Antenna configuration parameter optimization methods, devices, and storage media

Abstract

This application provides an antenna configuration parameter optimization method, apparatus, and storage medium, including: incrementally updating a first prediction model based on first antenna configuration parameters corresponding to an optimization region and the score of the first antenna configuration parameters to obtain a second prediction model; obtaining second antenna configuration parameters based on the second prediction model; determining whether to use the second antenna configuration parameters as a first target configuration parameter; and if the second antenna configuration parameters are used as the first target configuration parameter, then distributing the second antenna configuration parameters. By incrementally updating the first prediction model based on the first antenna configuration parameters and their scores, new antenna configuration parameters are obtained based on the updated prediction model. Using this method, the updated prediction model can quickly converge to antenna configuration parameters with higher scores, thereby improving the optimization efficiency of antenna configuration parameters.

Description

Technical Field

[0001] This application relates to the field of communication technology, and in particular to an antenna configuration parameter optimization method, apparatus and storage medium. Background Technology

[0002] Wireless communication network planning involves designing a reasonable and feasible wireless network layout (usually deploying sites in a cellular pattern) based on customer requirements for network quality (coverage, interference, capacity) and the topography and user distribution characteristics of the planning area, with minimal investment to meet customer needs. In real-world scenarios, differences between network planning and the actual physical environment, changes in urban construction and user development, and varying configuration requirements for scenario-specific parameters can lead to network quality issues such as weak coverage, overlapping coverage, and unbalanced load. Operators need to optimize existing network parameters to address these quality problems.

[0003] Network parameter optimization typically involves adjusting radio frequency parameters to control site coverage, enhancing the quality, capacity, and rate of three-dimensional coverage for road segments or all users, and fully ensuring user accessibility, mobility, and experience.

[0004] The new characteristics of 5G (5th generation wireless systems) present significant technical challenges to network parameter optimization. For Massive Multi-input Multi-output (MIMO) antennas, which provide broadcast beamweights for different coverage scenarios, the radio frequency (RF) parameters change from three (physical azimuth, physical downtilt, and electronic downtilt) for ordinary antennas to six (physical azimuth, physical downtilt, digital azimuth, digital downtilt, horizontal beamwidth, and vertical beamwidth) for Massive MIMO antennas. Therefore, with the introduction of Massive MIMO antennas, the adjustable parameters increase from three to six, resulting in an exponential increase in the adjustable combination space of RF parameters. Furthermore, in 5G scenarios, the combination of macro and micro base stations leads to closer proximity and stronger coupling between stations, resulting in a more complex network environment. New features such as MIMO and Coordinated Multi-Point Processing (CoMP) networking further complicate the network structure. In such scenarios, simulation platforms based solely on basic electromagnetic wave propagation formulas suffer from low simulation model accuracy, making it difficult to accurately identify quality issues in the network and assess the differences in network quality across different parameter combinations.

[0005] Existing technologies construct simulation models based on data from existing networks (electronic maps, antenna files, engineering parameter data, measurement reports (MR) / drive test (DT) data). These simulation models evaluate how network quality changes with varying parameter combinations. Based on the simulation model's evaluation of the current parameter combinations, poor-quality areas in the network are identified, and problem cells requiring parameter adjustments are determined. A genetic algorithm is then used to optimize the parameter combinations for these problem cells until convergence or the maximum number of iterations is reached. The optimized antenna parameters are then configured for these problem cells, and the solution is deployed to the existing network. However, existing technologies require rebuilding the simulation model and re-initializing the optimization algorithm for each new round of measurement data. This results in long end-to-end runtime and low optimization efficiency. Summary of the Invention

[0006] This application discloses an antenna configuration parameter optimization method, apparatus, and storage medium, which can improve the efficiency of finding the optimal antenna configuration parameters.

[0007] In a first aspect, embodiments of this application provide an antenna configuration parameter optimization method, comprising: incrementally updating a first prediction model based on a first antenna configuration parameter corresponding to an optimization region and a score of the first antenna configuration parameter to obtain a second prediction model; obtaining a second antenna configuration parameter based on the second prediction model; determining whether to use the second antenna configuration parameter as a first target configuration parameter; and if the second antenna configuration parameter is used as the first target configuration parameter, then distributing the second antenna configuration parameter.

[0008] The above-mentioned incremental update of the first prediction model based on the first antenna configuration parameters corresponding to the optimization region and the score of the first antenna configuration parameters to obtain the second prediction model can be understood as: updating the calculation formula, parameters, or attributes of the first prediction model based on the first antenna configuration parameters corresponding to the optimization region and the score of the first antenna configuration parameters, thereby obtaining the second prediction model. In contrast, a full update generates the second prediction model based on the first antenna configuration parameters, the score of the first antenna configuration parameters, and the antenna configuration parameters and scores used before generating the first prediction model. That is to say, a full update does not generate the second prediction model based on the already generated first prediction model; it does not establish a correlation between the first and second prediction models. Incremental updates, compared to full updates, update the already generated first prediction model based on the first antenna configuration parameters and the score of the first antenna configuration parameters to obtain a new prediction model.

[0009] In this embodiment, the first prediction model is incrementally updated based on the first antenna configuration parameters and their scores, and new antenna configuration parameters are obtained from the updated prediction model. This incremental update allows the prediction model to quickly converge to antenna configuration parameters with higher scores, thus improving the optimization efficiency of antenna configuration parameters. Compared to existing full-scale updates, this solution effectively improves delivery quality and efficiency.

[0010] One implementation involves determining whether to use the second antenna configuration parameter as the first target configuration parameter by checking if a preset condition is met. For example, this can be based on whether the iteration count N1 has been reached, or whether the second antenna configuration parameter has been obtained N2 consecutive times (i.e., the same antenna configuration parameter is obtained in N2 iterations), thereby determining whether to use the second antenna configuration parameter as the first target configuration parameter. Here, N1 and N2 are both positive integers. If the iteration count N1 has been reached, or the second antenna configuration parameter has been obtained N2 consecutive times (at which point, the first antenna configuration parameter and the second antenna configuration parameter are the same), then the second antenna configuration parameter is determined to be used as the first target configuration parameter.

[0011] In one implementation, if the second antenna configuration parameters are not used as the first target configuration parameters, the second prediction model is incrementally updated based on the second antenna configuration parameters and the scores of the second antenna configuration parameters to obtain a third prediction model; and the third antenna configuration parameters are obtained based on the third prediction model.

[0012] A third prediction model is obtained by incrementally updating the second prediction model based on the second antenna configuration parameters and their scores. This method, through continuous updates, allows the updated prediction model to quickly converge to antenna configuration parameters with higher scores, thereby improving the efficiency of antenna configuration parameter optimization and effectively enhancing delivery quality and efficiency.

[0013] As one implementation, the method further includes: acquiring measurement data and simulation data, wherein the measurement data is data obtained after the second target configuration parameters are issued, and the simulation data is data obtained based on the second target configuration parameters and the first simulation model; correcting the first simulation model based on the measurement data and simulation data to obtain a second simulation model; and obtaining a score for the first antenna configuration parameters based on the first antenna configuration parameters and the second simulation model.

