition model (preferably adopting a semantic segmentation model)Taking the boundaries of the corresponding topography/relief areas, storing the extracted vector boundary data, and the set of vector boundary areas can be expressed as area i 。
And a third step of: intersection operation of topographic/geomorphic boundary and grid boundary
1) Geographical rasterization by adopting rectangle or hexagon, setting the radius length of the grid as d, and the grid area is expressed as AreB i 。
2) List of topographic/geomorphic boundaries area i With grid boundary list AreaB i Performing intersection operation to generate new irregular grid AreaX i ：
AreaX i ＝AreaE i ∩AreaB i
3) For the generated irregular grid AreaX i Encoding to generate AreaX i Id (region identification) to reduce complexity in model training.
Fourth step: based on generated AreaX i Mobile network fingerprint positioning model learning
1) By comparing the latitude and longitude information of each sampling point in the MDT data to which AreaX the latitude and longitude information falls i Within the region, the boundary information AreaX of the irregular grid is added for each sampling information i . Preferably adding AreaX i Id instead of AreaX i 。
2) And inputting the MDT data added with the irregular grid information column as a training data set into a machine learning model, and executing fingerprint positioning model training. (the application is not limited to what machine learning model is employed for model training).
Fifth step: application of fingerprint positioning model
And acquiring MR data of the mobile network, and applying a trained fingerprint positioning model to realize fingerprint positioning of the mobile network.
The method for establishing the mobile network fingerprint positioning model comprises the steps of firstly collecting MDT data of a mobile network; then, geographic information system GIS topography/landform data of the target area is obtained, wherein the GIS topography/landform data comprises a topography/landform boundary list extracted based on different topography/landforms; performing geographic rasterization on the target area to obtain a grid boundary list; finally, carrying out intersection operation on the topography/landform boundary list and the grid boundary list to generate an irregular grid region; and obtaining a fingerprint positioning model according to the MDT data and the irregular grid area. According to the application, an irregular grid area is automatically established through the landform/landform boundary and the grid boundary to replace the original halving grid method, and the method is applied to fingerprint positioning model training and application, so that the positioning accuracy of the fingerprint positioning model can be improved, the operation amount of model training is reduced, and the problems of low positioning accuracy and high model training cost of the fingerprint positioning model caused by excessive halving of the scenes in the grid and no obvious data characteristic of the grid boundary in the existing fingerprint positioning method based on the grid are solved.
Example 2:
as shown in fig. 7, the present embodiment provides a device for establishing a mobile network fingerprint positioning model, which is configured to execute the method for establishing a mobile network fingerprint positioning model, including:
the data acquisition module 11 is used for acquiring MDT data of the mobile network;
the topography/landform acquisition module 12 is connected with the data acquisition module 11 and is used for acquiring Geographic Information System (GIS) topography/landform data of a target area, wherein the GIS topography/landform data comprises a topography/landform boundary list extracted based on different topography/landforms;
the geographic rasterization module 13 is connected with the terrain/landform acquisition module 12 and is used for carrying out geographic rasterization on the target area to obtain a grid boundary list;
the irregular grid generating module 14 is connected with the geographic rasterization module 13 and is used for performing intersection operation on the terrain/landform boundary list and the grid boundary list to generate an irregular grid region;
and the fingerprint positioning model building module 15 is connected with the irregular grid generating module 14 and is used for obtaining a fingerprint positioning model according to the MDT data and the irregular grid area.
Optionally, the terrain/topography acquisition module 12 includes:
the map information acquisition unit is used for acquiring GIS map information of the target area;
the boundary information extraction unit is used for extracting the topographic/geomorphic boundary information of different topographic/geomorphic features in the GIS map information based on the image recognition model;
the storage unit is used for storing the extracted topographic/geomorphic boundary information according to a preset format to obtain the topographic/geomorphic boundary list.
Optionally, the geographic rasterization module 13 is specifically configured to:
and carrying out geographic rasterization on the target area by adopting a preset rectangle or hexagon to obtain a grid boundary list.
Optionally, the fingerprint positioning model building module 15 includes:
the region identification adding unit is used for adding a corresponding region identification for each sampling point according to the irregular grid region in which each sampling point in the MDT data falls;
and the model training unit is used for inputting the MDT data added with the area identifier as a data set into a preset machine learning model for training to obtain the fingerprint positioning model.
Optionally, the area identifier adding unit specifically includes:
the coding unit is used for coding all the generated irregular grid areas to obtain area identifiers corresponding to each irregular grid area;
a column adding unit, configured to add a column of irregular grid information column to the MDT data;
and the column assignment unit is used for adding the area identifier corresponding to each sampling point in the irregular grid information column according to the irregular grid area in which each sampling point in the MDT data falls.
