Design method and system for a dual-band Yagi antenna

Abstract

This application relates to the field of antenna technology, and in particular to a design method and system for a dual-band Yagi antenna. The method includes constructing an initial antenna model based on the design requirements of the dual-band Yagi antenna; selecting several groups of antenna design parameters as input samples in the antenna design space and inputting them into the initial antenna model for simulation to obtain the antenna model response corresponding to each input sample; calculating the coupling compensation value based on each input sample and its corresponding antenna model response, and using the input sample as the input, its corresponding antenna model response and the coupling compensation value as the output to train a preset model to obtain an antenna performance prediction model; designing an objective function, searching for antenna design parameters and inputting them into the model, calculating the objective function value according to the output, and determining the optimal antenna design parameters based on whether the objective function value meets the preset optimization criteria. This application can specifically design for the dual-band Yagi antenna to determine the optimal antenna parameters that meet the requirements of dual-band coordination.

Description

Technical Field

[0001] This application relates to the field of antenna technology, and particularly to a design method and system for a dual-band Yagi antenna. Background Art

[0002] In modern wireless communication, Internet of Things, and 5G application scenarios, the ability to support transmission on two predetermined operating frequency bands simultaneously can significantly improve the flexibility and performance of the system. For example, different frequency bands may have their own advantages: one frequency band can achieve longer-distance coverage, while the other frequency band can provide higher data transmission rates or better anti-interference capabilities; in addition, through dual-band collaborative operation, load balancing can be achieved, reducing the risk of congestion and interference in a single frequency band. The Yagi antenna has always been regarded as an ideal choice for long-distance communication and directional transmission due to its simple structure, convenient installation, high forward gain, and excellent directivity. To meet the dual-band operation requirements, the dual-drive oscillator method is usually adopted. Based on the traditional Yagi antenna structure, two independent drive oscillators are set at the front end of the antenna, corresponding to different target frequency bands respectively, and the matching reflectors and directors are configured to achieve simultaneous resonance and efficient radiation of the antenna at two frequency bands.

[0003] After adopting the dual-drive oscillator method, the basic hardware structure of the dual-band Yagi antenna is determined, and the key issue turns to how to accurately determine the parameters such as the optimal size, spacing, and number of each component, such as reflectors, two independent drive oscillators, and directors, to ensure that the antenna can achieve the best performance at both frequency bands. In the prior art, the following method is usually adopted: First, an initial antenna model is constructed, and several sample antenna parameters are collected in the design parameter space. Then, these parameters are input into the initial antenna model, and the corresponding antenna response data is obtained through simulation. Next, the mapping relationship between the sample antenna parameters and the antenna response data is established. Finally, combined with the multi-objective optimization algorithm, parameter screening is carried out for each performance index to determine the optimal antenna parameters that meet the design requirements.

[0004] However, directly applying the above traditional method to the dual-band Yagi antenna constituted by the dual-drive oscillator method has obvious deficiencies. Since the dual-band Yagi antenna requires simultaneous operation at two frequency bands, there may be cross-coupling and mutual interference between the two frequency bands, and its performance response is no longer a simple single-frequency band problem, but rather a complex mapping from design parameters to the responses of two frequency bands, and there is even a compensatory effect of mutual influence. In this case, the general antenna parameter optimization method is difficult to accurately capture and handle the coupling problem between frequency bands, resulting in the optimization result not being able to take into account the best performance of each frequency band at the same time. Therefore, there is an urgent need to develop a method specifically for the design of dual-band Yagi antennas to determine the optimal antenna parameters that meet the dual-band collaborative requirements. Summary of the Invention

[0005] The present application provides a design method and system for a dual - frequency Yagi antenna, which can be specifically designed for a dual - frequency Yagi antenna to determine the optimal antenna parameters that meet the requirements of dual - frequency cooperation. The present application provides the following technical solutions:

[0006] In a first aspect, the present application provides a design method for a dual - frequency Yagi antenna, and the method includes:

[0007] Construct an initial antenna model based on a predefined operating frequency band and the design requirements of the dual - frequency Yagi antenna;

[0008] Adopt a hierarchical clustering sampling method to select several groups of antenna design parameters as input samples in the antenna design space and input them into the initial antenna model for simulation to obtain the antenna model response corresponding to each input sample;

[0009] Calculate the coupling compensation value based on each input sample and its corresponding antenna model response, and the coupling compensation value is used to measure the mutual coupling degree between the two operating frequency bands of the dual - frequency Yagi antenna;

[0010] Use the input sample as the input, its corresponding antenna model response and the coupling compensation value as the output to train a preset neural network model to obtain an antenna performance prediction model;

[0011] Design an objective function based on the antenna model response and the coupling compensation value, search for antenna design parameters and input them into the antenna performance prediction model, calculate the objective function value according to the model output, and determine whether the preset optimization standard is met based on the objective function value to determine the optimal antenna design parameters.

[0012] In a specific feasible implementation, the constructing an initial antenna model based on a predefined operating frequency band and the design requirements of the dual - frequency Yagi antenna includes:

[0013] Define the two target operating frequency bands of the dual - frequency Yagi antenna, denoted as and ;

[0014] The initial antenna model is used to describe the basic structure of the dual - frequency Yagi antenna. The dual - frequency Yagi antenna includes two independent driven elements, a reflector, and one or more directors;

[0015] The antenna design parameters include the lengths and relative spacings of the two driven elements, the length of the reflector and its actual distance from the driven elements, the number of directors, their lengths, and their spacings from the driven elements.

[0016] In a specific feasible implementation, the adopting a hierarchical clustering sampling method to select several groups of antenna design parameters as input samples in the antenna design space and input them into the initial antenna model for simulation includes:

[0017] Perform sensitivity analysis on the antenna design parameters and stratify the antenna design parameters based on the sensitivity weights;

[0018] Construct multi-dimensional parameter intervals based on the stratification results, and select representative samples using a clustering method within each multi-dimensional interval;

[0019] Integrate the selected representative samples into a candidate sample set , and perform weighted random sampling based on the sampling probability to obtain input samples.

