Antenna and feeder obstruction determination method, device, equipment, and product based on user distribution

Abstract

This invention provides a method, apparatus, device, and product for determining antenna obstruction based on user distribution. The method includes: acquiring user distribution data of a target cell; acquiring collection operating parameters and theoretical operating parameters of the target cell; inputting first input data into a trained first prediction model, inputting second input data into a trained second prediction model, acquiring first antenna azimuth correction data output by the first prediction model and second antenna azimuth correction data output by the second prediction model, wherein the first input data includes user distribution data and theoretical operating parameters, and the second input data includes user distribution data and collection operating parameters; and determining whether there is obstruction between the base station antenna feeder corresponding to the target cell and the target cell based on the difference between the first antenna azimuth correction data and the second antenna azimuth correction data. This invention can improve the accuracy of antenna obstruction identification results.

Description

Technical Field

[0001] This invention relates to the field of communication technology, and in particular to a method, apparatus, device, and product for determining antenna and feeder obstruction based on user distribution. Background Technology

[0002] Identifying antenna obstruction issues (such as tall buildings or billboards obstructing antenna signals between the base station antenna and the cell) plays a significant role in network planning and optimization. Current technologies often rely on building and base station engineering parameters for identification, using simple spatial trigonometric functions and considering factors like station height, orientation, and GIS information of buildings to determine obstruction. However, this approach depends on accurate base station and building information. Rapid urban development leads to outdated building maps, incomplete data acquisition via web crawlers, and frequent antenna optimization in the field, further hindering timely updates to existing base station engineering parameters. These factors all negatively impact identification accuracy, resulting in low accuracy for current antenna obstruction identification methods. Summary of the Invention

[0003] This invention provides a method, apparatus, device, and product for determining antenna occlusion based on user distribution, in order to solve the problem of low accuracy in antenna occlusion identification results in the prior art and improve the accuracy of antenna occlusion identification results.

[0004] This invention provides a method for determining antenna obstruction based on user distribution, comprising:

[0005] Obtain user distribution data for the target cell, wherein the user distribution data reflects the geographical distribution of users in the target cell when they access the base station corresponding to the target cell;

[0006] Acquire the collection engineering parameter data and theoretical engineering parameter data of the target cell. The collection engineering parameter data reflects the base station antenna feeder parameters of the target cell obtained based on the MR data of the target cell. The theoretical engineering parameter data reflects the base station antenna feeder design parameters of the target cell.

[0007] The first input data is input into the trained first prediction model, and the second input data is input into the trained second prediction model. The first antenna azimuth correction data output by the first prediction model and the second antenna azimuth correction data output by the second prediction model are obtained. The first input data includes the user distribution data and the theoretical engineering parameter data, and the second input data includes the user distribution data and the collection engineering parameter data.

[0008] Based on the difference between the first antenna azimuth correction data and the second antenna azimuth correction data, determine whether there is an obstruction between the base station antenna feeder corresponding to the target cell and the target cell;

[0009] The first prediction model and the second prediction model are trained based on multiple sets of first training data and multiple sets of second training data, respectively. The first training data includes sample first input data and first antenna azimuth label, and the second training data includes sample second input data and second antenna azimuth label.

[0010] According to the antenna obstruction determination method based on user distribution provided by the present invention, the step of obtaining user distribution data of the target cell includes:

[0011] Obtain the latitude and longitude information of each sampling user;

[0012] The user distribution data is obtained based on the latitude and longitude information of each of the sampled users.

[0013] According to the antenna obstruction determination method based on user distribution provided by the present invention, the step of obtaining the user distribution data based on the latitude and longitude information of each sampled user includes:

[0014] Initialize the mean and covariance matrices of each Gaussian distribution;

[0015] The probability that the sampled user belongs to each Gaussian distribution is determined based on the latitude and longitude information of the sampled user;

[0016] The mean and covariance matrix of each Gaussian distribution are updated based on the probability that the sampled user belongs to each Gaussian distribution.

[0017] Repeat the step of determining the probability that the sampled user belongs to each Gaussian distribution based on the latitude and longitude information of the sampled user until the mean and covariance matrix of each Gaussian distribution converges or the number of updates of the mean and covariance matrix of each Gaussian distribution reaches a preset number.

[0018] The mean of each Gaussian distribution is obtained as the center point of the user distribution, and the user distribution data is determined based on the center point.

[0019] According to the antenna obstruction determination method based on user distribution provided by the present invention, the step of obtaining the latitude and longitude information of each sampled user includes:

[0020] Data matching is performed based on the 4G MR data and 5G MRO data of the target cell to obtain the latitude and longitude information of the sampled user.

[0021] According to the antenna obstruction determination method based on user distribution provided by the present invention, the step of inputting first input data into a trained first prediction model, inputting second input data into a trained second prediction model, and obtaining first antenna azimuth correction data output by the first prediction model and second antenna azimuth correction data output by the second prediction model includes:

[0022] The input data is input into multiple first prediction models respectively, and the first antenna azimuth correction data output by each first prediction model is obtained. Each first prediction model corresponds to an antenna downtilt angle.

