CN113840225A - Terminal positioning method, device and storage medium - Google Patents
Terminal positioning method, device and storage medium Download PDFInfo
- Publication number
- CN113840225A CN113840225A CN202010579830.XA CN202010579830A CN113840225A CN 113840225 A CN113840225 A CN 113840225A CN 202010579830 A CN202010579830 A CN 202010579830A CN 113840225 A CN113840225 A CN 113840225A
- Authority
- CN
- China
- Prior art keywords
- data
- terminal
- information
- longitude
- training
- 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 40
- 238000012549 training Methods 0.000 claims abstract description 122
- 239000013598 vector Substances 0.000 claims description 64
- 239000011159 matrix material Substances 0.000 claims description 30
- 238000005259 measurement Methods 0.000 claims description 12
- 238000010276 construction Methods 0.000 claims description 8
- 238000004590 computer program Methods 0.000 claims description 5
- 238000010801 machine learning Methods 0.000 abstract description 3
- 238000010586 diagram Methods 0.000 description 12
- 239000000284 extract Substances 0.000 description 4
- 238000000605 extraction Methods 0.000 description 4
- 230000006870 function Effects 0.000 description 4
- 238000012545 processing Methods 0.000 description 4
- 238000004891 communication Methods 0.000 description 2
- 238000005516 engineering process Methods 0.000 description 2
- 238000012986 modification Methods 0.000 description 2
- 230000004048 modification Effects 0.000 description 2
- 230000008447 perception Effects 0.000 description 2
- 230000009286 beneficial effect Effects 0.000 description 1
- 238000004364 calculation method Methods 0.000 description 1
- 230000007547 defect Effects 0.000 description 1
- 238000012423 maintenance Methods 0.000 description 1
- 238000004519 manufacturing process Methods 0.000 description 1
- 239000000463 material Substances 0.000 description 1
- 238000012360 testing method Methods 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
-
- 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
- H04W24/10—Scheduling measurement reports ; Arrangements for measurement reports
-
- 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
Landscapes
- Engineering & Computer Science (AREA)
- Computer Networks & Wireless Communication (AREA)
- Signal Processing (AREA)
- Software Systems (AREA)
- Theoretical Computer Science (AREA)
- Computer Vision & Pattern Recognition (AREA)
- Evolutionary Computation (AREA)
- Medical Informatics (AREA)
- Data Mining & Analysis (AREA)
- Physics & Mathematics (AREA)
- Computing Systems (AREA)
- General Engineering & Computer Science (AREA)
- General Physics & Mathematics (AREA)
- Mathematical Physics (AREA)
- Artificial Intelligence (AREA)
- Position Fixing By Use Of Radio Waves (AREA)
- Mobile Radio Communication Systems (AREA)
Abstract
The disclosure provides a terminal positioning method, a device and a storage medium, wherein the method comprises the following steps: dividing MR data into MR data groups corresponding to all access cells according to the access cells of the terminal, extracting first MR data carrying latitude and longitude information of the terminal to generate a training sample, and training a terminal position prediction model; predicting second MR data which does not carry the terminal longitude and latitude information by using a trained terminal position prediction model and adding the predicted terminal longitude and latitude information; the method, the device and the medium can establish a corresponding terminal position prediction model for each access cell, can independently train, adjust or replace, and cannot influence other terminal position prediction models during adjustment, training or replacement; the terminal position prediction model is trained by using a machine learning method according to the collected training samples of different cells, manual adjustment is not needed, the cost can be reduced, the positioning accuracy is high, and the prediction time is short.
Description
Technical Field
The present invention relates to the field of communications technologies, and in particular, to a terminal positioning method, apparatus, and storage medium.
Background
The accurate positioning of the terminal position is beneficial to improving the use perception of the terminal user in the important area, and the limited resources can be rapidly put in the important area. At present, due to the limitation of terminal functions, only a small number of terminals report longitude and latitude through a radio measurement report MRO when the terminals are outdoors, the coverage degree is low, in addition, a terminal APP can encrypt the position information of a user and does not have a user number, so that an operator cannot acquire the position of the user, and the user perception real situation of an important area cannot be known. The existing terminal positioning technology has the following problems: the 4G network has low positioning accuracy due to the loss of related parameters caused by technical defects; the DT and CQT tests are required to be carried out periodically, and a large amount of manpower and material resources are consumed; each position record needs triangulation to perform complex calculation, the matching time is long during prediction, and the performance of equipment is consumed.
