CN118395877B - Method, device, equipment and medium for optimally designing microstrip patch antenna - Google Patents

Abstract

The application discloses an optimization design method, a device, equipment and a medium of a microstrip patch antenna, which relate to the technical field of antenna design, and the method comprises the following steps: when antenna parameters are received, determining the parameter types corresponding to the antenna parameters; determining a forward prediction model or a reverse inversion model based on the parameter type, wherein the forward prediction model is a hybrid model; the antenna parameters are processed based on the forward prediction model or the reverse inversion model, and the target S parameter spectrum curve or the target structure parameter of the antenna is determined, so that the technical problems that the error of predicting the performance parameter curve of the antenna based on the structure parameter through experimental data in the related technology is large and the accuracy is low are effectively solved, the technical effects of shortening the antenna design period and improving the antenna design efficiency and accuracy are achieved.

Description

Optimization design method, device, equipment and medium of microstrip patch antenna

Technical Field

The present application relates to the field of antenna design technology, and in particular to an optimization design method, device, equipment and medium for a microstrip patch antenna.

Background Art

In the field of wireless communications, antennas are devices that send and receive electromagnetic waves. Their performance directly affects the quality of the entire communication system, and antenna design has attracted more and more attention. Faced with the rapid development of wireless communications, it is necessary not only to propose various new antennas that meet the design indicators, but also to provide a method that can quickly and accurately predict antenna performance or design antenna structure solutions, which undoubtedly poses a great challenge to antenna designers.

In the relevant technology, the main antenna design methods used mainly include empirical methods and numerical methods. The empirical method is mainly based on the designer's experience and previous successful cases to design antennas, which is suitable for some simple antenna structures or common application scenarios; the numerical method uses computer simulation and numerical calculation technology to numerically solve the electromagnetic field of the antenna, so as to design an antenna that meets the requirements. Common numerical methods include finite element method, time domain integral equation method, time domain finite difference method, etc. These methods are mainly used to design relatively simple antenna structures. However, these methods have limitations in terms of long design cycle and low accuracy. Based on this, with the development of deep learning technology, people have begun to apply it to the field of antenna design. For example, machine learning is used in the design of multi-band rectangular spiral microstrip antennas and rectangular microstrip patch antennas. Most of these methods are limited to obtaining the performance parameter curve of the antenna from the structural parameters.

However, these methods are limited to predicting the performance parameter curve of the antenna based on the structural parameters through experimental data, and the experimental results have large errors and low accuracy.

Summary of the invention

The embodiments of the present application provide a method, device, equipment and medium for optimizing the design of a microstrip patch antenna, thereby solving the technical problem in the related art that the performance parameter curve of the antenna is predicted based on structural parameters using experimental data with large errors and low accuracy, thereby achieving the technical effect of shortening the antenna design cycle and improving the antenna design efficiency and accuracy.

The embodiment of the present application provides an optimization design method for a microstrip patch antenna, and the optimization design method for a microstrip patch antenna includes:

Upon receiving the antenna parameter, determining a parameter type corresponding to the antenna parameter;

Determining a forward prediction model or a reverse inversion model based on the parameter type, wherein the forward prediction model is a hybrid model;

The antenna parameters are processed based on the forward prediction model or the reverse inversion model to determine a target S parameter spectrum curve or a target structural parameter of the antenna.

Optionally, when receiving the antenna parameter, the step of determining the parameter type corresponding to the antenna parameter includes:

Upon receiving an input value corresponding to an antenna parameter, determining the input value;

According to the judgment result, it is determined that the parameter type of the antenna parameter is a structural parameter or an S-parameter spectrum curve.

Optionally, the step of determining a forward prediction model or a reverse inversion model based on the parameter type, wherein the forward prediction model is a hybrid model, comprises:

When the parameter type is a structural parameter, determining that the forward prediction model is a long short-term memory model hybrid attention mechanism model;

When the parameter type is an S-parameter spectrum curve, the reverse inversion model is determined to be a converter model.

Optionally, the step of processing the antenna parameters based on the forward prediction model or the reverse inversion model to determine a target S parameter spectrum curve or a target structural parameter of the antenna includes:

Determine the target S-parameter spectrum curve of the antenna based on the forward prediction model processing structure parameters; or,

The target structural parameters of the antenna are determined by processing the S-parameter spectrum curve based on the inverse inversion model.

