Antenna planning method, system, device and medium

By encoding the antenna layout planning into a population and using genetic algorithms and proxy models to optimize, the problem that antenna position change in the prior art cannot change the network coverage rate is solved, and more efficient antenna planning and optimization are achieved.

CN118890632BActive Publication Date: 2024-11-29CENT SOUTH UNIV
View PDF 2 Cites 0 Cited by

Patent Information

Application Number
CN202411394832.6
Authority / Receiving Office
CN · China
Patent Type
Patents(China)
Current Assignee / Owner
Filing Date
2024-10-08
Publication Date
2024-11-29
Estimated Expiration
2044-10-08

AI Technical Summary

Technical Problem

In the existing antenna distribution planning technology, there is a problem of changing the antenna position but not changing the network coverage rate, and the problem of changing the number of antennas or other variables causing complex antenna optimization when looking for the optimal antenna configuration.

Method used

An antenna planning method is proposed. By encoding the antenna layout planning into a population represented by real-number code, the cross-mutation operator in the genetic algorithm generates offspring populations, and optimizes them based on the preset proxy model, and gradually adjusts the parent population to Improve network coverage and algorithm efficiency.

Benefits of technology

It effectively improves the algorithm efficiency of network coverage and antenna planning methods, solves the problems of arrangement invariance and optimization complexity, and can better meet practical application needs.

✦ Generated by Eureka AI based on patent content.

Smart Images

  • Figure CN118890632B_ABST
    Figure CN118890632B_ABST
Patent Text Reader

Abstract

The present application relates to an antenna planning method, system, device and medium. The method encodes the antenna layout plan into a population represented by real number coding, generates a child population by using a crossover mutation operator from an initialized parent population, optimizes the child individuals using a local search strategy based on the prediction results of a preset proxy model, thereby determining the optimal parent population based on the parent population and the child individuals, and finally selecting the optimal individual from the optimal parent population to obtain the corresponding antenna layout plan. The method solves the problems of ignoring permutation invariance and difficulty in handling non-fixed dimensional solutions by providing multiple optimal antenna layout plans, and can effectively improve network coverage and the algorithm efficiency of the antenna planning method, thereby better meeting the needs of practical applications.
Need to check novelty before this filing date? Find Prior Art

Description

Technical Field

[0001] The embodiments of the present application relate to the field of antenna planning technology, and in particular, to an antenna planning method, system, device and medium. Background Art

[0002] The growth of mobile devices and applications has increased the demand for mobile data traffic, which in turn has accelerated the development of 5G technology. Currently, improving 5G network coverage has become the key to popularizing and stabilizing 5G services.

[0003] In-Building Distributed Antenna Systems (IB-DAS) are widely deployed as a popular technology to improve mobile communication services in buildings. However, existing antenna distribution planning technologies have problems such as the arrangement invariance that changing the antenna position cannot change the network coverage, and the complexity of antenna optimization caused by changes in the number of antennas or other variables in finding the optimal antenna configuration. Summary of the invention

[0004] The following is a summary of the subject matter described in detail herein. This summary is not intended to limit the scope of the claims.

[0005] The main purpose of the embodiments of the present disclosure is to propose an antenna planning method, system, device and medium, which can effectively improve the network coverage and the algorithm efficiency of the antenna planning method.

[0006] A first aspect of an embodiment of the present application provides an antenna planning method, the method comprising:

[0007] When an Nth parent population is obtained, a first performance indicator of an individual of the Nth parent population is calculated; if N is 1, the Nth parent population is generated based on initial antenna positions of multiple antennas in a layout scenario; if N is a positive integer greater than 1, the Nth parent population is generated based on an N-1th parent population;

[0008] Obtaining an Nth generation population according to the crossover mutation of the Nth parent population, and calculating a second performance index of individuals in the Nth generation population;

[0009] Optimizing the individuals of the Nth generation population according to the second performance index of the individuals of the Nth generation population, and calculating the third performance index of the individuals of the Nth generation population after optimization;

[0010] According to the individuals of the Nth offspring population with the highest third performance index, the individuals of the Nth parent population with the lowest first performance index are adjusted to obtain the N+1th parent population;

[0011] If the N+1th parent population meets the preset condition, the final antenna positions of the multiple antennas in the layout scenario are generated according to the individuals of the N+1th parent population.

[0012] The embodiment of the present application provides an antenna planning method, which encodes the antenna layout plan as a population represented by real number coding, generates a child population by using a crossover mutation operator from an initialized parent population, optimizes the child individuals using a local search strategy based on the prediction results of a preset proxy model, thereby determining the optimal parent population based on the parent population and the child individuals, and finally selecting the optimal individual from the optimal parent population to obtain the corresponding antenna layout plan. This method solves the problems of ignoring permutation invariance and difficulty in handling non-fixed dimensional solutions by providing multiple optimal antenna layout plans, and can effectively improve network coverage and the algorithm efficiency of the antenna planning method, thereby better meeting the needs of actual applications.

[0013] In some embodiments of the present application, before calculating the first performance indicator of the individuals of the Nth parent population when the Nth parent population is obtained, the method further includes:

[0014] Clustering the multiple antennas according to their initial antenna positions in the layout scenario to obtain multiple antenna classes;

[0015] One antenna is selected from each of the antenna classes to form a first parent population.

[0016] In some embodiments of the present application, the calculating the first performance index of the individuals of the Nth parent population includes:

[0017] Obtaining the network signal strength of the individuals of the Nth parent population by using the layout scenario parameters corresponding to the layout scenario and the antenna parameters by using ray tracing;

[0018] The network coverage of the individuals in the Nth parent population is determined according to the network signal strength of the individuals in the Nth parent population.

[0019] In some embodiments of the present application, the first performance indicator includes network coverage, and the obtaining of the Nth child population according to the crossover mutation of the Nth parent population includes:

[0020] Place all individuals of the Nth parent population on the target coverage plane respectively;

[0021] Randomly splitting the target coverage plane to obtain multiple target coverage sub-planes;

[0022] The target coverage sub-planes are randomly spliced ​​to obtain the Nth child population corresponding to the Nth parent population.

[0023] In some embodiments of the present application, the second performance indicator includes network coverage, and the calculation of the second performance indicator of the individuals of the Nth generation population includes:

[0024] Inputting the antenna positions corresponding to the individuals of the Nth parent population into a preset proxy model, obtaining the network coverage of the individuals of the Nth child population output by the preset proxy model,

[0025] The proxy model is trained based on the antenna position corresponding to an individual in at least one parent population among the first N-1 parent populations.