Disclosure of Invention
In view of the above, an object of the present invention is to provide a method, an apparatus and a storage medium for positioning a terminal.
According to a first aspect of the present disclosure, there is provided a terminal positioning method, including: acquiring measurement report MR data sent by a terminal, and dividing the MR data into MR data groups corresponding to access cells according to the access cells of the terminal; extracting first MR data carrying longitude and latitude information of a terminal and second MR data not carrying longitude and latitude position information from the MR data group, generating a training sample based on the first MR data and taking the second MR data as predicted target data; constructing a terminal position prediction model corresponding to the access cell, and training the terminal position prediction model by using the training sample; and predicting the longitude and latitude information of the terminal corresponding to the second MR data by using the trained terminal position prediction model, and adding the predicted longitude and latitude information of the terminal into the second MR data.
Optionally, the generating training samples based on the first MR data comprises: extracting information of the access cell from the first MR data; wherein the information of the access cell comprises: timing advance TA and first reference signal received power RSRP; extracting information of a neighbor cell from the first MR data; wherein the information of the neighbor cell includes: the second reference signal received power RSRP, the physical cell identifier PCI and the carrier frequency point number Earfcn; generating a neighbor cell identity based on a combination of the PCI and the Earfcn; generating feature information corresponding to the first MR data; wherein the feature information includes: the TA, the first RSRP, the second RSRP, and the neighbor cell identity; and generating the training sample according to the characteristic information.
Optionally, the generating the training sample according to the feature information includes: generating training feature vectors corresponding to the feature information; extracting terminal longitude and latitude information from first MR data corresponding to the training feature vector; and setting label information corresponding to the training characteristic vector based on the extracted longitude and latitude information of the terminal, and forming the training characteristic matrix by using the training characteristic vector to serve as the training sample.
Optionally, the setting of label information corresponding to the training feature vector based on the extracted longitude and latitude information of the terminal, and the forming of the training feature matrix using the training feature vector includes: acquiring first longitude and latitude information of the extracted longitude and latitude information of the terminal; setting label information corresponding to the training feature vector based on the first degree information, and forming a first feature matrix by using the training feature vector; setting label information corresponding to the training feature vector based on the first latitude information, and forming a second feature matrix by using the training feature vector.
Optionally, the terminal location prediction model includes: a longitude prediction model and a latitude prediction model; training the terminal position prediction model using the training samples comprises: the longitude prediction model is trained using the first feature matrix and the latitude prediction model is trained using the second feature matrix.
Optionally, the predicting, by using the trained terminal location prediction model, the terminal longitude and latitude information corresponding to the second MR data includes: predicting second longitude information corresponding to the second MR data using the trained longitude prediction model, and predicting second latitude information corresponding to the second MR data using the trained latitude prediction model.
Optionally, generating a predicted feature vector based on the second MR data, wherein a format of the predicted feature vector is the same as a format of the training feature vector; and inputting the predicted feature vector into the trained longitude prediction model and latitude prediction model respectively, and acquiring the second longitude information and the second latitude information respectively.
Optionally, the terminal location prediction model includes: LightGBM model.
According to a second aspect of the present disclosure, there is provided a terminal positioning device comprising: the report collection module is used for acquiring measurement report MR data sent by a terminal and dividing the MR data into MR data groups corresponding to the access cells according to the access cells of the terminal; the sample construction module is used for extracting first MR data carrying terminal longitude and latitude information and second MR data not carrying longitude and latitude position information from the MR data group, generating a training sample based on the first MR data and taking the second MR data as prediction target data; the model training module is used for constructing a terminal position prediction model corresponding to the access cell and training the terminal position prediction model by using the training sample; and the position prediction module is used for predicting the longitude and latitude information of the terminal corresponding to the second MR data by using the trained terminal position prediction model and adding the predicted longitude and latitude information of the terminal into the second MR data.