[0023] The input data is input into multiple second prediction models respectively, and the azimuth correction data of each second antenna output by each second prediction model is obtained. Each second prediction model corresponds to an antenna downtilt angle.

[0024] According to the antenna obstruction determination method based on user distribution provided by the present invention, the step of determining whether there is obstruction between the base station antenna feeder corresponding to the target cell and the target cell based on the difference between the first antenna azimuth correction data and the second antenna azimuth correction data includes:

[0025] Obtain the mean square error between the first data group and the second data group, wherein the first data group includes each of the first antenna azimuth correction data and the second data group includes each of the second antenna azimuth correction data.

[0026] Based on the mean square error, it is determined whether there is any obstruction between the base station antenna feeder corresponding to the target cell and the target cell.

[0027] The present invention also provides a device for determining antenna obstruction based on user distribution, comprising:

[0028] The distribution data acquisition module is used to acquire user distribution data of the target cell, wherein the user distribution data reflects the geographical distribution of users in the target cell when they access the base station corresponding to the target cell.

[0029] The parameter data acquisition module is used to acquire the collected parameter data and theoretical parameter data of the target cell. The collected parameter data reflects the base station antenna feeder parameters of the target cell obtained based on the MR data of the target cell, and the theoretical parameter data reflects the base station antenna feeder design parameters of the target cell.

[0030] An azimuth correction module is used to input first input data into a trained first prediction model, input second input data into a trained second prediction model, and obtain first antenna azimuth correction data output by the first prediction model and second antenna azimuth correction data output by the second prediction model. The first input data includes the user distribution data and the theoretical engineering parameter data, and the second input data includes the user distribution data and the collection engineering parameter data.

[0031] The judgment module is used to determine whether there is an obstruction between the base station antenna feeder corresponding to the target cell and the target cell based on the difference between the first antenna azimuth correction data and the second antenna azimuth correction data;

[0032] The first prediction model and the second prediction model are trained based on multiple sets of first training data and multiple sets of second training data, respectively. The first training data includes sample first input data and first antenna azimuth label, and the second training data includes sample second input data and second antenna azimuth label.

[0033] The present invention also provides an electronic device, including a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein the processor executes the computer program to implement the antenna occlusion determination method based on user distribution as described above.

[0034] The present invention also provides a non-transitory computer-readable storage medium having a computer program stored thereon, which, when executed by a processor, implements the antenna occlusion determination method based on user distribution as described above.

[0035] The present invention also provides a computer program product, including a computer program that, when executed by a processor, implements the antenna occlusion determination method based on user distribution as described above.

[0036] The present invention provides a method, apparatus, device, and product for determining antenna obstruction based on user distribution. By acquiring user distribution data of a target cell, and utilizing a trained first and second prediction model, the method calibrates the theoretical operating parameters reflecting the base station antenna design parameters corresponding to the target cell, and the antenna azimuth data in the base station antenna design parameters calculated based on the target cell's MR data, based on the user distribution data. This yields first antenna azimuth correction data corresponding to the target cell's user distribution data under an ideal, unobstructed state, and second antenna azimuth correction data corresponding to the target cell's user distribution data under actual conditions. The difference between the second antenna azimuth correction data and the first antenna azimuth correction data is used to determine whether obstruction exists between the antenna and the target cell. In this process, user distribution data is used to calibrate the antenna azimuth data of the antenna, thereby identifying antenna obstruction and improving the accuracy of antenna obstruction identification results. Attached Figure Description

[0037] To more clearly illustrate the technical solutions in this invention or the prior art, the drawings used in the description of the embodiments or the prior art will be briefly introduced below. Obviously, the drawings described below are some embodiments of this invention. For those skilled in the art, other drawings can be obtained from these drawings without creative effort.

[0038] Figure 1 This is a flowchart illustrating a method for determining antenna occlusion based on user distribution provided by the present invention.

[0039] Figure 2 This is a schematic diagram of antenna occlusion based on building information identification.

[0040] Figure 3 This is a schematic diagram illustrating the principle of a user-distributed antenna occlusion determination method provided by the present invention.

[0041] Figure 4 This is a schematic diagram of the process of obtaining user latitude and longitude information in a user distribution-based antenna occlusion judgment method provided by the present invention.

[0042] Figure 5 This invention provides a spatial distribution map of sampled users in a method for determining antenna occlusion based on user distribution.

[0043] Figure 6 This is a schematic diagram of the prediction model in a user distribution-based antenna occlusion judgment method provided by the present invention.

[0044] Figure 7 This is a schematic diagram illustrating the process of outputting antenna correction results in an antenna occlusion judgment method based on user distribution provided by the present invention.

[0045] Figure 8 This is a schematic diagram illustrating the effect of a user-distribution-based antenna occlusion determination method provided by the present invention.