[0014] This application embodiment uses measurement data obtained after the second target configuration parameters are issued, and simulation data obtained by simulation based on the second target configuration parameters, to correct the simulation model, thereby reducing the error of the simulation model, making it closer to the actual implementation effect of the live network, and thus improving the accuracy of the simulation.

[0015] As one implementation, the above-mentioned correction of the first simulation model based on the measurement data and simulation data to obtain the second simulation model includes: correcting the link loss of the first simulation model based on the measurement data and simulation data to obtain the second simulation model.

[0016] By correcting the link loss in the simulation model, the accuracy of the simulation can be improved.

[0017] As one implementation, the above-described correction of the link loss of the first simulation model based on the measurement data and simulation data includes: obtaining a first reference signal received power for each of the plurality of grids based on the measurement data; obtaining a second reference signal received power for each of the plurality of grids based on the simulation data of the first simulation model; obtaining a Kalman matrix for each grid based on the first reference signal received power and the second reference signal received power for each grid; and correcting the link loss of the first simulation model based on the first reference signal received power, the second reference signal received power, and the Kalman matrix for each grid.

[0018] The embodiments of this application use Kalman filtering-based coverage evaluation and correction techniques to correct errors in the simulation model, making the simulation evaluation more accurate.

[0019] As one implementation, the method further includes: acquiring the stored first prediction model.

[0020] By storing the first prediction model, it can be directly retrieved when incrementally updating the first prediction model, thereby improving the efficiency of antenna configuration parameter optimization.

[0021] As one implementation method, the method also includes storing the corrected link loss. This approach allows the corrected link loss to be directly obtained during simulation using a simulation model, improving simulation efficiency and consequently, the efficiency of antenna configuration parameter optimization.

[0022] As one implementation method, the method also includes storing a second simulation model. By storing the second simulation model, it is possible to directly obtain and utilize the second simulation model during subsequent simulations, thereby improving simulation efficiency.

[0023] As one implementation, the method further includes: obtaining an initial prediction model based on multiple sets of historical antenna configuration parameters corresponding to the optimization region and the score of each set of historical antenna configuration parameters, so as to obtain the first prediction model based on the initial prediction model.

[0024] Secondly, embodiments of this application provide an antenna configuration parameter optimization device, comprising: a first model generation module, configured to incrementally update a first prediction model based on the first antenna configuration parameters corresponding to the optimization region and the score of the first antenna configuration parameters to obtain a second prediction model; a first parameter generation module, configured to obtain a second antenna configuration parameter based on the second prediction model; a judgment module, configured to determine whether to use the second antenna configuration parameter as a first target configuration parameter; and a parameter determination module, configured to send the second antenna configuration parameter if it is used as the first target configuration parameter.

[0025] In this embodiment, the first prediction model is incrementally updated based on the first antenna configuration parameters and their scores, resulting in new antenna configuration parameters. This method improves the prediction accuracy of the prediction model by updating it, thereby increasing the efficiency of antenna configuration parameter optimization and effectively enhancing both delivery quality and efficiency.

[0026] In one embodiment, the apparatus further includes: a second model generation module, configured to incrementally update the second prediction model based on the second antenna configuration parameters and the score of the second antenna configuration parameters to obtain a third prediction model if the second antenna configuration parameters are not used as the first target configuration parameters; and a second parameter generation module, configured to obtain the third antenna configuration parameters based on the third prediction model.

[0027] In one embodiment, the device further includes a scoring determination module, configured to: acquire measurement data and simulation data, wherein the measurement data is data obtained after the second target configuration parameters are sent out, and the simulation data is data obtained based on the second target configuration parameters and the first simulation model; correct the first simulation model based on the measurement data and the simulation data to obtain a second simulation model; and obtain a score for the first antenna configuration parameters based on the first antenna configuration parameters and the second simulation model.

[0028] In one embodiment, the apparatus further includes an acquisition module for: acquiring the stored first prediction model.

[0029] In one implementation, the device further includes a third model generation module, configured to: obtain an initial prediction model based on multiple sets of historical antenna configuration parameters corresponding to the optimization region and the score of each set of historical antenna configuration parameters, so as to obtain the first prediction model based on the initial prediction model.

[0030] Thirdly, this application provides an antenna configuration parameter optimization device, including a processor and a memory; wherein the memory is used to store program code, and the processor is used to call the program code to execute the above-described method.

[0031] Fourthly, this application provides a computer storage medium including computer instructions that, when executed on an electronic device, cause the electronic device to perform a method as provided in any possible implementation of the first aspect.

[0032] Fifthly, embodiments of this application provide a computer program product that, when run on a computer, causes the computer to perform the method provided in any possible implementation of the first aspect.

[0033] In a sixth aspect, embodiments of this application provide a chip system applied to an electronic device; the chip system includes one or more interface circuits and one or more processors; the interface circuits and the processors are interconnected via lines; the interface circuits are used to receive signals from the memory of the electronic device and send the signals to the processor, the signals including computer instructions stored in the memory; when the processor executes the computer instructions, the electronic device executes the method.

[0034] It is understood that the apparatus described in the second aspect, the apparatus described in the third aspect, the computer storage medium described in the fourth aspect, the computer program product described in the fifth aspect, or the chip system described in the sixth aspect are all used to perform the method provided in any of the first aspects. Therefore, the beneficial effects they can achieve can be referred to the beneficial effects in the corresponding methods, and will not be repeated here. Attached Figure Description

[0035] The accompanying drawings used in the embodiments of this application are described below.

[0036] Figure 1a This is a schematic diagram of a scenario for an antenna configuration parameter optimization system provided in an embodiment of this application;

[0037] Figure 1b This is a schematic diagram of an antenna configuration parameter optimization method provided in an embodiment of this application;

[0038] Figure 1cThis is a schematic diagram of an antenna configuration parameter optimization method provided in an embodiment of this application;

[0039] Figure 2 This is a flowchart illustrating an antenna configuration parameter optimization method provided in an embodiment of this application;

[0040] Figure 3 This is a flowchart illustrating another antenna configuration parameter optimization method provided in an embodiment of this application;

[0041] Figure 4 This is a flowchart illustrating another optimization method provided in an embodiment of this application;

[0042] Figure 5 This is a schematic diagram of the structure of an antenna configuration parameter optimization device provided in an embodiment of this application;

[0043] Figure 6 This is a schematic diagram of another antenna configuration parameter optimization device provided in the embodiments of this application. Detailed Implementation

[0044] The embodiments of this application are described below with reference to the accompanying drawings. The terminology used in the implementation section of this application is for explaining specific embodiments only and is not intended to limit the scope of this application.

[0045] The embodiments of this application can be applied to 5G or other future networks, such as 6G.

[0046] Reference Figure 1aThe diagram illustrates a scenario of an antenna configuration parameter optimization system provided in this embodiment. This system includes a computing server 101, a network management server 102, and a base station 103. The network management server 102 can specifically be an Operation and Maintenance Center (OMC). This application uses an OMC as an example for illustration, but it should be understood that the network management server 102 can also be an Operations Support System (OSS), etc. The computing server 101 sends a measurement data input request to the OMC. Upon receiving the request, the OMC controls the base station 103 to begin collecting data. The base station 103 begins measuring and controlling the antennas in the communication network and reports the measurement data to the OMC in real time. After collecting the measurement data, the OMC imports it into the computing server 101 to achieve data reporting. The computing server 101 performs antenna configuration parameter optimization processing based on the measurement data, obtains the optimized antenna configuration parameters, and then sends a configuration command to the OMC. These antenna configuration parameters can also be called antenna radio frequency (RF) parameters. OMC converts the optimized RF parameters into Man Machine Language (MML) instructions and adjusts the RF parameters of the base station antenna, thus enabling the distribution of the adjusted parameters.

