A base station latitude and longitude estimation method and apparatus

Abstract

This disclosure provides a method and apparatus for estimating the latitude and longitude of a base station. The method includes acquiring measurement and statistical MR data of multiple cells covered by the base station to be estimated. The MR data includes at least one of the following: the identifier of the base station to be estimated, the received reference signal strength (RSRP), and the timing advance (TA). The MR data is preprocessed to obtain a set of MR sample points for the base station to be estimated. Based on RSRP or TA, a distance calculation model and an objective function are established. The distance calculation model is used to calculate the distance between the multiple MR sample points and the base station to be estimated. Based on the distance calculation model and the objective function, a gradient descent algorithm is used to determine the longitude and latitude of the base station to be estimated. This method reduces the amount of data required for solving the problem by preprocessing the MR data to obtain the set of MR sample points. The gradient descent algorithm is used to iteratively solve the objective function to obtain the longitude and latitude of the base station to be estimated, improving the speed of calculating the latitude and longitude of the base station and the accuracy of the measurement results.

Description

Technical Field

[0001] This disclosure relates to the field of wireless communication, and in particular to a method and apparatus for estimating the latitude and longitude of a base station. Background Technology

[0002] In the field of wireless communication technology, the latitude and longitude of a base station are crucial engineering parameters, greatly aiding in network planning, signal coverage, and troubleshooting. Common techniques for calculating base station latitude and longitude involve transforming MR (Radar) measurement data into planar coordinates, solving for the transformed data, and then converting the results back to base station latitude and longitude. Alternatively, MR data can be combined with triangulation to estimate the base station's latitude and longitude. However, these methods suffer from high computational complexity and low efficiency due to the coordinate transformation involved, numerous matrix inverse operations, and the limitations of triangulation, which relies on the cell azimuth parameter. Furthermore, the inaccuracies introduced by cell azimuth recording deviations further complicate matters. Summary of the Invention

[0003] This disclosure provides a method and apparatus for estimating the latitude and longitude of a base station, in order to solve the problems in related technologies, reduce the amount of data used for solving, and improve the speed of solving the latitude and longitude of the base station and the accuracy of the measurement results.

[0004] The first aspect of this disclosure proposes a method for estimating the latitude and longitude of a base station. The method includes: acquiring measurement statistics (MR) data of multiple cells covered by the base station to be estimated, wherein the MR data includes at least one of the following: the identifier of the base station to be estimated, the received reference signal strength (RSRP), and the timing advance (TA); preprocessing the MR data to obtain a set of MR sample points for the base station to be estimated; establishing a distance calculation model and an objective function based on RSRP or TA, wherein the distance calculation model is used to calculate the distance between the multiple MR sample points and the base station to be estimated; and determining the longitude and latitude of the base station to be estimated using a gradient descent algorithm based on the distance calculation model and the objective function.

[0005] In some embodiments of this disclosure, preprocessing MR data to obtain multiple MR sample points for multiple cells includes: using an anomaly filtering algorithm to remove abnormal data from the MR data; allocating the MR data to a grid of a preset size to obtain multiple MR sample points, wherein the grid information of each MR sample point includes the longitude and latitude of the MR sample point; and determining the set of MR sample points for the base station to be estimated based on the multiple MR sample points.

[0006] In some embodiments of this disclosure, determining the MR sample point set of the base station to be estimated based on multiple MR sample points includes: determining the MR sample point set corresponding to each of multiple cells; determining the intersection of the MR sample point sets corresponding to each of the multiple cells as the MR sample point set of the base station to be estimated; and determining the MR sample point set corresponding to each of the multiple cells as the MR sample point set of the base station to be estimated when the intersection of the MR sample point sets corresponding to each of the multiple cells is an empty set.

[0007] In some embodiments of this disclosure, the MR data further includes at least one of the following: Physical Cell Identifier (PCI), Cell Antenna Frequency (arfcn), Sub-cell User Longitude, and Sub-cell User Latitude.

[0008] In some embodiments of this disclosure, establishing a distance calculation model based on RSRP or TA includes: using a deep neural network with a fully connected layer containing multiple hidden layers to fit the relationship between input data and output distance based on RSRP or TA to obtain a distance calculation model, wherein the input data includes a set of MR sample points of the base station to be estimated.

[0009] In some embodiments of this disclosure, when RSRP is used to establish a distance calculation model, the deep neural network consists of a fully connected layer containing four hidden layers, and when TA is used to establish a distance calculation model, the deep neural network consists of a fully connected layer containing two hidden layers.

[0010] In some embodiments of this disclosure, establishing an objective function based on RSRP or TA includes: determining the distance d between any two points according to the semi-sine formula; letting the longitude and latitude of the base station to be estimated be , and determining the distance D between the base station to be estimated and the MR sample point according to the longitude and latitude of the MR sample point corresponding to the base station to be estimated; wherein, the objective function is used to minimize the error between d and D.

[0011] In some embodiments of this disclosure, the determination of the longitude and latitude of the base station to be estimated using the gradient descent algorithm based on the distance calculation model and objective function includes: determining the initial longitude and initial latitude of the base station to be estimated; determining the gradient descent step size; iterating from the initial longitude and initial latitude of the base station to be estimated with the gradient descent step size; and determining the longitude and latitude corresponding to when the algorithm reaches the maximum number of iterations or the error meets the preset conditions as the longitude and latitude of the base station to be estimated.

[0012] A second aspect of this disclosure provides a base station latitude and longitude estimation device, comprising: an acquisition unit for acquiring measurement statistics MR data of multiple cells covered by a base station to be estimated, wherein the MR data includes at least one of the following: the identifier of the base station to be estimated, the received reference signal strength (RSRP), and the timing advance (TA); a first processing unit for preprocessing the MR data to obtain a set of MR sample points for the base station to be estimated; a second processing unit for establishing a distance calculation model and an objective function based on RSRP or TA, wherein the distance calculation model is used to calculate the distance between the multiple MR sample points and the base station to be estimated; and a third processing unit for determining the longitude and latitude of the base station to be estimated using a gradient descent algorithm based on the distance calculation model and the objective function.

[0013] A third aspect of this disclosure provides a communication device, comprising: at least one processor; and a memory communicatively connected to the at least one processor; wherein the memory stores instructions executable by the at least one processor, the instructions being executed by the at least one processor to enable the at least one processor to perform the methods described in the first aspect of this disclosure.

[0014] A fourth aspect of this disclosure provides a non-transitory computer-readable storage medium storing computer instructions, wherein the computer instructions are used to cause a computer to perform the methods described in the first aspect of this disclosure.