[0026] In some embodiments of the present application, the step of optimizing the individuals of the Nth generation population according to the second performance index of the individuals of the Nth generation population includes:

[0027] Determine, according to the network coverage rates of the individuals in the Nth generation population, the individual with the highest network coverage rate in the Nth generation population as the best generation individual;

[0028] confirming a plurality of antenna positions in the parent population that do not belong to the initial antenna position in the best offspring individual;

[0029] The coverage contributions corresponding to multiple antenna positions in the parent population that do not belong to the initial antenna position in each of the best offspring individuals are calculated, and the best offspring individuals are adjusted according to the coverage contributions to obtain optimized individuals of the Nth offspring population.

[0030] In some embodiments of the present application, generating final antenna positions of multiple antennas in a layout scenario according to individuals of the N+1th parent population includes:

[0031] Calculating a fourth performance index of individuals in the N+1th parent population;

[0032] Determine, according to the fourth performance index of the individuals in the N+1th parent population, the individual with the highest fourth performance index among the individuals in the N+1th parent population;

[0033] The final antenna positions of the multiple antennas in the layout scenario are generated according to the individual with the highest fourth performance index among the individuals of the N+1th parent population.

[0034] A second aspect of an embodiment of the present invention provides an antenna planning system, the system comprising:

[0035] A calculation module, configured to calculate a first performance indicator of an individual of the Nth parent population when an Nth parent population is obtained; if N is 1, the Nth parent population is generated based on initial antenna positions of multiple antennas in a layout scenario; if N is a positive integer greater than 1, the Nth parent population is generated based on an N-1th parent population;

[0036] A generation module, used to obtain an Nth generation population according to the crossover mutation of the Nth parent generation population, and calculate a second performance indicator of individuals of the Nth generation population;

[0037] An optimization module, used for optimizing the individuals of the Nth generation population according to the second performance index of the individuals of the Nth generation population, and calculating the third performance index of the individuals of the Nth generation population after optimization;

[0038] An evaluation module, configured to adjust the individuals of the Nth parent population with the lowest first performance index according to the individuals of the Nth child population with the highest second performance index, to obtain an N+1th parent population;

[0039] An updating module is used to generate final antenna positions of multiple antennas in a layout scenario according to individuals of the N+1th parent population if the N+1th parent population meets a preset condition.

[0040] A third aspect of an embodiment of the present invention provides an electronic device, comprising: at least one control processor and a memory for communicating with the at least one control processor; the memory stores instructions that can be executed by the at least one control processor, and the instructions are executed by the at least one control processor so that the at least one control processor can execute the above-mentioned antenna planning method.

[0041] A fourth aspect of an embodiment of the present invention provides a computer-readable storage medium, wherein the computer-readable storage medium stores computer-executable instructions, and the computer-executable instructions are used to enable a computer to execute the above-mentioned antenna planning method.

[0042] It can be understood that the beneficial effects of the second to fourth aspects compared with the related art are the same as the beneficial effects of the first aspect compared with the related art. Please refer to the relevant description in the first aspect, and no further details will be given here. BRIEF DESCRIPTION OF THE DRAWINGS

[0043] In order to more clearly illustrate the technical solutions in the embodiments of the present application, the drawings required for use in the embodiments or related technical descriptions will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the embodiments of the present application. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying creative labor.

[0044] Figure 1 It is a flowchart of an antenna planning method provided in an embodiment of the present application;

[0045] Figure 2 It is a schematic diagram of clustering multiple antennas in a local scene provided by an embodiment of the present application;

[0046] Figure 3 It is a schematic diagram of ray tracing model training provided in an embodiment of the present application;

[0047] Figure 4 is a schematic diagram of segmenting and splicing a target coverage plane provided by an embodiment of the present application;

[0048] Figure 5 It is a schematic diagram of the execution of the proxy model provided in the embodiment of the present application;

[0049] Figure 6 is a schematic diagram of optimizing antenna positions according to coverage contribution provided by an embodiment of the present application;

[0050] Figure 7 It is a flow chart of an implementation method of an antenna planning method provided in an embodiment of the present application;

[0051] Figure 8 It is a structural diagram of an antenna planning system provided in an embodiment of the present application;

[0052] Fig. 9 It is a schematic diagram of the structure of an electronic device provided in an embodiment of the present application. DETAILED DESCRIPTION

[0053] The embodiments of the present application are described in detail below, and examples of the embodiments are shown in the accompanying drawings, wherein the same or similar reference numerals throughout represent the same or similar elements or elements having the same or similar functions. The embodiments described below with reference to the accompanying drawings are exemplary and are only used to explain the present application, and cannot be understood as limiting the present application.

[0054] In the description of this application, if there is a description of first, second, etc., it is only for the purpose of distinguishing technical features, and cannot be understood as indicating or implying relative importance or implicitly indicating the number of the indicated technical features or implicitly indicating the order of the indicated technical features.

[0055] In the description of the present application, it should be understood that the descriptions involving orientation, such as the orientation or positional relationship indicated as up, down, etc., are based on the orientation or positional relationship shown in the drawings, and are only for the convenience of describing the present application and simplifying the description, and do not indicate or imply that the device or element referred to must have a specific orientation, be constructed and operated in a specific orientation, and therefore should not be understood as a limitation on the present application.

[0056] In the description of this application, it should be noted that, unless otherwise clearly defined, terms such as setting, installing, connecting, etc. should be understood in a broad sense, and technicians in the relevant technical field can reasonably determine the specific meaning of the above terms in this application based on the specific content of the technical solution.

[0057] The exponential growth of mobile devices and applications has significantly increased the demand for mobile data traffic, which has accelerated the development of 5G technology. Improving 5G network coverage has become the key to popularizing and stabilizing 5G services. However, existing antenna distribution planning technologies have problems such as the invariance of the arrangement that changing the antenna position cannot change the network coverage, and the complexity of antenna optimization caused by changes in the number of antennas or other variables in finding the optimal antenna configuration.

[0058] For this purpose, refer to Figure 1 The embodiment of the present application provides an antenna planning method, which is applied to a controller. The controller may be a server, an electronic device, or a mobile terminal. The method is applicable to 5G antenna planning, but it may also be applicable to 3G antenna planning and 4G antenna planning. No specific limitation is made here. The method includes the following steps S110 to S150:

[0059] Step S110: When the Nth parent population is obtained, calculate the first performance indicator of the individuals of the Nth parent population; if N is 1, the Nth parent population is generated based on the initial antenna positions of multiple antennas in the layout scenario; if N is a positive integer greater than 1, the Nth parent population is generated based on the N-1th parent population.

[0060] In this step, the initial antenna position is randomly initialized. One purpose of this embodiment is to generate the Nth parent population based on the initial antenna positions of multiple antennas in the layout scenario, where when N=1, it is the first parent population, that is, the initial parent population. Each parent population includes multiple population individuals, and each individual represents an antenna layout plan. When N>1, it is a new population evolved from the previous generation (that is, the N-1th generation), and the new population is obtained by performing genetic operations such as selection, crossover (recombination) and mutation on the N-1th population or using an evolutionary algorithm.