[0046] Figure 9 This is a schematic diagram of the antenna occlusion determination device based on user distribution provided by the present invention.

[0047] Figure 10 This is a schematic diagram of the structure of the electronic device provided by the present invention. Detailed Implementation

[0048] To make the objectives, technical solutions, and advantages of this invention clearer, the technical solutions of this invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some, not all, of the embodiments of this invention. All other embodiments obtained by those skilled in the art based on the embodiments of this invention without creative effort are within the scope of protection of this invention.

[0049] In existing technologies, based on building and base station engineering parameter information, a simple spatial trigonometric function is used to determine whether the antenna feeder is obstructed, based on factors such as station height, orientation, and GIS information of the building. For example... Figure 2 As shown, if the angle between the two rays of building A is within the horizontal lobe of the antenna (approximately 60 degrees for 5G and 40 degrees for 4G), and the distance does not exceed the set value (300 meters for 2.6G, 800 meters for 700M, and 400 meters for 4G), then the antenna is considered to cover building A. If the angle between the antenna and the rays of other buildings between building A and building C overlaps by more than 80%, then building C is considered to be obstructing the horizontal defense line. However, this method relies on accurate base station operating parameters and building information. Urban development is rapid, building information may not be updated in a timely manner, data obtained by web crawlers may be incomplete, and there are many optimizations to the antenna feeders on-site, resulting in untimely updates to the operating parameters of the existing base stations. These factors all affect the recognition results, leading to low accuracy in the current antenna feeder obstruction recognition.

[0050] To address the low accuracy of antenna occlusion identification results in existing technologies, this invention provides a method, apparatus, device, and product for antenna occlusion judgment based on user distribution, thereby improving the accuracy of antenna occlusion identification results.

[0051] The following is combined Figure 1 The antenna occlusion determination method based on user distribution provided by this invention is described as follows: Figure 1 As shown, the present invention provides a method for determining antenna occlusion based on user distribution, comprising the following steps:

[0052] S110. Obtain user distribution data for the target cell. The user distribution data reflects the geographical location distribution of users in the target cell when they access the base station corresponding to the target cell.

[0053] S120. Obtain the collected working parameter data and theoretical working parameter data of the target cell. The collected working parameter data reflects the base station antenna feeder parameters of the target cell obtained based on the MR data of the target cell. The theoretical working parameter data reflects the base station antenna feeder design parameters of the target cell.

[0054] S130. Input the first input data into the trained first prediction model, input the second input data into the trained second prediction model, and obtain the first antenna azimuth correction data output by the first prediction model and the second antenna azimuth correction data output by the second prediction model. The first input data includes user distribution data and theoretical engineering parameter data, and the second input data includes user distribution data and collection engineering parameter data.

[0055] S140. Based on the difference between the first antenna azimuth correction data and the second antenna azimuth correction data, determine whether there is an obstruction between the base station antenna feeder corresponding to the target cell and the target cell.

[0056] The method provided by this invention acquires user distribution data of a target cell and uses a trained first and second prediction model to calibrate the theoretical engineering parameters reflecting the base station antenna feeder design parameters corresponding to the target cell and the antenna azimuth angle data in the base station antenna feeder design parameters calculated based on the MR data of the target cell, respectively. This allows for the acquisition of first antenna azimuth angle correction data corresponding to the user distribution data of the target cell under an ideal unobstructed state and second antenna azimuth angle correction data corresponding to the user distribution data of the target cell under actual conditions. The difference between the second antenna azimuth angle correction data and the first antenna azimuth angle correction data is used to determine whether there is obstruction between the antenna feeder and the target cell. In this process, user distribution data is used to calibrate the antenna azimuth angle data of the antenna feeder, thereby identifying antenna feeder obstruction and improving the accuracy of antenna feeder obstruction identification results.

[0057] The user distribution data of the target cell reflects the geographical distribution of users accessing the base station antenna feeder corresponding to the target cell. This data can be obtained by analyzing the geographical locations of the users' communication terminals when they access the base station. Since there are many users in the target cell, in one possible implementation of this invention, instead of obtaining the geographical locations of all users within the target cell, users are sampled, and user distribution data reflecting the user distribution of the target cell is obtained based on the geographical locations of each sampled user. That is, obtaining the user distribution data of the target cell includes:

[0058] Obtain the latitude and longitude information of each sampling user;

[0059] User distribution data is obtained based on the latitude and longitude information of each sampled user.

[0060] Specifically, the latitude and longitude information of each sampling user is obtained, including:

[0061] Data matching is performed based on 4G MR data and 5G MRO data of the target cell to obtain the latitude and longitude information of the sampled users.