[0015] In summary, the base station latitude and longitude estimation method proposed in this disclosure includes: acquiring measurement statistics MR data of multiple cells covered by the base station to be estimated, wherein the MR data includes at least one of the following: the identifier of the base station to be estimated, the reference signal received strength (RSRP), and the timing advance (TA); preprocessing the MR data to obtain a set of MR sample points for the base station to be estimated; establishing a distance calculation model and an objective function based on RSRP or TA, wherein the distance calculation model is used to calculate the distance between the multiple MR sample points and the base station to be estimated; and determining the longitude and latitude of the base station to be estimated using a gradient descent algorithm based on the distance calculation model and the objective function. This method reduces the amount of data required for solving the problem by preprocessing the MR data to obtain the set of MR sample points; and improves the speed of solving for the base station's latitude and longitude and the accuracy of the measurement results by establishing an objective function based on the set of MR sample points and iteratively solving the objective function using a gradient descent algorithm.

[0016] It should be understood that the above general description and the following detailed description are exemplary and explanatory only, and are not intended to limit this disclosure. Attached Figure Description

[0017] The accompanying drawings, which are incorporated in and form part of this specification, illustrate embodiments consistent with this disclosure and, together with the description, serve to explain the principles of this disclosure, and are not intended to unduly limit this disclosure.

[0018] Figure 1 This is a schematic diagram of an application scenario provided by an embodiment of the present disclosure;

[0019] Figure 2 A flowchart illustrating a base station latitude and longitude estimation method provided in this embodiment of the disclosure;

[0020] Figure 3 A flowchart illustrating another base station latitude and longitude estimation method provided in this disclosure embodiment;

[0021] Figure 4 A flowchart illustrating yet another method for estimating the latitude and longitude of a base station provided in this embodiment of the present disclosure;

[0022] Figure 5 A flowchart illustrating another method for estimating the latitude and longitude of a base station provided in this embodiment of the present disclosure;

[0023] Figure 6 This is a schematic diagram of the structure of a distance calculation model provided in an embodiment of the present disclosure;

[0024] Figure 7 A flowchart illustrating a base station latitude and longitude estimation method provided in this embodiment of the disclosure;

[0025] Figure 8 A flowchart illustrating another base station latitude and longitude estimation method provided in this disclosure embodiment;

[0026] Figure 9 This is a schematic diagram of the structure of a base station latitude and longitude estimation device provided in an embodiment of the present disclosure;

[0027] Figure 10 This is a schematic diagram of the structure of a communication device provided in an embodiment of this disclosure. Detailed Implementation

[0028] Embodiments of this disclosure are described in detail below. Examples of these embodiments are illustrated in the accompanying drawings, wherein the same or similar reference numerals denote the same or similar elements or elements having the same or similar functions throughout. The embodiments described below with reference to the accompanying drawings are exemplary and intended to explain this disclosure, and should not be construed as limiting this disclosure.

[0029] In the field of wireless communication technology, the latitude and longitude of a base station are important engineering parameters that greatly help in communication engineering such as network planning, signal coverage, and fault diagnosis. However, how to quickly and accurately obtain the latitude and longitude of a base station is a major technical challenge.

[0030] The following is a brief introduction to several solutions in related technologies:

[0031] Option A: Convert the latitude and longitude in the MR data into planar coordinates, solve iteratively using the Taylor method, and then convert the results back into latitude and longitude.

[0032] Option B: Convert the latitude and longitude in the MR data into planar coordinates, solve using the logarithmic path loss model, and then convert the results back into latitude and longitude.

[0033] Option C: Combine MR data with three-point positioning to estimate the latitude and longitude of the base station; or use the latitude and longitude of the centroid of each cell in the same station to calculate the intersection of the azimuth rays of each cell to estimate the latitude and longitude of the base station.

[0034] The above three solutions have the following drawbacks:

[0035] Solution A: When using Taylor's method for iterative solution, this solution requires solving the inverse of the gradient matrix, which has high computational complexity and low efficiency when the amount of data is large.

[0036] Option B: When using the logarithmic path loss model, this option requires a series of transformations and simplifications to find the optimal solution, which is complex and involves a large number of matrix inverse operations, resulting in low solution efficiency.

[0037] Option C: When using three-point positioning, only the coordinates of three points are used for position estimation, which selects too few points and results in a large error; or it relies on the cell azimuth angle parameter when solving, which introduces the recording deviation of the cell azimuth angle, thus causing errors in the solution results and has a large limitation.

[0038] To address the problems existing in related technologies, this disclosure proposes a base station latitude and longitude estimation method. This method eliminates the need to convert the latitude and longitude in MR data into planar coordinates. Instead, it directly uses the latitude and longitude in the MR data to establish the objective function and solves it using the gradient descent algorithm. This method is more efficient and convenient. Furthermore, it utilizes the overlapping coverage area of ​​co-located cells to further narrow the search range, thereby improving the accuracy and efficiency of the solution.

[0039] Before introducing the detailed solution of this disclosure, the application scenario of this disclosure solution will be described first. Figure 1 This is an application scenario diagram of the base station latitude and longitude estimation method in one embodiment. For example... Figure 1 As shown, this application scenario includes base station 101 and user 102, where user 102 is a user within the cell coverage of base station 101. By obtaining the MR data of user 102, the latitude and longitude of base station 101 can be estimated using the method described in this disclosure.

[0040] The present disclosure will now be described in further detail with reference to the accompanying drawings and specific embodiments.

[0041] Figure 2 A flowchart illustrating a base station latitude and longitude estimation method provided in an embodiment of this disclosure. Figure 2 As shown, steps 201-204 are included.

[0042] Step 201: Obtain the measurement statistics MR data of multiple cells covered by the base station to be estimated.

[0043] In some embodiments, the method of obtaining the measurement statistics MR data of multiple cells covered by the base station to be estimated is not limited, such as obtaining it through high-level measurement or periodic measurement.

[0044] In some embodiments, MR data includes at least one of the following: base station identifier (eNodeBID), reference received signal strength (RSRP), and timing advance (TA).

[0045] In some embodiments, the MR data further includes at least one of the following: Physical Cell Identifier (PCI), Cell Antenna Frequency (arfcn), Sub-cell User Longitude, and Sub-cell User Latitude.

[0046] For example, by obtaining the original measurement report (MRO) data of the base station and parsing the MRO, the MR data of fields such as eNodeBID, PCI, arfcn, longitude, latitude, RSRP, and TA of multiple cells covered by the base station can be obtained.

[0047] Step 202: Preprocess the MR data to obtain the MR sample point set of the base station to be estimated.

[0048] In some embodiments, MR data is preprocessed to remove outliers and improve the reliability of the MR data.

[0049] In some embodiments, preprocessing of MR data further includes rasterizing the MR data after removing outliers to obtain a set of MR sample points for the base station to be estimated.