[0047] The aforementioned computing server 101 optimizes antenna configuration parameters based on the measurement data, which can be referred to as follows: Figure 1bAs shown. First, the computing server 101 establishes a simulation model based on the electromagnetic wave propagation characteristics of the input data (measurement data). When the RF parameters change, it can simulate and evaluate the changed Reference Signal Received Power (RSRP) value. Then, based on the RSRP threshold and the simulated RSRP value, it determines the problem cell in the optimization area. Next, it uses an optimization algorithm to optimize the RF parameters of the problem cell. During the optimization process, it can evaluate the scores of different antenna configuration parameters through simulation. Based on the antenna configuration parameters tried during the RF parameter optimization process and the corresponding scores, a prediction model is trained. The current optimal antenna configuration parameters can be obtained from this prediction model. By inputting these optimal antenna configuration parameters into the simulation model, a score for these antenna configuration parameters can be obtained. The prediction model is then incrementally updated based on these antenna configuration parameters and their scores. The computing server 101 also stores the updated prediction model. By continuously updating the prediction model until the algorithm corresponding to the prediction model converges or reaches the required number of iterations, the optimal antenna configuration parameters corresponding to the current prediction model are used as the target configuration parameters, and then the network management server 102 distributes these target configuration parameters. For specific steps on how to use and update the prediction model, please refer to [link / reference]. Figure 1c As shown, the details are as follows:

[0048] The computing server 101 determines whether a prediction model is stored. If no prediction model is stored, the computing server 101 reads multiple sets of historical antenna configuration parameters and the score of each set of antenna configuration parameters from the storage, and generates a prediction model, which can be called the initial prediction model. If a prediction model is stored, the stored prediction model is read directly.

[0049] Furthermore, the computing server 101 obtains the currently predicted optimal antenna configuration parameters based on the prediction model, and obtains a score for the antenna configuration parameters based on the simulation model. When the algorithm corresponding to the prediction model fails to converge or reaches the required number of iterations, the computing server 101 incrementally updates the prediction model based on the antenna configuration parameters and the score of the antenna configuration parameters until the algorithm corresponding to the prediction model converges or reaches the required number of iterations. The computing server 101 then uses the optimal antenna configuration parameters corresponding to the current prediction model as the target configuration parameters and stores the prediction model.

[0050] As one implementation method, the computing server 101 also corrects the above simulation model based on the RSRP value obtained from the simulation and the RSRP value obtained from the measurement, so as to obtain a corrected simulation model.

[0051] Furthermore, the computing server 101 also stores the above simulation model, the problem cell, and the antenna configuration parameters and corresponding scores tried during the above RF parameter optimization process.

[0052] The above system is illustrated using the example of a computing server 101 optimizing antenna configuration parameters and a network management server 102 distributing antenna configuration parameters. Alternatively, the computing server 101 and the network management server 102 can be integrated into one unit to directly optimize and distribute antenna configuration parameters.

[0053] It should be understood that Figure 1a , Figure 1b and Figure 1c The specific details of the antenna configuration parameter optimization methods involved will be further elaborated below.

[0054] The implementation process of the antenna configuration parameter optimization method provided in the embodiments of this application will be described below. (Refer to...) Figure 2 The diagram shown is a flowchart illustrating an antenna configuration parameter optimization method provided in an embodiment of this application. This embodiment uses a computing server executing the antenna configuration parameter optimization method as an example for illustration. The antenna configuration parameter optimization method includes steps 201-204, as detailed below:

[0055] 201. Based on the first antenna configuration parameters corresponding to the optimization region and the score of the first antenna configuration parameters, the first prediction model is incrementally updated to obtain the second prediction model;

[0056] The aforementioned optimization area can be the area corresponding to multiple communication cells controlled by the base station and optimized periodically.

[0057] The aforementioned first antenna configuration parameter can be any antenna configuration parameter obtained by the computing server during the process of determining the target configuration parameters. This first antenna configuration parameter may include one or more of the following: physical azimuth, physical downtilt, digital azimuth, digital downtilt, horizontal beamwidth, or vertical beamwidth, etc.

[0058] The score for the aforementioned first antenna configuration parameter is used to characterize the quality (specifically, signal quality) of the network corresponding to that first antenna configuration parameter. It should be understood that this score can also be referred to as a label, etc., and this solution does not specifically limit it to that term.

[0059] The above-mentioned incremental update of the first prediction model based on the first antenna configuration parameters corresponding to the optimization region and the score of the first antenna configuration parameters to obtain the second prediction model can be understood as: updating the calculation formula, parameters, or attributes of the first prediction model based on the first antenna configuration parameters corresponding to the optimization region and the score of the first antenna configuration parameters, thereby obtaining the second prediction model. In contrast, a full update generates the second prediction model based on the first antenna configuration parameters, the score of the first antenna configuration parameters, and the antenna configuration parameters and scores used before generating the first prediction model. That is to say, a full update does not generate the second prediction model based on the already generated first prediction model; it does not establish a correlation between the first and second prediction models. Incremental updates, compared to full updates, update the already generated first prediction model based on the first antenna configuration parameters and the score of the first antenna configuration parameters to obtain a new prediction model.

[0060] Specifically, based on the antenna configuration parameter set X N And the corresponding antenna configuration parameter score set F N Generate the first prediction model described above. The antenna configuration parameter set X is as follows. N ={x i Let ,i∈N} represent multiple sets of antenna configuration parameters x i The set of ratings; the set of ratings F N ={f(x i ),i∈N}, representing the score f(x) corresponding to each set of parameters in the multiple sets of antenna configuration parameters. i The set of ) is described below. The incremental update of the first prediction model mentioned above is introduced below.

[0061] As one implementation method, we will take the first prediction model as a probabilistic prediction model as an example for explanation.

[0062] First, define any two antenna configuration parameters x. i ,x j The covariance k(x) between i ,x j The following conditions must be met:

[0063]

[0064] Among them, antenna configuration parameter x i ,x j All belong to the antenna configuration parameter set X N α and θ are both non-zero coefficients.

[0065] Then, maximize the score F under these two hyperparameters α and θ. N The probability of occurrence is calculated by maximizing the marginal log-likelihood, which can be expressed as:

[0066]

[0067] Wherein, p(F N |α,θ) represents F under the corresponding hyperparameters α and θ. N The probability of occurrence.

[0068] The hyperparameters α and θ are obtained by maximizing the marginal log-likelihood as described above, and then k(x) is calculated accordingly. i ,x j The calculated covariance matrix K can be expressed as:

[0069]

[0070] Gain matrix F N It can be represented as:

[0071]

[0072] The parameters of the first prediction model include the covariance matrix K and the gain matrix F. N .