According to a third aspect of the present disclosure, there is provided a terminal positioning device comprising: a memory; and a processor coupled to the memory, the processor configured to perform the method as described above based on instructions stored in the memory.
According to a fourth aspect of the present disclosure, there is provided a computer readable storage medium storing computer instructions for execution by a processor to perform the method as described above.
According to the terminal positioning method, the terminal positioning device and the storage medium, MR data are divided into MR data groups corresponding to access cells according to the access cells of a terminal, first MR data carrying longitude and latitude information of the terminal are extracted to generate training samples, and a terminal position prediction model is trained; predicting second MR data which does not carry the terminal longitude and latitude information by using a trained terminal position prediction model and adding the predicted terminal longitude and latitude information; establishing a corresponding terminal position prediction model for each access cell, wherein the terminal position prediction model can be independently trained, adjusted or replaced, and other terminal position prediction models cannot be influenced during adjustment, training or replacement; the machine learning method is utilized to generate different training samples to train the terminal position prediction model, model parameters can be automatically adjusted, manual adjustment is not needed, cost can be reduced, positioning accuracy is high, prediction time is short, and consumption of equipment performance can be reduced.
Drawings
In order to more clearly illustrate the embodiments of the present disclosure or the technical solutions in the prior art, the drawings needed to be used in the description of the embodiments or the prior art will be briefly described below, it is obvious that the drawings in the following description are only some embodiments of the present disclosure, and other drawings can be obtained by those skilled in the art without inventive exercise.
Fig. 1 is a schematic flow chart diagram illustrating one embodiment of a terminal location method according to the present disclosure;
fig. 2 is a schematic flow chart of generating training samples in an embodiment of a terminal location method according to the present disclosure;
FIG. 3 is a block diagram of one embodiment of a terminal positioning device according to the present disclosure;
FIG. 4 is a block diagram representation of a sample construction module in one embodiment of a terminal positioning device according to the present disclosure;
fig. 5 is a block diagram of another embodiment of a terminal positioning device according to the present disclosure.
Detailed Description
The present disclosure now will be described more fully hereinafter with reference to the accompanying drawings, in which exemplary embodiments of the disclosure are shown. The technical solutions in the embodiments of the present disclosure will be clearly and completely described below with reference to the drawings in the embodiments of the present disclosure, and it is obvious that the described embodiments are only a part of the embodiments of the present disclosure, and not all of the embodiments. All other embodiments, which can be derived by a person skilled in the art from the embodiments disclosed herein without making any creative effort, shall fall within the protection scope of the present disclosure.
The terms "first", "second", and the like are used hereinafter only for descriptive distinction and not for other specific meanings.
Fig. 1 is a schematic flowchart of an embodiment of a terminal positioning method according to the present disclosure, as shown in fig. 1:
In one embodiment, the terminal may be a mobile phone, a tablet computer, or the like. The wireless access network side sends measurement control information to the terminal, and the terminal receives the measurement of the control information and sends measurement Report (measurement Report) MR data to the wireless access network side. The MR data records wireless measurement information such as access (serving) cell/neighbor cell ID, signal received power (RSRP), signal received quality (RSRQ), angle of arrival (AOA), transmit Power Headroom (PHR), etc. of the terminal during the service maintenance process. The MR data may be data automatically acquired by the LTE base station device, and is grouped according to the area code of the access cell in the MR data, and the MR data is divided into MR data groups corresponding to the access cells.
In the MR data groups corresponding to the access cells, according to whether the MR data carry the longitude and latitude information of the terminal or not, first MR data and second MR data are obtained in the MR data groups, a training sample is generated based on the first MR data, the second MR data is used as prediction target data, and the training sample can be a training feature matrix or the like.
And 103, constructing a terminal position prediction model corresponding to the access cell, and training the terminal position prediction model by using the training sample.
A terminal location prediction model can be constructed for each MR data set, i.e. for each access cell. The terminal position prediction model may be various, for example, the terminal position prediction model is a Light Gradient Boosting Machine (Light Gradient elevator) model or the like. The terminal position prediction model can be trained by using the training samples through the existing training method, and the optimal parameters of the model can be automatically learned.
