Antenna routing generation method and device, and computer program product

Abstract

The invention discloses an antenna routing generation method and device and a computer program product, relates to the technical field of wireless communication, and discloses an antenna routing generation method, which comprises the following steps: after the number of iterations of a population reaches a preset number of iterations, detecting that the fitness of a target individual in the population is within a preset variation range; according to individuals in a to-be-subdivided area, target grids in an antenna routing area are determined, and the to-be-subdivided area is used for caching antenna routing of which the fitness is greater than preset fitness in the iteration process; and dividing the target grid to generate new antenna routing, thereby realizing diversity of the antenna routing, and reducing the difference between the finally searched target individual and the design target.

Description

Technical Field

[0001] This application relates to the field of wireless communication technology, and in particular to methods, devices and computer programs for generating antenna traces. Background Technology

[0002] In automated antenna design schemes, the antenna routing area is first divided into multiple grids. Each grid includes two states: the presence of an antenna radiator and the absence of an antenna radiator. Therefore, a specific grid-based antenna routing design scheme can be represented as a Boolean matrix. Depending on the number of grids, an initial population of antenna routing patterns is generated. An optimization algorithm then selects the target individual from this initial population as the final output antenna routing pattern. However, in the aforementioned design schemes, because the initial population contains a limited number of individuals, searching for the target individual from this finite population often results in a significant discrepancy between the final target individual and the design objective. Summary of the Invention

[0003] The main purpose of this application is to provide a method, device and computer program product for generating antenna traces, which aims to achieve diversity in antenna traces and reduce the difference between the final target and the design goal.

[0004] To at least achieve the above objectives, this application proposes a method for generating antenna traces. The method includes: after the number of iterations of the population reaches a preset number of iterations, detecting that the fitness of a target individual in the population is within a preset range of variation; determining a target grid in the antenna trace area based on the individuals in the subdivided region, wherein the subdivided region is used to cache antenna traces with fitness greater than a preset fitness during the iteration process; dividing the target grid to generate new antenna traces.

[0005] This application also proposes an electronic device, the device comprising: a memory, a processor, and a computer program stored in the memory and executable on the processor, the computer program being configured to implement the steps of the antenna trace generation method described above.

[0006] This application also provides a computer program product, which includes a computer program that, when executed by a processor, implements the steps of the antenna trace generation method described above.

[0007] This application also proposes an antenna trace generation apparatus, the antenna trace generation apparatus comprising:

[0008] The target mesh determination module is used to determine the target mesh in the antenna routing area after the population has reached a preset number of iterations and the fitness of the target individual in the population is within a preset range of variation. The target mesh is determined based on the individuals in the subdivided region. The subdivided region is used to cache antenna routings with fitness greater than the preset fitness during the iteration process. The target mesh division module is used to divide the target mesh and generate new antenna routings.

[0009] This application discloses an antenna routing generation method. After the number of iterations of the population reaches a preset number, the fitness of the target individual in the population is detected to be within a preset range of variation. The target grid is determined based on the individuals cached in the subdivision region in each iteration, and the target grid is divided. Since the grid division will generate new antenna routing, the diversity of antenna routing in the population is enriched, thereby reducing the difference between the final target individual and the design target. Attached Figure Description

[0010] The accompanying drawings, which are incorporated in and form part of this specification, illustrate embodiments consistent with this application and, together with the description, serve to explain the principles of this application.

[0011] To more clearly illustrate the technical solutions in the embodiments of this application or the prior art, the drawings used in the description of the embodiments or the prior art will be briefly introduced below. Obviously, for those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0012] Figure 1 A flowchart illustrating the method for generating antenna traces in this application (Example 1);

[0013] Figure 2 This is a schematic diagram showing the distribution of antenna traces in the antenna trace area of ​​this application;

[0014] Figure 3 This is a schematic diagram showing the distribution of the new antenna traces in the antenna trace area after meshing in this application;

[0015] Figure 4 A flowchart illustrating the antenna trace generation method of this application, provided in Embodiment 3;

[0016] Figure 5 A detailed flowchart is provided for Embodiment 3 of the antenna trace generation method of this application;

[0017] Figure 6 This is another detailed flowchart illustrating the antenna trace generation method of this application in Embodiment 3;

[0018] Figure 7The flowchart of the antenna trace generation method in Embodiment 4 of this application is provided;

[0019] Figure 8 This is a simplified flowchart illustrating the antenna trace generation method of this application;

[0020] Figure 9 This is a schematic diagram of the module structure of the antenna trace generation device of this application;

[0021] Figure 10 This is a schematic diagram of the structure of the electronic device of this application.

[0022] The purpose, features, and advantages of this application will be further explained in conjunction with the embodiments and with reference to the accompanying drawings. Detailed Implementation

[0023] It should be understood that the specific embodiments described herein are merely illustrative of the technical solutions of this application and are not intended to limit this application.

[0024] To better understand the technical solution of this application, a detailed description will be provided below in conjunction with the accompanying drawings and specific implementation methods.

[0025] The main solution of this application embodiment is: after the number of iterations of the population reaches a preset number of iterations, the fitness of the target individual in the population is detected to be within a preset range of variation. Based on the individuals in the subdivided region, the target grid in the antenna routing region is determined. The subdivided region is used to cache antenna routing with fitness greater than the preset fitness during the iteration process. The target grid is divided to generate new antenna routing.

[0026] In this embodiment, for ease of description, the following description uses an electronic device as the execution subject.

[0027] Existing technologies use optimization algorithms to select target individuals from the initial population as the final output antenna trace. Because the initial population is finite, searching for the target individual from this finite pool often results in a significant discrepancy between the final target individual and the intended design.

[0028] This application provides a solution that, after the number of iterations in the population reaches a preset number, detects that the fitness of the target individual in the population is within a preset range of variation, determines the target grid based on the individuals cached in the subdivision region in each iteration, and performs the division of the target grid. Since the grid division generates new antenna traces, it enriches the diversity of antenna traces in the population, thereby reducing the difference between the final target individual and the design target.

[0029] It should be noted that the executing entity in this embodiment can be an electronic device with antenna routing generation, network communication, and program execution functions, such as a tablet computer, personal computer, or mobile phone, or a computer service device capable of performing the above functions. The following description uses an electronic device as an example to illustrate this embodiment and the subsequent embodiments.

[0030] Based on this, embodiments of this application provide a method for generating antenna traces, referring to... Figure 1 , Figure 1 This is a flowchart illustrating the first embodiment of the antenna trace generation method of this application.

[0031] In this embodiment, the method for generating antenna traces may include steps S10 to S20:

[0032] Step S10: After the number of iterations of the population reaches the preset number of iterations, it is detected that the fitness of the target individual in the population is within the preset range of variation. Based on the individuals in the subdivided region, the target grid in the antenna trace region is determined. The subdivided region is used to cache antenna traces with fitness greater than the preset fitness during the iteration process.

[0033] It should be noted that the preset number of iterations refers to the number of iterations of the population, which is the number of loops or repetitions executed by the algorithm. It represents the number of times the population is repeatedly executed during the calculation process. It is determined during the model training phase, so it can be a fixed setting. For example, the preset number of iterations can be set to 100 times or any other applicable number.

[0034] It should be noted that the population refers to a group or set operated on by the algorithm, containing multiple individuals, each typically representing a type of antenna trace within the antenna trace area. These individuals evolve and optimize through the algorithm's iterative process, ultimately determining the target individual. This target individual can improve signal transmission efficiency and performance, thereby enhancing communication between wireless terminals.

[0035] In one feasible implementation, the algorithm described above can be a genetic algorithm.

[0036] In genetic algorithms, the population consists of multiple individuals, each with a chromosome representing its genotype, typically represented using binary encoding or other methods. These chromosomes are combined and altered through gene crossover, mutation, and selection operations in the genetic algorithm to generate new individuals. In each generation, the fitness of an individual is evaluated according to a fitness function, and then a selection operation is used to choose individuals with high fitness as parents for the next generation, continuing the iterative optimization process. Through continuous iterative optimization, the iteration stops when the selected individual meets the required criteria, and that individual is output as the optimal individual.

[0037] In another feasible implementation, the algorithm described above can also be the whale algorithm or the wolf pack algorithm, etc.