[0020] In a specific feasible implementation, the performing sensitivity analysis on the antenna design parameters and stratifying the antenna design parameters based on the sensitivity weights includes:

[0021] On the basis of the initial antenna model, make a small perturbation to each antenna design parameter and record the change amplitude of the return loss. For each antenna design parameter , define its sensitivity coefficient as follows:

[0022] ;

[0023] where is the change amplitude of the return loss after perturbation, is the change amount of the antenna design parameter;

[0024] Normalize the sensitivity coefficient to the sensitivity weight as follows:

[0025] ;

[0026] where is the number of types of antenna design parameters;

[0027] Stratify the antenna design parameters according to the sensitivity weights. For the antenna design parameters with sensitivity weights greater than the preset threshold, identify them as key parameters and divide their parameter intervals into 10 layers. For the antenna design parameters with sensitivity weights less than or equal to the preset threshold, identify them as secondary parameters and divide their parameter intervals into 2 layers.

[0028] In a specific feasible implementation, the integrating the selected representative samples into a candidate sample set , and performing weighted random sampling based on the sampling probability to obtain input samples includes;

[0029] Integrate the representative samples selected from all multi-dimensional intervals into the final candidate sample set, and calculate the representative score of each candidate sample using the following formula :

[0030] ;

[0031] wherein, is the value of the candidate sample in the dimension, and are the mean and standard deviation of the th parameter in the candidate sample set respectively, is the sensitivity weight of the th parameter, is the sample and its distance from the cluster center it belongs to, is the scaling coefficient to adjust the influence of the clustering distance;

[0032] Calculate the sampling probability of each sample as follows:

[0033] ;

[0034] Based on the calculated sampling probability perform weighted random sampling on all candidate samples, and use the obtained representative sample set as the input sample.

[0035] In a specific feasible implementation, the calculating the coupling compensation value based on each input sample and its corresponding antenna model response includes:

[0036] The antenna model response includes the frequency band response and the frequency band response. The frequency band response includes the antenna return loss, gain and standing wave ratio at . The frequency band response includes

[0037] The calculation formula of the coupling compensation value is as follows:

[0038] ;

[0039] wherein, and respectively represent the return loss at the working frequency bands and , and respectively represent the antenna gains at the working frequency bands and , and respectively represent the return loss at the working frequency bands and The standing wave ratio under is a positive constant, and is the hyperbolic tangent function.

[0040] In a specific feasible implementation, the design objective function based on the antenna model response and the coupling compensation value includes:

[0041] The objective function is as follows:

[0042] ;

[0043] Wherein, is the coupling compensation value output by the antenna performance prediction model, is the adjustment coefficient, and are respectively the performance errors in the working frequency bands and The calculation methods are as follows:

[0044] ;

[0045] ;

[0046] Wherein, and are respectively the return loss outputs by the antenna performance prediction model in the working frequency bands and The antenna gain outputs by the antenna performance prediction model in the working frequency bands and are respectively and The standing wave ratio outputs by the antenna performance prediction model in the working frequency bands and are respectively and The preset return loss target values in the working frequency bands and are respectively and The preset antenna gain target values in the working frequency bands and are respectively and The preset standing wave ratio target values in the working frequency bands and are respectively and The preset standing wave ratio target values.

[0047] In a second aspect, the present application provides a design system for a dual - frequency Yagi antenna, adopting the following technical solutions:

[0048] A design system for a dual - frequency Yagi antenna, comprising:

[0049] An initial antenna model construction module, configured to construct an initial antenna model based on a predefined operating frequency band and the design requirements of the dual - frequency Yagi antenna;

[0050] An antenna design parameter simulation module, configured to select several groups of antenna design parameters as input samples in the antenna design space by using a hierarchical clustering sampling method and input them into the initial antenna model for simulation, so as to obtain the antenna model response corresponding to each input sample;

[0051] A coupling compensation value calculation module, configured to calculate a coupling compensation value based on each input sample and its corresponding antenna model response, where the coupling compensation value is used to measure the mutual coupling degree between the two operating frequency bands of the dual - frequency Yagi antenna;

[0052] A prediction model training module, configured to use the input samples as inputs, and the corresponding antenna model responses and coupling compensation values as outputs to train a preset neural network model to obtain an antenna performance prediction model;

[0053] An antenna design parameter determination module, configured to design an objective function based on the antenna model response and the coupling compensation value, search for antenna design parameters and input them into the antenna performance prediction model, calculate the objective function value according to the model output, and determine whether the preset optimization criterion is met based on the objective function value, so as to determine the optimal antenna design parameters.

[0054] In a third aspect, the present application provides an electronic device, where the device includes a processor and a memory; a program is stored in the memory, and the program is loaded and executed by the processor to implement a design method for a dual - frequency Yagi antenna as described in the first aspect.

[0055] In a fourth aspect, the present application provides a computer - readable storage medium, where a program is stored in the storage medium, and the program is used to implement a design method for a dual - frequency Yagi antenna as described in the first aspect when executed by a processor.

[0056] In summary, the beneficial effects of the present application at least include:

[0057] (1) An optimization method based on adaptive weight hierarchical clustering sampling is proposed. Compared with the traditional Latin hypercube sampling method, it can optimize design parameters more accurately. This method first conducts a sensitivity analysis of antenna design parameters and assigns different weights according to the influence degree of each parameter on the antenna performance. For parameters with higher sensitivity, more levels are divided to explore the key parameter space that has a greater impact on the antenna performance with a finer resolution, thus effectively improving the pertinence of the sampling strategy. Through this method of sensitivity weighting and hierarchical clustering sampling, the exploration accuracy of the area affecting the key performance can be significantly improved, avoiding the problem of insufficient attention to key parameters in traditional sampling methods. This method can more effectively identify the important factors affecting the antenna performance, reduce the exploration of the invalid design space, and thus achieve a more accurate design optimization process, ensuring that the obtained antenna design parameters achieve the optimal performance effect.