[0050] In some embodiments, the method of preprocessing MR data is not limited, such as preprocessing by density clustering algorithm, preprocessing by 3-sigma principle, etc.

[0051] In some embodiments, the MR sample point set with the estimated base station is used to determine the longitude and latitude of the cell. The specific definition of the MR sample point set for the base station is explained in the following embodiments and will not be repeated here.

[0052] Step 203: Based on RSRP or TA, establish a distance calculation model and objective function.

[0053] In some embodiments, a distance calculation model and objective function can be established based on RSRP or TA to calculate the distance between multiple MR sample points and the base station to be estimated.

[0054] In some embodiments, the distance calculation model is used to calculate the distance between multiple MR sample points and the base station to be estimated.

[0055] In some embodiments, a distance calculation model can be established to construct the objective function.

[0056] It should be understood that, in different embodiments, the distance calculation model and objective function established based on RSRP or TA may be the same or different, and this disclosure does not limit them.

[0057] Step 204: Based on the distance calculation model and objective function, the gradient descent algorithm is used to determine the longitude and latitude of the base station to be estimated.

[0058] In some embodiments, the longitude and latitude of the base station to be estimated are determined by iteratively solving a distance calculation model and an objective function using a gradient descent algorithm.

[0059] In summary, the base station latitude and longitude estimation method proposed in this disclosure includes: acquiring measurement and statistical MR data of multiple cells covered by the base station; the MR data includes at least one of the following: base station identifier, Reference Received Signal Strength (RSRP), and Timing Advance (TA); preprocessing the MR data to obtain a set of MR sample points for the base station; establishing a distance calculation model and an objective function based on RSRP or TA; the distance calculation model is used to calculate the distance between the multiple MR sample points and the base station to be estimated; and determining the longitude and latitude of the base station to be estimated using a gradient descent algorithm based on the distance calculation model and the objective function. This method avoids planar coordinate transformation by directly establishing the distance calculation model and objective function based on the MR data, and improves the solution speed and accuracy by using the gradient descent algorithm.

[0060] Figure 3 A flowchart of a base station latitude and longitude estimation method proposed in this disclosure is further shown. Based on Figure 2 The illustrated embodiments further explain, Figure 3 It includes the following steps.

[0061] Step 301: Use an anomaly filtering algorithm to remove abnormal data from the MR data.

[0062] In some embodiments, anomaly filtering algorithms are used to remove outlier data from the MR data to improve the reliability of the MR data and thus improve the accuracy of the method.

[0063] In some embodiments, the type of abnormal data in MR data is not limited, such as abnormal data location of longitude or latitude data, abnormal data value of RSRP and TA data, etc.

[0064] In some embodiments, the method for determining outlier data in MR data is not limited. For example, data exceeding a preset value range may be identified as outlier data; or outlier data that is far from the data center may be identified as outlier data.

[0065] In some embodiments, the anomaly filtering algorithm used is not limited, and may include, for example, density clustering algorithms, 3-sigma principles, etc.

[0066] For example, if the location of longitude or latitude data is abnormal, density clustering algorithm can be used to remove outliers with abnormal location distribution; if the values ​​of RSRP and TA data are abnormal, the 3-sigma principle can be used to filter out abnormal data.

[0067] Step 302: Distribute the MR data into a grid of a preset size to obtain multiple MR sample points.

[0068] In some embodiments, multiple MR sample points are obtained by distributing MR data into a grid of a preset size, which facilitates the determination of the set of MR sample points for the base station to be estimated.

[0069] In some embodiments, the grid can be a square grid divided based on MR data. For example, if the acquired MR data is MR data within a 200×200 square meter area, the grid can be preset to 20×20 square meters to divide the area where the MR data is acquired into 100 grids. This disclosure does not limit the size of the preset grid, which can be set according to the actual coverage area of ​​the base station to be estimated.

[0070] In some embodiments, the raster information of each MR sample point includes the longitude and latitude of the MR sample point, wherein the longitude and latitude of the MR sample point are the longitude and latitude of the user under the cell corresponding to the MR sample point.

[0071] In some embodiments, the raster information for each MR sample point also includes the RSRP and / or TA corresponding to that MR sample point.

[0072] In some embodiments, a grid may include one or more MR sample points, or it may not include MR sample points. This disclosure does not limit the correspondence between MR sample points and preset grids.

[0073] For example, each MR data point can be assigned to a square grid of a preset size of 20*20 meters to obtain multiple MR sample points.

[0074] Step 303: Determine the set of MR sample points for the base station to be estimated based on multiple MR sample points.

[0075] In some embodiments, a set of MR sample points for a base station is determined based on multiple MR sample points to facilitate the calculation of the base station's latitude and longitude.

[0076] In some embodiments, MR data with the same base station number can be identified by the base station identifier in the MR data, so as to obtain MR sample points of all users under the same base station.

[0077] In some embodiments, the cell corresponding to an MR sample point can be determined using the PCI in the MR data.

[0078] Optionally, in some embodiments, a base station may have multiple cells, and the intersection of the MR sample point sets corresponding to each of the multiple cells is determined as the MR sample point set of the base station. If the intersection of the MR sample point sets corresponding to each of the multiple cells is an empty set, the MR sample point set corresponding to each of the multiple cells is determined as the MR sample point set of the base station.

[0079] Optionally, in some embodiments, a base station has only one cell, and the set of MR sample points corresponding to that cell is determined as the set of MR sample points of the base station.

[0080] In summary, based on the base station latitude and longitude estimation method proposed in this disclosure, an anomaly filtering algorithm is used to remove abnormal data from the MR data; the MR data is allocated to a grid of a preset size to obtain multiple MR sample points, wherein the grid information of each MR sample point includes the longitude and latitude of the MR sample point; based on multiple MR sample points, the MR sample point set of the base station is determined, thereby improving the accuracy of the MR data and the accuracy of the base station latitude and longitude estimation.

[0081] Figure 4 A flowchart of a base station latitude and longitude estimation method proposed in this disclosure is further shown. Based on Figure 3 The illustrated embodiments further explain, Figure 4 It includes the following steps.

[0082] Step 401: Determine the MR sample point set corresponding to each of the multiple cells.

[0083] In some embodiments, a base station may have multiple cells, and the MR sample point set of the base station can be determined by determining the MR sample point set corresponding to each of the multiple cells.

[0084] In some embodiments, the cell corresponding to an MR sample point can be determined using the PCI in the MR data.

[0085] For example, let's take three cells to be estimated: Cell1, Cell2, and Cell3. The MR sample point set corresponding to Cell1 can be determined using the PCI in the MR data as {Grid1, Grid2, Grid3, ..., Grid...}. n1 The set of MR sample points contained in}, where Cell2 corresponds to the set of MR sample points {Grid2, Grid4, Grid5, ..., Grid}. n2 The set of MR sample points contained in}, where Cell3 corresponds to the set of MR sample points {Grid1, Grid2, Grid4, ..., Grid}. n3} contains a set of MR sample points, where Grid represents a preset grid.