[0038] Whale Algorithm: The whale algorithm is an emerging heuristic optimization algorithm inspired by the group behavior and migration patterns of whales. This algorithm simulates the intelligent strategies whales employ during food searching and migration to solve optimization problems. The whale group in the algorithm represents candidate solutions in the solution space, and each whale can be considered a potential solution. Whales communicate and cooperate to find the optimal solution. The whale algorithm ensures a broad search within the solution space while also deeply exploring potential optimal solutions by balancing exploration and exploitation. The algorithm can explore globally and refine its approach around the current optimal solution through local search strategies, improving the quality of the solution. The whale algorithm typically incorporates parameters (such as migration speed and leader whale) that can be dynamically adjusted based on the characteristics of the problem to optimize performance.

[0039] The implementation process of the whale algorithm includes: randomly generating a group of whales as the initial solution. In each iteration, each whale updates its state based on its current position and the positions of other whales, simulating whale migration and group cooperation. The fitness of each whale is evaluated according to the objective function of the problem, and the better whale is selected as the seed whale for the next round. The position of the entire whale group is updated based on the selected seed whale and the positions of other whales. When a predetermined stopping condition is met (such as reaching the maximum number of iterations or converging to a sufficiently good solution), the algorithm terminates and outputs the optimal antenna routing.

[0040] The Whale Algorithm, with its biomimetic characteristics and excellent global search capabilities, can solve complex antenna routing optimization problems.

[0041] Wolf Pack Algorithm: The wolf pack algorithm is a heuristic optimization algorithm inspired by the hunting behavior and social structure of wolves. This algorithm simulates the group and individual behaviors of a wolf pack, searching for target individuals through group cooperation and individual competition. In the wolf pack algorithm, the wolves represent candidate solutions in the solution space, and the wolves search for the optimal solution through cooperation and competition. A wolf pack typically has a leader and multiple followers; the leader wolf is responsible for guiding the entire pack towards a better solution. Each wolf in the pack adjusts its search direction based on its current position and the positions of other wolves, thereby progressively improving the quality of the solution.

[0042] The implementation process of the wolf pack algorithm includes: First, a pack of wolves is randomly generated as the initial solution. In each iteration, the wolves update their positions based on their current position and the positions of other wolves. The fitness of each wolf is evaluated according to the objective function of the optimization problem, and the better wolf is selected as the seed wolf for the next round. The overall position of the wolf pack is updated based on the selected seed wolf and the positions of other wolves. The algorithm terminates when a predetermined stopping condition is met (such as reaching the maximum number of iterations or converging to a sufficiently good solution).

[0043] The wolf pack algorithm, by simulating the intelligent behavior of a wolf pack and combining strategies of cooperation and competition, demonstrates good performance and convergence in solving complex optimization problems.

[0044] This embodiment, as well as other embodiments, uses a genetic algorithm as an example.

[0045] It should be noted that the target individual is the optimal individual in this iteration, that is, the optimal antenna routing determined after this iteration. The individual with the highest fitness in this iteration can be identified as the target individual.

[0046] It should be noted that fitness is typically used as a metric to evaluate the quality of each individual in the solution space. During each iteration, the algorithm calculates the fitness of each individual in the population and uses this fitness to determine the target individual for each iteration.

[0047] It should be noted that the preset range of variation can be set according to the actual situation, and it can be obtained during the algorithm model training phase.

[0048] In one feasible implementation, the fitness of a target individual in the population being within a preset range indicates that the target individual determined after each iteration is the same, and the fitness of that target individual has not changed. For example, after 100 iterations, the target individual determined in each iteration is A, and the fitness of A in each iteration is within the preset range, meaning the fitness of A has not changed in each iteration. The target individual and its corresponding fitness in each iteration can be recorded. After the number of iterations reaches a preset number, it is checked whether the target individual determined after each iteration is the same and whether its fitness is within the preset range. If the target individual determined after each iteration is the same and its fitness is within the preset range, a new antenna trace needs to be generated.

[0049] It should be noted that the subdivided region is used to cache antenna traces with a fitness greater than a preset fitness during the iteration process. In other words, the subdivided region serves as storage space for antenna traces with a fitness greater than the preset fitness during the iteration process. The preset fitness can be determined based on actual conditions; it can be obtained during the algorithm model training phase and set to a fixed value. The subdivided region can cache multiple individual antenna traces.

[0050] It should be noted that the antenna routing area refers to the area where the antenna routing is installed.

[0051] It should be noted that the antenna trace area can be pre-divided into multiple grids according to a preset grid division rule, and the divided grids are sequentially numbered. Each grid has two states: the presence of an antenna radiator and the absence of an antenna radiator, which can be represented by different identifiers. For example, "1" indicates the presence of an antenna radiator in the grid, and "0" indicates the absence of an antenna radiator in the grid. Before the population begins iterative updates, each grid can be randomly assigned a different identifier. Since each grid corresponds to a different identifier, multiple different individuals can be generated, that is, multiple different antenna traces can be randomly generated. The randomly generated antenna traces form the initial population.

[0052] It should be noted that since the antenna trace area may be regular or irregular, the size and shape of each grid obtained by dividing the antenna trace area can be the same or different, and the shape can be regular or irregular.

[0053] In one feasible implementation, when the antenna trace area is irregular, the meshing rule can be as follows: mark the antenna trace area with X and Y axes, and divide the antenna trace area into a mesh along the coordinate axis directions. The mesh size is not limited, but each mesh must have two mutually perpendicular lines to facilitate subsequent secondary meshing. Irregular parts are incorporated into the adjacent mesh with the largest contact surface, ensuring that all meshes ultimately do not intersect and completely cover the antenna trace area. The resulting meshes will have different shapes and sizes.

[0054] In another feasible implementation, when the antenna trace area is regular, the grid division rule can also be: to divide the antenna trace area into rows and columns at equal intervals, so that each grid obtained by the division has the same shape and size, which facilitates the subsequent secondary division of the grid.

[0055] It should be noted that the target mesh is one or more of the multiple meshes after the antenna trace area is divided. Compared with the related technologies that divide all meshes, this application can selectively determine one or more target meshes and divide the selected target meshes, which can reduce mesh complexity and subsequent model training costs. Moreover, the target meshes are generally meshes with significant electromagnetic field characteristics, or meshes that need further optimization. Dividing only the target meshes can improve the accuracy of the final determined antenna traces.

[0056] In this embodiment, the target individual may be the same or different in each iteration update.

[0057] During each iteration, it is determined whether the target individual meets the set performance criteria. If it does not, the individuals in the population are updated, and the next iteration begins. If, after a preset number of iterations, the target individual still does not meet the criteria and the fitness of the target individual in each iteration remains within a preset range, it indicates that the individuals in the population are concentrated in a local region. The algorithm may find the optimal solution within this local region and cannot escape it, i.e., it is trapped in a local optimum. This leads to inaccurate target individuals being identified. The reason for the population being trapped in a local optimum may be the limitation of the number of individuals in the initial population. Therefore, after the population has reached the preset number of iterations, if the fitness of the target individual in the population is detected to be within the preset range, the target grid in the antenna routing area is determined based on the individuals in the subdivided region. The target grid is then divided, and new antenna routing is generated. Since target grid division can generate new antenna routing, it enriches the diversity of individuals in the initial population, thereby reducing the difference between the final target individual and the design target.

[0058] In one feasible implementation, since the subdivided region contains multiple antenna traces with fitness greater than a preset fitness during the iteration process, the target mesh in the antenna trace area can be determined by the multiple antenna traces with fitness greater than the preset fitness cached in the subdivided region.

[0059] In another feasible implementation, electromagnetic simulation information of each individual element in the subdivided region can be obtained, and a target grid can be selected in the antenna trace area based on the electromagnetic simulation information corresponding to each individual element. The electromagnetic simulation information may include the maximum electric field point and / or the maximum current point. Since the electromagnetic simulation information can accurately reflect the distribution of the maximum electric field point and / or the maximum current point of each antenna trace, the target grid can be accurately selected from the antenna trace area.

[0060] Step S20: Divide the target mesh and generate new antenna traces.