[0062] like Figure 4 As shown, the latitude and longitude information of each sampling user can be obtained based on the 4G MR (Measurement Report) fingerprint database and the 5G inter-system measurement data table. Specifically, by using the neighbor cell information of 4G carried by 5G, the 4G MR fingerprint database is queried to obtain the 4G latitude and longitude information. The 4G latitude and longitude information is then backfilled into the 5G MRO (MR Original) sampling data with neighbor cell relationships, realizing 5G MRO rasterization processing, and obtaining the latitude and longitude information of the 5G user sampling points, such as... Figure 5 As shown.

[0063] In one possible implementation, the latitude and longitude information of each sampled user can be directly used as user distribution data. However, this would result in a large amount of data processing and would not directly reflect the user distribution. To further extract user distribution data that directly reflects the user distribution and has a smaller data volume, user distribution data is obtained based on the latitude and longitude information of each sampled user, including:

[0064] Initialize the mean and covariance matrices of each Gaussian distribution;

[0065] The probability of a sampled user belonging to a Gaussian distribution is determined based on the latitude and longitude information of the sampled user.

[0066] The mean and covariance matrix of each Gaussian distribution are updated based on the probability that the sampled user belongs to each Gaussian distribution.

[0067] Repeat the step of determining the probability that a sampled user belongs to each Gaussian distribution based on the latitude and longitude information of the sampled user until the mean and covariance matrix of each Gaussian distribution converges or the number of updates of the mean and covariance matrix of each Gaussian distribution reaches the preset number.

[0068] The mean of each Gaussian distribution is obtained as the center point of the user distribution, and the user distribution data is determined based on the center point.

[0069] In one implementation of the method provided by this invention, the latitude and longitude information of the sampled users is considered as a mixture of multiple Gaussian distributions, each representing a cluster. The data is fitted by maximizing the likelihood function to find the optimal cluster center, thereby obtaining user distribution data that reflects the specific distribution of users. Specifically, the latitude and longitude information corresponding to each sampled user is used as data points to construct a dataset. The mean (representing the cluster center) and covariance matrix of each Gaussian distribution are randomly initialized, and the probability (posterior probability) of each data point belonging to each Gaussian distribution is calculated. in, Representing data points The probability of belonging to the k-th Gaussian distribution. These are the weights of the k-th Gaussian distribution. This is the density function of a multivariate Gaussian distribution. Then, based on the probability that each data point belongs to a Gaussian distribution, the mean and covariance matrix are updated, as shown in the following formula:

[0070] ;

[0071] ;

[0072] .

[0073] Repeat the steps of calculating the probability of each data point belonging to each Gaussian distribution and updating the mean and covariance matrix until convergence (the mean and covariance matrix no longer change) or the preset number of iterations is reached. The final mean of each Gaussian distribution is used as the center point of the user cluster. User distribution data is determined based on the center point; specifically, the position of the center point relative to the base station of the target cell can be used as the user distribution data. In one possible implementation, the center point can be connected to the vertex of the base station of the target cell to obtain a connecting line, and the angle and start and end positions of the connecting line can be used as the user distribution data.

[0074] The antenna structure affects user distribution because users tend to communicate in locations with better signal strength. Theoretically, the signal is better in the direction the antenna azimuth points (also known as the direction of the power parameter normal), and users will be concentrated on both sides of this direction. As the distance from the antenna azimuth increases, the number of user sampling points should decrease. However, if there is obstruction, the direction in which users are concentrated will deviate from the direction the antenna azimuth points.

[0075] The method provided by this invention utilizes this principle to infer the antenna azimuth angle under unobstructed conditions based on user distribution data, using both collected engineering parameter data and theoretical engineering parameter data. Figure 3As shown, the antenna azimuth angle inferred from theoretical engineering parameter data and user distribution data corresponds to the normal engineering parameter normal direction, while the antenna azimuth angle inferred from collected engineering parameter data and user distribution data corresponds to the fuzzy engineering parameter normal direction. Based on the difference between the two, it is possible to identify whether there is obstruction between the target cell and the antenna feeder.

[0076] The theoretical engineering parameter data includes the parameters of the base station antenna feeder. This data reflects the design parameters of the base station antenna feeder corresponding to the target cell. The theoretical engineering parameter data can be obtained based on the antenna feeder design documents, such as the initial design documents or the design documents after on-site RF optimization. While the theoretical engineering parameter data represents the theoretical parameters of the base station antenna feeder corresponding to the current target cell, due to potential RF optimization during base station operation or environmental influences, the antenna azimuth angle in the theoretical engineering parameter data recorded in the design documents may not be accurate. The method provided in this invention corrects the antenna azimuth angle in the theoretical engineering parameter data based on user distribution data and theoretical engineering parameter data, under the assumption of no obstruction. Specifically, the first input data, including theoretical engineering parameter data and user distribution data, is input into the first prediction model. The first prediction model is trained based on multiple sets of first training data. Each set of first training data includes sample first input data and a first antenna azimuth label. The first training data is obtained based on cell data without obstruction. In other words, the first prediction model can output antenna azimuth correction data that makes the actual user distribution consistent with the user distribution data under unobstructed conditions, based on the input theoretical engineering parameter data and user distribution data.