[0073] According to the first antenna configuration parameter x best And the score f(x) of the first antenna configuration parameters best Update the first prediction model described above. The first antenna configuration parameter x... best It was obtained based on the first prediction model; please refer to the details. Figure 4 The descriptions in the embodiments described herein will not be repeated here.

[0074] Specifically, the updated covariance matrix K of the first prediction model can be expressed as:

[0075]

[0076] The new gain matrix F N It can be represented as: F N =F N +[f(x best (4)

[0077] By adding a row and a column to the original covariance matrix as shown in formula (1), the covariance matrix is ​​obtained as shown in formula (3). This is based on the original gain matrix F. N As shown in formula (2), add a row to obtain the gain matrix F. N As shown in formula (4), the incremental update of the first prediction model can be achieved based on the above method. The above is only one example, and it can also be other forms of update, which are not specifically limited in this scheme.

[0078] As another implementation method, we will illustrate this with an example where the first prediction model is a neural network prediction model. Based on the antenna configuration parameter set X... N And the corresponding antenna configuration parameter score set F N Generate sample pairs. The sample pairs take antenna configuration parameters as input data and the score of the antenna configuration parameters as the expected output. Through learning, a neural network is trained.

[0079] Neural networks satisfy the following conditions:

[0080] W,B = argmin W,B {WX N +BF N} (5)

[0081] Accordingly, the parameters of the first prediction model are the weight matrix W and the bias matrix B.

[0082] According to the first antenna configuration parameter x best And the score f(x) of the first antenna configuration parameters best Update the first prediction model described above. The first antenna configuration parameter x... best It was obtained based on the first prediction model; please refer to the details. Figure 4 The descriptions in the embodiments described herein will not be repeated here.

[0083] The updated sample pairs from the first prediction model can be represented as follows:

[0084] X N =X N +[x best ]

[0085] F N =F N +[f(x best )]

[0086] The updated weight matrix W and bias matrix B of the first prediction model can be expressed as follows:

[0087] W = W

[0088] B = argmin B {WX N +BF N} (6)

[0089] It should be noted that for the first prediction model, only the bias matrix is ​​updated, not the weight matrix. Based on the above method, incremental updates of the aforementioned neural network prediction model can be achieved.

[0090] The above examples illustrate incremental updates of models using two different models. However, incremental updates can also be applied to other types of models, and this approach does not impose specific limitations on them.

[0091] The first prediction model mentioned above can be obtained in the following way:

[0092] An initial prediction model is obtained based on multiple sets of historical antenna configuration parameters corresponding to the optimization area and the score of each set of historical antenna configuration parameters, and the first prediction model is obtained based on the initial prediction model.

[0093] In one implementation, the initial prediction model can be the first prediction model. The computing server simulates multiple sets of historical antenna configuration parameters to obtain scores for each set, and then obtains the first prediction model based on these multiple sets of historical antenna configuration parameters and their scores. Specifically, the historical antenna configuration parameter with the highest score among the multiple sets of historical antenna configuration parameters is identified by the first prediction model; this highest-scoring historical antenna configuration parameter is the first antenna configuration parameter. It should be understood that the multiple sets of historical antenna configuration parameters can be obtained from a preset set of antenna configuration parameters. Alternatively, the multiple sets of historical antenna configuration parameters can be randomly generated.

[0094] As another implementation, the initial prediction model can also be the first prediction model in time sequence corresponding to the optimization region. The computing server incrementally updates the initial prediction model, obtaining the first prediction model through continuous iterative updates. Specifically, the computing server simulates multiple sets of historical antenna configuration parameters to obtain scores for each set of historical antenna configuration parameters, and obtains the initial prediction model based on the multiple sets of historical antenna configuration parameters and the scores for each set of historical antenna configuration parameters. The first prediction model is obtained by continuously iteratively updating the initial prediction model. The specific implementation method of obtaining the first prediction model by continuously iteratively updating the initial prediction model can be found in the description of the iterative update of the first prediction model in the above embodiments, and the update principle is the same, so it will not be repeated here.

[0095] Optionally, before step 201, the method further includes step 200: obtaining the stored first prediction model. By storing the first prediction model, it can be directly obtained when incrementally updating the first prediction model, thereby improving the efficiency of antenna configuration parameter optimization.

[0096] Accordingly, after step 201, the method further includes step 2011: storing the second prediction model so that the second prediction model can be incrementally updated subsequently. Step 2011 can be after step 201 and before step 202, or it can be after step 202; this solution does not specifically limit this.

[0097] In other words, when the computing server obtains any prediction model, it also stores the prediction model for use in subsequent incremental updates.

[0098] Optionally, the computing server can obtain the score of the first antenna configuration parameters in step 201 above by correcting the simulation model based on measurement data and simulation data. Specifically, before step 201, the method also includes steps 3001-3003, which can be found in [reference missing]. Figure 3 The diagram shown is a flowchart illustrating an antenna configuration parameter optimization method provided in an embodiment of this application. The details are as follows:

[0099] 3001. Acquire measurement data and simulation data, wherein the measurement data is the data obtained after the second target configuration parameters are sent out, and the simulation data is the data obtained based on the second target configuration parameters and the first simulation model;

[0100] The aforementioned measurement data includes one or more of the following: electronic maps, antenna files, engineering parameter data representing current radio frequency parameters, and measured MR data or DT data, etc. Among them, electronic maps are maps stored and viewed digitally using computer technology. Antenna files are antenna lobe diagram files. MR data is a measurement report reported by the User Equipment (UE), including the primary serving cell identifier, primary serving cell RSRP, neighboring cell identifier, neighboring cell RSRP, and the beam IDs and corresponding RSRPs of each cell-level broadcast beam in the primary and neighboring cells. DT data is road test data, similar to Minimization of Drive-Test (MDT) data. MDT data is a measurement report reported by the UE with latitude and longitude information, and can be considered as MR data with latitude and longitude.

[0101] It should be noted that the second target configuration parameter is the target configuration parameter obtained by the computing server in the previous search, relative to the current target parameter (i.e., the first target configuration parameter in step 204 below). Specifically, after the computing server distributes the second target configuration parameter through the network management server, it obtains the aforementioned measurement data through the network management server. The computing server also obtains the aforementioned simulation data by inputting the second target configuration parameter into the first simulation model.

[0102] 3002. The first simulation model is corrected based on the measurement data and simulation data to obtain the second simulation model;

[0103] As one implementation method, the link loss of the first simulation model is corrected based on the measurement data and simulation data to obtain a second simulation model.

[0104] Specifically, the first reference signal received power of each grid in the plurality of grids is obtained based on the measurement data; the second reference signal received power of each grid in the plurality of grids is obtained based on the simulation data of the first simulation model; the Kalman matrix of each grid is obtained based on the first reference signal received power and the second reference signal received power of each grid; the link loss of the first simulation model is corrected based on the first reference signal received power, the second reference signal received power and the Kalman matrix of each grid to obtain the corrected link loss, and then the second simulation model is obtained.

[0105] In other words, the accuracy of the simulation can be improved by correcting the link loss in the simulation model.

[0106] Specifically, when the measurement data obtained after the second target configuration parameters are issued is acquired, the calculation server performs rasterization processing on the MR / DT data based on the latitude and longitude information in each MR / DT data in the measurement data, and then assigns it to multiple raster cells.