[0058] (2) By introducing the concept of coupling compensation value, the interference problem caused by mutual coupling between frequency bands in the dual-frequency working mode is further solved. The defined coupling compensation value comprehensively considers key performance parameters such as the return loss, gain, and voltage standing wave ratio of the antenna in different frequency bands, and quantitatively analyzes the coupling degree between the working frequency bands of the dual-frequency antenna through a specific calculation formula. In the antenna optimization design process, by incorporating the coupling compensation value into the performance evaluation system, the actual performance of the antenna in the dual-frequency working state can be more accurately reflected. By preferentially selecting design parameters with lower coupling degree and stable performance, the signal coupling and interference between different working frequency bands can be effectively reduced, thereby improving the working stability and overall performance of the antenna in the dual-frequency mode. This has significant technical advantages for ensuring signal quality and suppressing signal loss in dual-frequency communication applications.

[0059] By first constructing an initial antenna model that only determines the basic structure and geometric layout, then using the adaptive weight hierarchical clustering sampling method to select representative samples in the design parameter space, and obtaining the antenna performance response data at two operating frequency bands through electromagnetic simulation, calculating the coupling compensation value based on the antenna performance response data, comprehensively mapping the key performance parameters of the two frequency bands into a quantization index to reflect the performance balance and coordination degree between the frequency bands. Subsequently, using the input samples and their corresponding dual-frequency responses and coupling compensation values as outputs to train a preset neural network model, an antenna performance prediction model that can simultaneously predict the performance and coupling compensation value at two operating frequency bands is constructed. Finally, a multi-objective optimization algorithm is used to search for candidate solutions in the entire design parameter space, and an objective function is designed. The objective function not only considers the performance errors of the two frequency bands but also strengthens the constraint on insufficient coupling through an exponential penalty mechanism, thereby screening out one or more groups of optimal design parameters that not only meet the best performance requirements of each frequency band but also achieve good cooperative operation. Such a design not only makes up for the deficiency of traditional methods in capturing complex nonlinear interactions between frequency bands but also realizes the overall optimization of the cooperative performance of the dual-frequency antenna through the introduction of the coupling compensation value, effectively solving the problem of performance imbalance caused by frequency band cross-coupling and mutual interference during dual-frequency operation in the background technology.

[0060] The above description is only an overview of the technical solution of this application. In order to be able to understand the technical means of this application more clearly and implement it in accordance with the content of the specification, the following describes in detail with reference to the preferred embodiments of this application and the accompanying drawings. Brief Description of the Drawings

[0061] Figure 1 It is a flowchart of the design method of the dual-frequency Yagi antenna in the embodiment of this application.

[0062] Figure 2 It is an overall flowchart of the design method of the dual-frequency Yagi antenna in the embodiment of this application.

[0063] Figure 3 It is a diagram of the effect verification data of the design method of the dual-frequency Yagi antenna in the embodiment of this application.

[0064] Figure 4 It is a structural block diagram of the design system of the dual-frequency Yagi antenna in the embodiment of this application.

[0065] Figure 5 It is a block diagram of the electronic device for the design of the dual-frequency Yagi antenna in the embodiment of this application. Detailed Description of the Embodiment

[0066] The following further describes in detail the specific implementation manners of this application with reference to the drawings and embodiments. The following embodiments are used to illustrate this application but are not used to limit the scope of this application.

[0067] Optionally, taking the design method of the dual - band Yagi - Uda antenna provided in each embodiment of the present application as an example for an electronic device, the electronic device is a terminal or a server. The terminal can be a mobile phone, a computer, a tablet computer, etc. The type of the electronic device is not limited in this embodiment.

[0068] Referring to Figure 1 , which is a schematic flowchart of the design method of the dual - band Yagi - Uda antenna provided in an embodiment of the present application. The method at least includes the following steps:

[0069] Step S101: Construct an initial antenna model based on a predefined operating frequency band and the design requirements of the dual - band Yagi - Uda antenna.

[0070] In step S101, first, two target operating frequency bands of the dual - band Yagi - Uda antenna are predefined, denoted as and respectively. Subsequently, an initial antenna model of the dual - band Yagi - Uda antenna is constructed based on the design idea of the dual - driven oscillator method. The initial antenna model is used to describe the basic structure of the dual - band Yagi - Uda antenna. The dual - band Yagi - Uda antenna includes two independent driven oscillators, a reflector, and one or more directors.

[0071] Among them, each driven oscillator corresponds to the operating frequency band and respectively. In its preliminary design, the length of the driven oscillator is set according to the half - wavelength of its respective operating frequency band. The reflector is located behind the driven oscillator and is usually slightly longer than the driven oscillator in design. The directors are arranged in front of the driven oscillator, and their quantity and size are reserved as adjustable parameters. After the initial antenna model is constructed, only the basic framework and geometric layout of the dual - band Yagi - Uda antenna are determined, and the antenna design parameters are not yet determined. The antenna design parameters include the lengths of the two driven oscillators and their relative spacing, the length of the reflector and its actual distance from the driven oscillator, the quantity, length, and spacing between the directors and the driven oscillator.

[0072] It should be noted that although the predefined operating frequency bands and provide a theoretical basis for the preliminary design of the driven oscillator, that is, the ideal length of each driven oscillator is half - wavelength, but in actual design, the exact dimensions of the driven oscillator still need to be further optimized. Because the actual antenna performance is affected not only by the operating frequency band, but also by various factors such as the cross - coupling between the two driven oscillators, the antenna environment, manufacturing errors, and the interaction with the reflector and director configurations. Therefore, although theoretically the length of the driven oscillator can be determined by half - wavelength, its specific dimensions still need to be fine - tuned through subsequent simulations and optimizations to achieve the best actual effect.

[0073] In addition, in this application, the dual-drive oscillators adopt a symmetric layout. Therefore, the distance between the reflector and the two drive oscillators is defined as the distance from the reflector to the symmetric center of the drive oscillators. Similarly, the distance between the director and the two drive oscillators is also the distance from the director to the symmetric center of the drive oscillators.

[0074] Step S102: Use the hierarchical clustering sampling method to select several groups of antenna design parameters in the antenna design space as input samples and input them into the initial antenna model for simulation to obtain the antenna model response corresponding to each input sample.