Optionally, the machine learning model is an extreme gradient lift tree XGBoost model.
Optionally, the apparatus further comprises:
and the fingerprint positioning module is used for inputting the acquired measurement report MR data without longitude and latitude information into the fingerprint positioning model so as to realize fingerprint positioning.
Example 3:
referring to fig. 8, the present embodiment provides an apparatus for building a mobile network fingerprint positioning model, which includes a memory 21 and a processor 22, wherein the memory 21 stores a computer program, and the processor 22 is configured to execute the computer program to perform the method for building a mobile network fingerprint positioning model in embodiment 1.
The memory 21 is connected to the processor 22, the memory 21 may be a flash memory, a read-only memory, or other memories, and the processor 22 may be a central processing unit or a single chip microcomputer.
Example 4:
the present embodiment provides a computer-readable storage medium having a computer program stored thereon, which when executed by a processor, implements the method for establishing a mobile network fingerprint positioning model in embodiment 1 described above.
Computer-readable storage media include volatile or nonvolatile, removable or non-removable media implemented in any method or technology for storage of information such as computer-readable instructions, data structures, computer program modules or other data. Computer-readable storage media includes, but is not limited to, RAM (Random Access Memory ), ROM (Read-Only Memory), EEPROM (Electrically Erasable Programmable Read Only Memory, charged erasable programmable Read-Only Memory), flash Memory or other Memory technology, CD-ROM (Compact Disc Read-Only Memory), digital Versatile Disks (DVD) or other optical disk storage, magnetic cassettes, magnetic tape, magnetic disk storage or other magnetic storage devices, or any other medium which can be used to store the desired information and which can be accessed by a computer.
In summary, the method, the device and the medium for establishing the mobile network fingerprint positioning model provided by the embodiment of the application collect MDT data of the mobile network; then, geographic information system GIS topography/landform data of the target area is obtained, wherein the GIS topography/landform data comprises a topography/landform boundary list extracted based on different topography/landforms; performing geographic rasterization on the target area to obtain a grid boundary list; finally, carrying out intersection operation on the topography/landform boundary list and the grid boundary list to generate an irregular grid region; and obtaining a fingerprint positioning model according to the MDT data and the irregular grid area. According to the application, an irregular grid area is automatically established through the landform/landform boundary and the grid boundary to replace the original halving grid method, and the method is applied to fingerprint positioning model training and application, so that the positioning accuracy of the fingerprint positioning model can be improved, the operation amount of model training is reduced, and the problems of low positioning accuracy and high model training cost of the fingerprint positioning model caused by excessive halving of the scenes in the grid and no obvious data characteristic of the grid boundary in the existing fingerprint positioning method based on the grid are solved.
It is to be understood that the above embodiments are merely illustrative of the application of the principles of the present application, but not in limitation thereof. Various modifications and improvements may be made by those skilled in the art without departing from the spirit and substance of the application, and are also considered to be within the scope of the application.
Claims (10)
1. A method for establishing a mobile network fingerprint positioning model, the method comprising:
collecting MDT data of a mobile network;
acquiring Geographic Information System (GIS) topography/landform data of a target area, wherein the GIS topography/landform data comprises a topography/landform boundary list extracted based on different topography/landforms;
performing geographic rasterization on the target area to obtain a grid boundary list;
performing intersection operation on the topography/landform boundary list and the grid boundary list to generate an irregular grid region;
and obtaining a fingerprint positioning model according to the MDT data and the irregular grid area.
2. The method according to claim 1, wherein the obtaining geographic information system GIS terrain/topography data of the target area specifically comprises:
acquiring GIS map information of a target area;
extracting topographic/geomorphic boundary information of different topographic/geomorphic features in the GIS map information based on an image recognition model;
and storing the extracted topographic/geomorphic boundary information according to a preset format to obtain the topographic/geomorphic boundary list.
3. The method according to claim 1, wherein the performing geographic rasterization on the target area to obtain a grid boundary list specifically includes:
and carrying out geographic rasterization on the target area by adopting a preset rectangle or hexagon to obtain a grid boundary list.
4. The method according to claim 1, wherein the deriving a fingerprint positioning model from the MDT data and the irregular grid region comprises:
adding a corresponding region identifier for each sampling point according to an irregular grid region in which each sampling point in the MDT data falls;
and inputting the MDT data added with the area identifier as a data set into a preset machine learning model for training to obtain the fingerprint positioning model.
5. The method according to claim 4, wherein the adding a corresponding region identifier for each sampling point according to the irregular grid region in which each sampling point in the MDT data falls, specifically includes:
coding all the generated irregular grid areas to obtain area identifiers corresponding to each irregular grid area;
adding a column of irregular grid information column into the MDT data;
and adding an area identifier corresponding to each sampling point in the irregular grid information column according to the irregular grid area in which each sampling point in the MDT data falls.