Optionally, the step of processing structural parameters based on the forward prediction model to determine the target S-parameter spectrum curve of the antenna includes:

The gating unit inside the long short-term memory model captures the long-term dependencies in the structural parameters;

Determining the time dependency and parameter information corresponding to the structural parameters based on the long-term dependency;

Based on the attention mechanism, weighted summation is performed on the hidden states of the long-term dependency to extract the attention features;

The target S-parameter spectrum curve of the antenna is predicted based on the time dependency, the parameter information, and the feature of interest.

Optionally, the step of processing the S-parameter spectrum curve based on the inverse inversion model to determine the target structural parameters of the antenna includes:

Based on the S-parameter spectrum curve as an input sequence of an encoder in a converter model;

Based on the self-attention mechanism, the input sequence is modeled for correlation to determine parameter features;

The parameter features are mapped to the structural parameter space of the antenna based on the output layer of the converter model to determine the target structural parameters.

Optionally, after the step of processing the antenna parameters based on the forward prediction model or the reverse inversion model to determine the target S parameter spectrum curve or target structural parameters of the antenna, the method further comprises:

Performing data dimension reduction on the antenna parameters and extracting key features;

Learning the nonlinear relationship between the structural parameters and the antenna performance based on the key features and the back propagation algorithm;

The reverse inversion model is trained based on the nonlinear relationship.

In addition, the present application also proposes an optimization design device for a microstrip patch antenna, the optimization design device for a microstrip patch antenna comprising:

A selection module, configured to determine, upon receiving antenna parameters, a parameter type corresponding to the antenna parameters; and determine a forward prediction model or a reverse inversion model based on the parameter type, wherein the forward prediction model is a hybrid model;

A forward spectrum curve prediction module, used to process structural parameters based on a forward prediction model to determine the target S-parameter spectrum curve of the antenna;

The inverse structural parameter inversion module is used to process the S parameter spectrum curve based on the inverse inversion model to determine the target structural parameters of the antenna.

In addition, the present application also proposes an optimization design device for a microstrip patch antenna, which includes a memory, a processor, and an optimization design program for a microstrip patch antenna stored in the memory and executable on the processor. When the processor executes the optimization design program for the microstrip patch antenna, the steps of the optimization design method for the microstrip patch antenna as described above are implemented.

In addition, the present application also proposes a computer-readable storage medium, on which is stored an optimization design program for a microstrip patch antenna. When the optimization design program for a microstrip patch antenna is executed by a processor, the steps of the optimization design method for a microstrip patch antenna as described above are implemented.

One or more technical solutions provided in the embodiments of the present application have at least the following technical effects or advantages:

Since the method adopts the method of determining the parameter type corresponding to the antenna parameter when the antenna parameter is received; determining the forward prediction model or the reverse inversion model based on the parameter type, wherein the forward prediction model is a hybrid model; processing the antenna parameters based on the forward prediction model or the reverse inversion model to determine the target S parameter spectrum curve or the target structural parameter of the antenna, the technical problem in the related art that the performance parameter curve of the antenna is predicted based on the structural parameters through experimental data has a large error and a low accuracy rate, and the technical effect of shortening the antenna design cycle and improving the antenna design efficiency and accuracy rate is achieved.

BRIEF DESCRIPTION OF THE DRAWINGS

FIG1 is a schematic diagram of a flow chart of a first embodiment of a method for optimizing the design of a microstrip patch antenna of the present application;

FIG2 is a flow chart of step S210 in Embodiment 2 of the optimization design method for a microstrip patch antenna of the present application;

FIG3 is a flow chart of a third embodiment of the optimization design method of the microstrip patch antenna of the present application;

FIG4 is a schematic diagram of the overall process involved in the optimization design of the microstrip patch antenna in Embodiment 3 of the optimization design method of the microstrip patch antenna of the present application;

FIG5 is a schematic diagram of the hardware structure involved in an embodiment of the optimization design device for the microstrip patch antenna of the present application.