[0075] In step S102, first, select several groups of antenna design parameters in the antenna design space as input samples, and input the input samples into the initial antenna model. Use an electromagnetic simulation tool to perform simulation and solution on the initial antenna model of the input samples, so as to obtain the antenna model response corresponding to the input samples. The antenna model response includes but is not limited to antenna return loss, gain, and voltage standing wave ratio.

[0076] In implementation, input the input samples into the initial antenna model, and use an electromagnetic simulation tool to perform dual-frequency joint simulation and solution on the initial antenna model of the input samples. That is, the simulation and solution of each input sample need to be carried out simultaneously in the operating frequency band and to obtain the antenna model responses in two frequency bands. For each input sample, the antenna model response includes the frequency band response and the frequency band response. Specifically, that is the antenna return loss, gain, and voltage standing wave ratio under and the antenna return loss, gain, and voltage standing wave ratio under

[0077] In addition, preferably, when selecting several groups of antenna design parameters in the antenna design space as input samples, the prior art usually uses the Latin hypercube sampling method for sample selection. This method can evenly cover the parameter space, but in the dual-frequency Yagi-Uda antenna design problem with high dimensions and significant nonlinearity and interaction between parameters, it is often difficult to fully capture the key performance sensitive areas, resulting in possible omission of important candidate solutions in subsequent optimization. Therefore, to make up for the deficiency of the Latin hypercube sampling in representing the key areas, this application proposes a new adaptive weight hierarchical clustering sampling method. By performing sensitivity analysis on each design parameter, different importance weights are assigned to each parameter, focusing on the key variables affecting the dual-frequency performance. Using hierarchical and clustering techniques to partition the design parameter space, and then selecting representative samples from each area, so as to ensure that the samples evenly cover the entire design space and focus on the high-sensitivity areas, avoiding repeated sampling.

[0078] Among them, the specific method of the adaptive weight hierarchical clustering sampling method is as follows:

[0079] Step S1021: Conduct sensitivity analysis on the antenna design parameters and perform hierarchical partitioning on the antenna design parameters based on the sensitivity weights.

[0080] Specifically, based on the initial antenna model, a small perturbation is made to each antenna design parameter. In this application, it is , and the change amplitude of the return loss is recorded. For each antenna design parameter , its sensitivity coefficient is defined as follows:

[0081] ;

[0082] where is the change amplitude of the return loss after perturbation, and is the change amount of the antenna design parameter.

[0083] It should be noted that the return loss directly reflects the radiation efficiency and matching degree of the antenna, and is the core performance index that best reflects the design quality. In a dual - frequency Yagi - Uda antenna, ensuring that the return loss under both operating frequency bands is simultaneously lower than - 10 dB is the core requirement.

[0084] Subsequently, the sensitivity coefficient is normalized to the sensitivity weight as follows:

[0085] ;

[0086] where is the number of types of antenna design parameters. The sensitivity weight reflects the relative influence degree of each antenna design parameter on the antenna performance. The higher the sensitivity weight, the greater the influence of the parameter on the antenna performance.

[0087] Finally, based on the sensitivity weights, hierarchical partitioning is performed on the antenna design parameters. For the antenna design parameters with sensitivity weights greater than the preset threshold, they are identified as key parameters and their parameter intervals are divided into more layers. In this application, it is 10 layers. For the antenna design parameters with sensitivity weights less than or equal to the preset threshold, they are identified as secondary parameters and their parameter intervals are divided into fewer layers. In this application, it is 2 layers. The above - mentioned hierarchical strategy can, to a certain extent, ensure more refined exploration of the key antenna design parameters while reducing the sampling complexity for the antenna design parameters with less influence.

[0088] Step S1022: Construct multi - dimensional parameter intervals based on the hierarchical results, and use the clustering method to select representative samples within each multi - dimensional interval.

[0089] Specifically, after conducting sensitivity analysis on each antenna design parameter and stratifying according to the sensitivity weights, each parameter has been divided into several layers. All Combining the layer divisions of the antenna design parameters can divide the entire design parameter space into small multi-dimensional intervals, where is the number of layers of the th antenna design parameter. For example, if there are 3 key parameters (10 layers) and 2 secondary parameters (2 layers), then a total of small intervals can be divided. The multi-dimensional interval is the discretization result of the antenna design parameter space. Each interval covers a set of similar parameter values, ensuring finer subdivision in the highly sensitive area, so as to more fully capture the impact of key parameter changes on performance indicators such as return loss. For each generated multi-dimensional interval, all possible candidate parameter points within the interval are extracted through coarse sampling. If there are many candidate points within the interval, the clustering algorithm is used to divide these points into several sub-clusters, and the point closest to the cluster center is selected from each sub-cluster as the representative sample. If the number of candidate points within the interval is small, then all points are directly selected or the regional center value is used as the representative sample. In this way, the number of layers set in the layering directly determines the number of divisions of the multi-dimensional interval, and clustering within each interval can avoid repeated sampling in the area where parameter values are similar, ensuring that the selected representative samples can evenly cover the entire design space and focus on capturing the change information in the highly sensitive area.

[0090] Optionally, the k-means or hierarchical clustering algorithm is used in this application as the specific clustering algorithm to divide points into several sub-clusters. Other types of clustering algorithms can also be used. This application does not limit the specific type of clustering algorithm.

[0091] Step S1023: Integrate the selected representative samples into a candidate sample set , and perform weighted random sampling based on the sampling probability to obtain input samples.

[0092] Specifically, integrate the representative samples selected from all multi-dimensional intervals into the final candidate sample set. Then, use the following formula to calculate the representative score of each candidate sample :

[0093] ;

[0094] where is the value of the candidate sample in the th dimension, and are the mean and standard deviation of the th parameter in the candidate sample set respectively, is the sensitivity weight of the th parameter, is the sample and its belonging cluster center The distance, is the scaling coefficient for adjusting the influence of the clustering distance. In the above formula, first, the standardized deviation degree of the samples in each dimension from the mean of all candidate samples is calculated and weighted by the sensitivity weights, so that the parameters with greater influence on the antenna performance can receive higher attention. This helps to ensure more comprehensive exploration in the critical parameter region. Second, the exponential term is used to penalize the samples that are closer to their respective clustering centers, thus encouraging sampling dispersion and avoiding excessive redundant samples, so that the selected samples can represent both the critical regions and cover the diversity of the parameter space. The formula multiplies the geometric mean by the exponential penalty term to combine the deviation degree of each parameter with the uniqueness of the sample in the clustering, thereby generating a comprehensive representative score, enabling the sampling process to focus on the performance-sensitive regions while avoiding the bias caused by local extrema, and helping to obtain a more balanced and representative set of candidate samples.