[0077] The collected operational parameter data is the base station antenna feeder parameter calculated based on the MR data of the target cell. The MR data of the target cell is data collected by the target terminal. In other words, the MR data of the target cell reflects the actual situation of users in the target cell. Based on the MR data of the target cell, the base station antenna feeder parameters can be inferred. The collected operational parameter data can also be called fuzzy operational parameter data. The method provided by this invention corrects the antenna azimuth angle in the collected operational parameter data based on user distribution data and collected operational parameter data. Specifically, a second input data including collected operational parameter data and user distribution data is input into a second prediction model. The second prediction model is trained based on multiple sets of second training data. Each set of second training data includes sample second input data and a second antenna azimuth angle label. The second training data can be obtained based on cells with or without obstruction. It can be seen that the collected operational parameter data already reflects the base station antenna feeder parameters corresponding to the actual situation of the corresponding users, and the user distribution data also reflects the base station antenna feeder parameters corresponding to the actual situation of the corresponding users. Through the second prediction model, the antenna azimuth angle correction data corresponding to the actual state of the cell can be obtained.

[0078] As can be seen from the preceding explanation, the first prediction model corrects the theoretical antenna azimuth angle based on user distribution data under the theoretical state of no obstruction. The first antenna azimuth angle correction data output is the antenna azimuth angle of the corresponding user distribution data under the theoretical state of no obstruction. The second prediction model corrects the antenna azimuth angle inferred from the actual MR data based on user distribution data. The output is the antenna azimuth angle of the corresponding user distribution data under the actual state. Therefore, the closer the second antenna azimuth angle correction data output by the second prediction model is to the first antenna azimuth angle correction data output by the first prediction model, the lower the probability of obstruction.

[0079] like Figure 6 As shown, in one possible implementation, both the first and second prediction models can adopt the Conv-LSTM model structure. During training, the sample data is divided into a standard set and a test set, where the ratio of the standard set to the test set is K, and the ratio corresponding to the test set is (1-K). The dataset is divided into a standard set train_set and a test set, i.e., standard set train_set = dataset * K, test_set = dataset * (1-K).

[0080] The main data processing components included in the Conv-LSTM model are:

[0081] Input Gate:

[0082] ;

[0083] Forget Gate:

[0084] ;

[0085] Update Gate:

[0086] ;

[0087] Cell Update:

[0088] ;

[0089] Output Gate:

[0090] ;

[0091] Hidden State Update:

[0092] ;

[0093] in, , , , These represent the outputs of the input gate, forget gate, update gate, and output gate, respectively. Represents the memory cells at the current moment. This indicates the hidden state at the current moment. denoted as the input at the current time step, W, U, and b represent the weight matrix, recursive weight matrix, and bias term, respectively, σ represents the Sigmoid function, ⊙ represents element-wise multiplication, and tanh represents the hyperbolic tangent function.

[0094] Antenna downtilt angle also has a certain impact on user distribution. In one possible implementation, to improve the accuracy of antenna obstruction identification, the method provided by this invention sets corresponding first and second prediction models for multiple antenna downtilt angles. That is, the first input data is input into the trained first prediction model, and the second input data is input into the trained second prediction model. The method then obtains the first antenna azimuth correction data output by the first prediction model and the second antenna azimuth correction data output by the second prediction model. This includes:

[0095] The input data is fed into multiple first prediction models respectively, and the first antenna azimuth correction data output by each first prediction model is obtained. Each first prediction model corresponds to an antenna downtilt angle.

[0096] The input data is fed into multiple second prediction models to obtain the second antenna azimuth correction data output by each second prediction model. Each second prediction model corresponds to an antenna downtilt angle.

[0097] By training prediction models for different downtilt angle scenarios, corresponding antenna azimuth correction data can be output for different antenna downtilt angle scenarios, providing more data information and improving the accuracy of the result of judging whether the antenna feed is blocked.

[0098] Determining whether the antenna feeder system to be judged has obstructions based on the difference between the first antenna azimuth correction data and the second antenna azimuth correction data includes:

[0099] Obtain the mean square error between the first data group and the second data group. The first data group includes the azimuth correction data of each first antenna, and the second data group includes the azimuth correction data of each second antenna.

[0100] The presence of obstruction in the antenna feeder system to be judged is determined based on the mean square error.

[0101] The antenna azimuth angle under the assumption of no obstruction is corrected under different downtilt angle scenarios to obtain a set of first antenna azimuth angle correction data. The actual antenna azimuth angle is then corrected under different downtilt angle scenarios to obtain a set of second antenna azimuth angle correction data. The mean square error (MSE) of these two sets of data is obtained. If the MSE is within a preset range, no obstruction is determined. In one possible implementation, if the MSE is below 3dB, no obstruction is determined; otherwise, obstruction exists.