[0061] The layout scenario includes corresponding layout scenario parameters and antenna parameters. The layout scenario parameters include building floor plan information, for example, the length of the building floor plan and width , the layout information of the building, and the number of grids that evenly divide the building. Antenna parameters include the antenna installation height and the upper and lower limits of the number of antennas used.

[0062] In this embodiment, as a preferred embodiment, the N-1th population is processed by genetic operation to obtain the Nth population. Genetic operation is the core component of the genetic algorithm, which simulates the evolutionary process in nature. It searches for the optimal solution through comprehensive operations such as selection, crossover and mutation. It can maintain a balance in the search process, and can quickly approach a good solution (through selection) while avoiding falling into local optimality (through crossover and mutation). This allows the algorithm to effectively find the global optimal solution or approximate optimal solution in a larger solution space, and can also help the algorithm adapt to changes in the environment. When the objective function or constraints change, the algorithm can still find a new optimal solution.

[0063] The following explains the specific training steps for the first parent population:

[0064] Step S210: clustering multiple antennas according to their initial antenna positions in the layout scenario to obtain multiple antenna classes.

[0065] Step S220: Select an antenna from each antenna class to form a first parent population.

[0066] First, the initial antenna positions of multiple antennas in the layout scenario are clustered and analyzed to obtain multiple antenna classes. Antenna positions with similar characteristics can be classified into one class through cluster analysis to avoid repeated consideration of similar antenna layout schemes in subsequent evolutionary calculations, which helps to improve the efficiency of the algorithm. Each antenna class contains a certain number of antenna positions, and then one antenna is selected from each antenna class to be added to the first parent population to form the first parent population, ensuring that the initial population has a high diversity, which helps to explore more solution spaces and avoid premature convergence to the local optimal solution for subsequent evolutionary calculations.

[0067] Clustering is an unsupervised learning method that aims to group similar objects into the same category. A variety of clustering algorithms can be selected according to specific needs and data characteristics. Clustering analysis is used to generate initial antenna positions and construct the first parent population. This can significantly improve the efficiency and quality of wireless communication system layout optimization, while simplifying the problem scale and enhancing the robustness and adaptability of the algorithm.

[0068] Combination Figure 2 As shown, in one implementation of this embodiment, by randomly initializing A layout scenario includes the initial antenna positions of multiple antennas in the layout scenario, and then the initial antenna positions of multiple antennas in the layout scenario are clustered into Antenna class, and for each Antenna Class Generation A solution is obtained by Randomly select an antenna position in the class to form the first parent population , the population size is ,in, , , are the maximum and minimum values ​​of the number of antenna positions respectively. The population size is determined according to the layout scenario. The number of antenna positions is one of the performance indicators of each individual population. When optimizing, adjusting or replacing according to the performance indicators of the individual population in the future, it is necessary to compare the number of antenna positions, that is, compare other performance indicators between the individuals in the population when the number is the same, so as to obtain antenna layout scenarios with different numbers of antennas, and then select the optimal antenna layout plan according to actual needs to determine the final antenna position.

[0069] In this embodiment, the random initialization Layout scenes and save them in collections S For each antenna position, the path loss of the corresponding grid is obtained through the ray tracing model, so as to obtain the signal strength of the antenna in each grid, and then the signal strength and signal strength threshold are used to calculate the signal strength of the antenna. To determine whether each grid is successfully covered, when the signal strength is greater than or equal to the threshold, the grid is successfully covered; otherwise, it means no coverage. The network coverage rate is determined by dividing the number of successfully covered grids by the total number of grids.

[0070] This embodiment provides the N A training process in multiple iterations of the parent population. S middle The antenna positions are clustered into The number of antenna positions that need to be evaluated is reduced, thereby improving computational efficiency. In addition, K-means clustering can help disperse antenna positions and avoid over-concentration in a small area, thereby improving the uniformity of network coverage. A random antenna position is selected in the class, resulting in The parent generation individuals are composed of antenna positions, thereby increasing the diversity of the parent population, which is conducive to the genetic algorithm to explore more possible solution spaces in subsequent iterations; repeat the above process Second, formed by The parent population consists of ; Then for each individual, calculate the maximum signal strength of all antennas in the individual on each grid, if the signal strength is greater than or equal to the threshold , the grid is successfully covered; otherwise, it means the grid is not covered. Among them, the key parameters in the clustering method (such as the number of repetitions of K-means M and signal strength threshold) can be flexibly adjusted according to actual conditions to adapt to different optimization goals and constraints. In this embodiment, the individual network coverage is obtained by dividing the number of successfully covered grids by the total number of grids. By calculating the maximum signal strength of each grid and judging whether it is covered according to the threshold, the network coverage can be evaluated more accurately, which can not only capture the actual situation of network coverage, but also avoid affecting the quality of the entire network due to insufficient signal strength at a single point.

[0071] In some embodiments, in step S110, the N The first performance index of the individuals of the parent population includes the following steps S310 to S320:

[0072] Step S310: Using the layout scene parameters and antenna parameters corresponding to the layout scene, obtain the first N The network signal strength of individuals in the parent population.

[0073] Step S320: N The network signal strength of the individuals in the parent population determines the N The network coverage of individuals in the parent population.

[0074] In this step, the first performance indicator includes network coverage, and the second performance indicator is obtained by using the ray tracing technology through the layout scenario parameters and antenna parameters corresponding to the layout scenario. N The network signal strength of the individuals in the parent population is used to determine the network coverage.

[0075] like Figure 3 As shown, in this embodiment, the performance of antenna layout planning, especially network signal strength and network coverage, is evaluated by a ray tracing model, including using ray tracing technology to obtain the first N The network signal strength of the individuals in the parent population is used to obtain the signal strength distribution corresponding to each antenna position, so as to calculate the network coverage of the corresponding area for each individual (i.e., an antenna layout plan). In the figure, there are three ray tracing models, each of which corresponds to a set of antenna positions. The three different antenna models can be simulations for different antenna layouts, or simulations performed at different time points or environments. A set of signal strength distribution diagrams are displayed below each ray tracing model. The network coverage of the corresponding model is calculated according to the calculation formula in the figure, and then the antenna layout is optimized according to the evaluation results to improve user experience and service quality.

[0076] Ray tracing is a physical simulation technology that can be used to simulate signal propagation in indoor environments, thereby improving the accuracy and reliability of indoor positioning. It can be used to predict the propagation path and strength of electromagnetic waves in a specific environment. In the field of wireless communications, ray tracing can be used to simulate signal propagation in complex environments, thereby calculating the signal strength at a specific location.

[0077] Step S120: N The parent population undergoes crossover mutation and obtains N The offspring population, and calculate the N A second performance indicator for individuals in the offspring population.