[0095] Subsequently, the sampling probability of each sample is calculated as follows:

[0096] ;

[0097] Finally, based on the calculated sampling probabilities all candidate samples are weighted and randomly sampled, and finally a representative sample set for subsequent dual-frequency joint simulation is obtained as the input samples. After calculating the representative score the sampling probabilities are then calculated through normalization to ensure that the sum of the sampling probabilities of all candidate samples is 1, so that the weighted random sampling method can be directly used to select the final input samples. This method not only ensures that high-score samples are preferentially selected but also guarantees the reasonable coverage of the overall samples.

[0098] Step S103: Calculate the coupling compensation value based on each input sample and its corresponding antenna model response. The coupling compensation value is used to measure the mutual coupling degree between the two operating frequency bands of the dual-frequency Yagi antenna.

[0099] In step S103, based on each input sample and its corresponding antenna model response in step S102, the coupling compensation value is calculated. The coupling compensation value is a quantitative index for measuring the mutual coupling degree between the two operating frequency bands and of the dual-frequency Yagi antenna, reflecting the balance and coordination between the performances of the two frequency bands. The coupling compensation value is calculated by the following formula:

[0100] ;

[0101] Where, and respectively represent the return loss at the operating frequency bands and ; and respectively represent the antenna gain at the operating frequency bands and ; and respectively represent the voltage standing wave ratio (VSWR) at the operating frequency bands and ; is a positive constant used to prevent the denominator from being zero, is the hyperbolic tangent function, whose value range is . In this application, the positive value range is taken to ensure that the coupling compensation value falls within .

[0102] In the above formula, the numerator part represents the difference in return loss between the two frequency bands. Return loss is an important indicator to measure the antenna input matching and radiation efficiency. The greater the difference, the more serious the performance imbalance and cross-coupling problems between the two frequency bands. The first part of the denominator represents the geometric mean of the gains of the two frequency bands, reflecting the overall radiation efficiency of the antenna. The latter part multiplied at the same time represents the average voltage standing wave ratio of the two frequency bands, which is used to reflect the matching situation. When the antenna design performance is good, the denominator is large, making the ratio brought by the same return loss difference lower; on the contrary, when the performance is poor, the denominator is small, and the compensation value drops more significantly. Finally, the hyperbolic tangent function is used to perform non-linear compression on the above ratio to ensure that the output value is stably between 0 and 1 and has a smooth transition characteristic. Based on the traditional weighted summation or difference comparison, this formula introduces the geometric mean and the average voltage standing wave ratio as comprehensive performance indicators and maps them with a non-linear function. It not only retains the information of the interaction between key performance indicators but also limits the output within through normalization. In summary, the formula structure embodies the idea of "overall performance adjustment - local difference amplification", that is, when the overall performance is good, the penalty for the difference in a single frequency band is small; while when the overall performance is poor, even a small difference will cause a large change in the coupling compensation value, reflecting a more sensitive coupling effect.

[0103] Specifically, the coupling compensation value represents the coordination degree of the two frequency bands in terms of return loss, gain and voltage standing wave ratio under the current antenna design parameters. The coupling compensation value The closer the value is to 1, the more balanced and non-interfering the performance of the two frequency bands is, and the closer the antenna design is to the optimal state of dual-frequency coordination. The closer the value is to 0, the greater the performance difference or coupling interference between the two frequency bands, and compensation and optimization are required in the design. The formula comprehensively considers the echo loss difference, gain level, and voltage standing wave ratio, and can comprehensively reflect the matching and radiation performance of the antenna in the dual-frequency working state. Using the hyperbolic tangent function ensures that small differences within a certain range do not cause excessive fluctuations, while large differences can quickly reflect the imbalance between frequency bands, which is in line with the characteristics of the non-linear effect in actual antenna design. By normalizing each item and calculating the ratio, and then mapping with the exponential function or hyperbolic tangent function, it is ensured that the formula output is stable and easy to interpret, which can be comprehensively considered as an important constraint index or part of the objective function in the subsequent optimization process.

[0104] Step S104: Use the input sample as the input, and the corresponding antenna model response and coupling compensation value as the output to train a preset neural network model to obtain an antenna performance prediction model.

[0105] In step S104, several input samples are used as the input, that is, several sets of numerical antenna design parameters, including the lengths of two driven oscillators and their relative spacing, the length of the reflector and its actual distance from the driven oscillator, the number of directors, the length, and the spacing from the driven oscillator. Use the antenna model responses corresponding to several input samples and the calculated coupling compensation values as the output to train a preset neural network model to obtain an antenna performance prediction model. The trained antenna performance prediction model can output the corresponding frequency bands according to the input antenna design parameters. response, frequency band response and coupling compensation value.

[0106] Optionally, in this application, a fully connected feedforward neural network, that is, FNN, is selected as the type of the preset neural network model. Since the input is a set of numerical antenna design parameters, and the parameters are numerical features with a fixed dimension and have no local correlation in space or time, FNN is the most direct and efficient neural network architecture for processing non-linear numerical mapping, especially suitable for processing complex non-linear relationships between input features and output responses. Other types of neural network models can also be selected, and this application does not limit the type of the preset neural network model.

[0107] Step S105: Design an objective function based on the antenna model response and coupling compensation value, search for antenna design parameters and input them into the antenna performance prediction model, calculate the objective function value according to the model output, and determine whether the preset optimization standard is reached based on the objective function value to determine the optimal antenna design parameters.