[0102] Furthermore, based on the second antenna azimuth correction data, antenna and feeder optimization suggestions can also be output, specifically, such as... Figure 7 As shown, the azimuth angles obtained from the existing network MR data of each cell are corrected to obtain the calibrated normal angle (i.e., the second antenna azimuth angle correction data). The deviation between the second antenna azimuth angle correction data corresponding to each cell and the standard technical parameter normal angle (the actual antenna azimuth angle) is calculated, and the antenna feeder correction result is output. The antenna feeder correction result is an optimization suggestion for the antenna feeder, which may include adjusting the azimuth angle, adjusting the base station address location, etc.

[0103] After obtaining the judgment result of whether there is occlusion, the judgment result, the community location raster data, the building layer information, the GIS-based information of surrounding occluding buildings, and the optimization suggestions can be output together.

[0104] The method provided by this invention was experimentally verified using a real residential community as an example. The results showed a root mean square error of 4.3 dB, indicating the presence of occlusion. Combined with weak coverage grid data, the building containing the weak coverage grid was located. Figure 8 As shown, there is a weak coverage grid within the angle range between the first antenna azimuth correction data and the second antenna azimuth correction data. Based on the on-site inspection, there is indeed severe obstruction.

[0105] The antenna obstruction determination device based on user distribution provided by the present invention will be described below. The antenna obstruction determination device based on user distribution described below can be referred to in correspondence with the antenna obstruction determination method based on user distribution described above. Figure 9 As shown, the antenna obstruction determination device based on user distribution provided by the present invention includes:

[0106] The distribution data acquisition module 910 is used to acquire user distribution data of the target cell. The user distribution data reflects the geographical distribution of users in the target cell when they access the base station corresponding to the target cell.

[0107] The engineering parameter data acquisition module 920 is used to acquire the collected engineering parameter data and theoretical engineering parameter data of the target cell. The collected engineering parameter data reflects the base station antenna feeder parameters of the target cell obtained based on the MR data of the target cell, and the theoretical engineering parameter data reflects the base station antenna feeder design parameters of the target cell.

[0108] The azimuth correction module 930 is used to input the first input data into the trained first prediction model, input the second input data into the trained second prediction model, and obtain the first antenna azimuth correction data output by the first prediction model and the second antenna azimuth correction data output by the second prediction model. The first input data includes user distribution data and theoretical engineering parameter data, and the second input data includes user distribution data and collection engineering parameter data.

[0109] The judgment module 940 is used to determine whether there is an obstruction between the base station antenna feeder corresponding to the target cell and the target cell based on the difference between the first antenna azimuth correction data and the second antenna azimuth correction data.

[0110] The first prediction model and the second prediction model are trained based on multiple sets of first training data and multiple sets of second training data, respectively. The first training data includes sample first input data and first antenna azimuth label, and the second training data includes sample second input data and second antenna azimuth label.

[0111] Figure 10 An example is a schematic diagram of the physical structure of an electronic device, such as... Figure 10As shown, the electronic device may include: a processor 1010, a communication interface 1020, a memory 1030, and a communication bus 1040. The processor 1010, communication interface 1020, and memory 1030 communicate with each other via the communication bus 1040. The processor 1010 can call logic instructions in the memory 1030 to execute a user distribution-based antenna obstruction judgment method. This method includes: acquiring user distribution data of the target cell, which reflects the geographical location distribution of users accessing the base station corresponding to the target cell; acquiring collection parameter data and theoretical parameter data of the target cell, where the collection parameter data reflects the base station antenna feeder parameters of the target cell obtained based on the target cell's MR data, and the theoretical parameter data reflects the base station antenna feeder design parameters of the target cell; inputting first input data into a trained first prediction model, and inputting second input data into a trained second prediction model to obtain... The first prediction model outputs first antenna azimuth correction data, and the second prediction model outputs second antenna azimuth correction data. The first input data includes user distribution data and theoretical operating parameters, and the second input data includes user distribution data and data collection operating parameters. Based on the difference between the first antenna azimuth correction data and the second antenna azimuth correction data, it is determined whether there is obstruction between the base station antenna feeder corresponding to the target cell and the target cell. The first prediction model and the second prediction model are trained based on multiple sets of first training data and multiple sets of second training data, respectively. The first training data includes sample first input data and first antenna azimuth labels, and the second training data includes sample second input data and second antenna azimuth labels.

[0112] Furthermore, the logical instructions in the aforementioned memory 1030 can be implemented as software functional units and, when sold or used as independent products, can be stored in a computer-readable storage medium. Based on this understanding, the technical solution of the present invention, in essence, or the part that contributes to the prior art, or a part of the technical solution, can be embodied in the form of a software product. This computer software product is stored in a storage medium and includes several instructions to cause a computer device (which may be a personal computer, server, or network device, etc.) to execute all or part of the steps of the methods described in the various embodiments of the present invention. The aforementioned storage medium includes various media capable of storing program code, such as USB flash drives, portable hard drives, read-only memory (ROM), random access memory (RAM), magnetic disks, or optical disks.