[0061] In this embodiment, after determining the target grid, the target grid is divided. Since the target grid may be regular or irregular, the rules for dividing the target grid are similar to those for dividing the antenna trace area, and will not be repeated here. That is, there are no restrictions on the shape, size, and number of grids obtained by dividing the target grid.

[0062] In this embodiment, after the target grid is divided, each sub-grid resulting from the secondary division of the target grid is numbered in the same way as described above, and each sub-grid is randomly assigned a different state. Based on the corresponding state, a new antenna trace is generated for each sub-grid after the division, compared to the original grid without secondary division. It should be noted that the new antenna trace is different from the initial antenna trace.

[0063] For example, for each type of antenna trace, the state of the antenna trace in each grid of the antenna trace area can be quantified to generate a 1*N Boolean matrix corresponding to the antenna trace. The nth number in the matrix represents the antenna state of the nth grid of the antenna trace area. "1" indicates that there is an antenna radiator in the grid, and "0" indicates that there is no antenna radiator in the grid.

[0064] For example, refer to Figure 2 , Figure 2 This represents one type of antenna trace, where gray indicates the presence of an antenna radiator within the grid, and white indicates the absence of an antenna radiator within the grid. By analyzing... Figure 2 Quantization yields the Boolean matrix representation of the antenna trace as follows:

[0065] [1 1 1 1 0 1 0 1 0 1 0 0 0 1 1 1 1 1 0 1].

[0066] After selecting the target mesh, draw perpendicular lines on the two mutually perpendicular edges of the target mesh to divide the target mesh into four new sub-mesh.

[0067] During quantization, the data in the undivided grid remains in the same position as before. The data in the first row and first column of the four sub-grids of the target grid are placed in the original position. The first row and second column of the four newly divided sub-grids of each target grid are added to the end of all data in sequence. When adding data, the order of arrangement is as follows: for the sub-grids in the target grid, compare the rows first, then the columns. The smaller the row, the larger the row, and the smaller the column, the larger the column.

[0068] For example, refer to Figure 3 , Figure 3 This represents the new antenna routing after the target mesh has been divided. (This is achieved through...) Figure 3 Quantization is performed, and the new antenna trace quantization representation is obtained as follows:

[0069]

[0070] The Boolean matrix representation corresponding to this new antenna trace is determined as follows:

[0071] [1 1 0 0 1 0 0 0 0 1 1 1 0 1 1 0 1 1 1 0 1 1 1 0 1 1 1 0 1 1 0 0 1].

[0072] In this embodiment, after the number of iterations of the population reaches a preset number of iterations, it can detect that the fitness of the target individual in the population is within a preset range of variation. The target grid is determined based on the individuals cached in the subdivision region in each iteration, and the target grid is divided. Since the grid division will generate new antenna traces, it enriches the diversity of antenna traces in the population, thereby reducing the difference between the final target individual and the design target.

[0073] Based on the first embodiment of this application, in the second embodiment of this application, the content that is the same as or similar to that in the first embodiment described above can be referred to the above description and will not be repeated hereafter. Based on this, before determining the target grid in the antenna trace area according to the individuals in the subdivided region in step S10, step S01 is also included:

[0074] Step S01: After each iteration of the population, the antenna traces with fitness greater than the preset fitness are cached in the subdivision region.

[0075] In one feasible implementation, during each iteration update of the population, the fitness of each individual in the population is calculated, and antenna traces with fitness greater than a preset fitness are cached in the subdivision region. Here, multiple antenna traces with fitness greater than the preset fitness can be set for each iteration, and these antenna traces may be the same or different in each iteration. Therefore, after each iteration update, the antenna traces with fitness greater than the preset fitness are compared with the cached antenna traces in the subdivision region. If they are the same, one of the identical antenna traces is deleted to avoid duplicate caching and occupying storage space in the subdivision region; if they are different, the corresponding antenna trace is cached in the subdivision region.

[0076] In another feasible implementation, after each iteration, each individual in the population can be sorted in descending order of fitness, and individuals with a fitness level greater than or equal to a preset sort number can be cached in the subdivision region. The individual with the highest fitness in the sorted sequence has the highest fitness. The preset sort number can be set based on the number of individuals in the population, for example, set to 50% of the total number of individuals. Assuming there are 10 individuals in the population, individuals with a sort number greater than or equal to 5 can be cached in the subdivision region.

[0077] In another feasible implementation, after dividing the target mesh and generating new antenna traces, individuals in the antenna trace area can be deleted so that antenna traces with fitness greater than the preset fitness can be cached in the subsequent new round of iteration updates. This avoids interference from historical cached data to the new round of iteration updates, thereby improving the accuracy of subsequent iteration updates.

[0078] In this embodiment, after each iteration, antenna traces with fitness greater than a preset fitness are cached in the subdivision region. Since the subdivision region caches the better individuals in the population, the target grid can be accurately determined based on the individuals in the subdivision region, thereby improving the accuracy of the target grid.

[0079] Based on the above embodiments of this application, in the third embodiment of this application, the content that is the same as or similar to the above embodiments can be referred to the above description, and will not be repeated hereafter. On this basis, refer to Figure 4 In step S10, determining the target grid in the antenna trace area based on the individuals in the subdivided region includes steps A11 to A12:

[0080] Step A11: Obtain the current distribution and electric field distribution of each individual in the subdivided region;

[0081] It is important to note that in antenna design, the current distribution and electric field distribution of the antenna traces are among the key factors determining antenna performance. The current distribution of the antenna traces directly affects the antenna's radiation characteristics. Generally, current circulation occurs in the antenna traces during operation. The electric field distribution describes the intensity and direction of the electric field, directly reflecting the shape of the antenna radiation and the selectivity of the antenna's operating frequency.

[0082] It should be noted that the current distribution determines the electric field distribution. By adjusting the layout, shape, and size of the antenna traces, the current distribution can be altered, thereby indirectly adjusting the electric field distribution. Electromagnetic field simulation software can be used to accurately simulate and optimize the current and electric field distributions of the antenna traces to achieve optimal antenna performance.

[0083] In one feasible implementation, electromagnetic simulation can be performed on each individual in advance to obtain the current distribution and electric field distribution corresponding to each individual, and a mapping relationship between each individual and the current distribution and electric field distribution corresponding to that individual can be established. In subsequent use, the current distribution and electric field distribution corresponding to each individual in the subdivided area can be quickly obtained based on this mapping relationship.

[0084] Step A12: Select the target grid in the antenna trace area based on the current and electric field distribution of each individual.

[0085] In one feasible implementation, step A12 includes: determining the maximum current point, current variation, current path, electric field polarization type, electric field direction, and electric field intensity for each individual based on the current distribution and electric field distribution of each individual; and selecting a target grid in the antenna trace area based on the maximum current point, current variation, current path, electric field polarization type, electric field direction, and electric field intensity for each individual. The following will provide a detailed description of the maximum current point, current variation, current path, electric field polarization type, electric field direction, and electric field intensity for each individual:

[0086] Maximum current point: In some antenna types, such as half-wavelength antennas, the current typically reaches its maximum value at the center or node of the trace.

[0087] Current variation: The current varies along the length of the trace, typically decreasing from the point of maximum current towards both ends. This distribution determines the antenna's radiation pattern and directivity.

[0088] The path of current: The path of current inside the trace determines how electromagnetic waves are radiated from the antenna, thus affecting the antenna's gain and radiation efficiency.

[0089] Electric field polarization: describes the direction and polarization type of the electric field, such as horizontal polarization, vertical polarization, etc., which directly affects the radiated power of the antenna in different directions.

[0090] Electric field strength: The electric field strength at different locations and in different directions varies with the operating frequency. This is an important indicator that needs to be optimized based on the antenna's design parameters and materials.

[0091] In this embodiment, through the above implementation method, since each individual can determine the maximum current point, current change, current path, electric field polarization type, electric field direction and electric field strength corresponding to each individual based on the current distribution and electric field distribution of each individual, these indicators affect the radiation characteristics and frequency response of the antenna, etc. By selecting the target grid from the antenna trace area through these indicators, the performance of the finally determined antenna trace can be improved.

[0092] In one feasible implementation, refer to Figure 5 Step A12 includes steps A121 to A123:

[0093] Step A121: Based on the current distribution and electric field distribution of each individual, determine the distribution of the maximum electric field point and the maximum current point of each individual in the antenna trace area.