DETAILED DESCRIPTION

In the related art, the antenna design methods mainly used include empirical methods and numerical methods. The empirical method mainly designs antennas based on the designer's experience and previous successful cases, and is suitable for some simple antenna structures or common application scenarios; the numerical method uses computer simulation and numerical calculation technology to numerically solve the electromagnetic field of the antenna, so as to design an antenna that meets the requirements. Common numerical methods include finite element method, time domain integral equation method, time domain finite difference method, etc. These methods are mainly used to design relatively simple antenna structures. However, these methods have limitations in terms of long design cycle and low accuracy. Based on this, with the development of deep learning technology, people have begun to apply it to the field of antenna design. For example, machine learning is used for the design of multi-band rectangular spiral microstrip antennas and rectangular microstrip patch antennas. Most of these methods are limited to obtaining the performance parameter curve of the antenna from the structural parameters. However, these methods are limited to predicting the performance parameter curve of the antenna based on the structural parameters through experimental data, and the experimental errors are large and the accuracy is low. The main technical solution adopted in the embodiment of the present application is: when receiving the antenna parameters, determining the parameter type corresponding to the antenna parameters;

Determining a forward prediction model or a reverse inversion model based on the parameter type, wherein the forward prediction model is a hybrid model;

The antenna parameters are processed based on the forward prediction model or the reverse inversion model to determine the target S parameter spectrum curve or target structural parameters of the antenna, thereby achieving the technical effect of shortening the antenna design cycle and improving the antenna design efficiency and accuracy.

In order to better understand the above technical solution, exemplary embodiments of the present application will be described in more detail below with reference to the accompanying drawings. Although exemplary embodiments of the present application are shown in the accompanying drawings, it should be understood that the present application can be implemented in various forms and should not be limited by the embodiments described herein. On the contrary, these embodiments are provided in order to enable a more thorough understanding of the present application and to fully convey the scope of the present application to those skilled in the art.

Embodiment 1 of the present application discloses an optimization design method for a microstrip patch antenna. Referring to FIG. 1 , the optimization design method for a microstrip patch antenna includes:

Step S110: upon receiving antenna parameters, determining parameter types corresponding to the antenna parameters.

In this embodiment, the antenna parameters are related parameters of the microstrip patch antenna, which are divided into structural parameters and spectrum curves according to the parameter type. The microstrip patch antenna is a commonly used planar antenna structure. It consists of a metal patch and a ground plane, separated by an insulating medium in the middle. The structural parameters include the size, shape and position of the patch, and the thickness of the medium. The spectrum curve of a common microstrip patch antenna can be different according to different design parameters, generally including center frequency, bandwidth and radiation characteristics. The structural parameters include: patch size: the length, width and thickness of the patch; patch shape: common shapes include rectangle, circle, ellipse, etc.; patch position: the position of the patch relative to the ground plane. The characteristics of the spectrum curve include: center frequency: the operating frequency center of the microstrip patch antenna; bandwidth: the frequency range that the antenna can cover; standing wave ratio: used to describe the matching of the input port, a lower standing wave ratio indicates better matching performance; radiation characteristics: the gain, radiation pattern, radiation power and other related parameters of the antenna in different directions.

As an optional implementation, step S110 includes: when receiving an input value corresponding to an antenna parameter, judging the input value; and determining, according to the judgment result, whether the parameter type of the antenna parameter is a structural parameter or a spectrum curve.

The structural parameters or S parameters of the microstrip patch antenna are input into the model, and the input values are judged. There are two results of the judgment. If the input value is the structural parameter of the microstrip patch antenna, a forward S parameter spectrum curve prediction is performed; otherwise, if the input is an S parameter spectrum curve, a reverse structural parameter inversion is performed. Among them, S parameters refer to scattering parameters, also known as network parameters or matrix description parameters. S parameters are used to describe the relationship between the voltage and current between the input and output of a circuit or device at a certain frequency, and are important parameters in the transmission process of electromagnetic waves. For microstrip patch antennas, common S parameters are S11 and S21, which represent the relationship between the incident wave and the reflected wave and the relationship between the incident wave and the transmitted wave, respectively. S11 represents the reflection coefficient at the antenna input port, that is, the proportion of the signal emitted from the antenna input port that is partially reflected back. S21 represents the transmission coefficient at the antenna input port, that is, the proportion of the signal emitted from the antenna input port that is transmitted to the output port through the antenna.