[0113] On the other hand, the present invention also provides a computer program product, which includes a computer program that can be stored on a non-transitory computer-readable storage medium. When the computer program is executed by a processor, the computer can execute the above-mentioned antenna feeder obstruction judgment method based on user distribution. The antenna feeder obstruction judgment method based on user distribution includes: acquiring user distribution data of a target cell, wherein the user distribution data reflects the geographical location distribution of users in the target cell when accessing the base station corresponding to the target cell; acquiring collection parameter data and theoretical parameter data of the target cell, wherein the collection parameter data reflects the base station antenna feeder parameters of the target cell obtained based on the MR data of the target cell, and the theoretical parameter data reflects the base station antenna feeder design parameters of the target cell; and inputting the first input data into the trained... A prediction model is formed by inputting second input data into a trained second prediction model to obtain first antenna azimuth correction data output by the first prediction model and second antenna azimuth correction data output by the second prediction model. The first input data includes user distribution data and theoretical operating parameters, and the second input data includes user distribution data and collection operating parameters. Based on the difference between the first antenna azimuth correction data and the second antenna azimuth correction data, it is determined whether there is obstruction between the base station antenna feeder corresponding to the target cell and the target cell. The first prediction model and the second prediction model are trained based on multiple sets of first training data and multiple sets of second training data, respectively. The first training data includes sample first input data and first antenna azimuth labels, and the second training data includes sample second input data and second antenna azimuth labels.

[0114] In another aspect, the present invention also provides a non-transitory computer-readable storage medium storing a computer program thereon, which, when executed by a processor, implements the aforementioned antenna feeder obstruction judgment method based on user distribution. The antenna feeder obstruction judgment method based on user distribution includes: acquiring user distribution data of a target cell, the user distribution data reflecting the geographical location distribution of users accessing the base station corresponding to the target cell; acquiring collection parameter data and theoretical parameter data of the target cell, the collection parameter data reflecting the base station antenna feeder parameters of the target cell obtained based on the MR data of the target cell, and the theoretical parameter data reflecting the base station antenna feeder design parameters of the target cell; inputting first input data into a trained first prediction model, and inputting second input data into... The data is fed into the trained second prediction model to obtain the first antenna azimuth correction data output by the first prediction model and the second antenna azimuth correction data output by the second prediction model. The first input data includes user distribution data and theoretical operating parameters, and the second input data includes user distribution data and collection operating parameters. Based on the difference between the first antenna azimuth correction data and the second antenna azimuth correction data, it is determined whether there is obstruction between the base station antenna feeder corresponding to the target cell and the target cell. The first prediction model and the second prediction model are trained based on multiple sets of first training data and multiple sets of second training data, respectively. The first training data includes sample first input data and first antenna azimuth labels, and the second training data includes sample second input data and second antenna azimuth labels.

[0115] The device embodiments described above are merely illustrative. The units described as separate components may or may not be physically separate. The components shown as units may or may not be physical units; that is, they may be located in one place or distributed across multiple network units. Some or all of the modules can be selected to achieve the purpose of this embodiment according to actual needs. Those skilled in the art can understand and implement this without any creative effort.

[0116] Through the above description of the embodiments, those skilled in the art can clearly understand that each embodiment can be implemented by means of software plus necessary general-purpose hardware platforms, and of course, it can also be implemented by hardware. Based on this understanding, the above technical solutions, in essence or the part that contributes to the prior art, can be embodied in the form of a software product. This computer software product can be stored in a computer-readable storage medium, such as ROM / RAM, magnetic disk, optical disk, etc., and includes several instructions to cause a computer device (which may be a personal computer, server, or network device, etc.) to execute the methods described in the various embodiments or some parts of the embodiments.

[0117] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention, and not to limit them; although the present invention has been described in detail with reference to the foregoing embodiments, those skilled in the art should understand that modifications can still be made to the technical solutions described in the foregoing embodiments, or equivalent substitutions can be made to some of the technical features; and these modifications or substitutions do not cause the essence of the corresponding technical solutions to deviate from the spirit and scope of the technical solutions of the embodiments of the present invention.

Claims

1. A method for determining antenna occlusion based on user distribution, characterized in that, include: Obtain user distribution data for the target cell, wherein the user distribution data reflects the geographical distribution of users in the target cell when they access the base station corresponding to the target cell; Acquire the collection engineering parameter data and theoretical engineering parameter data of the target cell. The collection engineering parameter data reflects the base station antenna feeder parameters of the target cell obtained based on the MR data of the target cell. The theoretical engineering parameter data reflects the base station antenna feeder design parameters of the target cell. The first input data is input into the trained first prediction model, and the second input data is input into the trained second prediction model. The first antenna azimuth correction data output by the first prediction model and the second antenna azimuth correction data output by the second prediction model are obtained. The first input data includes the user distribution data and the theoretical engineering parameter data, and the second input data includes the user distribution data and the collection engineering parameter data. Based on the difference between the first antenna azimuth correction data and the second antenna azimuth correction data, determine whether there is an obstruction between the base station antenna feeder corresponding to the target cell and the target cell; The first prediction model and the second prediction model are trained based on multiple sets of first training data and multiple sets of second training data, respectively. The first training data includes sample first input data and first antenna azimuth label, and the second training data includes sample second input data and second antenna azimuth label.