[0086] Step 402: The intersection of the MR sample point sets corresponding to each of the multiple cells is determined as the MR sample point set of the base station to be estimated.

[0087] In some embodiments, the intersection of the MR sample point sets corresponding to multiple cells is determined as the MR sample point set of the base station to be estimated, so as to reduce the amount of MR sample point data and thereby improve the calculation speed of the base station latitude and longitude estimation method.

[0088] For example, let's take a base station to be estimated with three cells: Cell1, Cell2, and Cell3. The MR sample point set corresponding to Cell1 is the set of MR sample points contained in {Grid1, Grid2, Grid3}. The MR sample point set corresponding to Cell2 is the set of MR sample points contained in {Grid2, Grid3, Grid4}. The MR sample point set corresponding to Cell3 is the set of MR sample points contained in {Grid2, Grid3, Grid5}. Then, the intersection of the MR sample point sets corresponding to Cell1, Cell2, and Cell3, {Grid2, Grid3}, is determined as the MR sample point set of the base station to be estimated.

[0089] Step 403: If the intersection of the MR sample point sets corresponding to multiple cells is an empty set, determine the MR sample point set corresponding to multiple cells as the MR sample point set of the base station to be estimated.

[0090] In some embodiments, if the intersection of the MR sample point sets corresponding to multiple cells is an empty set, the MR sample point sets corresponding to multiple cells are determined as the MR sample point set of the base station to ensure that there is sufficient sample data when calculating the latitude and longitude of the base station.

[0091] For example, let's take a base station to be estimated as having three cells: Cell1, Cell2, and Cell3. The MR sample point set corresponding to Cell1 is the set of MR sample points contained in {Grid1, Grid2, Grid3}, the MR sample point set corresponding to Cell2 is the set of MR sample points contained in {Grid4, Grid5, Grid6}, and the MR sample point set corresponding to Cell3 is the set of MR sample points contained in {Grid7, Grid8, Grid9}. Therefore, the set of MR sample points contained in {Grid1, Grid2, Grid3, Grid4, Grid5, Grid6, Grid7, Grid8, Grid9} can be determined as the MR sample point set of the base station to be estimated.

[0092] Optionally, in some embodiments, if there is only one cell under the base station to be estimated, the set of MR sample points corresponding to that cell can be determined as the set of MR sample points of the base station to be estimated.

[0093] In summary, based on the base station latitude and longitude estimation method proposed in this disclosure, the MR sample point sets corresponding to multiple cells are determined; the intersection of the MR sample point sets corresponding to multiple cells is determined as the MR sample point set of the base station to be estimated; if the intersection of the MR sample point sets corresponding to multiple cells is an empty set, the MR sample point set corresponding to multiple cells is determined as the MR sample point set of the base station to be estimated. This allows for MR sample point filtering, reducing the amount of MR sample point data and improving the speed of base station latitude and longitude calculation.

[0094] Figure 5 A flowchart of a base station latitude and longitude estimation method proposed in this disclosure is further shown. Based on Figure 2 The illustrated embodiments further explain, Figure 5 It includes the following steps.

[0095] Step 501: Establish a distance calculation model based on RSRP or TA.

[0096] In some embodiments, a distance calculation model is obtained by fitting the relationship between input data and output distance using a deep neural network with a fully connected layer containing multiple hidden layers based on RSRP or TA.

[0097] In some embodiments, the input data is a set of MR sample points of the base station to be estimated, namely the longitude and latitude, RSRP and / or TA of the MR sample points.

[0098] In some embodiments, when using TA to build a distance calculation model, the deep neural network consists of a fully connected layer containing two hidden layers.

[0099] Specifically, in some embodiments, the distance calculation model established using TA has the following network structure: Figure 6 As shown.

[0100] The following explanation uses the acquisition of 3×125089 MR data points as an example.

[0101] The first layer of the network structure, layer 1, is the input layer. This input layer has 3*125089 MR data points. Each MR data point contains 3 fields: the longitude of the MR point, the latitude of the MR point, and the TA corresponding to the MR point.

[0102] The number of hidden neurons (Hidden Size, HS) in the second layer (layer 2) of this network structure, which is the first hidden layer (HL), is 250178.

[0103] The third layer of this network structure, i.e. the second layer HL, has an HS value of 375267 and an input of [MR coordinates, TA], where the MR coordinates include the longitude and latitude of the MR point.

[0104] The fourth layer of this network structure, namely the output layer, outputs the distance from the MR point to the base station to be estimated.

[0105] Specifically, ReLU can be used as the activation function before the first hidden layer to increase the non-linearity of the distance calculation model; eLU can be used as the activation function in the second layer to accelerate the gradient descent of the Adam optimizer; and dropout can be used as the activation function before the output layer to randomly eliminate some neurons during training, thereby preventing overfitting.

[0106] It should be understood that this disclosure does not limit the neural network libraries supported by the distance calculation model built using TA, such as Keras, PyTorch, TensorFlow, etc.

[0107] In some embodiments, when using RSRP to build a distance calculation model, the deep neural network consists of a fully connected layer containing four hidden layers.

[0108] The following explanation uses the acquisition of 3*125089 MR data points as an example.

[0109] The HS values ​​for the first layer HL are 250178, the second layer HL's are 375267, the third layer HL's are 250178, and the fourth layer HL's are 175125. The input to the fourth layer is [MR coordinates, RSRP, ARFCN], and the output of the output layer is the distance from the MR point to the base station to be estimated.

[0110] In some embodiments, the distance from the MR point to the base station to be estimated obtained by the above distance calculation model is more accurate than the distance obtained by empirical formulas. Taking a dataset of 24,580 MR points collected in a city during the same time period as an example, the test results are as follows:

[0111] MAE(m) MSE(m) Empirical formula 30.0 1025.5 Machine learning fitting 12.1 210.9

[0112] Among them, MAE (Mean Absolute Error) and MSE (Mean Squared Error) are two important indicators for evaluating the accuracy of the calculation results. The smaller the MAE and MSE values, the more accurate the calculation results. The above results show that the distance from the MR point to the base station to be estimated obtained using the distance calculation model is more accurate, thus improving the accuracy of base station latitude and longitude estimation.

[0113] In summary, the base station latitude and longitude estimation method proposed in this disclosure uses a deep neural network with fully connected layers containing multiple hidden layers, based on RSRP or TA, to fit the relationship between the input data and the output distance to obtain a distance calculation model. The input data includes a set of MR sample points of the base station to be estimated. This method improves the accuracy of base station latitude and longitude estimation by establishing a distance calculation model and then using an aggregation calculation model to determine the distance from the MR points to the base station to be estimated.