[0107] For each grid cell, the reference signal received power (RSRP) of its most recently measured value is calculated. Specifically, this is done by removing some edge data from the measurements and averaging the remaining measurements. For example, for each grid cell, multiple measurement data at multiple times are obtained: MR = {mr} i}, i∈N′; each mr i The information includes the measured rsrp i Multiple MR signals at multiple times i Summary, according to each MR i rsrp i Sort by size from largest to smallest, and remove part of the MR (Mr / Mill) value. i For example, removing the MR located in the last 5% i This leads to the formation of new MR new The RSRP measurement value for each grid cell can be expressed as:

[0108]

[0109] For each grid cell, calculate its measurement variance R:

[0110]

[0111] Meanwhile, for each grid, simulation is performed based on the second target configuration parameters to obtain the simulated value RSRP' of the reference signal received power for each grid.

[0112] For each grid cell, its simulation variance is obtained. Specifically, the simulation variance P of that grid cell is obtained by acquiring the simulation values ​​of the surrounding eight grid cells. - .

[0113] The simulation value of the eight grids surrounding this grid can be represented as RSRP'. i ,i∈[1-8];

[0114] Calculate the mean of the simulated values ​​for this grid and the eight surrounding grids:

[0115] Calculate the simulation variance of the raster:

[0116] This solution uses 8 grids as an example, but other settings are possible, and this solution does not impose any specific limitations on this.

[0117] For each grid cell, calculate its Kalman matrix:

[0118]

[0119] For each grid cell, calculate its corrected reference signal received power RSRP based on its Kalman matrix:

[0120] RSRP”=RSRP’+Kal(RSRP-RSRP’)

[0121] Calculate the corrected link loss for each grid based on the corrected RSRP.

[0122]

[0123] This embodiment reduces simulation errors by employing Kalman filtering (multi-data source fusion) to calculate the corrected Linkloss for each grid cell, taking into account both the simulated RSRP value and the newly fed-back measured RSRP value from the live network. By combining the simulated RSRP value with the latest measured RSRP, simulation accuracy is improved.

[0124] As one implementation method, for each grid cell, its simulation variance P is updated and calculated. - ′=(I-Kal)P - I is the identity matrix. The computational server stores the simulation variance P. - This is so that it can be used during the next simulation model calibration.

[0125] The second simulation model is then obtained based on the corrected link loss described above.

[0126] 3003. The score of the first antenna configuration parameters is obtained based on the first antenna configuration parameters and the second simulation model.

[0127] The reference signal received power (RSRP) of each grid after correction is calculated by inputting the first antenna configuration parameters into the above-mentioned corrected simulation model, thereby obtaining a score for the first antenna configuration parameters. As one implementation method, this scheme calculates the proportion of grids in the entire optimized region whose RSRP values ​​are greater than the RSRP threshold, and records this as the score for the first antenna configuration parameters.

[0128] In this embodiment, the simulation model is corrected based on the measurement data obtained from the distribution of the second target configuration parameters and the simulation data obtained from simulating the second target configuration parameters, thereby obtaining a simulation model with higher simulation accuracy. This improves the accuracy and precision of antenna configuration parameter scoring during the optimization of target configuration parameters, and further enhances the efficiency of antenna configuration parameter optimization.

[0129] 202. Obtain the second skyline configuration parameters based on the second prediction model;

[0130] Specifically, the computing server obtains the second antenna configuration parameter by inputting multiple arbitrary antenna configuration parameters into the second prediction model. This second antenna configuration parameter is the optimal configuration parameter obtained by the second prediction model. For example, the second antenna configuration parameter is the configuration parameter with the highest score among the multiple arbitrary antenna configuration parameters.

[0131] 203. Determine whether to use the second antenna configuration parameters as the first target configuration parameters;

[0132] Specifically, the computing server determines whether to use the second antenna configuration parameter as the first target configuration parameter by determining whether a preset condition is met.

[0133] For example, the decision to use the second antenna configuration parameter as the first target configuration parameter is based on whether the iteration count N1 has been reached, or whether the second antenna configuration parameter has been obtained N2 consecutive times (i.e., the same antenna configuration parameter is obtained in N2 iterations). Here, N1 and N2 are both positive integers. If the iteration count N1 has been reached, or the second antenna configuration parameter has been obtained N2 consecutive times (in which case, the first antenna configuration parameter and the second antenna configuration parameter are the same), then the computing server determines to use the second antenna configuration parameter as the first target configuration parameter.

[0134] 204. If the second antenna configuration parameters are used as the first target configuration parameters, then the second antenna configuration parameters shall be sent out.

[0135] Once the computing server uses the second antenna configuration parameter as the first target configuration parameter, it distributes this first target configuration parameter (i.e., the second antenna configuration parameter) to the network management server. In other words, the computing server sends the first target configuration parameter to the network management server so that the network management server can distribute it accordingly.

[0136] In one implementation, if the second antenna configuration parameter is not used as the first target configuration parameter, the calculation server incrementally updates the second prediction model based on the second antenna configuration parameter and its score to obtain a third prediction model; and obtains the third antenna configuration parameter based on the third prediction model. Here, not using the second antenna configuration parameter as the first target configuration parameter means that a preset condition has not been met, such as not reaching the current iteration number N1, or not obtaining the second antenna configuration parameter consecutively N2 times.

[0137] Accordingly, the computing server determines whether to use the third antenna configuration parameter as the first target configuration parameter; if the third antenna configuration parameter is used as the first target configuration parameter, the third antenna configuration parameter is distributed through the network management server.

[0138] The computing server repeats the above steps until it obtains the first target configuration parameters, which are then distributed through the network management server.

[0139] It should be noted that this application uses a computing server as an example for illustration. The network management server and the computing server can also be integrated into one unit. Accordingly, if the aforementioned second antenna configuration parameters are used as the first target configuration parameters, the computing server will directly distribute these second antenna configuration parameters.

[0140] In this embodiment, the first prediction model is incrementally updated based on the first antenna configuration parameters and their scores, and new antenna configuration parameters are obtained from the updated prediction model. This incremental update allows the prediction model to quickly converge to antenna configuration parameters with higher scores, thus improving the optimization efficiency of antenna configuration parameters. Compared to existing full-scale updates, this solution effectively improves delivery quality and efficiency.

[0141] The following is a detailed description of the specific implementation process of the antenna configuration parameter optimization method provided in the embodiments of this application. (Refer to...) Figure 4The diagram shown is a flowchart illustrating an antenna configuration parameter optimization method provided in an embodiment of this application. It includes steps 401-406, as detailed below:

[0142] 401. Obtain the first measurement data corresponding to the optimized region;

[0143] The computing server can obtain the first measurement data corresponding to the optimized area through the network management server. The network management server collects this first measurement data by controlling the base station.

[0144] The first measurement data is the initial measurement data corresponding to the optimization region. In other words, the first measurement data is the measurement data corresponding to the optimization region before optimization has begun.

[0145] 402. Determine the third target configuration parameters based on the first measurement data, and distribute the third target configuration parameters through the network management server;

[0146] The computing server determines the cells that require antenna configuration parameter optimization based on the first measurement data, and then determines the third target configuration parameters. Accordingly, when the first prediction model in the aforementioned embodiment is the initial prediction model, the third target configuration parameter is the second target configuration parameter in the aforementioned embodiment.