[0108] In step S105, it is first necessary to construct an objective function that simultaneously considers the performance errors in two operating frequency bands and the coupling and synergy effects between the frequency bands. The objective function is as follows:

[0109] ;

[0110] Among them, is the coupling compensation value output by the antenna performance prediction model, is a positive adjustment coefficient used to enhance the penalty effect on insufficient coupling, and are the performance errors in the operating frequency bands and respectively, and the calculation methods are as follows:

[0111] ;

[0112] ;

[0113] Among them, and are the return losses output by the antenna performance prediction model in the operating frequency bands and respectively, and are the antenna gains output by the antenna performance prediction model in the operating frequency bands and respectively, and are the voltage standing wave ratios output by the antenna performance prediction model in the operating frequency bands and respectively, and are the preset return loss target values in the operating frequency bands and respectively, and are the preset antenna gain target values in the operating frequency bands and respectively, and are the preset voltage standing wave ratio target values in the operating frequency bands and respectively.

[0114] ​​​​​In the design of the above objective function, the first term represents the mean of the performance errors of the two frequency bands, which can comprehensively reflect the deviation of the antenna design parameters in each frequency band. The exponential term acts as a non-linear amplifier. When the coupling effect between the two frequency bands is good, that is, the coupling compensation value is close to 1, the exponential term tends to 1 and has little influence on the objective function. When the coupling compensation value is low, the exponential term increases rapidly, thus significantly increasing the overall objective function value. This design not only pursues the optimal performance of a single frequency band during the optimization process but also forces the responses between the two frequency bands to have high consistency and synergy.

[0115] After constructing the objective function, an existing multi-objective optimization algorithm is used to perform iterative search in the antenna design parameter space, which includes all adjustable parameters, such as the sizes and relative spacing of the two driven oscillators, the size of the reflector and its distance from the driven oscillator, as well as the number, size of the directors and their spacing from the driven oscillator. In each iteration, the candidate design parameters are input into the antenna performance prediction model trained in step S104 to obtain the corresponding frequency band response, frequency band response and coupling compensation value, and based on this, the value of the objective function is calculated. When the value of the objective function is lower than the preset optimization criterion, that is, less than or equal to the preset threshold, the performance errors in all frequency bands are within the acceptable range and the coupling compensation value is close enough to 1, it is considered to meet the requirement of dual-frequency collaborative operation, and thus the candidate design parameters are determined as the optimal antenna design parameters.

[0116] Optionally, in this application, NSGA-II or the multi-objective particle swarm algorithm is selected as the multi-objective optimization algorithm, and this application does not limit the specific type of the multi-objective optimization algorithm.

[0117] It should be noted that during the optimization process, usually more than one set of optimal design parameters will be obtained. Instead, several sets of candidate parameters will be screened out, and these parameters all meet or are close to the preset optimization criterion, thus forming an optimal solution set. Each set of design parameters in these solution sets can achieve low performance errors and high frequency band synergy in the two operating frequency bands.

[0118] In implementation, the benefits of the objective function design are as follows: First, it clearly integrates the performance deviations of the two frequency bands and the collaborative operation effect between the frequency bands. The performance is measured by the weighted root mean square error, and at the same time, the exponential penalty mechanism strengthens the penalty for insufficient coupling, ensuring that the frequency band synergy will not be sacrificed during the optimization process; Second, this design can adaptively adjust the influence of each part on the overall objective function, with high flexibility and engineering applicability. Finally, the overall objective function structure is simple and creative, which not only meets the multi-objective requirements of dual-frequency performance optimization but also ensures that the design parameters can achieve balanced collaboration in the actual operation of the antenna, providing strong theoretical support for subsequent actual verification and application.

[0119] In summary, in combination with Figure 2 , in this application, an initial antenna model that only determines the basic structure and geometric layout is first constructed. Then, through the adaptive weight hierarchical clustering sampling method, representative samples are selected in the design parameter space, and the antenna performance response data at two operating frequency bands are obtained by electromagnetic simulation. Based on the antenna performance response data, the coupling compensation value is calculated, and the key performance parameters of the two frequency bands are comprehensively mapped into a quantization index to reflect the performance balance and coordination degree between the frequency bands. Subsequently, using the input samples and their corresponding dual-frequency responses and coupling compensation values as outputs, a preset neural network model is trained to construct an antenna performance prediction model that can simultaneously predict the performance and coupling compensation value at two operating frequency bands. Finally, through a multi-objective optimization algorithm, candidate solutions are searched in the entire design parameter space, and an objective function is designed. The objective function not only considers the performance errors of the two frequency bands but also strengthens the constraint on insufficient coupling through an exponential penalty mechanism, thereby screening out one or more groups of optimal design parameters that not only meet the best performance requirements of each frequency band but also achieve good collaborative operation. Such a design not only makes up for the deficiency of traditional methods that cannot capture the complex nonlinear interactions between frequency bands but also realizes the overall optimization of the collaborative performance of the dual-frequency antenna through the introduction of the coupling compensation value, effectively solving the problem of performance imbalance caused by frequency band cross-coupling and mutual interference during dual-frequency operation in the background technology.

[0120] In addition, preferably, in combination with Figure 3 , to verify the technical effects of this application, a simulation verification test is carried out. As shown in the figure, a set of example data describes the objective function values obtained by using three different methods in 20 optimization iterations. The lower the numerical value of the objective function value, the closer the performance is to the preset target. The first one is the traditional conventional technology, that is, the prior art described in the background technology. The second one is the solution that only adopts the objective function designed in this application but does not introduce the coupling compensation value. The third one is the complete solution of this application, which includes both the design of the objective function and the introduction of the coupling compensation value. It can be seen from the data that as the number of iterations increases, the objective function values of all methods show a downward trend, but there are obvious differences in the decline speed and the final value. The optimization effect of the traditional conventional technology is the worst, and its objective function value is always relatively high. The solution that only adopts the designed objective function has achieved better improvement, but still fails to fully solve the problem of cross-interference between frequency bands. However, after introducing the coupling compensation value in the complete solution of this application, the objective function value is significantly reduced and finally reaches about 97.1, proving its obvious advantages in achieving the performance balance and collaborative operation of the two frequency bands. The data of this comparative line chart intuitively proves that the method of this application can achieve a better collaborative effect between frequency bands while maintaining low-error performance, thereby obtaining the optimal antenna design parameters that better meet the preset requirements.