[0114] Figure 7 A flowchart of a base station latitude and longitude estimation method proposed in this disclosure is further shown. Based on Figure 2 The illustrated embodiments further explain, Figure 7 It includes the following steps.

[0115] Step 701: Determine the distance d between any two points according to the semi-versus formula.

[0116] In some embodiments, the distance d between any two points is determined according to the semi-sine formula, so as to determine the distance D between the base station to be estimated and the MR sample points.

[0117] Specifically, according to the semi-versus formula, the formula for calculating the distance d between any two points, given their latitude and longitude (x, y) and (x0, y0), is as follows:

[0118]

[0119] Where R is the Earth's radius.

[0120] Step 702: Let the longitude and latitude of the base station to be estimated be (x0, y0). Based on the longitude and latitude of the MR sample point corresponding to the base station to be estimated, determine the distance D between the base station to be estimated and the MR sample point.

[0121] In some embodiments, the longitude and latitude of the base station to be estimated are set as (x0, y0), and the distance D between the base station to be estimated and the MR sample point is determined based on the longitude and latitude of the MR sample point corresponding to the base station to be estimated, so as to establish the objective function.

[0122] In some embodiments, let the longitude and latitude of the base station to be estimated be (x0, y0), and let the longitude and latitude of the MR points in the MR sample point set of the base station to be estimated be (x0, y0). i y i (i = 1, 2, ..., N), where N is the number of MR points. Let D be the distance between the base station to be estimated and the MR sample points. Then, in the ideal case, D = d, that is:

[0123]

[0124] Simplifying this expression, we get:

[0125]

[0126] By minimizing the difference of squares, D and d are made as close as possible, i.e.:

[0127]

[0128] Then we can let equation 1.4 be the objective function.

[0129] In some embodiments, the objective function is used to minimize the error between d and D.

[0130] In summary, according to the base station latitude and longitude estimation method proposed in this disclosure, the distance d between any two points is determined based on the semi-versus formula; let the longitude and latitude of the base station to be estimated be (x0, y0), and determine the distance D between the base station to be estimated and the MR sample point based on the longitude and latitude of the corresponding MR sample point; wherein, the objective function is used to minimize the error between d and D. The method of this disclosure improves the accuracy of base station latitude and longitude estimation by establishing an objective function to reduce the error between the obtained latitude and longitude of the base station to be estimated and the actual latitude and longitude of the base station.

[0131] Figure 8 A flowchart of a base station latitude and longitude estimation method proposed in this disclosure is further shown. Based on Figure 2 The illustrated embodiments further explain, Figure 8 It includes the following steps.

[0132] Step 801: Determine the initial longitude and initial latitude of the base station to be estimated.

[0133] In some embodiments, the average longitude and latitude of the MR points can be determined as the initial longitude and initial latitude of the base station to be estimated, but this is not limited to this. Alternatively, the median longitude and median latitude of the MR points can be determined as the initial longitude and initial latitude of the base station to be estimated. This disclosure does not limit the method of determining the initial longitude and initial latitude of the base station to be estimated.

[0134] Step 802: Determine the gradient descent step size.

[0135] In some embodiments, the lower the determined gradient descent step size, the slower the convergence speed (i.e., the more iterations) when using the gradient descent algorithm for iteration, but the better the convergence result (i.e., the more accurate the calculation result); conversely, the higher the determined gradient descent step size, the faster the convergence speed (i.e., the fewer iterations) when using the gradient descent algorithm for iteration, but the worse the convergence result (i.e., the larger the calculation result error).

[0136] In some embodiments, the gradient descent step size may be determined to be 1, or it may be determined to be other values, which are not limited in this disclosure.

[0137] In some embodiments, the gradient descent step size may be determined based on the accuracy requirements of the latitude and longitude of the base station to be estimated, or it may be determined based on the computing power of the device executing this scheme. This disclosure does not limit this.

[0138] Step 803: Starting from the initial longitude and initial latitude of the base station to be estimated, iterate using the gradient descent step size.

[0139] In some embodiments, the iterative formula for iterating with the gradient descent step size is as follows:

[0140]

[0141]

[0142] Among them, (x 0，k y 0，k (x) represents the longitude and latitude of the base station to be estimated in the k-th iteration. 0，0 y 0，0 Let ) represent the initial longitude and initial latitude of the base station to be estimated, and α represent the gradient descent step size. Let J(x0, y0) be the objective function, which is the partial derivative of Equation 1.4 with respect to x0 and y0 respectively.

[0143] Step 804: The longitude and latitude corresponding to when the algorithm reaches the maximum number of iterations or the error meets the preset conditions are determined as the longitude and latitude of the base station to be estimated.

[0144] In some embodiments, the longitude and latitude corresponding to the algorithm reaching the maximum number of iterations can be determined as the longitude and latitude of the base station to be estimated by presetting the maximum number of iterations.

[0145] In some embodiments, the longitude and latitude corresponding to the error meeting the preset conditions can be determined as the longitude and latitude of the base station to be estimated.

[0146] In some embodiments, the preset condition can be that the difference between the longitude and / or latitude corresponding to two adjacent iterations is less than the minimum allowable longitude and / or latitude error, which can be determined, for example, by the following formula:

[0147] ||x 0，k+1 -x 0，k ||≤ε (Equation 2.3)

[0148] Where, x 0，k+1 Let x represent the longitude of the base station to be estimated at the (K+1)th iteration. 0，k Let ε represent the longitude of the base station to be estimated at the Kth iteration, and ε be the minimum allowable longitude error.

[0149] In summary, based on the base station latitude and longitude estimation method proposed in this disclosure, the initial longitude and initial latitude of the base station to be estimated are determined; the gradient descent step size is determined; starting from the initial longitude and initial latitude of the base station to be estimated, iterations are performed with the gradient descent step size; the longitude and latitude corresponding to when the algorithm reaches the maximum number of iterations or the longitude error meets the preset conditions are determined as the longitude and latitude of the base station to be estimated. This achieves the determination of the longitude and latitude of the base station to be estimated, improving the accuracy of the measurement results.

[0150] Therefore, this solution has the following beneficial effects:

[0151] 1. The objective function is directly established based on the longitude and latitude in the MR data, eliminating the need for plane coordinate transformation and improving the solution speed for estimating the longitude and latitude of the base station.

[0152] 2. This scheme performs a simple transformation of the objective function through basic function transformation. The simplified objective function can be directly solved using the gradient descent algorithm. The optimal solution obtained is the latitude and longitude of the base station to be estimated, which improves the solution speed of the latitude and longitude of the base station to be estimated.