Step S120: determining a forward prediction model or a reverse inversion model based on the parameter type, wherein the forward prediction model is a hybrid model.

In this embodiment, the forward prediction model is used to determine the S parameter spectrum curve of the antenna as the target S parameter spectrum curve according to the input structural parameters, and the reverse inversion model is used to reversely invert the structural parameters of the antenna according to the input S parameter spectrum curve.

As an optional implementation, when the parameter type is a structural parameter, the forward prediction model is determined to be a long short-term memory model mixed attention mechanism model; when the parameter type is an S parameter spectrum curve, the reverse inversion model is determined to be a converter model.

Determine the parameter type, such as structural parameters and S-parameter spectrum curves. If the parameter type is structural parameters, the forward prediction model is a long short-term memory (LSTM) model mixed with an attention mechanism model. This model combines the LSTM model and the attention mechanism to predict the input structural parameters and output the relevant performance of the antenna. If the parameter type is an S-parameter spectrum curve, the reverse inversion model is a transformer model. This model can convert the input S-parameter spectrum curve into the structural parameters of the antenna, thereby achieving reverse inversion. Among them, the forward prediction model is an LSTM model combined with an Attention model, and the reverse inversion model is a Transformer model.

Step S130: Process the antenna parameters based on the forward prediction model or the reverse inversion model to determine a target S-parameter spectrum curve or a target structural parameter of the antenna.

In this embodiment, for the forward prediction model (long short-term memory model mixed attention mechanism model): input: known antenna structure parameters, prediction: use the hybrid model to process the input structure parameters to obtain the target S parameter spectrum curve of the antenna, output: target S parameter spectrum curve of the antenna.

For the reverse inversion model (converter model): Input: known S-parameter spectrum curve of the antenna, Inversion: use the converter model to process the input S-parameter spectrum curve to obtain the target structural parameters of the antenna, Output: target structural parameters of the antenna.

Since the method adopts the method of determining the parameter type corresponding to the antenna parameter when the antenna parameter is received; determining the forward prediction model or the reverse inversion model based on the parameter type, wherein the forward prediction model is a hybrid model; processing the antenna parameters based on the forward prediction model or the reverse inversion model to determine the target S parameter spectrum curve or the target structural parameter of the antenna, the technical problem in the related art that the performance parameter curve of the antenna is predicted based on the structural parameters through experimental data has a large error and a low accuracy rate, and the technical effect of shortening the antenna design cycle and improving the antenna design efficiency and accuracy rate is achieved.

Based on the first embodiment, the second embodiment of the present application proposes an optimization design method for a microstrip patch antenna, and step S130 includes:

Step S210, processing structural parameters based on the forward prediction model to determine the target S parameter spectrum curve of the antenna; or,

Step S220: Processing the S-parameter spectrum curve based on the inverse inversion model to determine the target structural parameters of the antenna.

In this embodiment, a known antenna structural parameter data set is collected, or the structural parameters are determined according to the input values, including information such as the size and material of the antenna, and the corresponding target S parameter spectrum curve data set. The data set is preprocessed, including data cleaning, normalization and other operations, to facilitate model training and prediction. The structural parameters of the antenna are processed using a forward prediction model (such as a long short-term memory model hybrid attention mechanism model in deep learning), and the target S parameter spectrum curve of the antenna is predicted by learning the relationship between the structural parameters and the target S parameter spectrum curve. Model evaluation and tuning are performed, and the model training is performed using a training set and a validation set, and the performance is evaluated on the test set, and the parameters and structure of the model are adjusted to obtain more accurate prediction results. A known antenna S parameter spectrum curve data set is collected, or the S parameter spectrum curve is determined according to the input value, and a corresponding antenna structural parameter data set is also required. Data preprocessing is performed, including data cleaning, normalization and other operations, to facilitate model training and inversion. The S parameter spectrum curve is processed using an inverse inversion model (such as a converter model), and the target structural parameters of the antenna are inverted by learning the relationship between the S parameter spectrum curve and the structural parameters. Perform model evaluation and tuning, use the training set and validation set to train the model, and perform performance evaluation on the test set, and adjust the model's hyperparameters and structure to obtain more accurate inversion results.