[0094] It should be noted that determining the maximum current point and the maximum electric field point is a crucial step in antenna design for evaluating antenna performance and safety. The distribution of the maximum electric field point and maximum current point for each individual antenna within the antenna trace area, based on its current and electric field distribution, can be represented as: determining the grid of the maximum electric field point and maximum current point for each individual antenna within the antenna trace area.

[0095] In one feasible implementation, electromagnetic simulation software can be used to determine the maximum electric field and maximum current points for each individual element: the current density distribution can be calculated within the antenna structure. The simulation software generates heatmaps or 2D / 3D images of the current density, showing the distribution of current on the conductor surface or traces. By analyzing these images, the locations of maximum current density can be found. Furthermore, the electric field strength and distribution are calculated using the electromagnetic simulation software; the electric field strength distribution is typically displayed as contour maps or 3D images. By analyzing these images, the regions with the maximum electric field strength can be found. Simulation software provides a powerful tool for calculating and visualizing these parameters, enabling rapid analysis of performance at different structures and frequencies during the design phase.

[0096] In one feasible implementation, the maximum electric field and maximum current points for each individual antenna can also be determined through experimental measurements: measurements are taken near the actual antenna structure using current probes or sensors. These measurements directly display the distribution of current density. Comparing and verifying the measurement results with simulation data helps confirm the location of the maximum current point. Furthermore, real-time measurements are performed around the antenna using electric field detectors or sensors. These measurements directly display the peak location of the electric field strength. Comparing the experimental measurement results with simulation data verifies the accuracy of the simulation model and confirms the location of the point of maximum electric field strength. Experimental measurements are crucial for verifying simulation results and evaluating performance in real-world environments, especially for applications requiring high precision or specific applications.

[0097] Using the above methods, the maximum current point and maximum electric field point in the antenna structure can be effectively determined, and the design can be optimized and adjusted accordingly.

[0098] Step A122: Based on the distribution of the maximum electric field point and the maximum current point in the antenna trace area, count the total number of times that the maximum electric field point and / or the maximum current point are distributed in each grid in the antenna trace area.

[0099] It should be noted that each individual has a corresponding maximum electric field point and maximum current point. The maximum electric field point and maximum current point for the same individual may be located in the same grid within the antenna trace area, or they may be located in different grids within the antenna trace area. Furthermore, the maximum electric field point and maximum current point for different individuals may be located in the same grid within the antenna trace area, or they may be located in different grids within the antenna trace area. Therefore, the total number of times the maximum electric field point and / or maximum current point is distributed across each grid within the antenna trace area can be counted.

[0100] It should be noted that the total number of times refers to the total number of times the maximum electric field point is distributed in each grid, or the total number of times the maximum current point is distributed in each grid, or the total number of times the maximum electric field point and the maximum current point are distributed in each grid.

[0101] Step A123: Determine the target grid based on the total number of times each grid has a maximum electric field point and / or a maximum current point.

[0102] In one feasible implementation, step A123 includes: determining the grid with the largest total number of distributions of maximum electric field points and / or maximum current points as the target grid.

[0103] For example: After the above algorithm iteration stops, electromagnetic simulation is performed on each individual cached in the subdivision region to obtain the current distribution and electric field distribution of each individual cached in the subdivision region. The grid where the maximum current point and the maximum electric field point of each individual are located is marked. After all the traces are marked, the number of times the maximum current point and / or the maximum electric field point appears in each grid is counted. The grid with the most occurrences of the maximum current point and / or the maximum electric field point is determined as the target grid, and the target grid is further subdivided.

[0104] For example, after electromagnetic simulation of trace 1 [1 0 0 1 0 0 0 0 1 1 0 1], the mesh numbers corresponding to the maximum current point and / or the maximum electric field point are 2 and 5 respectively. Then, mark it as [0 1 0 0 1 0 0 0 0 0 0 0]. Perform this operation on each individual in the subdivided region. Finally, count the number of times the maximum current point and / or the maximum electric field point appears in each mesh. Then, perform a secondary subdivision on the mesh with the most occurrences of the maximum current point and / or the maximum electric field point.

[0105] In another feasible implementation, step S123 includes: the grid with the total number of times the maximum electric field point and / or the maximum current point is distributed is greater than a preset number, which can be determined as the target grid. In this case, the number of target grids can be one or more. By dividing the determined multiple target grids, more new antenna traces are finally obtained, which enriches the diversity of antenna traces.

[0106] In this embodiment, the distribution of the maximum electric field point and the maximum current point corresponding to each individual in the antenna trace area is determined. Based on the distribution, the total number of times the maximum electric field point and / or the maximum current point is distributed in each grid in the antenna trace area is counted. Then, the target grid is determined based on the total number of times. Since the maximum electric field point and the maximum current point are not only used to evaluate the physical and electrical performance of the antenna structure, they can also guide the design optimization and material selection to ensure that the antenna has good performance and long-term stability under various operating conditions.

[0107] Reference Figure 6 Step S10, determining the target grid in the antenna trace area based on the individuals in the subdivided region, may further include steps B11 to B13:

[0108] Step B11: Determine the distribution of the antenna radiators of each individual in the antenna trace area within the subdivided region.

[0109] It should be noted that each individual antenna corresponds to a specific antenna trace, and each antenna trace may contain one or more antenna radiators depending on the actual design requirements. The antenna trace can be quantized within the antenna trace area to obtain the distribution of each individual antenna radiator within the antenna trace area. The quantization method is the same as in the first embodiment and will not be elaborated upon here.

[0110] Step B12: Based on the distribution of antenna radiators in the antenna trace area, count the total number of times that each grid in the antenna trace area contains an antenna radiator.

[0111] It should be noted that each individual has a corresponding antenna radiator. Since the initial antenna routing is randomly generated, it is uncertain which grid in the antenna routing area the antenna radiator belongs to. That is, each grid in the antenna routing area may have an antenna radiator, or only some grids may have antenna radiators. Furthermore, there may be a situation where a certain grid has antenna radiators from multiple individuals. Therefore, the total number of times antenna radiators are distributed in each grid of the antenna routing area can be counted.

[0112] It should be noted that the total number of times an antenna radiator is distributed in each grid within the antenna trace area is counted.

[0113] For example, suppose there are five groups of individuals in the region to be subdivided. After quantization, each group of individuals is represented by a different Boolean matrix:

[0114]

[0115] In each of the above Boolean matrices, the nth number represents the antenna state of the nth grid in the antenna trace area; "1" indicates that an antenna radiator exists in that grid, and "0" indicates that an antenna radiator does not exist in that grid. By statistically analyzing the distribution of each grid, the total number of times an antenna radiator exists in each grid distribution can be obtained. The statistical result can be expressed as:

[0116] [2 3 1 0 3 3 4 2 3 2 1 2 2 3 4 0 2 2 4 3].

[0117] Step B13: Determine the target grid based on the total number of times antenna radiators are distributed in each grid.

[0118] In this embodiment, the target grid is determined after counting the total number of times antenna radiators are distributed in each grid.

[0119] In one feasible implementation, the grid with the highest total number of antenna radiators can be identified as the target grid.

[0120] In another feasible implementation, to increase the diversity of antenna routing, grids with a total number of occurrences greater than a preset number can be identified based on the above statistical results, and these grids can be designated as target grids. For example, the preset number of occurrences can be set to 20%, and the top 20% of grids with the highest total number of occurrences can be designated as target grids.

[0121] In another feasible implementation, to further improve the accuracy of antenna routing, if multiple grids have a total number of iterations exceeding a preset number, the cutoff is made to one iteration more than the preset number. If the previous preset number of iterations all have the same total number of iterations, the cutoff is made to that number. For example, the preset number of iterations can be set to 20%. Since the number of iterations in the 20% is 3, the cutoff is made to 4 iterations, which is one iteration more than 3, according to the rules for subdividing the area. The number of iterations of 4 corresponds to... Figure 2 The second row, second column; the third row, fifth column; and the fourth row, fourth column are used to subdivide these three grids, resulting in the following: Figure 3 As shown.