And step 104, predicting the longitude and latitude information of the terminal corresponding to the second MR data by using the trained terminal position prediction model, and adding the predicted longitude and latitude information of the terminal into the second MR data.
And predicting the longitude and latitude information of the terminal corresponding to the second MR data through the trained terminal position prediction model, and automatically completing the longitude and latitude information in the MR data lacking the terminal longitude and latitude information to realize the positioning of all terminals.
In one embodiment, information of an access cell is extracted from the first MR data, and the information of the access cell includes timing advance ta (timing advance), first Reference Signal Receiving Power RSRP (Reference Signal Receiving Power), and the like. And extracting information of the neighboring cells from the first MR data, wherein the information of the neighboring cells comprises a second Reference Signal Received Power (RSRP), a Physical Cell Identity (PCI) of the neighboring cells, a carrier frequency point number Earfcn of the neighboring cells and the like. A neighbor cell identification is generated based on a combination of the PCI and Earfcn. Generating characteristic information corresponding to the first MR data, wherein the characteristic information comprises TA, first RSRP, second RSRP, adjacent cell identification and the like, and generating training samples according to the characteristic information.
Fig. 2 is a schematic flowchart of generating training samples in an embodiment of a terminal location method according to the present disclosure, as shown in fig. 2:
And 203, setting label information corresponding to the training characteristic vectors based on the extracted longitude and latitude information of the terminal, and forming a training characteristic matrix by using the training characteristic vectors to serve as a training sample.
In one embodiment, the training feature matrix may be constructed using first MR data corresponding to the same access cell, i.e. a corresponding training feature matrix is generated for one MR data set. Generating the feature information corresponding to the first MR data includes: a Timing Advance (TA) of an access cell (which can be a main access cell), a first RSRP of the access cell, and second RSRPs of a plurality of adjacent cells arranged according to occurrence times, and an adjacent cell identifier; the adjacent cell identifier is a combination of an adjacent cell physical cell identifier PCI and an adjacent cell carrier frequency point number Earfcn, and is used for identifying the adjacent cell.
Generating a training feature vector corresponding to the feature information, the elements of the training feature vector comprising: TA, a first RSRP, a plurality of second RSRPs; the training feature vector has a dimension of 2+ the number of the plurality of second RSRPs. Each second RSRP corresponds to a neighboring cell identifier (PCI combined with Earfcn), and the correspondence between the element (element value is the second RSRP) in the training feature vector and the neighboring cell identifier is stored, or the neighboring cell identifier may be added to the element corresponding to the second RSRP in the training feature vector, that is, the element value of the training feature vector may be the neighboring cell identifier + the second RSRP.
Extracting terminal longitude and latitude information from first MR data corresponding to the training feature vectors, setting label information corresponding to the training feature vectors based on the extracted terminal longitude and latitude information, forming all the training feature vectors into a training feature matrix, and training a terminal position prediction model (LightGBM model) by using the training feature matrix. The terminal position prediction model can be trained by using the training feature matrix by using the existing training method.
In one embodiment, first longitude information and first latitude information of the longitude and latitude information of the terminal extracted from the first MR data are acquired, label information corresponding to a training feature vector is set based on the first longitude information, namely the first longitude information is used as a label, and the training feature vector is used for forming a first feature matrix; label information corresponding to the training feature vector is set based on the first latitude information, i.e., the first latitude information is used as a label, and the training feature vector is used to constitute a second feature matrix.
The terminal position prediction model comprises a longitude prediction model and a latitude prediction model, and the longitude prediction model and the latitude prediction model are LightGBM models and the like. The longitude prediction model is trained using the first feature matrix and the latitude prediction model is trained using the second feature matrix.
The number of the first MR data corresponding to different access cells and the position of the terminal have great difference, the specific values and the numbers of the adjacent cells PCI and the adjacent cells Earfcn in the first MR data are different, and the dimensionality of the training feature vectors is different, so that two training models, namely a longitude prediction model and a latitude prediction model, are constructed for each access cell.