Optionally, referring to FIG. 2 , step S210 includes:

Step S211, capturing the long-term dependency in the structural parameters based on the gating unit inside the long short-term memory model;

Step S212, determining the time dependency and parameter information corresponding to the structural parameters based on the long-term dependency;

Step S213, performing weighted summation on the hidden states of the long-term dependency relationship based on the attention mechanism to extract attention features;

Step S214: predicting the target S-parameter spectrum curve of the antenna based on the time dependency, the parameter information, and the focus feature.

As an optional implementation, the structural parameter sequence is input into a long short-term memory (LSTM) model as input. The gated unit inside the LSTM model can capture the long-term dependency in the input sequence by adaptively selectively memorizing and forgetting some historical information. The LSTM model is used for training to learn the long-term dependency in the structural parameter sequence. According to the trained LSTM model, the hidden state information at each moment is obtained. The attention weight is calculated for these hidden states to capture important time dependencies. In the time dimension, the hidden state at each moment is multiplied and summed with the corresponding attention weight to obtain the attention feature. The attention mechanism is used to perform weighted summation on the hidden states to extract the attention feature. The attention mechanism can assign different weights according to the importance of the hidden state to capture the key information in the long-term dependency. The attention feature is combined with the time dependency and parameter information. The combined feature is used as input, and the target S parameter spectrum curve is predicted through a prediction model, such as a multi-layer perceptron (MLP) model. The prediction model is trained and optimized so that it can accurately predict the target S parameter spectrum curve of the antenna.

As another optional implementation, when using the LSTM+Attention model to predict the forward S-parameter spectrum curve, the LSTM network can effectively capture the long-term dependencies in the sequence data through its internal gating unit and maintain the memory of the previous information in the sequence. The LSTM network will receive the structural parameter sequence of the microstrip patch antenna as input and extract the features of these parameters through sequence learning. These features contain the time dependency and related information in the structural parameter sequence so that the model can predict the future S-parameter spectrum curve. By introducing the Attention mechanism, the model can perform weighted summation on the LSTM hidden state, thereby further extracting important features, so that more attention can be paid to relevant information during the prediction process, thereby improving the accuracy of the prediction.

Optionally, before step S210, the model is trained. When training the LSTM+Attention model, the S-parameter spectrum curve and structural parameters of the microstrip patch antenna are input into the model, wherein the input is the structural parameters of the microstrip patch antenna, and the S-parameter spectrum curve is the label; the LSTM network can effectively capture the complex correlation characteristics between the S-parameter sequence data and the structural parameters, and efficiently predict the S-parameter spectrum curve corresponding to a given microstrip patch antenna structure, with a lower mean square error than other traditional neural network algorithms.

Optionally, step S220 includes:

Step S221, using the S-parameter spectrum curve as an input sequence of an encoder in a converter model;

Step S222, performing correlation modeling on the input sequence based on a self-attention mechanism to determine parameter features;

Step S223: Mapping the parameter features to the structural parameter space of the antenna based on the output layer of the converter model to determine the target structural parameters.

As an optional implementation, based on the step of using the S-parameter spectrum curve as an input sequence of the encoder in the converter model, the S-parameter spectrum curve is converted into a sequence form as an input of the converter model. The S-parameter spectrum curve may be a vector sequence containing parameter values at different frequencies, or a time series, depending on the specific application. Based on the self-attention mechanism, the input sequence is modeled for correlation and the parameter features are determined. The input sequence is modeled using the self-attention mechanism to capture the correlation between different elements in the sequence. The self-attention mechanism can adaptively assign different weights according to the importance of each element in the input sequence to obtain the parameter features. Based on the output layer of the converter model, the parameter features are mapped to the structural parameter space of the antenna, and the step of determining the target structural parameters is to use the output layer of the converter model to map the parameter features to the structural parameter space of the antenna. The output layer can be a fully connected layer that maps the parameter features to the target structural parameters. When training the converter model, sample data with the target structural parameters need to be used for training to optimize the parameters of the output layer.