[0122] In this embodiment, the distribution of the antenna radiators of each individual in the subdivided region in the antenna trace area is determined. Based on this distribution, the total number of times each grid in the antenna trace area contains an antenna radiator is counted. Then, the target grid is determined based on the total number of times each grid contains an antenna radiator. By evaluating the importance or contribution of each grid in the antenna radiation design, grids that may have significant electromagnetic field characteristics or grids that need further optimization are identified, thereby improving the accuracy of the target grid.

[0123] Based on the first embodiment of this application, in the fourth embodiment of this application, the same or similar content as the above embodiments can be referred to the above description, and will not be repeated hereafter. On this basis, in each population iteration, the target individual for each iteration is determined. The purpose of determining the target individual is to guide the algorithm towards the optimal solution to the problem, optimize the performance and efficiency of the algorithm, and provide feedback to improve the algorithm's execution strategy. In this embodiment, a method for determining the target individual is provided, which may include step A110:

[0124] Step A110: Determine the target individual based on the actual antenna reflection coefficient of each individual in the population at the target frequency band and the target antenna reflection coefficient.

[0125] It's important to note that a target frequency band refers to a specific range of frequencies designed or planned for transmitting or receiving signals in wireless communication or radio spectrum management. In wireless communication, devices need to transmit and receive data within a specific frequency range. The target frequency band determines the frequency range the device will use to ensure communication reliability and performance. When designing wireless devices, the selection of the target frequency band usually takes into account the device's technical compatibility. Different wireless technologies may operate in different frequency bands, therefore, the determination of the target frequency band affects the device's design and market adaptability. Furthermore, choosing a suitable target frequency band can help optimize the performance of the wireless system. Certain frequency bands may have better transmission characteristics (such as penetration ability, transmission distance, etc.) or avoid interference with other frequency bands; therefore, for a specific application, the selection of the target frequency band can directly affect the system's performance.

[0126] It should be noted that the antenna reflection coefficient describes the degree of signal loss caused by reflection during signal transmission. It is commonly used to quantify the antenna's reflection capability, i.e., how much input power the antenna reflects back to the transmitter or receiver.

[0127] It should be noted that the actual antenna reflection coefficient is the actual measured antenna reflection coefficient, while the target antenna reflection coefficient is the standard antenna reflection coefficient.

[0128] In one feasible implementation, the fitness of each individual can be determined based on the actual antenna reflection coefficient of each individual in the population at the target frequency band and the target antenna reflection coefficient, and then the target individual can be determined based on the fitness of each individual.

[0129] In another feasible implementation, refer to Figure 7 Step A110 may include steps A111 to A115:

[0130] Step A111: Obtain the actual antenna reflection coefficient of each individual in the population in the target frequency band, wherein the target frequency band includes multiple frequency points, and each frequency point has an actual antenna reflection coefficient and a target antenna reflection coefficient;

[0131] In one feasible implementation, the actual antenna reflection coefficient can be determined by measuring the reflected power at the antenna port. For example, by using a reflection measurement device connected to the antenna port or by using a network analyzer, the actual antenna reflection coefficient of each individual in the target frequency band can be obtained, thereby improving the accuracy of the actual antenna reflection coefficient of each individual.

[0132] In another feasible implementation, each individual in the population can be input into a pre-trained model to obtain the actual antenna transmission coefficient corresponding to each individual, thereby improving the efficiency and accuracy of obtaining the actual antenna reflection coefficient corresponding to each individual.

[0133] In one feasible implementation, during model training, the target antenna reflection coefficient corresponding to each individual can be obtained in advance through simulation, and a mapping relationship between each individual and its corresponding antenna reflection coefficient can be established. In subsequent use, the target antenna reflection coefficient corresponding to each individual can be directly obtained.

[0134] In another feasible implementation, the target frequency band includes multiple frequency points. It is necessary to obtain the actual antenna reflection coefficient and the target antenna reflection coefficient for each frequency point. The actual antenna reflection coefficient and the target antenna reflection coefficient for each frequency point may be the same or different. The fitness needs to be determined based on the values ​​of the actual antenna reflection coefficient and the target antenna reflection coefficient. Specifically, the values ​​when the actual antenna reflection coefficient and the target antenna reflection coefficient for the frequency point are the same, and the values ​​when they are different, can be obtained. The fitness is then determined based on both the values ​​when they are the same and the values ​​when they are different.

[0135] Step A112: Detect the first type of frequency point within the target frequency band where the actual antenna reflection coefficient is the same as the target antenna reflection coefficient, and determine the preset value as the value corresponding to the first type of frequency point;

[0136] Step A113: Detect the second type of frequency point where the actual antenna reflection coefficient is different from the target antenna reflection coefficient within the target frequency band, and determine the difference between the actual antenna reflection coefficient and the target antenna reflection coefficient as the value of the second type of frequency point;

[0137] Step A114: Obtain the fitness of each individual based on the average of the values ​​of the first and second frequency points.

[0138] It should be noted that the first type of frequency point refers to the frequency point corresponding to the same reflection coefficient of the actual antenna as that of the target antenna; multiple such points can be set for one. The second type of frequency point refers to the frequency point corresponding to the different reflection coefficients of the actual antenna and the target antenna; multiple such points can also be set for one. The preset value can be a fixed value, for example, the preset value can be set to 0.

[0139] In one feasible implementation, the fitness of each individual can be obtained by adding the values ​​of each first type frequency point and each second type frequency point together and then averaging them.

[0140] The above method enables targeted analysis of different frequency points within an individual, determines the corresponding values ​​for each frequency point, and determines the fitness based on the corresponding values ​​for each frequency point. This takes into account the differences between different frequency points and improves the accuracy of the individual's fitness.

[0141] Step A115: Determine the target individual based on fitness.

[0142] It should be noted that the lower the fitness, the better the individual. Therefore, the individual with the lowest fitness can be selected as the target individual and recorded.

[0143] In this embodiment, the fitness of each individual is determined based on the actual antenna reflection coefficient and the target antenna reflection coefficient at the target frequency. Then, each individual is identified as a target individual based on its corresponding fitness. Since the antenna reflection coefficient is used to quantify the antenna's reflection capability, determining the fitness of an individual through the antenna reflection coefficient ensures that the identified target individual possesses superior reflection capability and performance.

[0144] This application also provides another method for identifying target individuals, which may include step B110:

[0145] Step B110: The target individual is determined by the antenna routing optimization model, wherein the antenna routing optimization model is trained based on the correspondence between the preset antenna routing samples and the preset antenna reflection coefficient samples by a genetic algorithm.

[0146] In this embodiment, all individuals in the population can be input into the antenna routing optimization model. The antenna routing optimization model calculates the actual antenna reflection coefficient of each individual in the target frequency band. Based on the actual antenna reflection coefficient and the target antenna reflection coefficient, the fitness of each individual is determined. Then, the target individual is determined based on the fitness and the target individual is output.

[0147] In one feasible implementation, M preset antenna routing samples are randomly generated, each preset antenna routing sample being a 1*N Boolean matrix. The nth number in the matrix represents the antenna state of the nth grid, where 1 indicates the presence of an antenna radiator in that grid, and 0 indicates the absence of an antenna radiator in that grid. The M samples represent the randomly generated M types of preset antenna routing samples. These generated preset antenna routing samples are then simulated or measured to obtain the preset antenna reflection coefficient samples for each preset antenna routing sample within the target frequency band. Finally, the correspondence between the preset antenna routing samples and the preset antenna reflection coefficient samples is used to train the initial antenna routing optimization model, resulting in the final antenna routing optimization model.

[0148] In this dataset, the input information is X, represented as X = [x1, x2, ..., xn, ..., xN]. To distinguish between the actual antenna reflection coefficient output by the network and the preset antenna reflection coefficient, the actual antenna reflection coefficient output by the network is defined as Yc, represented as Yc = [Yc1, Yc2, ..., Yck..., YcK], where K is the number of frequency points in the target frequency band. The preset antenna reflection coefficient is defined as Y, represented as Y = [Y1, Y2, ..., Yk..., YK].

[0149] In this embodiment, the target individual is determined by an antenna routing optimization model, thereby improving the accuracy and acquisition efficiency of the target individual.