[0078] In this step, the second performance indicator includes network coverage. N The individuals in the parent population are paired and new individuals are generated through crossover operations. The individuals generated after crossover are mutated to introduce new genetic mutations. The individuals after crossover and mutation are then combined to form the first N The offspring population, then N Each individual in the offspring population calculates its second performance indicator.

[0079] In this embodiment, randomly select N Individuals are selected from the parent population for pairing, and a crossover operation is performed on the randomly selected individuals after pairing. The crossover operation refers to exchanging some gene fragments of a pair of individuals to generate new individuals. There are many forms of crossover, such as single-point crossover, multi-point crossover, and uniform crossover. In the antenna layout planning process, crossover involves exchanging some antenna positions in two individuals, and then performing a mutation operation on the new individuals generated after the crossover, that is, randomly changing some antenna positions in the individuals. One purpose of this embodiment is to introduce new genetic variations through mutation operations, increase the diversity of the population, and prevent premature convergence to the local optimal solution. The new individuals generated after the crossover and mutation operations are then collected to form the first N During the processing, the offspring population keeps the population size unchanged, that is, the number of offspring population is the same as that of parent population.

[0080] Then, for the NEach individual in the offspring population calculates the second performance index of the corresponding individual. The second performance index can be other important performance indicators besides network coverage, such as cost, power consumption, installation difficulty, etc. In this embodiment, network coverage is preferred. One purpose of this embodiment is to comprehensively evaluate the overall performance of the individual, so as to guide the next step of selection and optimization. Moreover, through crossover and mutation operations, new genetic information can be introduced into the offspring population, increasing the diversity of the population and helping to avoid the algorithm from falling into a local optimal solution. Among them, the crossover operation helps to quickly combine excellent features, and the mutation operation introduces new mutation points. The combination of the two can accelerate the convergence process of the algorithm and improve the search efficiency.

[0081] Finally, by calculating the second performance index of individuals in the offspring population, the performance of individuals can be comprehensively evaluated, the quality of optimization results can be improved, the robustness of the algorithm can be enhanced, and complex multi-objective optimization requirements can be adapted.

[0082] In some embodiments, in step S120, according to N The parent population undergoes crossover mutation and obtains N The offspring population includes the following steps S410 to S430:

[0083] Step S410: N All individuals of the parent population are placed on the target coverage plane.

[0084] Step S420: randomly divide the target coverage plane to obtain multiple target coverage sub-planes.

[0085] Step S430: randomly splice each target coverage sub-plane to obtain the first N The parent population corresponds to N Offspring population.

[0086] In this step, you first need to N All individuals in the parent population are placed on the target coverage plane, and then the different target coverage planes are randomly divided into multiple target coverage sub-planes, and the target coverage sub-planes obtained by the division are randomly spliced ​​to form a new layout plan, thereby obtaining the first N Offspring population.

[0087] In this embodiment, first, N The antenna position of each individual in the parent population is placed on the target coverage plane, that is, all antenna positions in each individual (i.e., an antenna layout solution) are projected onto the same plane. Then, the target coverage plane is randomly divided into two parts, and the two parts of the different target coverage planes are spliced ​​together to form a new child solution, that is, a part of the antenna positions are taken from two different individuals and combined into a new individual. If the first NIf the antenna positions contained in the individuals in the parent population are less than the predetermined value, the antenna positions are randomly added to the corresponding individuals to meet the predetermined value. On the contrary, if the antenna positions contained in the individuals exceed the predetermined value, the antenna positions are randomly reduced to reach the predetermined number. N In the parent population, for each individual, an antenna position is randomly selected and changed to increase the diversity of the population. In addition, by randomly adding or reducing antenna positions, the problem of non-fixed dimensional solutions commonly found in genetic algorithms is solved, increasing the flexibility and applicability of the algorithm.

[0088] like Figure 4 As shown, in this embodiment, first N The antenna of each individual in the parent population is placed on the target coverage plane, and then the target coverage plane is randomly divided into two parts. Then the two parts of the different target coverage planes are spliced ​​into the child solution, which can generate diverse individuals, thereby increasing the diversity of the population, helping to avoid the algorithm from falling into the local optimal solution, improving the ability of global search, and solving the problem that ordinary operators are difficult to handle non-fixed dimensional solutions. N The antenna position of each individual in the parent population is less than , add random On the contrary, when N The number of antenna positions for each individual in the parent population exceeds , reducing the randomness to For the mutation operator, in N The antenna position of each individual in the parent population is randomly changed to ensure that it remains effective when facing problems of different scales. This not only improves the flexibility and adaptability of the algorithm, but also enhances its robustness and convergence speed. It is suitable for optimization problems of antenna layout planning of various scales.

[0089] In some embodiments, in step S120, the N The second performance indicator of individuals in the offspring population includes the following steps S510:

[0090] Step S510: N The antenna positions corresponding to the individuals in the parent population are input into the preset proxy model to obtain the first output of the preset proxy model. N The network coverage of individuals in the offspring population, where the agent model is based on the previous N The antenna position corresponding to the individual in at least one parent population of -1 parent population is obtained by training.

[0091] In this step, the second performance indicator includes network coverage. NThe antenna position of each individual in the parent population is input into the preset proxy model, and the proxy model is used to predict the antenna position of each individual to obtain the network coverage of each individual. Using the proxy model instead of the real physical simulation or experiment can significantly reduce the calculation time, and the proxy model can quickly provide prediction results, thereby accelerating the optimization process, which not only improves the efficiency and speed of the genetic algorithm, but also enhances the generalization ability and robustness of the algorithm. In addition, by learning the patterns in the historical data, the proxy model can accurately predict the performance of the new antenna layout scheme in terms of network coverage, thereby guiding the genetic algorithm to better search and optimize.

[0092] The specific training steps of the proxy model are explained below:

[0093] This embodiment provides a training process in multiple iterations of the proxy model. Construct a training set, and the input parameters of each training sample are antenna positions, the output is The network coverage corresponding to the antenna position. The training set consists of two parts, one of which is The other part is randomly generated individuals, each of which is The antenna positions are randomly selected In addition to the first generation of proxy models, the proxy models are incrementally trained based on the previous generation of models, that is, the proxy models are based on the previous generation of models. N -1 parent populations are trained with the antenna positions corresponding to the individuals in at least one parent population. By combining the parent individuals and the randomly generated individuals to construct the training set, the generalization ability of the proxy model can be enhanced, so that it can better adapt to different types of antenna layouts. Moreover, each iteration will improve on the existing knowledge and gradually approach the optimal solution.