[0153] 3. This scheme utilizes the characteristic that cells under the same base station have overlapping coverage areas, reducing the data size of the MR sample point set. This improves both the accuracy of the estimation results and the efficiency of the solution. For single cells or base stations without overlapping coverage areas, the gradient descent algorithm can still be used to solve for the latitude and longitude of the base station to be estimated, making this scheme highly applicable.

[0154] 4. This solution establishes a distance calculation model through TA / RSRP, which is more accurate than the distance estimated using empirical path loss formulas, thus improving the accuracy of solving the latitude and longitude results of the base station to be estimated.

[0155] Figure 9 This is a schematic diagram of the structure of a base station latitude and longitude estimation device 900 provided in an embodiment of this disclosure. Figure 9 As shown, the base station latitude and longitude estimation device includes:

[0156] The acquisition unit 910 is used to acquire measurement and statistical MR data of multiple cells covered by the base station to be estimated. The MR data includes at least one of the following: the identifier of the base station to be estimated, the reference received signal strength RSRP, and the timing advance TA.

[0157] The first processing unit 920 is used to preprocess the MR data to obtain the MR sample point set of the base station to be estimated.

[0158] The second processing unit 930 is used to establish a distance calculation model and an objective function based on RSRP or TA. The distance calculation model is used to calculate the distance between multiple MR sample points and the base station to be estimated.

[0159] The third processing unit 940 is used to determine the longitude and latitude of the base station to be estimated based on the distance calculation model and objective function, using the gradient descent algorithm.

[0160] According to the base station latitude and longitude estimation device proposed in this disclosure, the method acquires measurement and statistical MR data of multiple cells covered by the base station to be estimated. The MR data includes at least one of the following: the identifier of the base station to be estimated, the reference signal received strength (RSRP), and the timing advance (TA). The MR data is preprocessed to obtain a set of MR sample points for the base station to be estimated. Based on RSRP or TA, a distance calculation model and an objective function are established. The distance calculation model is used to calculate the distance between the multiple MR sample points and the base station to be estimated. Based on the distance calculation model and the objective function, a gradient descent algorithm is used to determine the longitude and latitude of the base station to be estimated. This method reduces the amount of data required for solving the problem by preprocessing the MR data to obtain the set of MR sample points. By establishing an objective function based on the set of MR sample points and iteratively solving the objective function using a gradient descent algorithm, the longitude and latitude of the base station to be estimated are obtained, thus improving the solution speed and the accuracy of the measurement results.

[0161] In some embodiments of this disclosure, the acquisition unit 910 is used to acquire measurement statistics MR data of multiple cells covered by the base station to be estimated. The MR data includes at least one of the following: the identifier of the base station to be estimated, the reference received signal strength RSRP, and the timing advance TA.

[0162] In some embodiments of this disclosure, the first processing unit 920 is used to preprocess the MR data to obtain a set of MR sample points of the base station to be estimated.

[0163] In some embodiments of this disclosure, the first processing unit 920 is further configured to: use an anomaly filtering algorithm to remove abnormal data from the MR data; allocate the MR data to a grid of a preset size to obtain multiple MR sample points, wherein the grid information of each MR sample point includes the longitude and latitude of the MR sample point; and determine the set of MR sample points of the base station to be estimated based on the multiple MR sample points.

[0164] In some embodiments of this disclosure, the first processing unit 920 is further configured to: determine the MR sample point set corresponding to each of the multiple cells; determine the intersection of the MR sample point sets corresponding to each of the multiple cells as the MR sample point set of the base station to be estimated; the intersection of the MR sample point sets corresponding to each of the multiple cells is an empty set, and the MR sample point set corresponding to each of the multiple cells is determined as the MR sample point set of the base station to be estimated.

[0165] In some embodiments of this disclosure, the second processing unit 930 is used to establish a distance calculation model and an objective function based on RSRP or TA. The distance calculation model is used to calculate the distance between multiple MR sample points and the base station to be estimated.

[0166] In some embodiments of this disclosure, the MR data further includes at least one of the following: Physical Cell Identifier (PCI), Cell Antenna Frequency (arfcn), Sub-cell User Longitude, and Sub-cell User Latitude.

[0167] In some embodiments of this disclosure, the second processing unit 930 is further configured to fit the relationship between the input data and the output distance using a deep neural network with a fully connected layer containing multiple hidden layers, based on RSRP or TA, to obtain a distance calculation model, wherein the input data includes a set of MR sample points of the base station to be estimated.

[0168] In some embodiments of this disclosure, when RSRP is used to establish a distance calculation model, the deep neural network consists of a fully connected layer containing four hidden layers, and when TA is used to establish a distance calculation model, the deep neural network consists of a fully connected layer containing two hidden layers.

[0169] In some embodiments of this disclosure, the second processing unit 930 is further configured to: determine the distance d between any two points according to the semi-sine formula; let the longitude and latitude of the base station to be estimated be , and determine the distance D between the base station to be estimated and the MR sample point according to the longitude and latitude of the MR sample point corresponding to the base station to be estimated; wherein, the objective function is used to minimize the error between d and D.

[0170] In some embodiments of this disclosure, the third processing unit 940 is used to determine the longitude and latitude of the base station to be estimated using a gradient descent algorithm based on a distance calculation model and an objective function.

[0171] In some embodiments of this disclosure, the third processing unit 940 is further configured to: determine the initial longitude and initial latitude of the base station to be estimated; determine the gradient descent step size; iterate from the initial longitude and initial latitude of the base station to be estimated with the gradient descent step size; and determine the longitude and latitude corresponding to when the algorithm reaches the maximum number of iterations or the longitude error meets the preset conditions as the longitude and latitude of the base station to be estimated.

[0172] It should be noted that the base station latitude and longitude estimation device provided in the above embodiments is only illustrated by the division of the above program modules when performing base station latitude and longitude estimation. In actual applications, the above processing can be assigned to different program modules as needed, that is, the internal structure of the base station latitude and longitude estimation device can be divided into different program modules to complete all or part of the processing described above.

[0173] Since the apparatus provided in this embodiment corresponds to the methods provided in the above embodiments, the implementation of the methods is also applicable to the apparatus provided in this embodiment, and will not be described in detail in this embodiment.

[0174] The methods and apparatus provided in the embodiments of this application have been described above. To implement the functions of the methods provided in the embodiments of this application, the electronic device may include a hardware structure and software modules, and may implement the above functions in the form of a hardware structure, software modules, or a hardware structure plus software modules. One of the above functions may be executed in the form of a hardware structure, software modules, or a hardware structure plus software modules.