6. The method of claim 4, wherein the machine learning model is an extreme gradient lifted tree XGBoost model.
7. The method of claim 1, wherein after the deriving a fingerprint positioning model from the MDT data and the irregular grid region, the method further comprises:
and inputting the acquired measurement report MR data without longitude and latitude information into the fingerprint positioning model to realize fingerprint positioning.
8. The device for establishing the fingerprint positioning model of the mobile network is characterized by comprising the following components:
the data acquisition module is used for acquiring MDT data of the mobile network;
the geographic/geomorphic acquisition module is connected with the data acquisition module and is used for acquiring Geographic Information System (GIS) geographic/geomorphic data of a target area, wherein the GIS geographic/geomorphic data comprises a geographic/geomorphic boundary list extracted based on different geographic/geomorphic;
the geographic rasterization module is connected with the topography/landform acquisition module and is used for carrying out geographic rasterization on the target area to obtain a grid boundary list;
the irregular grid generation module is connected with the geographic rasterization module and is used for carrying out intersection operation on the topographic/geomorphic boundary list and the grid boundary list to generate an irregular grid region;
and the fingerprint positioning model building module is connected with the irregular grid generating module and is used for obtaining a fingerprint positioning model according to the MDT data and the irregular grid area.
9. A mobile network fingerprint positioning model building device, comprising a memory and a processor, the memory storing a computer program, the processor being arranged to run the computer program to implement a mobile network fingerprint positioning model building method according to any of claims 1-7.
10. A computer readable storage medium, characterized in that the computer readable storage medium has stored thereon a computer program which, when executed by a processor, implements a method of establishing a mobile network fingerprint positioning model according to any of claims 1-7.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202311140325.5A CN117241374A (en) | 2023-09-05 | 2023-09-05 | Method, device and medium for establishing mobile network fingerprint positioning model |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202311140325.5A CN117241374A (en) | 2023-09-05 | 2023-09-05 | Method, device and medium for establishing mobile network fingerprint positioning model |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN117241374A true CN117241374A (en) | 2023-12-15 |
Family
ID=89097681
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202311140325.5A Pending CN117241374A (en) | 2023-09-05 | 2023-09-05 | Method, device and medium for establishing mobile network fingerprint positioning model |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN117241374A (en) |
-
2023
- 2023-09-05 CN CN202311140325.5A patent/CN117241374A/en active Pending
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| US11206555B2 (en) | Method for implementing antenna azimuth correction based on user data | |
| CN109493119B (en) | POI data-based urban business center identification method and system | |
| CN110956412B (en) | Flood dynamic assessment method, device, medium and equipment based on real-scene model | |
| CN111107556B (en) | Signal coverage quality evaluation method and device of mobile communication network | |
| CN106951828B (en) | Urban area function attribute identification method based on satellite images and network | |
| CN111291145B (en) | Mapping method, device and storage medium of wireless hotspot and interest point | |
| CN113505769B (en) | Target detection method and vehicle throwing and dripping identification method applying same | |
| CN107392252A (en) | Computer deep learning characteristics of image and the method for quantifying perceptibility | |
| CN110298687B (en) | Regional attraction assessment method and device | |
| CN110798804A (en) | Indoor positioning method and device | |
| Gervasoni et al. | Convolutional neural networks for disaggregated population mapping using open data | |
| CN114722944A (en) | Point cloud precision determination method, electronic device and computer storage medium | |
| CN112381078B (en) | Elevated-based road identification method, elevated-based road identification device, computer equipment and storage medium | |
| US20240356663A1 (en) | Methods and apparatus for determining above ground coverage of a mobile telecommunications network | |
| CN104662540A (en) | Method and apparatus for region sampling and estimation in location based networks | |
| CN115861408A (en) | Unmanned aerial vehicle road surface pit inspection method based on laser point tracking and application thereof | |
| CN113141570A (en) | Underground scene positioning method and device, computing equipment and computer storage medium | |
| CN111491255A (en) | Backfill longitude and latitude positioning and expanding method, system and storage medium | |
| CN112770326B (en) | LoRa gateway deployment method, device, equipment and storage medium | |
| CN111080080B (en) | Village geological disaster risk prediction method and system | |
| CN117241374A (en) | Method, device and medium for establishing mobile network fingerprint positioning model | |
| CN105848177B (en) | A kind of method and system obtaining antenna attitude information | |
| US20180181879A1 (en) | Outdoor localization using hotspot labels | |
| CN117062220A (en) | Mobile network fingerprint positioning method, device and readable storage medium | |
| CN114867033B (en) | Cell position determining method and device |
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 |