[0094] like Figure 5 As shown in the figure, a neural network for antenna layout operation is constructed as a proxy model, and the input of the proxy model is a set of antenna positions , then each element in the antenna layout Independently through the multi-layer perceptron , and mapped to Then, the summation operator is executed , and input it into another multi-layer perceptron to provide predicted network coverage , by using a multi-layer perceptron to process the antenna positions independently and aggregating the results through a summation operation, the evaluation of the antenna layout during the optimization process can be simplified, making the model easier to understand and implement, and then the network training is performed. During the training process, the accuracy of the proxy model in predicting network coverage is improved by minimizing the following loss function, thereby more accurately guiding the optimization of the antenna layout, and finally updating the network parameters according to the training set:

[0095] ;

[0096] in, represents the size of the training set, Represents a multilayer perceptron and Parameters.

[0097] Step S130: N The second performance index of individuals in the offspring population is optimized N The individuals of the offspring population are calculated and the optimized N The third performance indicator of the individuals in the offspring population.

[0098] In this step, the first N The individuals of the offspring population are optimized. The optimization can be improved by local search algorithms (such as gradient descent, simulated annealing, etc.) or other heuristic methods to improve the performance of individuals. One goal of this embodiment is to improve the N The second performance indicator of the individuals in the offspring population is to improve the network coverage of the offspring individuals, thereby calculating the optimized N The third performance index of the individuals in the offspring population is used to evaluate the optimized N Individuals of the offspring population.

[0099] In some embodiments, in step S130, the second performance index of the individuals in the Nth generation population is optimized. N The individuals of the offspring population include the following steps S610 to S630:

[0100] Step S610: N The network coverage of individuals in the offspring population determines the N The individual with the highest network coverage in the offspring population is taken as the best offspring individual.

[0101] Step S620: confirm multiple antenna positions in the parent population that do not belong to the initial antenna position in the best offspring individuals.

[0102] Step S630: Calculate the coverage contributions corresponding to multiple antenna positions in the parent population that do not belong to the initial antenna position in each best offspring individual, and adjust the best offspring individual according to the coverage contribution to obtain the optimized first N Individuals of the offspring population.

[0103] In this step, according to N The second performance indicator of the individuals in the offspring population is the network coverage, which determines the N The individual with the highest network coverage in the offspring population is taken as the best offspring individual. By confirming multiple antenna positions in the parent population that do not belong to the initial antenna position in the best offspring individual, the coverage contribution corresponding to the multiple antenna positions in the parent population that do not belong to the initial antenna position in each best offspring individual is calculated, and the best offspring individual is adjusted according to the coverage contribution to obtain the optimized individual of the Nth offspring population.

[0104] The specific calculation steps of coverage contribution are explained below:

[0105] First, identify the individual with the highest network coverage rate among the individuals in the Nth generation population (expressed as .for Not in layout scene collection Each antenna position in , the signal strength on each grid is obtained through the ray tracing model, and then the The coverage contribution of each antenna position in the calculation is calculated, and the antenna position with the smallest coverage contribution is adjusted. First, establish the definition of coverage contribution:

[0106] ;

[0107] in, Indicates from Remove After the antenna position, if Smaller, it means The coverage contribution is low; is the network coverage function.

[0108] Then use the scene collection from the layout The closest of Antenna positions created The neighborhood of ).for Each antenna position in ,Will In Replace with , to obtain , and based on Evaluate performance. ,but Replaced by .

[0109] like Figure 6 As shown, in one implementation of this embodiment, first, for each , determine the individual with the highest network coverage rate among the individuals in the Nth generation population (expressed as , that is, having The progeny population at antenna positions The individual with the largest network coverage among all individuals.

[0110] Then, for Not in Each antenna position in , the corresponding signal strength is obtained through the ray tracing model , using a ray tracing model to accurately evaluate the signal strength at each antenna position, making adjustments more precise and avoiding errors that may be caused by rough estimates, and adjusting the signal strength based on its corresponding network coverage. The performance of each antenna is evaluated in Adjust the antenna position with the minimum coverage contribution By identifying and replacing antenna positions with low coverage contribution, the antenna layout can be gradually optimized to improve the overall network coverage. Closest of Antenna positions created The neighborhood of ), through the concept of neighborhood, it is possible to search near the current optimal solution and find better alternatives, thereby enhancing the optimization effect. Each antenna position in ,we will In Replace with , to obtain , Figure 6 It is a kind of antenna cross variation mode. Figure 6 (a) Has the minimum coverage contribution rate, Figure 6 (b) quilt Replace, and based on Evaluate performance. ,but Replaced by This embodiment optimizes the antenna layout by evaluating the coverage contribution of the antenna position and combining the neighborhood search strategy, which not only improves the network coverage but also enhances the robustness and adaptability of the algorithm, and can effectively reduce unnecessary antenna deployment, optimize resource allocation, and improve the overall performance of the network.

[0111] Step S140: According to the third performance index, the N The individuals in the offspring population have the lowest performance index. N The individuals of the parent population are adjusted to obtain N +1 parent population.

[0112] In this step, the offspring with the best performance on the third performance indicator (i.e. N The individuals in the offspring population) are used to replace or modify the individuals in the previous generation with the worst performance index (i.e., the N The individual in the parent population), thereby improving the overall performance of the population and gradually improving the individuals with poor performance in the population.

[0113] In one implementation of this embodiment, from N Parent population and N Selection among offspring populations antenna layout planning scheme as the new parent population, and find the N Parent population Medium network coverage The individual with the smallest value (expressed as ), the third highest performance index N The network coverage of individuals in the offspring population is expressed as ,when , then by Replace with To update .

[0114] Step S150: If N +1 The parent population meets the preset conditions, then according to the N +1 The individuals of the parent population generate the final antenna positions of multiple antennas in the layout scenario.

[0115] In this step, in each iteration, the first N +1 Whether the parent population meets the conditions, where the preset condition can be any defined stopping criteria, such as reaching a certain fitness threshold, limiting the number of iterations, or the rate of change of the fitness value is lower than a certain threshold. N +1 The parent population meets the preset conditions and determines the N+1 The best individual in the parent population, the best individual is preferably the individual with the highest network coverage, and the final antenna positions of multiple antennas in the layout scenario are obtained based on the selected best individual, and then the actual layout of the antennas is implemented.

[0116] In some embodiments, in step S150, according to N +1 The individuals of the parent population generate the final antenna positions of multiple antennas in the layout scenario, including the following steps S710 to S730:

[0117] Step S710: Calculate N +1 The fourth performance indicator of the individuals in the parent population.

[0118] Step S720: N +1 The fourth performance index of the individuals in the parent population determines the N +1 The individual with the fourth highest performance index among the individuals in the parent population.

[0119] Step S730: N The individual with the fourth highest performance index among the individuals in the +1 parent population generates the final antenna positions of the multiple antennas in the layout scenario.

[0120] In this step, the fourth performance indicator includes network coverage, and the calculation determines the N +1 The individual with the highest network coverage in the parent population obtains the final antenna positions of multiple antennas in the layout scenario based on the selected optimal individual, thereby implementing the antenna layout planning.