[0147] Step 402 may include steps 4021-4023, as follows:

[0148] 4021. Construct a simulation model for grid RSRP evaluation based on the first measurement data;

[0149] The computing server uses the aforementioned first measurement data to establish a simulation model based on the electromagnetic wave propagation characteristics. This simulation model includes link loss.

[0150] The propagation characteristics of electromagnetic waves satisfy the following formula:

[0151] Reference signal received power RSRP = AntennaGain + Power - Linkloss (7);

[0152] The aforementioned link loss is the link loss of each grid cell obtained after grating processing the first measurement data.

[0153] Specifically, based on the latitude and longitude information in each MR / DT data in the first measurement data, the MR / DT data is rasterized to obtain multiple grids, and then the RSRP of each grid is obtained. Based on the antenna file and engineering parameter data in the first measurement data, the antenna gain (AntennaGain) and power (Power) of each grid are obtained. Then, according to formula (7), the link loss of each grid can be obtained. The calculation server stores the link loss of each grid for subsequent use of the simulation model.

[0154] 4022. Identify the cells that need adjustment (problem cells) based on the first measurement data;

[0155] The computing server obtains the RSRP for each grid based on MR / DT data. Grids with RSRP below a preset threshold are defined as weak coverage grids. Weak coverage grids are clustered to obtain weak coverage areas. Cells whose latitude and longitude belong to weak coverage areas are defined as problem cells.

[0156] 4023. Obtain the third target configuration parameters corresponding to the problematic cell, and distribute the third target configuration parameters through the network management server.

[0157] Specifically, the computing server randomly generates a set of antenna configuration parameters x. i ,x i ∈X all X all This represents the global antenna configuration parameters, which is the complete set of optional antenna configuration parameters for the problem cell corresponding to the optimization area; in other words, it is the set of all optional parameters.

[0158] Regarding antenna configuration parameter x i Based on the antenna file and engineering parameter data, the AntennaGain and Power corresponding to each grid are obtained.

[0159] According to antenna configuration parameter x i The antenna configuration parameter x can be obtained from the above simulation model. i The RSRP value for each corresponding grid cell.

[0160] The percentage of rasters in the entire optimized region whose RSRP value is greater than the RSRP threshold is calculated and denoted as the score f(x). i Among them, the calculation server stores the antenna configuration parameter x. i And antenna configuration parameter x i The corresponding rating.

[0161] For multiple different antenna configuration parameters x i Repeat the above steps to obtain the scores corresponding to each group of antenna configuration parameters until the number of attempts N3 is met, and then output the antenna configuration parameter x with the highest score during the attempt. * and its evaluation score f(x) * ).

[0162] The highest-rated antenna configuration parameter x * This is the third target configuration parameter. This third target configuration parameter includes the configuration of each RF parameter for each antenna in the aforementioned problematic cell. The computing server distributes this third target configuration parameter to the live network via the network management server to adjust the RF parameters of the antennas in the actual communication network.

[0163] 403. Obtain the second measurement data corresponding to the optimized region, wherein the second measurement data is the data obtained after the third target configuration parameters are sent out;

[0164] The computing server obtains the second measurement data, i.e., the new measurement data, corresponding to the issuance of the third target configuration parameters, through the network management server. The computing server may send the second measurement data obtained after the issuance of the third target configuration parameters to the network management server after a preset time interval. This preset time interval can be several hours or several days, etc., and this solution does not specifically limit it. When the third target configuration parameter is the second target configuration parameter in the aforementioned embodiments, the second measurement data can correspondingly be the measurement data in the aforementioned embodiments.

[0165] The second measurement data may include only MR / DT data. It may also include electronic maps, antenna files, engineering parameter data representing the current RF parameters, measured MR data and / or DT data, etc. This solution does not impose specific limitations on this.

[0166] 404. Determine whether further optimization is needed based on the second measurement data;

[0167] The computing server, based on the new measurement data and the latitude and longitude information in each MR / DT data point, rasterizes the MR / DT data and assigns it to multiple grids generated in step 4021, thereby obtaining the RSRP of each grid. It then calculates the proportion of grids in the entire optimization area whose RSRP value is greater than the RSRP threshold. If this proportion meets a preset requirement, the optimization proceeds to step 406 to end; otherwise, if the proportion of grids in the entire optimization area whose RSRP value is greater than the RSRP threshold is less than the preset requirement, the optimization proceeds to step 405.

[0168] 405. If further optimization is needed, obtain the first target configuration parameter based on multiple sets of historical configuration parameters and the score of each set of historical configuration parameters, and then issue the first target configuration parameter.

[0169] As one implementation, the computing server obtains the first target configuration parameters based on the second prediction model; wherein, the computing server incrementally updates the first prediction model based on the first antenna configuration parameters and the score of the first antenna configuration parameters to obtain the second prediction model.

[0170] Specifically, 1) the computing server obtains an initial prediction model based on multiple sets of historical antenna configuration parameters tried during the first optimization process to obtain the aforementioned third target configuration parameters, and the score of each set of historical antenna configuration parameters, and stores the initial prediction model. For details, please refer to the description of obtaining the initial prediction model in step 201 of the aforementioned embodiments, which will not be repeated here.

[0171] The computing server then reads the stored initial prediction model and iteratively updates it to obtain the first prediction model, which is then stored. For details, please refer to the description of obtaining the first prediction model in step 201 of the aforementioned embodiments; it will not be repeated here.

[0172] 2) During subsequent updates, the computing server reads the stored first prediction model, obtains the first antenna configuration parameters based on the first prediction model, simulates the first antenna configuration parameters to obtain a score for the first antenna configuration parameters, and then incrementally updates the first prediction model based on the first antenna configuration parameters and the score of the first antenna configuration parameters to obtain the second prediction model. For details, please refer to the description of incrementally updating the first prediction model to obtain the second prediction model in step 201 of the aforementioned embodiments, which will not be repeated here.

[0173] There are multiple ways to obtain the first antenna configuration parameters based on the first prediction model. This application illustrates two implementation methods as examples.

[0174] As a first implementation method, we will take the first prediction model in the aforementioned embodiment as a probabilistic prediction model as an example for explanation. For any set of antenna configuration parameters x... k By configuring the antenna parameter x k The input is fed into the first prediction model to calculate the predicted score μ(x). k ) and the variance σ(x) of the predicted value k ):

[0175] k = [k(x k ,x1),k(x k ,x2)…,k(x k ,xN )]

[0176] μ(x k )=kK -1 F N

[0177] σ 2 (x k )=k(x k ,x k )-kK -1 k T

[0178] Based on this first prediction model, for each x among all global antenna configuration parameters... k ,x k ∈X all Based on the above formula, the predicted score μ(x) corresponding to each antenna configuration parameter can be calculated. k ) and the variance σ(x) of the predicted value k Alternatively, only a subset of antenna configuration parameters from all global antenna configuration parameters can be sampled and calculated; this scheme does not impose specific limitations on this. For each x in all global antenna configuration parameters... k According to the cumulative probability density formula, μ(x) predicted above... k ) and σ(x k The gain corresponding to the antenna configuration parameters can be calculated.