As another optional implementation, when the Transformer model is used for reverse structural parameter inversion, the main function of the encoder is feature extraction. The encoder in the Transformer model is composed of multiple layers of the same structure, each of which contains a self-attention mechanism and a feedforward neural network. In the reverse structural parameter inversion, the input is the S parameter spectrum curve of the microstrip patch antenna. The encoder in the Transformer model takes these S parameters as the input sequence and uses the self-attention mechanism to model the correlation of each element in the sequence to extract features. These features contain relevant information in the S parameter sequence and the dependencies between them. Finally, through the output layer of the Transformer model, the extracted features can be mapped to the structural parameter space of the antenna, thereby realizing reverse inversion.

Optionally, before step S220, the inverse inversion model is also included to be trained, and the training process is as follows: when training the Transformer model, the S parameter spectrum curve and structural parameters of the microstrip patch antenna are input into the model, wherein the input data is the S parameter spectrum curve and the structural parameters are labels; the Transformer model processes the structural parameter prediction task of the microstrip patch antenna in a sequence-to-sequence (seq2seq) manner, and can directly learn the complex relationship between the structural parameters from the S parameter curve; at the same time, the Transformer model introduces a global attention mechanism, so that the model can simultaneously pay attention to all positions in the S parameter spectrum curve sequence, so as to better capture the global dependency between the structural parameters, which is beneficial to explore the impact of the overall structure on the performance in the design of the microstrip patch antenna; and the self-attention mechanism in the Transformer model can dynamically adjust the weights according to the relationship between each structural parameter in the input sequence, so as to better capture the dependency between the structural parameters, which is very effective for predicting the structural parameters of the microstrip patch antenna.

In this embodiment, on the one hand, the LSTM+Attention model is used to quickly and efficiently predict the S parameter spectrum curve from the input structural parameters, which can save the time of simulation calculation by simulation software such as HFS and avoid possible operational errors in the simulation modeling process. On the other hand, the Transformer model is used to quickly and efficiently predict the antenna structure parameters from the input S parameter spectrum curve, which can save the time of trial and error inversion based on empirical methods and reduce the knowledge requirements of designers. Finally, the VAE+BPNN+SWO model is used to optimize the microstrip patch antenna in combination with the specific needs of the user, so that the designed microstrip antenna has better performance and can save a lot of iterative optimization time.

Based on the first or second embodiment, the third embodiment of the present application proposes an optimization design method for a microstrip patch antenna, referring to FIG. 3 , after step S130, including:

Step S310, performing data dimension reduction on the antenna parameters and extracting key features.

In this embodiment, the original antenna parameter data may be reduced in dimension by using common dimension reduction techniques (such as principal component analysis, feature selection, etc.) The reduced-dimensional data may contain fewer but more informative key features to reduce the complexity and computational complexity of the model.

Step S320: learning the nonlinear relationship between the structural parameters and the antenna performance based on the key features and a back propagation algorithm.

In this embodiment, the extracted key features are used as input to train a neural network model using a back propagation algorithm. The output of the neural network model is the target S parameter spectrum curve or target structural parameter of the antenna. The neural network can be a multi-layer perceptron or a reverse inversion model, i.e., a Transformer model, which learns nonlinear relationships through multiple hidden layers and activation functions.

Step S330: training the reverse inversion model based on the nonlinear relationship.

In this embodiment, the weights of the trained forward prediction model and the nonlinear relationship between the structural parameters and the antenna performance are used as the input and label of the reverse inversion model. The reverse inversion model is trained by the back propagation algorithm to learn the mapping from the antenna performance to the target structural parameters.

As an optional implementation, VAE+BPNN+SWO is used to complete the performance optimization design of the microstrip patch antenna. VAE (Variational Autoencoder) is used to learn the potential representation of the structural parameters of the microstrip patch antenna to reduce the data dimension and extract key features; BPNN (Backpropagation Neural Network) uses the low-dimensional potential representation learned by VAE to train the neural network model through the backpropagation algorithm to accurately predict the structural parameters of the microstrip patch antenna. Through the backpropagation algorithm, the model can learn the complex nonlinear relationship between the structural parameters and the performance of the microstrip patch antenna. WSO (Sine Wave Optimization) comprehensively considers multiple performance indicators of the microstrip patch antenna by weighted summation, thereby achieving multi-objective optimization in the optimization process. By adjusting the weights, a balance between different performance indicators can be achieved, resulting in a better microstrip patch antenna design.