[0121] In this embodiment, first, for the N +1 For each individual in the parent population, calculate the corresponding fourth performance index, that is, calculate the N +1 the network coverage of each individual in the parent population, and then determine the individual with the highest fourth performance index based on the calculated fourth performance index of all individuals. Finally, determine the final position of the antenna in the layout scenario based on the individual with the highest fourth performance index, so as to obtain better system performance and technical effects in actual deployment.

[0122] In one implementation of this embodiment, a ray tracing model is used to simulate signal strength, a proxy model is pre-built to fit those evaluation index functions with high computational cost, and the antenna layout planning scheme is encoded as a population represented by real number coding, that is, each individual (i.e., a layout scheme) is represented as a series of real values, each value corresponds to the position of an antenna or other related parameters, and then a crossover mutation operator is used to generate a descendant population, specifically including selecting individuals from the parent population, and generating new individuals through standard operations in the genetic algorithm (such as single-point crossover, uniform crossover, mutation, etc.), and then using the proxy model to predict the individuals in the descendant population, evaluate their performance indicators such as signal coverage, and use the efficiency of the proxy model to quickly screen out individuals with better performance. Then, based on the individuals initially screened, a local search is performed, that is, the individuals are optimized according to their individual performance, which helps to find a better solution in the neighborhood of the current solution, and then some representative individuals are selected in each generation of the descendant population, and the performance is evaluated by the signal strength obtained by the ray tracing model, and the proxy model is updated according to the signal strength to improve the accuracy of its prediction. Finally, the optimal individual solution set, that is, the best antenna layout scheme, is selected, and then the selected optimal individual solution set is decoded into an actual antenna layout planning scheme.

[0123] like Figure 7 As shown, in one implementation of the present embodiment, multiple antenna layout plans are obtained by randomly initializing antenna positions to obtain a parent population, and then a proxy model is constructed according to a training set through a ray tracing model to evaluate the antenna layout planning performance, and a parent solution is generated according to the parent population based on the ray tracing model, that is, an individual of the parent population, and then the parent solution is processed by methods such as crossover mutation to generate a child solution, that is, an individual of the child population, and then the performance of the child solution is evaluated according to the pre-constructed proxy model, and the child solution and the parent solution are locally searched based on the performance of the child solution and the parent solution, that is, the child solution and the parent solution are optimized, wherein the optimization process includes optimizing the individual by adding antenna positions for performance evaluation based on the ray tracing model, and then judging whether to end the iteration according to preset conditions, when the judgment result is that the iteration is not ended, continuing to update the parent solution and returning to the step of generating the child solution, and repeating the subsequent steps; when the judgment result is that the iteration is ended, the best layout plan is output according to the individual with the highest performance in the final parent solution.

[0124] like Figure 8 As shown, some embodiments of the present application provide an antenna planning system, the system includes a calculation module 810, a generation module 820, an optimization module 830, an evaluation module 840, and an update module 850, specifically:

[0125] The calculation module 810 is used to obtain the first N In the case of the parent population, calculate theN The first performance index of the individuals in the parent population; if N is 1, then N The parent population is generated based on the initial antenna positions of multiple antennas in the layout scenario. If N is a positive integer greater than 1, then N The parent population is based on N -1 Parent population generation;

[0126] Generating module 820, for N The parent population undergoes crossover mutation and obtains N The offspring population, and calculate the N A second performance indicator of the individuals in the offspring population;

[0127] Optimization module 830, for N The second performance index of individuals in the offspring population is optimized N The individuals of the offspring population are calculated and the optimized N The third performance index of individuals in the offspring population;

[0128] Evaluation module 840, for evaluating the N The individuals in the offspring population have the lowest performance index. N The individuals of the parent population are adjusted to obtain N +1 parent population;

[0129] Update module 850, for N +1 The parent population meets the preset conditions, then according to the N +1 The individuals of the parent population generate the final antenna positions of multiple antennas in the layout scenario.

[0130] In one implementation, the calculation module 810 is further used to: cluster the multiple antennas according to their initial antenna positions in the layout scenario to obtain multiple antenna classes; and select an antenna from each antenna class to form a first parent population.

[0131] In one embodiment, the calculation module 810 is further used to: obtain the network signal strength of individuals of the Nth parent population by using the layout scene parameters and antenna parameters corresponding to the layout scene by using ray tracing; N The network signal strength of the individuals in the parent population determines the network coverage of the individuals in the Nth parent population.

[0132] In one embodiment, the generating module 820 is further used to: NAll individuals of the parent population are placed on the target coverage plane; the target coverage plane is randomly divided to obtain multiple target coverage sub-planes; each target coverage sub-plane is randomly spliced ​​to obtain the first N The parent population corresponds to N Offspring population.

[0133] In one embodiment, the generating module 820 is further used to: N The antenna positions corresponding to the individuals in the parent population are input into the preset proxy model to obtain the first output of the preset proxy model. N The network coverage of individuals in the offspring population, where the agent model is based on the previous N The antenna position corresponding to the individual in at least one parent population of -1 parent population is obtained by training.

[0134] In one embodiment, the optimization module 830 is further configured to: N The network coverage of individuals in the offspring population determines the N The individual with the highest network coverage rate in the offspring population is taken as the best offspring individual; multiple antenna positions in the parent population that do not belong to the initial antenna position in the best offspring individual are confirmed; the coverage contribution corresponding to multiple antenna positions in the parent population that do not belong to the initial antenna position in each best offspring individual is calculated, and the best offspring individual is adjusted according to the coverage contribution to obtain the optimized N Individuals of the offspring population.

[0135] In one embodiment, the updating module 850 is further used to: calculate the N +1 The fourth performance index of the individuals in the parent population; N +1 The fourth performance index of the individuals in the parent population determines the N +1 The individual with the highest fourth performance index among the individuals in the parent population; N The individual with the fourth highest performance index among the individuals in the +1 parent population generates the final antenna positions of the multiple antennas in the layout scenario.

[0136] It should be noted that the antenna planning system provided in this embodiment and the above-mentioned antenna planning method are based on the same inventive concept, so the relevant content of the above-mentioned antenna planning method is also applicable to the content of the antenna planning system, and therefore, it will not be repeated here.

[0137] like Fig. 9 The embodiment of the present application further provides an electronic device, the electronic device comprising:

[0138] At least one hydrogen fuel cell;

[0139] at least one memory;

[0140] at least one processor;

[0141] at least one program;

[0142] The program is stored in the memory, and the processor executes at least one program to implement the large language model training method implemented in the present disclosure.

[0143] The electronic device may be any intelligent terminal including a mobile phone, a tablet computer, a personal digital assistant (PDA), a vehicle-mounted computer, etc.

[0144] The electronic device according to the embodiment of the present application is described in detail below.