The longitude prediction model is trained by using the first feature matrix, the latitude prediction model is trained by using the second feature matrix, parameters of the longitude prediction model and parameters of the latitude prediction model are automatically adjusted and optimized, the longitude prediction model and the latitude prediction model of each access cell cannot be influenced mutually, and if the longitude prediction model and the latitude prediction model need to be adjusted in the later period, only training data of the corresponding access cell need to be added for training and automatically adjusting the model parameters, manual adjustment is not needed, and the most appropriate longitude prediction model and latitude prediction model are trained for each cell.
In one embodiment, the second longitude information corresponding to the second MR data is predicted using a trained longitude prediction model, and the second latitude information corresponding to the second MR data is predicted using a trained latitude prediction model. A predicted feature vector is generated based on the second MR data, the format of the predicted feature vector being the same as the format of the training feature vector, i.e. the dimensions of the predicted feature vector and the training feature vector, the meaning of each element representation, etc. are the same.
And inputting the predicted feature vector into the trained longitude prediction model and latitude prediction model respectively, and acquiring second longitude information and second latitude information respectively. The latitude and longitude information of the terminal corresponding to the second MR data can be predicted by using the trained terminal position prediction model by using the existing method. And the terminal is positioned by predicting the position information of the terminal which does not upload the latitude and longitude information and completing the position information of the terminal on the second MR data which lacks the latitude and longitude information.
In one embodiment, as shown in fig. 3, the present disclosure provides a terminal location apparatus 30 comprising a report collection module 31, a sample construction module 32, a model training module 33, and a location prediction module 34. The report collection module 31 obtains measurement report MR data sent by the terminal, and divides the MR data into MR data groups corresponding to each access cell according to the access cell of the terminal. The sample construction module 32 extracts first MR data carrying the longitude and latitude information of the terminal and second MR data not carrying the longitude and latitude position information from the MR data group, generates a training sample based on the first MR data, and uses the second MR data as prediction target data.
The model training module 33 constructs a terminal position prediction model corresponding to the access cell, and trains the terminal position prediction model using the training samples. The position prediction module 34 predicts the terminal longitude and latitude information corresponding to the second MR data by using the trained terminal position prediction model, and adds the predicted terminal longitude and latitude information to the second MR data.
In one embodiment, as shown in FIG. 4, the sample construction module 32 includes: an information extraction unit 321, a feature construction unit 322, and a sample generation unit 323. The information extraction unit 321 extracts information of an access cell from the first MR data, the information of the access cell including: timing advance TA and first reference signal received power RSRP, etc.
The information extraction unit 321 extracts information of neighboring cells from the first MR data, the information of neighboring cells including the second reference signal received power RSRP, the physical cell identity PCI, the carrier frequency point number Earfcn, and the like, and the information extraction unit 321 generates a neighboring cell identity based on a combination of the PCI and the Earfcn.
The feature construction unit 322 generates feature information corresponding to the first MR data, the feature information including a TA, a first RSRP, a second RSRP, and a neighbor cell identifier. The sample generation unit 323 generates a training sample from the feature information.
In one embodiment, the sample generation unit 323 generates a training feature vector corresponding to the feature information, and extracts terminal latitude and longitude information from the first MR data corresponding to the training feature vector. The sample generation unit 323 sets label information corresponding to the training feature vectors based on the extracted terminal latitude and longitude information, and uses the training feature vectors to form a training feature matrix as a training sample.
The sample generating unit 323 acquires first longitude information and first latitude information of the extracted terminal longitude and latitude information, sets label information corresponding to a training feature vector based on the first longitude information, and forms a first feature matrix using the training feature vector. The sample generation unit 323 sets label information corresponding to a training feature vector based on the first latitude information, and composes a second feature matrix using the training feature vector.
In one embodiment, the terminal location prediction model includes a longitude prediction model and a latitude prediction model; the model training module 33 trains the longitude prediction model using the first feature matrix and trains the latitude prediction model using the second feature matrix. The location prediction module 34 predicts second longitude information corresponding to the second MR data using the trained longitude prediction model and predicts second latitude information corresponding to the second MR data using the trained latitude prediction model.
The location prediction module 34 generates a predicted feature vector based on the second MR data, the predicted feature vector having a format that is the same as a format of the training feature vector. The position prediction module 34 inputs the predicted feature vectors into the trained longitude prediction model and latitude prediction model, respectively, and obtains second longitude information and second latitude information, respectively.