[0179]

[0180] The f(x) * ) is the target configuration parameter x that was distributed to the live network in the previous round of parameter optimization. * The score corresponding to (i.e., the third objective configuration parameter). This gain EI(x) k The meaning of ) can be understood as: antenna configuration parameter x k The score f(x) k Compared to the antenna configuration parameters previously distributed to the live network, x * The score f(x) * We need to have even higher expectations.

[0181] Based on the antenna configuration parameters obtained above, EI(x) k Determine the antenna configuration parameter x corresponding to the largest EI. best That is, the first antenna configuration parameter x mentioned above best :

[0182]

[0183] By x bestThe input is given to the simulation model to obtain the corresponding score f(x). best ).

[0184] By determining whether the total number of iterations N1 has been reached, or whether the value of x has been obtained consecutively N2 times. best This is used to determine whether to stop the algorithm. If the iteration count N1 is reached, or if the x value is obtained consecutively for N2 iterations... best Then output the antenna configuration parameter x. best and its score f(x) best The N1 mentioned above can be, for example, 100, and the N2 mentioned above can be, for example, 10. This solution does not impose specific limitations on this.

[0185] As a second implementation method, we will take the example of a neural network prediction model in the aforementioned embodiment as an example.

[0186] For any set of antenna configuration parameters x k By configuring the antenna parameter x k The input is fed into the first prediction model to calculate the predicted score μ(x). k ) and the variance σ(x) of the predicted value k ):

[0187] μ(x k )=Wx k +B

[0188] σ 2 (x k ) = G

[0189] Where G is a non-zero constant.

[0190] Based on this first prediction model, for each x among all global antenna configuration parameters... k ,x k ∈X all Based on the above formula, the predicted score μ(x) corresponding to each antenna configuration parameter can be calculated. k ) and the variance σ(x) of the predicted value k Alternatively, only some antenna configuration parameters can be sampled and calculated; this solution does not impose specific limitations on this.

[0191] For each x in all global antenna configuration parameters k According to the cumulative probability density formula, μ(x) predicted above... k ) and σ(x k The gain corresponding to the antenna configuration parameters can be calculated.

[0192]

[0193] Where, f(x) * ) is the target configuration parameter x that was distributed to the live network in the previous round of parameter optimization. * The score corresponding to (i.e., the third target configuration parameter).

[0194] Based on the antenna configuration parameters obtained above, EI(x) k Determine the antenna configuration parameter x corresponding to the largest EI. best That is, the first antenna configuration parameter x mentioned above best :

[0195]

[0196] By x best The input is given to the simulation model to obtain the corresponding score f(x). best ).

[0197] By determining whether the total number of iterations N1 has been reached, or whether the value of x has been obtained consecutively N2 times. best This is used to determine whether to stop the algorithm. If the iteration count N1 is reached, or if the x value is obtained consecutively for N2 iterations... best Then output the antenna configuration parameter x. best and its score f(x) best The N1 mentioned above can be, for example, 100, and the N2 mentioned above can be, for example, 10. This solution does not impose specific limitations on this.

[0198] If the above conditions are not met, then according to x best and score f(x) best The first prediction model is updated to obtain the second prediction model. This incremental update can be referred to the description of step 201 in the foregoing embodiments, and will not be repeated here.

[0199] 406. If no further optimization is needed, end the optimization process.

[0200] The above explanation uses only two instances of issuing target configuration parameters as an example. It can also involve any number of other iterations. For instance, after issuing the first target configuration parameter in step 405, steps 403-406 can be repeated until the measurement data corresponding to the optimization area meets the preset requirements, at which point optimization stops.

[0201] In this embodiment, the prediction model is incrementally updated based on antenna configuration parameters and their scores, resulting in new antenna configuration parameters. This method, by continuously updating the prediction model incrementally, allows for rapid convergence to antenna configuration parameters with higher scores, thereby improving the efficiency of antenna configuration parameter optimization and effectively enhancing delivery quality and efficiency.

[0202] Furthermore, the antenna configuration parameter optimization method provided in this solution does not require rebuilding the simulation model when new measurement data is obtained, based on the aforementioned stored simulation model. Compared with existing technologies, this effectively improves the overall optimization efficiency.

[0203] Reference Figure 5 As shown in the figure, this application provides an antenna configuration parameter optimization device. The device includes a first model generation module 501, a first parameter generation module 502, a judgment module 503, and a parameter determination module 504, as detailed below:

[0204] The first model generation module 501 is used to incrementally update the first prediction model based on the first antenna configuration parameters corresponding to the optimization region and the score of the first antenna configuration parameters to obtain the second prediction model.

[0205] The first parameter generation module 502 is used to obtain the second skyline configuration parameters according to the second prediction model;

[0206] The judgment module 503 is used to determine whether to use the second antenna configuration parameters as the first target configuration parameters;

[0207] The parameter determination module 504 is used to send the second antenna configuration parameters if the second antenna configuration parameters are used as the first target configuration parameters.

[0208] In this embodiment, the first prediction model is incrementally updated based on the first antenna configuration parameters and their scores, resulting in new antenna configuration parameters. This method improves the prediction accuracy of the prediction model by updating it, thereby increasing the efficiency of antenna configuration parameter optimization and effectively enhancing both delivery quality and efficiency.

[0209] The device further includes: a second model generation module, used to obtain a third prediction model based on the second antenna configuration parameters, the score of the second antenna configuration parameters, and the second prediction model if the second antenna configuration parameters are not the first target configuration parameters; and a second parameter generation module, used to obtain third antenna configuration parameters based on the third prediction model.

[0210] The scoring of the first antenna configuration parameters is obtained based on the first antenna configuration parameters and the first simulation model. The device also includes a simulation model generation module, used to: acquire measurement data and simulation data, wherein the measurement data is data obtained after the second target configuration parameters are sent out, and the simulation data is data obtained based on the second target configuration parameters and the second simulation model; and correct the second simulation model based on the measurement data and simulation data to obtain the first simulation model.

[0211] The device further includes an acquisition module for: acquiring the stored first prediction model.

[0212] The device further includes a third model generation module, used to: obtain an initial prediction model based on multiple sets of historical antenna configuration parameters corresponding to the optimization area and the score of each set of historical antenna configuration parameters, so as to obtain the first prediction model based on the initial prediction model.

[0213] The specific implementation methods of the above modules can be found in the corresponding descriptions in the foregoing embodiments, and will not be repeated here.

[0214] This application also provides an antenna configuration parameter optimization device, such as... Figure 6 As shown, the antenna configuration parameter optimization device includes at least one processor 601, at least one memory 602, and at least one communication interface 603. The processor 601, the memory 602, and the communication interface 603 are connected through the communication bus and communicate with each other.

[0215] Processor 601 may be a general-purpose central processing unit (CPU), a microprocessor, an application-specific integrated circuit (ASIC), or one or more integrated circuits used to control the execution of programs in the above scheme.

[0216] Communication interface 603 is used to communicate with other devices or communication networks, such as Ethernet, Radio Access Network (RAN), Wireless Local Area Networks (WLAN), etc.