[0150] Furthermore, before determining the target individual through the antenna routing optimization model, a training process for the antenna routing optimization model is also included. The training process of the antenna routing optimization model includes: training the initial antenna routing optimization model using the correspondence between preset antenna routing samples and preset antenna reflection coefficient samples to obtain the actual antenna reflection coefficients of the preset antenna routing samples; obtaining the loss value between the actual antenna reflection coefficients of the preset antenna routing samples and the preset antenna reflection coefficients; if the loss value is found to be below the preset loss value, updating the preset antenna routing samples and preset antenna reflection coefficient samples, and continuing training using the correspondence between the updated preset antenna routing samples and updated preset antenna reflection coefficient samples; if the loss value is found to be above the preset loss value, outputting the antenna routing optimization model.

[0151] In a feasible implementation, after determining the preset antenna routing samples and preset antenna reflection coefficient samples, i.e., after determining the training dataset, it is also necessary to determine a loss function. This loss function is used to measure the difference between the actual antenna reflection coefficient Yc output by the initial antenna routing optimization model and the preset antenna reflection coefficient Y. The loss function here can be the mean squared error loss function, or it can be the mean absolute error loss function, Huber loss function, etc. Minimizing the loss value allows the actual antenna reflection coefficient output by the initial antenna routing optimization model to better fit the preset antenna reflection coefficient; the smaller the loss value, the better the accuracy of the neural network.

[0152] In one feasible implementation, the training process can be as follows: A randomly generated dataset is divided into a training set and a test set according to a certain ratio. The training set is used to train the initial antenna routing optimization model, while the test set is not used for training; only the loss value is calculated for the trained neural network. The fitting effect of the test set is ultimately used to represent the fitting effect of the neural network at this point. When training the neural network corresponding to the initial antenna routing optimization model, the sigmoid activation function and gradient descent method are selected to update the parameters in the neural network. After training is completed, all parameters in the antenna routing optimization model are fixed, and a genetic algorithm is used to determine the target individual for this antenna routing optimization model.

[0153] In this embodiment, by obtaining the loss value between the actual antenna reflection coefficient and the preset antenna reflection coefficient of the preset antenna routing sample, the difference between the actual antenna reflection coefficient and the preset antenna reflection coefficient of the model's current output can be determined by the loss value. By minimizing this loss value, the parameters of the model can be adjusted so that the prediction results of the model on the training data are closer to the preset values, thereby improving the generalization ability and prediction accuracy of the antenna routing optimization model for the optimal antenna routing.

[0154] Furthermore, after dividing the target mesh and generating new antenna traces, electromagnetic simulation is performed on the new antenna traces to obtain their antenna reflection coefficients. The antenna trace optimization model is then retrained using the correspondence between the new antenna traces and their antenna reflection coefficients.

[0155] In one feasible implementation, the trained dataset needs to be updated with new antenna routing, and the neural network corresponding to the antenna routing optimization model needs to be retrained and updated. After training, the data is re-introduced into the genetic algorithm for optimization. If the actual antenna reflection coefficient of the locally optimal individual still does not reach the preset antenna reflection coefficient, the data in the area to be subdivided is further subdivided to generate a new dataset for iterative optimization. As the number of secondary subdivisions increases, the convergence speed and efficiency of the antenna routing will tend to the optimal solution of the antenna scheme, until an antenna routing that reaches the preset antenna reflection coefficient is generated, at which point mesh subdivision stops.

[0156] In this embodiment, after generating a new antenna trace, the antenna trace optimization model is retrained using the correspondence between the new antenna trace and the antenna reflection coefficient of the new antenna trace, thereby improving the generalization ability and prediction accuracy of the antenna trace optimization model for the optimal antenna trace.

[0157] Based on the first embodiment of this application, in the fifth embodiment of this application, the content that is the same as or similar to the above embodiments can be referred to the above description, and will not be repeated hereafter. On this basis, this application also proposes an iterative update method for a population, which may include steps S210 to S230:

[0158] Step S210: After each iteration of the population, obtain the actual antenna reflection coefficient of the target individual;

[0159] Step S220: If the actual antenna reflection coefficient of the target individual does not meet the target antenna reflection coefficient, update the population and continue to iterate on the updated population.

[0160] Step S230: If the actual antenna reflection coefficient of the target individual meets the target antenna reflection coefficient, output the target individual.

[0161] It should be noted that the target antenna reflection coefficient can be the preset antenna reflection coefficient described in the above embodiments.

[0162] In this embodiment, the population identifies a target individual and obtains its actual antenna reflection coefficient at each iteration. If the actual antenna reflection coefficient of the target individual differs from the target antenna reflection coefficient determined by the antenna routing optimization model, the population is updated, and iteration continues on the updated population until the actual antenna reflection coefficient of the target individual satisfies the target antenna reflection coefficient. At this point, iteration stops, and the target individual is output.

[0163] In this process, the population is updated and iterated each time, and the number of iterations increases by one until the preset number of iterations is reached. When the actual antenna reflection coefficient of the target individual is different from the target antenna reflection coefficient determined by the antenna routing optimization model, and the fitness of the target individual is within the preset variation range, the target grid in the antenna routing area is determined, the target grid is divided, a new antenna routing is generated, and the antenna routing optimization model is retrained.

[0164] In this embodiment, the above method can solve the problem of increased grid partitioning and training costs caused by performing grid partitioning when the target antenna reflection coefficient is not met in each population iteration update, and can reduce grid partitioning and model training costs.

[0165] Furthermore, in a feasible implementation, if the actual antenna reflection coefficient of the target individual does not meet the target antenna reflection coefficient after each iteration update, the population is updated.

[0166] It should be noted that population updating can include one or more operations such as mutation, crossover, and selection. One or more of these operations can be performed simultaneously on the population to achieve updating. For example, mutating individuals in a population yields a new population, which is then used to update the existing population; and / or, crossovering individuals in a population yields a new population, which is then used to update the existing population; and / or, selection individuals in a population yields a new population, which is then used to update the existing population. The characteristics of each updating operation will be described in detail below:

[0167] (1) Mutation Operation: The mutation operation introduces new gene combinations or eigenvalues ​​by randomly altering genes or traits in the chromosomes of individuals, thereby increasing the diversity of the population. This helps avoid the algorithm getting trapped in local optima while exploring new possibilities in the search space. Furthermore, population diversity is crucial for avoiding premature convergence and improving global search capabilities during optimization. The mutation operation helps maintain the differences among individuals in the population, thus increasing the chance of discovering better solutions.

[0168] (2) Crossover operation: Crossover operation simulates the process of gene combination. By exchanging gene segments (chromosomal parts) between different individuals, different advantageous traits are combined to produce new individuals. This helps to inherit superior traits and share information, accelerating the optimization process. Moreover, by combining the advantageous traits of different individuals, crossover operation helps to accelerate the convergence of the population to a better solution, improving the efficiency and convergence speed of the algorithm.

[0169] (3) Selection Operation: The selection operation simulates the "survival of the fittest" principle in evolution, meaning that better individuals are more likely to survive and reproduce in the next generation. Common selection strategies include roulette wheel selection and competitive selection, which determine an individual's chances of survival and reproduction in the next generation based on its fitness (i.e., the quality of the solution). The selection operation ensures that only individuals with high fitness can continue to exist in the evolutionary process, thereby preserving and passing on excellent solutions to the next generation, further optimizing the overall performance of the population.

[0170] It should be noted that during the mutation operation, an individual in the population is randomly selected, and a 0-to-1 or 1-to-0 operation is randomly performed on a specific position of that individual. During the crossover operation, the current population is divided into pairs, forming the "old population." A specific position of each pair of individuals in the pair is swapped; this position can be a single point or multiple points, resulting in a new population. During the selection operation, the fitness of the old population and the corresponding new population is compared, and the individual with the better fitness is added to the new population. The resulting new population is then added to the next generation, and the iteration count is incremented by 1.

[0171] Furthermore, it should be noted that before the mutation operation, the individual with the best current fitness is retained. After the selection operation, the individual with the worst fitness in the new population is replaced with the previously retained individual with the best fitness.

[0172] In this embodiment, by performing mutation, crossover, and selection operations on the population in the manner described above, the optimal antenna routing can be found in the search space to simulate natural selection and genetic mechanisms.