Fig. 5 is a block diagram of another embodiment of a terminal positioning device according to the present disclosure. As shown in fig. 5, the apparatus may include a memory 51, a processor 52, a communication interface 53, and a bus 54. The memory 51 is used for storing instructions, the processor 52 is coupled to the memory 51, and the processor 52 is configured to execute the terminal positioning method implemented above based on the instructions stored in the memory 51.
The memory 51 may be a high-speed RAM memory, a non-volatile memory (non-volatile memory), or the like, and the memory 51 may be a memory array. The storage 51 may also be partitioned and the blocks may be combined into virtual volumes according to certain rules. The processor 52 may be a central processing unit CPU, or an application Specific Integrated circuit asic, or one or more Integrated circuits configured to implement the terminal location method of the present disclosure.
In one embodiment, the present disclosure provides a computer-readable storage medium storing computer instructions for a processor to perform a terminal positioning method as in any one of the above embodiments.
The present disclosure is described with reference to flowchart illustrations and/or block diagrams of methods, apparatus (systems) and computer program products according to embodiments of the disclosure. It will be understood that each flow and/or block of the flow diagrams and/or block diagrams, and combinations of flows and/or blocks in the flow diagrams and/or block diagrams, can be implemented by computer program instructions. These computer program instructions may be provided to a processor of a general purpose computer, special purpose computer, embedded processor, or other programmable data processing apparatus to produce a machine, such that the instructions, which execute via the processor of the computer or other programmable data processing apparatus, create means for implementing the functions specified in the flowchart flow or flows and/or block diagram block or blocks.
These computer program instructions may also be stored in a computer-readable memory that can direct a computer or other programmable data processing apparatus to function in a particular manner, such that the instructions stored in the computer-readable memory produce an article of manufacture including instruction means which implement the function specified in the flowchart flow or flows and/or block diagram block or blocks.
According to the terminal positioning method, the terminal positioning device and the storage medium in the embodiment, the MR data are divided into MR data groups corresponding to the access cells according to the access cells of the terminal, the first MR data carrying the longitude and latitude information of the terminal are extracted to generate training samples, and a terminal position prediction model is trained; predicting second MR data which does not carry the terminal longitude and latitude information by using a trained terminal position prediction model and adding the predicted terminal longitude and latitude information; establishing a corresponding terminal position prediction model for each access cell, wherein the terminal position prediction model can be independently trained, adjusted or replaced, and other terminal position prediction models cannot be influenced during adjustment, training or replacement; the terminal position prediction model is trained by using a machine learning method and generating different training samples, so that the model parameters can be automatically adjusted, manual adjustment is not needed, and the cost can be reduced; the longitude and latitude prediction model can be trained for each access cell, the positioning accuracy is low, the prediction time is short, and the consumption of the equipment performance can be reduced.
The method and system of the present disclosure may be implemented in a number of ways. For example, the methods and systems of the present disclosure may be implemented by software, hardware, firmware, or any combination of software, hardware, and firmware. The above-described order for the steps of the method is for illustration only, and the steps of the method of the present disclosure are not limited to the order specifically described above unless specifically stated otherwise. Further, in some embodiments, the present disclosure may also be embodied as programs recorded in a recording medium, the programs including machine-readable instructions for implementing the methods according to the present disclosure. Thus, the present disclosure also covers a recording medium storing a program for executing the method according to the present disclosure.
The description of the present disclosure has been presented for purposes of illustration and description, and is not intended to be exhaustive or limited to the disclosure in the form disclosed. Many modifications and variations will be apparent to practitioners skilled in this art. The embodiment was chosen and described in order to best explain the principles of the disclosure and the practical application, and to enable others of ordinary skill in the art to understand the disclosure for various embodiments with various modifications as are suited to the particular use contemplated.