[0121] Figure 4It is a structural block diagram of a design system for a dual - frequency Yagi antenna provided by an embodiment of the present application. The system at least includes the following modules:

[0122] An initial antenna model construction module, configured to construct an initial antenna model based on a predefined operating frequency band and the design requirements of a dual - frequency Yagi antenna;

[0123] An antenna design parameter simulation module, configured to select several groups of antenna design parameters as input samples in the antenna design space by using a hierarchical clustering sampling method and input them into the initial antenna model for simulation, so as to obtain the antenna model response corresponding to each input sample;

[0124] A coupling compensation value calculation module, configured to calculate a coupling compensation value based on each input sample and its corresponding antenna model response. The coupling compensation value is used to measure the mutual coupling degree between two operating frequency bands of the dual - frequency Yagi antenna;

[0125] A prediction model training module, configured to train a preset neural network model with the input samples as inputs, and the corresponding antenna model responses and coupling compensation values as outputs to obtain an antenna performance prediction model;

[0126] An antenna design parameter determination module, configured to design an objective function based on the antenna model response and the coupling compensation value, search for antenna design parameters and input them into the antenna performance prediction model, calculate the objective function value according to the model output, and determine whether the preset optimization standard is met based on the objective function value to determine the optimal antenna design parameters.

[0127] For related details, refer to the above - mentioned method embodiment.

[0128] Figure 5 It is a block diagram of an electronic device provided by an embodiment of the present application. The device at least includes a processor 401 and a memory 402.

[0129] The processor 401 may include one or more processing cores, such as a 4-core processor, an 8-core processor, etc. The processor 401 may be implemented in at least one hardware form of DSP (Digital Signal Processing), FPGA (Field-Programmable Gate Array), and PLA (Programmable Logic Array). The processor 401 may also include a main processor and a coprocessor. The main processor is a processor for processing data in the wake state, also known as the CPU (Central Processing Unit); the coprocessor is a low-power processor for processing data in the standby state. In some embodiments, the processor 401 may be integrated with a GPU (Graphics Processing Unit), and the GPU is responsible for rendering and drawing the content to be displayed on the display screen. In some embodiments, the processor 401 may further include an AI (Artificial Intelligence) processor, and the AI processor is used to process computational operations related to machine learning.

[0130] The memory 402 may include one or more computer-readable storage media, and the computer-readable storage media may be non-transitory. The memory 402 may further include high-speed random access memory and non-volatile memory, such as one or more disk storage devices and flash storage devices. In some embodiments, the non-transitory computer-readable storage media in the memory 402 is used to store at least one instruction, and the at least one instruction is used to be executed by the processor 401 to implement the design method of the dual-band Yagi antenna provided in the method embodiments of the present application.

[0131] In some embodiments, the electronic device may further optionally include: a peripheral device interface and at least one peripheral device. The processor 401, the memory 402, and the peripheral device interface may be connected through a bus or signal lines. Each peripheral device may be connected to the peripheral device interface through a bus, signal lines, or a circuit board. Schematically, the peripheral devices include but are not limited to: a radio frequency circuit, a touch display screen, an audio circuit, and a power supply, etc.

[0132] Of course, the electronic device may also include fewer or more components, and this embodiment does not limit this.

[0133] Optionally, the present application also provides a computer-readable storage medium, and a program is stored in the computer-readable storage medium, and the program is loaded and executed by the processor to implement the design method of the dual-band Yagi antenna in the above method embodiments.

[0134] Optionally, the present application also provides a computer product, which includes a computer-readable storage medium. A program is stored in the computer-readable storage medium and is loaded and executed by a processor to implement the design method of the dual-band Yagi antenna in the above method embodiment.

[0135] The technical features of the above embodiments can be combined arbitrarily. For the sake of concise description, not all possible combinations of the technical features in the above embodiments are described. However, as long as there is no contradiction in the combination of these technical features, it should be considered as the scope recorded in this specification.

[0136] The above embodiments only express several implementation manners of the present application, and the description is relatively specific and detailed, but it should not be understood as a limitation on the scope of the invention patent. It should be noted that for those of ordinary skill in the art, without departing from the concept of the present application, several modifications and improvements can still be made, and these all belong to the protection scope of the present application. Therefore, the protection scope of the patent of the present application shall be subject to the appended claims.

Claims

1. A design method for a dual-frequency Yagi antenna, characterized in that The method includes: Construct an initial antenna model based on the predefined operating frequency bands and the design requirements of the dual-band Yagi antenna; the constructing of the initial antenna model based on the predefined operating frequency bands and the design requirements of the dual-band Yagi antenna includes: defining two target operating frequency bands of the dual-band Yagi antenna, denoted respectively as and ; the initial antenna model is used to describe the basic structure of the dual-band Yagi antenna, and the dual-band Yagi antenna includes two independent driven elements, a reflector, and one or more directors; the antenna design parameters include the lengths of the two driven elements and their relative spacing, the length of the reflector and its actual distance from the driven element, the number of directors, the lengths, and the spacing between the directors and the driven element; Using a hierarchical clustering sampling method to select several groups of antenna design parameters in the antenna design space as input samples and input them into the initial antenna model for simulation to obtain the antenna model responses corresponding to each input sample; Calculating a coupling compensation value based on each input sample and its corresponding antenna model response, where the coupling compensation value is used to measure the mutual coupling degree between the two operating frequency bands of the dual-band Yagi antenna; The calculating the coupling compensation value based on each input sample and its corresponding antenna model response includes: The antenna model response includes frequency bands Response and frequency band Response, frequency band The response includes The antenna return loss, gain, and voltage standing wave ratio under the frequency band The response includes The antenna return loss, gain, and voltage standing wave ratio under it; Coupling compensation value The calculation formula is as follows: ; Among them, and respectively represent the return loss at the operating frequency bands and ; and respectively represent the antenna gain at the operating frequency bands and ; and respectively represent the voltage standing wave ratio at the operating frequency bands and ; is a positive constant, is the hyperbolic tangent function; Using the input sample as the input, its corresponding antenna model response and the coupling compensation value as the output to train a preset neural network model to obtain an antenna performance prediction model; Designing an objective function based on the antenna model response and the coupling compensation value, searching for antenna design parameters and inputting them into the antenna performance prediction model, calculating the objective function value according to the model output, and determining whether the preset optimization criterion is met based on the objective function value to determine the optimal antenna design parameters.