[0173] For example, to help understand the implementation flow of the antenna trace generation method obtained by combining this embodiment with the above embodiment one, please refer to... Figure 8 , Figure 8 A simplified flowchart of an antenna routing generation method is provided. The antenna routing optimization process uses a genetic algorithm, and the algorithm flow is shown below:

[0174] 1. Create the initial grid.

[0175] 2. Update the grid and generate the dataset.

[0176] 3. Train the neural network using the dataset to obtain the antenna routing optimization model.

[0177] 4. Initialize the population based on the latest grid and set the regions to be subdivided. The initial population generation method is the same as the input information method used to generate the dataset when training the neural network. The regions to be subdivided are for subsequent storage of target individuals.

[0178] 5. Calculate fitness and determine the optimal individual (i.e., the target individual). Specifically, input the population into the antenna routing optimization model to obtain the actual antenna reflection coefficient for each individual. When the actual antenna reflection coefficient of an individual is the same as that of the target antenna, the individual is considered as a first-class frequency point with a preset value. When the actual antenna reflection coefficient of an individual is different from that of the target antenna, the difference between the actual antenna reflection coefficient of the individual and that of the target antenna is determined as the data for the second-class frequency point. Then, based on the value of the first-class frequency point and the average value of the second-class frequency point, the fitness of the individual is obtained. The lower the fitness, the better the individual. Select the target individual and record it.

[0179] 6. Does it meet judgment condition 1? Condition 1 is: the actual antenna reflection coefficient of the individual meets the target antenna reflection coefficient; if yes, proceed to step 7. If no, proceed to step 8.

[0180] 7. Output the optimal antenna routing and end the process.

[0181] 8. Update the population, including one or more of the following operations: mutation, crossover, and selection. Continue execution after updating the population. 9.

[0182] 9. Cache antenna traces with fitness greater than the preset fitness into the subdivision area.

[0183] 10. Does condition 2 meet? Condition 2 is: the fitness of the target individual in the population is within the preset range of variation. If yes, proceed to step 11; otherwise, return to step 5.

[0184] 11. Determine the target grid based on the individuals cached in the subdivided region.

[0185] 12. Divide the target grid and return to Execute 2.

[0186] It should be noted that the above examples are only for understanding this application and do not constitute a limitation on the antenna routing method of this application. Any simple transformations based on this technical concept are within the protection scope of this application.

[0187] This application also provides an antenna trace generation apparatus, please refer to... Figure 9 The antenna trace generation device includes:

[0188] The target mesh determination module 10 is used to detect that the fitness of the target individual in the population is within a preset range after the number of iterations of the population reaches a preset number of iterations, and to determine the target mesh in the antenna trace area based on the individuals in the subdivided area. The subdivided area is used to cache antenna traces with fitness greater than the preset fitness during the iteration process.

[0189] The target mesh partitioning module 20 is used to partition the target mesh and generate new antenna traces.

[0190] For example, the antenna routing generation device further includes a caching module, which is used to cache antenna routings with fitness greater than a preset fitness to the subdivision region after each iteration of the population.

[0191] For example, the target mesh determination module 10 is also used to obtain the current distribution and electric field distribution of each individual in the subdivided region; and select the target mesh in the antenna trace area based on the current distribution and electric field distribution of each individual.

[0192] For example, the target grid determination module 10 is further configured to determine the distribution of the maximum electric field point and the maximum current point of each individual in the antenna trace area based on the current distribution and electric field distribution of each individual; count the total number of times the maximum electric field point and / or the maximum current point is distributed in each grid in the antenna trace area based on the distribution of the maximum electric field point and the maximum current point in the antenna trace area; and determine the target grid based on the total number of times the maximum electric field point and / or the maximum current point is distributed in each grid.

[0193] For example, the target grid determination module 10 is also used to determine the distribution of the antenna radiators of each individual in the antenna trace area in the subdivided area; based on the distribution of the antenna radiators in the antenna trace area, to count the total number of times that each grid in the antenna trace area has an antenna radiator; and based on the total number of times that each grid has an antenna radiator, to determine the target grid.

[0194] For example, the antenna routing generation device further includes a target individual determination module, which is further used to determine the target individual based on the actual antenna reflection coefficient of each individual in the population in the target frequency band and the target antenna reflection coefficient; or, the target individual is determined by an antenna routing optimization model, wherein the antenna routing optimization model is trained based on a genetic algorithm on the correspondence between preset antenna routing samples and preset antenna reflection coefficient samples.

[0195] For example, the target individual determination module is further used to obtain the actual antenna reflection coefficient of each individual in the population in the target frequency band, wherein the target frequency band includes multiple frequency points, and each frequency point has an actual antenna reflection coefficient and a target antenna reflection coefficient; detect a first type of frequency point in the target frequency band where the actual antenna reflection coefficient is the same as the target antenna reflection coefficient, and determine a preset value as the value corresponding to the first type of frequency point; detect a second type of frequency point in the target frequency band where the actual antenna reflection coefficient is different from the target antenna reflection coefficient, and determine the difference between the actual antenna reflection coefficient and the target antenna reflection coefficient as the value of the second type of frequency point; obtain the fitness of each individual based on the average of the values ​​of the first type of frequency point and the values ​​of the second type of frequency point; and determine the target individual based on the fitness.

[0196] For example, the antenna routing generation device further includes a training module for an antenna routing optimization model. This training module trains the initial antenna routing optimization model using the correspondence between preset antenna routing samples and preset antenna reflection coefficient samples to obtain the actual antenna reflection coefficient of the preset antenna routing samples; it also obtains the loss value between the actual antenna reflection coefficient of the preset antenna routing samples and the preset antenna reflection coefficient; if the loss value is detected to be below the preset loss value, it updates the preset antenna routing samples and the preset antenna reflection coefficient samples, and continues training using the correspondence between the updated preset antenna routing samples and the updated preset antenna reflection coefficient samples; and if the loss value is detected to be above the preset loss value, it outputs the antenna routing optimization model.

[0197] For example, the training module of the antenna routing optimization model is also used to perform electromagnetic simulation on the new antenna routing to obtain the antenna reflection coefficient of the new antenna routing; and to retrain the antenna routing optimization model by using the correspondence between the new antenna routing and the antenna reflection coefficient of the new antenna routing.

[0198] The antenna routing generation apparatus provided in this application, employing the antenna routing generation method described in the above embodiments, can achieve antenna routing diversity and reduce the technical effect of the discrepancy between the final target individual and the design target. Compared with the prior art, the beneficial effects of the antenna routing generation apparatus provided in this application are the same as those of the antenna routing generation method provided in the above embodiments, and other technical features in the antenna routing generation apparatus are the same as those disclosed in the methods of the above embodiments, and will not be repeated here.

[0199] This application provides an electronic device, which includes: at least one processor; and a memory communicatively connected to the at least one processor; wherein the memory stores instructions executable by the at least one processor, which are executed by the at least one processor to enable the at least one processor to perform the antenna trace generation method in Embodiment 1 above.

[0200] The following is for reference. Figure 10 The diagram illustrates a structural schematic of an electronic device suitable for implementing embodiments of this application. The electronic devices in these embodiments may include, but are not limited to, mobile terminals such as mobile phones, laptops, digital broadcast receivers, PDAs (Personal Digital Assistants), PADs (Portable Application Descriptions), PMPs (Portable Media Players), in-vehicle terminals (e.g., in-vehicle navigation terminals), and fixed terminals such as digital TVs and desktop computers. Figure 10The electronic device shown is merely an example and should not impose any limitation on the functionality and scope of use of the embodiments of this application.