[0217] Memory 602 may be a read-only memory (ROM) or other type of static storage device capable of storing static information and instructions, random access memory (RAM) or other type of dynamic storage device capable of storing information and instructions, or electrically erasable programmable read-only memory (EEPROM), compact disc read-only memory (CD-ROM) or other optical disc storage, optical disc storage (including compressed optical discs, laser discs, optical discs, digital versatile optical discs, Blu-ray discs, etc.), magnetic disk storage media or other magnetic storage devices, or any other medium capable of carrying or storing desired program code in the form of instructions or data structures and accessible by a computer, but not limited thereto. Memory may exist independently and be connected to the processor via a bus. Memory may also be integrated with the processor.

[0218] The memory 602 stores the application code for executing the above scheme, and its execution is controlled by the processor 601. The processor 601 executes the application code stored in the memory 602.

[0219] The code stored in memory 602 can execute one of the antenna configuration parameter optimization methods provided above.

[0220] This application also provides a chip system applied to an electronic device; the chip system includes one or more interface circuits and one or more processors; the interface circuits and the processors are interconnected via lines; the interface circuits are used to receive signals from the memory of the electronic device and send the signals to the processor, the signals including computer instructions stored in the memory; when the processor executes the computer instructions, the electronic device executes the method.

[0221] This application also provides a computer-readable storage medium storing instructions that, when executed on a computer or processor, cause the computer or processor to perform one or more steps of any of the above methods.

[0222] This application also provides a computer program product containing instructions. When the computer program product is run on a computer or processor, it causes the computer or processor to perform one or more steps of any of the methods described above.

[0223] In the above embodiments, implementation can be achieved entirely or partially through software, hardware, firmware, or any combination thereof. When implemented using software, it can be implemented entirely or partially as a computer program product. The computer program product includes one or more computer instructions. When the computer program instructions are loaded and executed on a computer, all or part of the processes or functions described in the embodiments of this application are generated. The computer can be a general-purpose computer, a special-purpose computer, a computer network, or other programmable device. The computer instructions can be stored in a computer-readable storage medium or transmitted through the computer-readable storage medium. The computer instructions can be transmitted from one website, computer, server, or data center to another website, computer, server, or data center via wired (e.g., coaxial cable, fiber optic, digital subscriber line) or wireless (e.g., infrared, wireless, microwave, etc.) means. The computer-readable storage medium can be any available medium that a computer can access or a data storage device such as a server or data center that integrates one or more available media. The available medium can be a magnetic medium (e.g., floppy disk, hard disk, magnetic tape), an optical medium (e.g., DVD), or a semiconductor medium (e.g., solid-state disk (SSD)).

[0224] In this application, "at least one" means one or more, and "more" means two or more. "And / or" describes the relationship between related objects, indicating that there can be three relationships. For example, A and / or B can mean: A exists alone, A and B exist simultaneously, or B exists alone, where A and B can be singular or plural.

[0225] Those skilled in the art will understand that all or part of the processes in the methods of the above embodiments can be implemented by a computer program instructing related hardware. This program can be stored in a computer-readable storage medium, and when executed, it can include the processes described in the above method embodiments. The aforementioned storage medium includes various media capable of storing program code, such as ROM or random access memory (RAM), magnetic disks, or optical disks.

[0226] The above description is merely a specific implementation of the embodiments of this application, but the protection scope of the embodiments of this application is not limited thereto. Any changes or substitutions within the technical scope disclosed in the embodiments of this application should be covered within the protection scope of the embodiments of this application. Therefore, the protection scope of the embodiments of this application should be determined by the protection scope of the claims.

Claims

1. A method for optimizing antenna configuration parameters, characterized in that, include: The first prediction model is incrementally updated based on the first antenna configuration parameters corresponding to the optimization region and the score of the first antenna configuration parameters to obtain the second prediction model. The second skyline configuration parameters are obtained based on the second prediction model; Determine whether to use the second antenna configuration parameters as the first target configuration parameters; If the second antenna configuration parameters are used as the first target configuration parameters, then the second antenna configuration parameters will be sent out.

2. The method according to claim 1, characterized in that, The method further includes: If the second antenna configuration parameter is not used as the first target configuration parameter, the second prediction model is incrementally updated based on the second antenna configuration parameter and the score of the second antenna configuration parameter to obtain the third prediction model; The third antenna configuration parameters are obtained based on the third prediction model.

3. The method according to claim 1 or 2, characterized in that, The method further includes: Acquire measurement data and simulation data, wherein the measurement data is obtained after the second target configuration parameters are sent out, and the simulation data is obtained based on the second target configuration parameters and the first simulation model; The first simulation model is corrected based on the measurement data and simulation data to obtain the second simulation model; The score of the first antenna configuration parameters is obtained based on the first antenna configuration parameters and the second simulation model.

4. The method according to any one of claims 1 to 3, characterized in that, The method further includes: Retrieve the stored first prediction model.

5. The method according to any one of claims 1 to 4, characterized in that, The method further includes: An initial prediction model is obtained based on multiple sets of historical antenna configuration parameters corresponding to the optimized region and the score of each set of historical antenna configuration parameters, so as to obtain the first prediction model based on the initial prediction model.

6. An antenna configuration parameter optimization device, characterized in that, include: The first model generation module is used to incrementally update the first prediction model based on the first antenna configuration parameters corresponding to the optimization region and the score of the first antenna configuration parameters to obtain the second prediction model. The first parameter generation module is used to obtain the second skyline configuration parameters based on the second prediction model; The determination module is used to determine whether to use the second antenna configuration parameters as the first target configuration parameters; The parameter determination module is used to send out the second antenna configuration parameters if the second antenna configuration parameters are used as the first target configuration parameters.

7. The apparatus according to claim 6, characterized in that, The device further includes: The second model generation module is used to incrementally update the second prediction model based on the second antenna configuration parameters and the score of the second antenna configuration parameters to obtain a third prediction model if the second antenna configuration parameters are not used as the first target configuration parameters. The second parameter generation module is used to obtain the third antenna configuration parameters based on the third prediction model.

8. The apparatus according to claim 6 or 7, characterized in that, The device further includes a scoring determination module for: Acquire measurement data and simulation data, wherein the measurement data is obtained after the second target configuration parameters are issued, and the simulation data is obtained based on the second target configuration parameters and the first simulation model; calibrate the first simulation model based on the measurement data and simulation data to obtain a second simulation model; obtain a score for the first antenna configuration parameters based on the first antenna configuration parameters and the second simulation model.

9. The apparatus according to any one of claims 6 to 8, characterized in that, The device further includes an acquisition module for: Retrieve the stored first prediction model.

10. The apparatus according to any one of claims 6 to 9, characterized in that, The device further includes a third model generation module, used for: An initial prediction model is obtained based on multiple sets of historical antenna configuration parameters corresponding to the optimized region and the score of each set of historical antenna configuration parameters, so as to obtain the first prediction model based on the initial prediction model.

11. An antenna configuration parameter optimization device, characterized in that, It includes a processor and a memory; wherein the memory is used to store program code, and the processor is used to call the program code to perform the method as described in any one of claims 1 to 5.

12. A computer-readable storage medium, characterized in that, The computer-readable storage medium stores a computer program that is executed by a processor to implement the method of any one of claims 1 to 5.

13. A computer program product, characterized in that, When the computer program product is run on a computer, it causes the computer to perform the method as described in any one of claims 1 to 5.

Images