Claims (11)
1. A terminal positioning method comprises the following steps:
acquiring measurement report MR data sent by a terminal, and dividing the MR data into MR data groups corresponding to access cells according to the access cells of the terminal;
extracting first MR data carrying longitude and latitude information of a terminal and second MR data not carrying longitude and latitude position information from the MR data group, generating a training sample based on the first MR data and taking the second MR data as predicted target data;
constructing a terminal position prediction model corresponding to the access cell, and training the terminal position prediction model by using the training sample;
and predicting the longitude and latitude information of the terminal corresponding to the second MR data by using the trained terminal position prediction model, and adding the predicted longitude and latitude information of the terminal into the second MR data.
2. The method of claim 1, the generating training samples based on the first MR data comprising:
extracting information of the access cell from the first MR data; wherein the information of the access cell comprises: timing advance TA and first reference signal received power RSRP;
extracting information of a neighbor cell from the first MR data; wherein the information of the neighbor cell includes: the second reference signal received power RSRP, the physical cell identifier PCI and the carrier frequency point number Earfcn;
generating a neighbor cell identity based on a combination of the PCI and the Earfcn;
generating feature information corresponding to the first MR data; wherein the feature information comprises the TA, the first RSRP, the second RSRP, and the neighbor cell identity;
and generating the training sample according to the characteristic information.
3. The method of claim 2, the generating the training sample from the feature information comprising:
generating training feature vectors corresponding to the feature information;
extracting terminal longitude and latitude information from first MR data corresponding to the training feature vector;
and setting label information corresponding to the training characteristic vector based on the extracted longitude and latitude information of the terminal, and forming the training characteristic matrix by using the training characteristic vector to serve as the training sample.
4. The method of claim 3, wherein the setting of label information corresponding to the training feature vectors based on the extracted terminal longitude and latitude information, and the forming of the training feature matrix using the training feature vectors comprises:
acquiring first longitude and latitude information of the extracted longitude and latitude information of the terminal;
setting label information corresponding to the training feature vector based on the first degree information, and forming a first feature matrix by using the training feature vector;
setting label information corresponding to the training feature vector based on the first latitude information, and forming a second feature matrix by using the training feature vector.
5. The method of claim 4, the terminal location prediction model comprising: a longitude prediction model and a latitude prediction model; training the terminal position prediction model using the training samples comprises:
the longitude prediction model is trained using the first feature matrix and the latitude prediction model is trained using the second feature matrix.
6. The method of claim 5, the predicting the terminal longitude and latitude information corresponding to the second MR data using the trained terminal location prediction model comprising:
predicting second longitude information corresponding to the second MR data using the trained longitude prediction model, and predicting second latitude information corresponding to the second MR data using the trained latitude prediction model.
7. The method of claim 6, further comprising:
generating a predicted feature vector based on the second MR data, wherein a format of the predicted feature vector is the same as a format of the training feature vector;
and inputting the predicted feature vector into the trained longitude prediction model and latitude prediction model respectively, and acquiring the second longitude information and the second latitude information respectively.
8. The method of any one of claims 1 to 7,
the terminal position prediction model includes: LightGBM model.
9. A terminal positioning device, comprising:
the report collection module is used for acquiring measurement report MR data sent by a terminal and dividing the MR data into MR data groups corresponding to the access cells according to the access cells of the terminal;
the sample construction module is used for extracting first MR data carrying terminal longitude and latitude information and second MR data not carrying longitude and latitude position information from the MR data group, generating a training sample based on the first MR data and taking the second MR data as prediction target data;
the model training module is used for constructing a terminal position prediction model corresponding to the access cell and training the terminal position prediction model by using the training sample;
and the position prediction module is used for predicting the longitude and latitude information of the terminal corresponding to the second MR data by using the trained terminal position prediction model and adding the predicted longitude and latitude information of the terminal into the second MR data.
10. A terminal positioning device, comprising:
a memory; and a processor coupled to the memory, the processor configured to perform the method of any of claims 1-8 based on instructions stored in the memory.
11. A computer readable storage medium having stored thereon computer program instructions which, when executed by a processor, implement the steps of the method of any one of claims 1 to 8.