[0145] Processor 1600, for obtaining the first N In the case of the parent population, calculate the N The first performance index of the individuals in the parent population; if N is 1, then N The parent population is generated based on the initial antenna positions of multiple antennas in the layout scenario. If N is a positive integer greater than 1, then N The parent population is based on N -1 Parent population generation.

[0146] In one embodiment, the processor 1600 is further configured to: N The parent population undergoes crossover mutation and obtains N The offspring population, and calculate the N A second performance indicator for individuals in the offspring population.

[0147] In one embodiment, the processor 1600 is further configured to: N The second performance index of individuals in the offspring population is optimized N The individuals of the offspring population are calculated and the optimized N The third performance indicator of the individuals in the offspring population.

[0148] In one embodiment, the processor 1600 is further configured to: N The individuals in the offspring population have the lowest performance index. N The individuals of the parent population are adjusted to obtain N +1 parent population.

[0149] In one embodiment, the processor 1600 is further configured to: N +1 The parent population meets the preset conditions, then according to the N +1 The individuals of the parent population generate the final antenna positions of multiple antennas in the layout scenario.

[0150] The processor 1600 may be implemented by a general-purpose central processing unit (CPU), a microprocessor, an application-specific integrated circuit (ASIC), or one or more integrated circuits, and is used to execute relevant programs to implement the technical solutions provided in the embodiments of the present disclosure;

[0151] The memory 1700 may be implemented in the form of a read-only memory (ROM), a static storage device, a dynamic storage device, or a random access memory (RAM). The memory 1700 may store an operating system and other application programs. When the technical solution provided in the embodiment of this specification is implemented by software or firmware, the relevant program code is stored in the memory 1700, and the processor 1600 calls and executes an antenna planning method of the embodiment of the present disclosure.

[0152] Input / output interface 1800, used to implement information input and output;

[0153] Communication interface 1900, used to realize communication interaction between the device and other devices, which can be realized through wired mode (such as USB, network cable, etc.) or wireless mode (such as mobile network, WIFI, Bluetooth, etc.);

[0154] Bus 2000 , which transmits information between various components of the device (e.g., processor 1600 , memory 1700 , input / output interface 1800 , and communication interface 1900 );

[0155] The processor 1600 , the memory 1700 , the input / output interface 1800 , and the communication interface 1900 are connected to each other in communication within the device via the bus 2000 .

[0156] The embodiment of the present disclosure further provides a storage medium, which is a computer-readable storage medium. The computer-readable storage medium stores computer-executable instructions, and the computer-executable instructions are used to enable a computer to execute the above-mentioned antenna planning method.

[0157] The memory, as a non-transient computer-readable storage medium, can be used to store non-transient software programs and non-transient computer executable programs. In addition, the memory may include a high-speed random access memory, and may also include a non-transient memory, such as at least one disk storage device, a flash memory device, or other non-transient solid-state storage device. In some embodiments, the memory may optionally include a memory remotely disposed relative to the processor, and these remote memories may be connected to the processor via a network. Examples of the above-mentioned network include, but are not limited to, the Internet, an intranet, a local area network, a mobile communication network, and combinations thereof.

[0158] The embodiments described in the embodiments of the present disclosure are intended to more clearly illustrate the technical solutions of the embodiments of the present disclosure and do not constitute a limitation on the technical solutions provided by the embodiments of the present disclosure. Those skilled in the art will appreciate that with the evolution of technology and the emergence of new application scenarios, the technical solutions provided by the embodiments of the present disclosure are also applicable to similar technical problems.

[0159] Those skilled in the art will appreciate that the technical solutions shown in the figures do not limit the embodiments of the present disclosure and may include more or fewer steps than shown in the figures, or a combination of certain steps, or different steps.

[0160] The device embodiments described above are merely illustrative, and the units described as separate components may or may not be physically separated, that is, they may be located in one place or distributed on multiple network units. Some or all of the modules may be selected according to actual needs to achieve the purpose of the solution of this embodiment.

[0161] Those skilled in the art will appreciate that all or some of the steps in the methods disclosed above, and the functional modules / units in the systems and devices may be implemented as software, firmware, hardware, or a suitable combination thereof.

[0162] The terms "first", "second", "third", "fourth", etc. (if any) in the specification of the present application and the above-mentioned drawings are used to distinguish similar objects, and are not necessarily used to describe a specific order or sequence. It should be understood that the data used in this way can be interchangeable where appropriate, so that the embodiments of the present application described herein can be implemented in an order other than those illustrated or described herein. In addition, the terms "including" and "having" and any of their variations are intended to cover non-exclusive inclusions, for example, a process, method, system, product or device comprising a series of steps or units is not necessarily limited to those steps or units clearly listed, but may include other steps or units that are not clearly listed or inherent to these processes, methods, products or devices.

[0163] It should be understood that in the present application, "at least one (item)" means one or more, and "plurality" means two or more. "And / or" is used to describe the association relationship of associated objects, indicating that three relationships may exist. For example, "A and / or B" can mean: only A exists, only B exists, and A and B exist at the same time, where A and B can be singular or plural. The character " / " generally indicates that the objects associated before and after are in an "or" relationship. "At least one of the following" or similar expressions refers to any combination of these items, including any combination of single or plural items. For example, at least one of a, b or c can mean: a, b, c, "a and b", "a and c", "b and c", or "a and b and c", where a, b, c can be single or multiple.

[0164] In the several embodiments provided in the present application, it should be understood that the disclosed devices and methods can be implemented in other ways. For example, the device embodiments described above are only schematic. For example, the division of units is only a logical function division. There may be other division methods in actual implementation, such as multiple units or components can be combined or integrated into another system, or some features can be ignored or not executed. Another point is that the mutual coupling or direct coupling or communication connection shown or discussed can be through some interfaces, indirect coupling or communication connection of devices or units, which can be electrical, mechanical or other forms.

[0165] The units described as separate components may or may not be physically separated, and the components shown as units may or may not be physical units, that is, they may be located in one place or distributed on multiple network units. Some or all of the units may be selected according to actual needs to achieve the purpose of the solution of this embodiment.

[0166] In addition, each functional unit in each embodiment of the present application may be integrated into one processing unit, or each unit may exist physically separately, or two or more units may be integrated into one unit. The above-mentioned integrated unit may be implemented in the form of hardware or in the form of software functional units.

[0167] If the integrated unit is implemented in the form of a software functional unit and sold or used as an independent product, it can be stored in a computer-readable storage medium. Based on this understanding, the technical solution of the present application is essentially or the part that contributes to the prior art or all or part of the technical solution can be embodied in the form of a software product, and the computer software product is stored in a storage medium, including multiple instructions to enable an electronic device (which can be a personal computer, server, or network device, etc.) to perform all or part of the steps of each embodiment method of the present application. The aforementioned storage medium includes: U disk, mobile hard disk, read-only memory (Read-Only Memory, ROM), random access memory (Random Access Memory, RAM), disk or optical disk and other media that can store programs.