[0201] like Figure 10 As shown, the electronic device may include a processing unit 1001 (e.g., a central processing unit, a graphics processing unit, etc.), which can perform various appropriate actions and processes according to a program stored in a read-only memory (ROM) 1002 or a program loaded from a storage device 1003 into a random access memory (RAM) 1004. The RAM 1004 also stores various programs and data required for the operation of the electronic device. The processing unit 1001, ROM 1002, and RAM 1004 are interconnected via a bus 1005. An input / output (I / O) interface 1006 is also connected to the bus. Typically, the following systems can be connected to the I / O interface 1006: input devices 1007 including, for example, touchscreens, touchpads, keyboards, mice, image sensors, microphones, accelerometers, gyroscopes, etc.; output devices 1008 including, for example, liquid crystal displays (LCDs), speakers, vibrators, etc.; storage devices 1003 including, for example, magnetic tapes, hard disks, etc.; and communication devices 1009. Communication device 1009 allows electronic devices to communicate wirelessly or wiredly with other devices to exchange data. While electronic devices with various systems are shown in the figures, it should be understood that implementation or possession of all the systems shown is not required. More or fewer systems may be implemented alternatively.

[0202] Specifically, according to the embodiments disclosed in this application, the processes described above with reference to the flowcharts can be implemented as computer software programs. For example, embodiments disclosed in this application include a computer program product comprising a computer program carried on a computer-readable medium, the computer program containing program code for performing the methods shown in the flowcharts. In such embodiments, the computer program can be downloaded and installed from a network via a communication device, or installed from storage device 1003, or installed from ROM 1002. When the computer program is executed by processing device 1001, it performs the functions defined in the methods of the embodiments disclosed in this application.

[0203] The electronic device provided in this application, employing the antenna routing generation method described in the above embodiments, can achieve antenna routing diversity and reduce the technical effect of the discrepancy between the final target and the design target. Compared with the prior art, the beneficial effects of the electronic device provided in this application are the same as those of the antenna routing generation method provided in the above embodiments, and other technical features of this electronic device are the same as those disclosed in the previous embodiment method, and will not be repeated here.

[0204] It should be understood that the various parts disclosed in this application can be implemented using hardware, software, firmware, or a combination thereof. In the description of the above embodiments, specific features, structures, materials, or characteristics can be combined in any suitable manner in one or more embodiments or examples.

[0205] The above are merely specific embodiments of this application, but the scope of protection of this application is not limited thereto. Any variations or substitutions that can be easily conceived by those skilled in the art within the scope of the technology disclosed in this application should be included within the scope of protection of this application. Therefore, the scope of protection of this application should be determined by the scope of the claims.

[0206] In addition, this application also provides a computer program product, which includes a computer program that, when executed by a processor, implements the steps of the antenna trace generation method described above.

[0207] It should be noted that, in this document, the terms "comprising," "including," or any other variations thereof are intended to cover non-exclusive inclusion, such that a process, method, article, or system that comprises a list of elements includes not only those elements but also other elements not expressly listed, or elements inherent to such a process, method, article, or system. Unless otherwise specified, an element defined by the phrase "comprising one..." does not exclude the presence of other identical elements in the process, method, article, or system that includes that element.

[0208] Through the above description of the embodiments, those skilled in the art can clearly understand that the methods of the above embodiments can be implemented by means of software plus necessary general-purpose hardware platforms. Of course, they can also be implemented by hardware, but in many cases the former is a better implementation method. Based on this understanding, the technical solution of this application, in essence, or the part that contributes to the prior art, can be embodied in the form of a software product. This computer software product is stored in a storage medium (such as ROM / RAM, magnetic disk, optical disk) as described above, and includes several instructions to cause a terminal device (which may be a mobile phone, computer, server, or network device, etc.) to execute the methods of the various embodiments of this application.

[0209] The above are merely preferred embodiments of this application and do not limit the patent scope of this application. Any equivalent structural or procedural transformations made using the content of this application's specification and drawings, or direct or indirect applications in other related technical fields, are similarly included within the patent protection scope of this application.

Claims

1. A method for generating antenna traces, characterized in that, The method includes: After the number of iterations of the population reaches a preset number of iterations, it is detected that the fitness of the target individual in the population is within a preset range of variation. Based on the individuals in the subdivided region, the target grid in the antenna trace region is determined. The subdivided region is used to cache antenna traces with fitness greater than the preset fitness during the iteration process. The target mesh is divided to generate new antenna traces.

2. The method as described in claim 1, characterized in that, Before determining the target grid in the antenna trace area based on the individuals in the subdivided region, the process also includes: After each iteration of the population, antenna traces with fitness greater than a preset fitness are cached in the subdivided region.

3. The method as described in claim 1 or 2, characterized in that, The process of determining the target grid in the antenna trace area based on the individuals in the subdivided region includes: Obtain the current distribution and electric field distribution of each individual in the subdivided region; The target grid is selected in the antenna trace area based on the current distribution and electric field distribution of each individual.

4. The method as described in claim 3, characterized in that, The step of selecting the target grid in the antenna trace area based on the current distribution and electric field distribution of each individual grid includes: Based on the current distribution and electric field distribution of each individual, determine the distribution of the maximum electric field point and the maximum current point corresponding to each individual in the antenna trace area; Based on the distribution of the maximum electric field point and the maximum current point in the antenna trace area, count the total number of times that the maximum electric field point and / or the maximum current point are distributed in each grid of the antenna trace area. The target grid is determined based on the total number of times the maximum electric field point and / or the maximum current point are distributed in each grid.

5. The method as described in claim 1 or 2, characterized in that, The process of determining the target grid in the antenna trace area based on the individuals in the subdivided region includes: Determine the distribution of the antenna radiators of each individual in the antenna trace area within the subdivided region; Based on the distribution of the antenna radiators in the antenna trace area, the total number of times the antenna radiators are distributed in each grid of the antenna trace area is counted. The target grid is determined based on the total number of times the antenna radiators are distributed in each grid.

6. The method as described in claim 1, characterized in that, The method further includes: The target individual is determined based on the actual antenna reflection coefficient of each individual in the population at the target frequency band and the target antenna reflection coefficient; Alternatively, the target individual can be determined through an antenna routing optimization model, wherein the antenna routing optimization model is trained based on a genetic algorithm to obtain the correspondence between preset antenna routing samples and preset antenna reflection coefficient samples.

7. The method as described in claim 6, characterized in that, The step of determining the target individual based on the actual antenna reflection coefficient of each individual in the population at the target frequency band and the target antenna reflection coefficient includes: Obtain the actual antenna reflection coefficient of each individual in the population in the target frequency band, wherein the target frequency band includes multiple frequency points, and each frequency point has an actual antenna reflection coefficient and a target antenna reflection coefficient; Detect the first type of frequency point in the target frequency band where the actual antenna reflection coefficient is the same as the target antenna reflection coefficient, and determine the preset value as the value corresponding to the first type of frequency point; Detect a second type of frequency point where the actual antenna reflection coefficient differs from the target antenna reflection coefficient within the target frequency band, and determine the difference between the actual antenna reflection coefficient and the target antenna reflection coefficient as the value of the second type of frequency point; The fitness of each individual is obtained based on the average of the values ​​of the first type of frequency points and the values ​​of the second type of frequency points. The target individual is determined based on the fitness.

8. The method as described in claim 6, characterized in that, Before determining the target individual using the antenna routing optimization model, the process also includes: The initial antenna routing optimization model is trained using the correspondence between the preset antenna routing sample and the preset antenna reflection coefficient sample to obtain the actual antenna reflection coefficient of the preset antenna routing sample. Obtain the loss value between the actual antenna reflection coefficient and the preset antenna reflection coefficient of the preset antenna trace sample; If the loss value is found to be less than the preset loss value, the preset antenna trace sample and the preset antenna reflection coefficient sample are updated, and the training continues using the correspondence between the updated preset antenna trace sample and the updated preset antenna reflection coefficient sample. If the loss value is detected to reach the preset loss value, the antenna routing optimization model is output.

9. The method as described in claim 6, characterized in that, After dividing the target mesh and generating new antenna traces, the process further includes: Electromagnetic simulation was performed on the new antenna trace to obtain the antenna reflection coefficient of the new antenna trace; The antenna routing optimization model is retrained using the correspondence between the new antenna routing and the antenna reflection coefficient of the new antenna routing.

10. An electronic device, characterized in that, The device includes: a memory, a processor, and a computer program stored in the memory and executable on the processor, the computer program being configured to implement the steps of the antenna trace generation method as described in any one of claims 1 to 9.

11. A computer program product, characterized in that, The computer program product includes a computer program that, when executed by a processor, implements the steps of the antenna trace generation method as described in any one of claims 1 to 9.

Images