[0175] Figure 10 This is a schematic diagram of the hardware composition structure of the communication device provided in the embodiments of this disclosure, such as... Figure 10 As shown, the communication device 1000 includes at least one processor 1002; and a memory 1001 communicatively connected to the at least one processor 1002; wherein the memory 1001 stores instructions executable by the at least one processor 1002, the instructions being executed by the at least one processor 1002 to implement the steps of the base station latitude and longitude estimation method described in the embodiments of this disclosure; or, the instructions being executed by the at least one processor 1002 to implement the steps of the base station latitude and longitude estimation method described in the embodiments of this disclosure.

[0176] It is understood that the communication device also includes a communication interface 1003. The various components in the communication device are coupled together via a bus system 1004. It is understood that the bus system 1004 is used to implement communication between these components. In addition to a data bus, the bus system 1004 also includes a power bus, a control bus, and a status signal bus. However, for clarity, in... Figure 10 The general labeled all buses as Bus System 1004.

[0177] It is understood that memory 1001 can be volatile memory or non-volatile memory, and may also include both volatile and non-volatile memory.

[0178] Both non-volatile and non-volatile memory are considered. Non-volatile memory can be read-only memory (ROM), programmable read-only memory (PROM), erasable programmable read-only memory (EPROM), electrically erasable programmable read-only memory (EEPROM), magnetic random access memory (FRAM), flash memory, magnetic surface memory, optical disc, or compact disc read-only memory (CD ROM); magnetic surface memory can be disk storage or magnetic tape storage. Volatile memory can be random access memory (RAM), which is used as an external cache. By way of example, but not limitation, many forms of RAM are available, such as Static Random Access Memory (SRAM), Synchronous Static Random Access Memory (SSRAM), Dynamic Random Access Memory (DRAM), Synchronous Dynamic Random Access Memory (SDRAM), Double Data Rate Synchronous Dynamic Random Access Memory (DDRSDRAM), Enhanced Synchronous Dynamic Random Access Memory (ESDRAM), SyncLink Dynamic Random Access Memory (SLDRAM), and Direct Rambus Random Access Memory (DRRAM). The memory 1001 described in this embodiment of the invention is intended to include, but is not limited to, these and any other suitable types of memory.

[0179] The methods disclosed in the above embodiments can be applied to or implemented by the processor 1002. The processor 1002 may be an integrated circuit chip with signal processing capabilities. In implementation, each step of the above method can be completed by integrated logic circuits in the hardware of the processor 1002 or by instructions in software form.

[0180] The processor 1002 described above can be a general-purpose processor, a DSP, or other programmable logic devices, discrete gate or transistor logic devices, discrete hardware components, etc. The processor 1002 can implement or execute the methods, steps, and logic block diagrams disclosed in the embodiments of this invention. The general-purpose processor can be a microprocessor or any conventional processor, etc. The steps of the methods disclosed in the embodiments of this invention can be directly manifested as execution by a hardware decoding processor, or execution by a combination of hardware and software modules in the decoding processor. The software modules can be located in a storage medium, which is located in the memory 1001. The processor 1002 reads the information in the memory 1001 and, in conjunction with its hardware, completes the steps of the aforementioned methods.

[0181] In an exemplary embodiment, the communication device may be implemented by one or more application-specific integrated circuits (ASICs), DSPs, programmable logic devices (PLDs), complex programmable logic devices (CPLDs), FPGAs, general-purpose processors, controllers, MCUs, microprocessors, or other electronic components to perform the aforementioned method.

[0182] This disclosure also provides a non-transitory computer-readable storage medium storing computer instructions, which, when executed, implement the steps of the base station latitude and longitude estimation method described in this embodiment of the invention.

[0183] It should be noted that the terms "first," "second," etc., used in the specification, claims, and accompanying drawings of this disclosure are used to distinguish similar objects and are not necessarily used to describe a specific order or sequence. It should be understood that such data can be interchanged where appropriate so that the embodiments of this disclosure described herein can be implemented in orders other than those illustrated or described herein. The embodiments described in the following exemplary embodiments do not represent all embodiments consistent with this disclosure. Rather, they are merely examples of apparatuses and methods consistent with some aspects of this disclosure as detailed in the appended claims.

[0184] In the description of this specification, the references to terms such as "one embodiment," "some embodiments," "illustrative embodiment," "example," "specific example," or "some examples," etc., indicate that a specific feature, structure, material, or characteristic described in connection with an embodiment or example is included in at least one embodiment or example of the present invention. In this specification, the illustrative expressions of the above terms do not necessarily refer to the same embodiment or example. Furthermore, the specific features, structures, materials, or characteristics described may be combined in any suitable manner in one or more embodiments or examples.

[0185] In some embodiments of this specification, the terms "MR sample point" and "MR point" may be used interchangeably.

[0186] Any process or method description in the flowchart or otherwise herein can be understood as representing a module, segment, or portion of code comprising one or more executable instructions for implementing a particular logical function or process, and the scope of the preferred embodiments of the invention includes additional implementations in which functions may be performed not in the order shown or discussed, including substantially simultaneously or in reverse order depending on the functions involved, as will be understood by those skilled in the art to which embodiments of the invention pertain.

[0187] The logic and / or steps represented in the flowchart or otherwise described herein, for example, can be considered as a sequenced list of executable instructions for implementing logical functions, and can be embodied in any computer-readable medium for use by, or in conjunction with, an instruction execution system, apparatus, or device (such as a computer-based system, a system including a processing module, or other system that can fetch and execute instructions from, an instruction execution system, apparatus, or device). For the purposes of this specification, "computer-readable medium" can be any means that can contain, store, communicate, propagate, or transmit programs for use by, or in conjunction with, an instruction execution system, apparatus, or device. More specific examples (a non-exhaustive list) of computer-readable media include: an electrical connection having one or more wires (control method), a portable computer disk drive (magnetic device), random access memory (RAM), read-only memory (ROM), erasable and editable read-only memory (EPROM or flash memory), fiber optic device, and portable optical disc read-only memory (CDROM). In addition, computer-readable media can even be paper or other suitable media on which programs can be printed, because programs can be obtained electronically, for example, by optically scanning paper or other media, followed by editing, interpreting or otherwise processing as necessary, and then stored in computer memory.

[0188] It should be understood that various parts of the embodiments of the present invention can be implemented in hardware, software, firmware, or a combination thereof. In the above embodiments, multiple steps or methods can be implemented in software or firmware stored in memory and executed by a suitable instruction execution system. For example, if implemented in hardware, as in another embodiment, it can be implemented using any one or a combination of the following techniques known in the art: discrete logic circuits having logic gates for implementing logical functions on data signals, application-specific integrated circuits (ASICs) having suitable combinational logic gates, programmable gate arrays (PGAs), field-programmable gate arrays (FPGAs), etc.