[0168] The above is a specific description of the preferred implementation of the embodiments of the present application, but the embodiments of the present application are not limited to the above-mentioned implementation methods. Technical personnel familiar with the field can also make various equivalent modifications or substitutions without violating the spirit of the embodiments of the present application. These equivalent modifications or substitutions are all included in the scope defined by the claims of the embodiments of the present application.

[0169] The embodiments of the present application are described in detail above in conjunction with the accompanying drawings, but the present application is not limited to the above embodiments. Various changes can be made within the knowledge scope of ordinary technicians in the relevant technical field without departing from the purpose of the present application.

Claims

1. An antenna planning method, characterized in that: The method comprises: When an Nth parent population is obtained, a first performance indicator of an individual of the Nth parent population is calculated; if N is 1, the Nth parent population is generated based on initial antenna positions of multiple antennas in a layout scenario; if N is a positive integer greater than 1, the Nth parent population is generated based on an N-1th parent population; Obtaining an Nth generation population according to the crossover mutation of the Nth parent population, and calculating a second performance index of individuals in the Nth generation population; Optimizing the individuals of the Nth generation population according to the second performance index of the individuals of the Nth generation population, and calculating the third performance index of the individuals of the Nth generation population after optimization; According to the individuals of the Nth offspring population with the highest third performance index, the individuals of the Nth parent population with the lowest first performance index are adjusted to obtain the N+1th parent population; If the N+1th parent population meets the preset conditions, generating final antenna positions of multiple antennas in the layout scenario according to individuals of the N+1th parent population; The first performance indicator includes network coverage, and the calculating of the first performance indicator of the individuals of the Nth parent population includes: Obtaining the network signal strength of the individuals of the Nth parent population by using the layout scenario parameters and antenna parameters corresponding to the layout scenario using ray tracing; Determining the network coverage of the individuals in the Nth parent population according to the network signal strength of the individuals in the Nth parent population; The step of obtaining the Nth generation population through crossover mutation of the Nth generation population comprises: Place all individuals of the Nth parent population on the target coverage plane respectively; Randomly splitting the target coverage plane to obtain multiple target coverage sub-planes; Randomly splicing each target coverage sub-plane to obtain the Nth child population corresponding to the Nth parent population; The second performance indicator includes network coverage, and the calculation of the second performance indicator of the individuals of the Nth generation population includes: Inputting the antenna positions corresponding to the individuals of the Nth parent population into a preset proxy model, obtaining the network coverage of the individuals of the Nth child population output by the preset proxy model, The proxy model is trained based on the antenna position corresponding to an individual in at least one parent population among the first N-1 parent populations.

2. The antenna planning method according to claim 1, characterized in that: Before calculating the first performance index of the individuals of the Nth parent population after the Nth parent population is obtained, the method further includes: Clustering the multiple antennas according to their initial antenna positions in the layout scenario to obtain multiple antenna classes; One antenna is selected from each of the antenna classes to form a first parent population.

3. The antenna planning method according to claim 1, characterized in that: Optimizing the individuals of the Nth generation population according to the second performance index of the individuals of the Nth generation population includes: Determine, according to the network coverage rates of the individuals in the Nth generation population, the individual with the highest network coverage rate in the Nth generation population as the best generation individual; confirming a plurality of antenna positions in the parent population that do not belong to the initial antenna position in the best offspring individual; The coverage contributions corresponding to multiple antenna positions in the parent population that do not belong to the initial antenna position in each of the best offspring individuals are calculated, and the best offspring individuals are adjusted according to the coverage contributions to obtain optimized individuals of the Nth offspring population.

4. The antenna planning method according to claim 1, characterized in that: Generating final antenna positions of multiple antennas in a layout scenario according to individuals of the N+1th parent population includes: Calculating a fourth performance index of individuals in the N+1th parent population; Determine, according to the fourth performance index of the individuals in the N+1th parent population, the individual with the highest fourth performance index among the individuals in the N+1th parent population; The final antenna positions of the multiple antennas in the layout scenario are generated according to the individual with the highest fourth performance index among the individuals of the N+1th parent population.

5. An antenna planning system, characterized in that: The system comprises: A calculation module, configured to calculate a first performance indicator of an individual of the Nth parent population when an Nth parent population is obtained; if N is 1, the Nth parent population is generated based on initial antenna positions of multiple antennas in a layout scenario; if N is a positive integer greater than 1, the Nth parent population is generated based on an N-1th parent population; A generation module, used to obtain an Nth generation population according to the crossover mutation of the Nth parent generation population, and calculate a second performance indicator of individuals of the Nth generation population; An optimization module, used for optimizing the individuals of the Nth generation population according to the second performance index of the individuals of the Nth generation population, and calculating the third performance index of the individuals of the Nth generation population after optimization; An evaluation module, configured to adjust the individuals of the Nth parent population with the lowest first performance index according to the individuals of the Nth child population with the highest second performance index, to obtain an N+1th parent population; An updating module, configured to generate final antenna positions of multiple antennas in a layout scenario according to individuals of the N+1th parent population if the N+1th parent population meets a preset condition; The first performance indicator includes network coverage, and the calculating of the first performance indicator of the individuals of the Nth parent population includes: Obtaining the network signal strength of the individuals of the Nth parent population by using the layout scenario parameters and antenna parameters corresponding to the layout scenario using ray tracing; Determining the network coverage of the individuals in the Nth parent population according to the network signal strength of the individuals in the Nth parent population; The step of obtaining the Nth generation population through crossover mutation of the Nth generation population comprises: Place all individuals of the Nth parent population on the target coverage plane respectively; Randomly splitting the target coverage plane to obtain multiple target coverage sub-planes; Randomly splicing each target coverage sub-plane to obtain the Nth child population corresponding to the Nth parent population; The second performance indicator includes network coverage, and the calculation of the second performance indicator of the individuals of the Nth generation population includes: Inputting the antenna positions corresponding to the individuals of the Nth parent population into a preset proxy model, obtaining the network coverage of the individuals of the Nth child population output by the preset proxy model, The proxy model is trained based on the antenna position corresponding to an individual in at least one parent population among the first N-1 parent populations.

6. An electronic device, characterized in that: It includes at least one controller and a memory for communicating with the controller; the memory stores instructions that can be executed by the at least one controller, and the instructions are executed by the at least one controller so that the at least one controller executes the antenna planning method as described in any one of claims 1 to 4.

7. A computer-readable storage medium, characterized in that: The computer-readable storage medium stores a computer program, and when the computer program is executed, the antenna planning method according to any one of claims 1 to 4 is implemented.

Citation Information

Patent Citations

  • Approximate optimal partial repetition code construction method

    CN114924907A

  • Multi-objective optimization-oriented circularly polarized microstrip antenna parameter optimization method and system

    CN116151096A