2. The antenna obstruction determination method based on user distribution according to claim 1, characterized in that, The acquisition of user distribution data for the target cell includes: Obtain the latitude and longitude information of each sampling user; The user distribution data is obtained based on the latitude and longitude information of each of the sampled users.

3. The antenna obstruction determination method based on user distribution according to claim 2, characterized in that, The step of obtaining the user distribution data based on the latitude and longitude information of each of the sampled users includes: Initialize the mean and covariance matrices of each Gaussian distribution; The probability that the sampled user belongs to each Gaussian distribution is determined based on the latitude and longitude information of the sampled user; The mean and covariance matrix of each Gaussian distribution are updated based on the probability that the sampled user belongs to each Gaussian distribution. Repeat the step of determining the probability that the sampled user belongs to each Gaussian distribution based on the latitude and longitude information of the sampled user until the mean and covariance matrix of each Gaussian distribution converges or the number of updates of the mean and covariance matrix of each Gaussian distribution reaches a preset number. The mean of each Gaussian distribution is obtained as the center point of the user distribution, and the user distribution data is determined based on the center point.

4. The antenna obstruction determination method based on user distribution according to claim 3, characterized in that, The acquisition of latitude and longitude information for each sampling user includes: Data matching is performed based on the 4G MR data and 5G MRO data of the target cell to obtain the latitude and longitude information of the sampled user.

5. The antenna obstruction determination method based on user distribution according to claim 1, characterized in that, The step of inputting the first input data into a trained first prediction model and the second input data into a trained second prediction model to obtain the first antenna azimuth correction data output by the first prediction model and the second antenna azimuth correction data output by the second prediction model includes: The input data is input into multiple first prediction models respectively, and the first antenna azimuth correction data output by each first prediction model is obtained. Each first prediction model corresponds to an antenna downtilt angle. The input data is input into multiple second prediction models respectively, and the second antenna azimuth correction data output by each second prediction model is obtained. Each second prediction model corresponds to an antenna downtilt angle.

6. The antenna obstruction determination method based on user distribution according to claim 5, characterized in that, The step of determining whether there is obstruction between the base station antenna feeder corresponding to the target cell and the target cell based on the difference between the first antenna azimuth correction data and the second antenna azimuth correction data includes: Obtain the mean square error between the first data group and the second data group, wherein the first data group includes each of the first antenna azimuth correction data and the second data group includes each of the second antenna azimuth correction data. Based on the mean square error, it is determined whether there is any obstruction between the base station antenna feeder corresponding to the target cell and the target cell.

7. A device for determining antenna obstruction based on user distribution, characterized in that, include: The distribution data acquisition module is used to acquire user distribution data of the target cell, wherein the user distribution data reflects the geographical distribution of users in the target cell when they access the base station corresponding to the target cell. The parameter data acquisition module is used to acquire the collected parameter data and theoretical parameter data of the target cell. The collected parameter data reflects the base station antenna feeder parameters of the target cell obtained based on the MR data of the target cell, and the theoretical parameter data reflects the base station antenna feeder design parameters of the target cell. An azimuth correction module is used to input first input data into a trained first prediction model, input second input data into a trained second prediction model, and obtain first antenna azimuth correction data output by the first prediction model and second antenna azimuth correction data output by the second prediction model. The first input data includes the user distribution data and the theoretical engineering parameter data, and the second input data includes the user distribution data and the collection engineering parameter data. The judgment module is used to determine whether there is an obstruction between the base station antenna feeder corresponding to the target cell and the target cell based on the difference between the first antenna azimuth correction data and the second antenna azimuth correction data; The first prediction model and the second prediction model are trained based on multiple sets of first training data and multiple sets of second training data, respectively. The first training data includes sample first input data and first antenna azimuth label, and the second training data includes sample second input data and second antenna azimuth label.

8. An electronic device comprising a memory, a processor, and a computer program stored in the memory and executable on the processor, characterized in that, When the processor executes the computer program, it implements the antenna occlusion determination method based on user distribution as described in any one of claims 1-6.

9. A non-transitory computer-readable storage medium having a computer program stored thereon, characterized in that, When the computer program is executed by the processor, it implements the antenna occlusion determination method based on user distribution as described in any one of claims 1-6.

10. A computer program product, comprising a computer program, characterized in that, When the computer program is executed by the processor, it implements the antenna occlusion determination method based on user distribution as described in any one of claims 1-6.

Images