[0189] Those skilled in the art will understand that all or part of the steps of the methods described in the above embodiments can be implemented by a program instructing related hardware, and the program can be stored in a computer-readable storage medium. When executed, the program includes one or a combination of the steps of the method embodiments.

[0190] Furthermore, the functional units in the various embodiments of the present invention can be integrated into a processing module, or each unit can exist physically separately, or two or more units can be integrated into a module. The integrated module can be implemented in hardware or as a software functional module. If the integrated module is implemented as a software functional module and sold or used as an independent product, it can also be stored in a computer-readable storage medium. The storage medium mentioned above can be a read-only memory, a disk, or an optical disk, etc.

[0191] Although embodiments of the present invention have been shown and described above, it is understood that the above embodiments are exemplary and should not be construed as limiting the present invention. Those skilled in the art can make changes, modifications, substitutions and variations to the above embodiments within the scope of the present invention.

Claims

1. A method for estimating the latitude and longitude of a base station, characterized in that, The method includes: Acquire measurement and statistical MR data of multiple cells covered by the base station to be estimated. The MR data includes at least one of the following: the identifier of the base station to be estimated, the reference received signal strength (RSRP), and the timing advance (TA). The MR data is preprocessed to obtain the MR sample point set of the base station to be estimated; Based on the RSRP or the TA, a distance calculation model and an objective function are established. The distance calculation model is used to calculate the distance between the plurality of MR sample points and the base station to be estimated. Based on the distance calculation model and the objective function, the longitude and latitude of the base station to be estimated are determined using the gradient descent algorithm. The preprocessing of the MR data to obtain multiple MR sample points from the multiple cells includes: An anomaly filtering algorithm is used to remove abnormal data from the MR data; The MR data is distributed into a grid of a preset size to obtain multiple MR sample points, wherein the grid information of each MR sample point includes the longitude and latitude of the MR sample point; wherein the MR data further includes at least one of: Physical Cell Identifier (PCI), Cell Antenna Frequency (arfcn), Sub-cell User Longitude (longitude), and Sub-cell User Latitude (latitude); wherein the longitude and latitude of the MR sample point are the sub-cell user longitude and sub-cell user latitude corresponding to that MR sample point; Based on the multiple MR sample points, the set of MR sample points for the base station to be estimated is determined; The step of establishing a distance calculation model based on the RSRP or the TA includes: Based on the RSRP or the TA, a deep neural network with fully connected layers containing multiple hidden layers is used to fit the relationship between the input data and the output distance to obtain the distance calculation model. The input data includes the MR sample point set of the base station to be estimated; The step of determining the set of MR sample points for the base station to be estimated based on the plurality of MR sample points includes: Determine the MR sample point set corresponding to each of the multiple cells; The intersection of the MR sample point sets corresponding to each of the multiple cells is determined as the MR sample point set of the base station to be estimated. The intersection of the MR sample point sets corresponding to each of the multiple cells is an empty set, and the MR sample point sets corresponding to each of the multiple cells are determined as the MR sample point set of the base station to be estimated.

2. The method according to claim 1, characterized in that, When the distance calculation model is established using the RSRP, the deep neural network consists of a fully connected layer containing four hidden layers. When the distance calculation model is established using the TA, the deep neural network consists of a fully connected layer containing two hidden layers.

3. The method according to claim 1, characterized in that, The establishment of the objective function based on the RSRP or the TA includes: Determine the distance d between any two points using the semi-versus formula; Let the longitude and latitude of the base station to be estimated be (x0, y0). Based on the distance calculation model and the RSRP or TA, determine the distance D between the base station to be estimated and the MR sample point. The objective function is used to minimize the error between d and D.

4. The method according to claim 1, characterized in that, The step of determining the longitude and latitude of the base station to be estimated using the gradient descent algorithm based on the distance calculation model and the objective function includes: Determine the initial longitude and initial latitude of the base station to be estimated; Determine the gradient descent step size; Starting from the initial longitude and initial latitude of the base station to be estimated, the iteration is performed with the gradient descent step size; The longitude and latitude corresponding to when the algorithm reaches the maximum number of iterations or the error meets the preset conditions are determined as the longitude and latitude of the base station to be estimated.

5. A base station latitude and longitude estimation device, characterized in that, The device includes: The acquisition unit is used to acquire measurement and statistical MR data of multiple cells covered by the base station to be estimated. The MR data includes at least one of the following: the identifier of the base station to be estimated, the reference received signal strength (RSRP), and the timing advance (TA). The first processing unit is used to preprocess the MR data to obtain the MR sample point set of the base station to be estimated. The second processing unit is used to establish a distance calculation model and an objective function based on the RSRP or the TA. The distance calculation model is used to calculate the distance between the plurality of MR sample points and the base station to be estimated. The third processing unit is used to determine the longitude and latitude of the base station to be estimated based on the distance calculation model and the objective function, using a gradient descent algorithm. The preprocessing of the MR data to obtain multiple MR sample points from the multiple cells includes: An anomaly filtering algorithm is used to remove abnormal data from the MR data; The MR data is distributed into a grid of a preset size to obtain multiple MR sample points, wherein the grid information of each MR sample point includes the longitude and latitude of the MR sample point; wherein the MR data further includes at least one of: Physical Cell Identifier (PCI), Cell Antenna Frequency (arfcn), Sub-cell User Longitude (longitude), and Sub-cell User Latitude (latitude); wherein the longitude and latitude of the MR sample point are the sub-cell user longitude and sub-cell user latitude corresponding to that MR sample point; Based on the multiple MR sample points, the set of MR sample points for the base station to be estimated is determined; The step of establishing a distance calculation model based on the RSRP or the TA includes: Based on the RSRP or the TA, a deep neural network with fully connected layers containing multiple hidden layers is used to fit the relationship between the input data and the output distance to obtain the distance calculation model. The input data includes the MR sample point set of the base station to be estimated; The step of determining the set of MR sample points for the base station to be estimated based on the plurality of MR sample points includes: Determine the MR sample point set corresponding to each of the multiple cells; The intersection of the MR sample point sets corresponding to each of the multiple cells is determined as the MR sample point set of the base station to be estimated. The intersection of the MR sample point sets corresponding to each of the multiple cells is an empty set, and the MR sample point sets corresponding to each of the multiple cells are determined as the MR sample point set of the base station to be estimated.

6. A communication device, characterized in that, include: At least one processor; as well as A memory communicatively connected to the at least one processor; wherein, The memory stores instructions that can be executed by the at least one processor to enable the at least one processor to perform the method of any one of claims 1 to 4.

7. A non-transitory computer-readable storage medium storing computer instructions, characterized in that, The computer instructions are used to cause the computer to perform the method according to any one of claims 1-4.

Images