Priority Applications (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN202010579830.XA CN113840225A (en) | 2020-06-23 | 2020-06-23 | Terminal positioning method, device and storage medium |
Applications Claiming Priority (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN202010579830.XA CN113840225A (en) | 2020-06-23 | 2020-06-23 | Terminal positioning method, device and storage medium |
Publications (1)
Publication Number | Publication Date |
---|---|
CN113840225A true CN113840225A (en) | 2021-12-24 |
Family
ID=78963988
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
CN202010579830.XA Pending CN113840225A (en) | 2020-06-23 | 2020-06-23 | Terminal positioning method, device and storage medium |
Country Status (1)
Country | Link |
---|---|
CN (1) | CN113840225A (en) |
Cited By (1)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
WO2024099084A1 (en) * | 2022-11-07 | 2024-05-16 | 大唐移动通信设备有限公司 | Data processing method, and apparatus |
Citations (4)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN101213855A (en) * | 2005-04-29 | 2008-07-02 | 艾利森电话股份有限公司 | Method and device relating to satellite-based positioning |
CN109688535A (en) * | 2018-12-13 | 2019-04-26 | 大唐软件技术股份有限公司 | Determine method, apparatus, electronic equipment and the medium in hotspot geographical location |
CN109996168A (en) * | 2017-12-28 | 2019-07-09 | ***通信集团吉林有限公司 | A kind of method and device obtaining terminal location |
US20200151613A1 (en) * | 2018-11-09 | 2020-05-14 | Lunit Inc. | Method and apparatus for machine learning |
-
2020
- 2020-06-23 CN CN202010579830.XA patent/CN113840225A/en active Pending
Patent Citations (4)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN101213855A (en) * | 2005-04-29 | 2008-07-02 | 艾利森电话股份有限公司 | Method and device relating to satellite-based positioning |
CN109996168A (en) * | 2017-12-28 | 2019-07-09 | ***通信集团吉林有限公司 | A kind of method and device obtaining terminal location |
US20200151613A1 (en) * | 2018-11-09 | 2020-05-14 | Lunit Inc. | Method and apparatus for machine learning |
CN109688535A (en) * | 2018-12-13 | 2019-04-26 | 大唐软件技术股份有限公司 | Determine method, apparatus, electronic equipment and the medium in hotspot geographical location |
Cited By (1)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
WO2024099084A1 (en) * | 2022-11-07 | 2024-05-16 | 大唐移动通信设备有限公司 | Data processing method, and apparatus |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
US10542519B2 (en) | Terminal positioning method and network device | |
CN109982366B (en) | Target value area analysis method, device, equipment and medium based on big data | |
CN106658399B (en) | Method for positioning mobile phone position based on mobile phone user position fingerprint | |
CN108024269B (en) | Method and device for sending and receiving cell measurement configuration information | |
CN109996168B (en) | Method and device for acquiring terminal position | |
CN111327450B (en) | Method, device, equipment and medium for determining quality difference reason | |
US20190200318A1 (en) | Supporting an update of stored information | |
CN110493720B (en) | Terminal positioning method and device and storage medium | |
CN102685671A (en) | Minimization drive test (MDT) result reporting and reporting instruction methods and devices | |
CN111372183B (en) | Method, device, equipment and storage medium for identifying poor terminal | |
CN105376089A (en) | Network planning method and device | |
CN103686818A (en) | Simulation test method and equipment | |
CN111586740A (en) | Method for configuring minimization of drive tests and base station | |
CN102457894A (en) | Cell switching method, measuring method, device and system | |
CN109996270B (en) | Network coverage analysis method, device, equipment and medium | |
CN108401222B (en) | Positioning method and device | |
CN113840225A (en) | Terminal positioning method, device and storage medium | |
CN113645625A (en) | Pseudo base station positioning method and device, electronic equipment and readable medium | |
CN113052308B (en) | Method for training target cell identification model and target cell identification method | |
EP3232719B1 (en) | Methods, apparatuses, system and computer-readable storage medium for processing information used for positioning | |
CN108632740B (en) | Positioning method and device of user equipment | |
CN114745289A (en) | Method, device, storage medium and equipment for predicting network performance data | |
WO2023018943A1 (en) | Determining geolocation of devices in a communication network | |
CN111491255B (en) | Backfill longitude and latitude positioning and expanding method, system and storage medium | |
CN114205737A (en) | Base station cell location identification method, device and server |
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 |