2. The design method of the dual-band Yagi antenna according to claim 1, characterized in that The using a hierarchical clustering sampling method to select several groups of antenna design parameters in the antenna design space as input samples and input them into the initial antenna model for simulation includes: Performing a sensitivity analysis on the antenna design parameters and hierarchically dividing the antenna design parameters based on the sensitivity weights; Constructing multi-dimensional parameter intervals based on the hierarchical results and using a clustering method to select representative samples within each multi-dimensional interval; Integrate the selected representative samples into a candidate sample set , and perform weighted random sampling based on the sampling probability to obtain input samples.

3. The design method of the dual-band Yagi antenna according to claim 2, characterized in that, The performing a sensitivity analysis on the antenna design parameters and hierarchically dividing the antenna design parameters based on the sensitivity weights includes: Based on the initial antenna model, make a small perturbation to each antenna design parameter and record the change amplitude of the return loss. For each antenna design parameter , define its sensitivity coefficient as follows: ； Among them, is the change amplitude of the return loss after perturbation, is the change amount of the antenna design parameters; Normalize the sensitivity coefficient to the sensitivity weight As follows: ； Among them, is the number of types of antenna design parameters; Hierarchically dividing the antenna design parameters according to the sensitivity weights. For the antenna design parameters with sensitivity weights greater than a preset threshold, they are identified as key parameters and their parameter intervals are divided into 10 layers. For the antenna design parameters with sensitivity weights less than or equal to the preset threshold, they are identified as secondary parameters and their parameter intervals are divided into 2 layers.

4. The design method of the dual-band Yagi antenna according to claim 2, characterized in that Integrating the selected representative samples into a candidate sample set , and performing weighted random sampling based on the sampling probability to obtain input samples including; Integrate the representative samples selected from all multi-dimensional intervals into the final candidate sample set, and calculate the representative score of each candidate sample using the following formula : ； Among them, is the value of the candidate sample in the dimension, and are the mean and standard deviation of the th parameter in the candidate sample set respectively, is the sensitivity weight of the th parameter, is the distance between the sample and its affiliated cluster center , is the scaling coefficient for adjusting the influence of the clustering distance; Calculate the sampling probability for each sample as follows: ； Based on the calculated sampling probability Perform weighted random sampling on all candidate samples, and use the obtained representative sample set as the input samples.

5. The design method of the dual-band Yagi antenna according to claim 1, characterized in that, The designing an objective function based on the antenna model response and the coupling compensation value includes: The objective function is as follows: ； Among them, is the coupling compensation value output by the antenna performance prediction model, is the adjustment coefficient, and are the performance errors under the working frequency bands and respectively, and the calculation method is as follows: ； ； Wherein, and are respectively the return losses at the operating frequency bands and output by the antenna performance prediction model. and are respectively the antenna gains at the operating frequency bands and output by the antenna performance prediction model. and are respectively the voltage standing wave ratios at the operating frequency bands and output by the antenna performance prediction model. and are respectively the preset return loss target values at the operating frequency bands and and are respectively the preset antenna gain target values at the operating frequency bands and and are respectively the preset voltage standing wave ratio target values at the operating frequency bands and ​​​ 6. A design system for a dual-frequency Yagi antenna, characterized in that, Including: An initial antenna model construction module, configured to construct an initial antenna model based on a predefined operating frequency band and the design requirements of the dual-band Yagi antenna; The construction of the initial antenna model based on the pre-defined operating frequency bands and the design requirements of the dual-band Yagi antenna includes: defining two target operating frequency bands of the dual-band Yagi antenna, denoted as and ; the initial antenna model is used to describe the basic structure of the dual-band Yagi antenna, which includes two independent driven elements, a reflector, and one or more directors; the antenna design parameters include the lengths of the two driven elements and their relative spacing, the length of the reflector and its actual distance from the driven element, the number of directors, their lengths, and their spacing from the driven element; An antenna design parameter simulation module, configured to use a hierarchical clustering sampling method to select several groups of antenna design parameters in the antenna design space as input samples and input them into the initial antenna model for simulation to obtain the antenna model responses corresponding to each input sample; A coupling compensation value calculation module, configured to calculate a coupling compensation value based on each input sample and its corresponding antenna model response, where the coupling compensation value is used to measure the mutual coupling degree between the two operating frequency bands of the dual-band Yagi antenna; The calculating the coupling compensation value based on each input sample and its corresponding antenna model response includes: The antenna model response includes frequency bands Response and frequency band Response, frequency band The response includes The antenna return loss, gain, and voltage standing wave ratio under the frequency band The response includes The antenna return loss, gain, and voltage standing wave ratio under it; Coupling compensation value The calculation formula is as follows: ； Among them, and respectively represent the return loss at the operating frequency bands and ; and respectively represent the antenna gain at the operating frequency bands and ; and respectively represent the voltage standing wave ratio at the operating frequency bands and ; is a positive constant, is the hyperbolic tangent function; A prediction model training module, configured to use the input sample as the input, its corresponding antenna model response and the coupling compensation value as the output to train a preset neural network model to obtain an antenna performance prediction model; An antenna design parameter determination module, which is used to design an objective function based on the antenna model response and the coupling compensation value, search for antenna design parameters and input them into the antenna performance prediction model, calculate the objective function value according to the model output, and determine whether the preset optimization standard is reached based on the objective function value, so as to determine the optimal antenna design parameters.

7. An electronic device, characterized in that, The device includes a processor and a memory; a program is stored in the memory, and the program is loaded and executed by the processor to implement a design method of a dual-band Yagi antenna according to any one of claims 1 to 5.

8. A computer-readable storage medium, characterized in that, A program is stored in the storage medium, and when the program is executed by a processor, it is used to implement a design method of a dual-band Yagi antenna according to any one of claims 1 to 5.

Images