In this embodiment, VAE (Variational Autoencoder): Variational Autoencoder is a neural network model for unsupervised learning, which consists of an encoder and a decoder. It can be used to learn and extract the potential feature representation of input data, and can be used to generate new samples similar to the original data. In the design of microstrip patch antennas, VAE can be used for dimensionality reduction and feature extraction, and learn key features from the original antenna parameters. BPNN (Backpropagation Neural Network): Backpropagation Neural Network is a commonly used machine learning algorithm that can be used to train a neural network model with multiple hidden layers. In the design of microstrip patch antennas, BPNN can be used to learn the nonlinear mapping relationship between input features and output targets, such as the relationship between learning structural parameters and antenna performance. SWO (Sine Wave Optimization): Sine wave optimization algorithm is a global optimization algorithm based on sine wave function, which is used to find the optimal solution to the optimization problem. In the design of microstrip patch antennas, the SWO algorithm can be used to search and optimize the values of antenna structural parameters to maximize antenna performance, such as maximizing efficiency or bandwidth.

Optionally, after step S130, the method further includes: comparing the mean square error (MSE) to determine whether the requirements are met, and optimizing the S parameter spectrum curve of the microstrip patch. Generally, if the S parameter of a microstrip patch antenna is below -10dB at the resonance point, it can be regarded as having good antenna performance. Further performance optimization is performed to minimize the return loss. It is set that when the return loss of the microstrip patch antenna at the resonance point is below -15dB, it is determined to meet the optimization completion condition.

Referring to FIG4 , the overall process of the optimization design method of the microstrip patch antenna is as follows: when the user inputs the structural parameters or S parameter spectrum curve of the microstrip patch antenna, when the input is the structural parameters, the forward S parameter spectrum curve prediction is performed, that is, the input structural parameters are given, and the LSTM+Attention model is used to output the S parameter spectrum curve. When the input is not the structural parameters but the S parameter spectrum curve, the Transformer model is used to output the structural parameters, and the HFS simulation is performed based on the output structural parameters to output the S parameter spectrum curve. At this time, it is determined whether it meets the requirements or reaches the maximum number of iterations. If not, the step of determining whether the input is a structural parameter is re-executed. If so, the S parameter spectrum curves output by the two, that is, the forward prediction model and the reverse inversion model, are compared, and the mean square error is used to judge the quality of the prediction. When the mean square error is within the acceptable range, it indicates that the prediction is successful, that is, no optimization is required, and the prediction result is output. If the mean square error is not within the acceptable range, it indicates that the prediction needs to be optimized. At this time, VAE+BPNN+SWO is used to optimize the performance of the predicted value to determine whether the optimization requirements are met or the maximum number of iterations is reached. If not, the optimization is re-executed. If so, the optimization result is output.

By performing data dimensionality reduction on the antenna parameters and extracting key features; learning the nonlinear relationship between the structural parameters and the antenna performance based on the key features and a back propagation algorithm; and training the inverse inversion model based on the nonlinear relationship, it is possible to determine the balance between different performance indicators based on user selection, thereby obtaining a better microstrip patch antenna design.

The present application also proposes an optimization design device for a microstrip patch antenna, referring to FIG5 , which is a schematic diagram of the structure of the optimization design device for a microstrip patch antenna in the hardware operating environment involved in the embodiment of the present application.

As shown in FIG5 , the optimization design device of the microstrip patch antenna may include: a processor 1001, such as a central processing unit (CPU), a communication bus 1002, a user interface 1003, a network interface 1004, and a memory 1005. Among them, the communication bus 1002 is used to realize the connection and communication between these components. The user interface 1003 may include a display screen (Display), an input unit such as a keyboard (Keyboard), and the optional user interface 1003 may also include a standard wired interface and a wireless interface. The network interface 1004 may optionally include a standard wired interface and a wireless interface (such as a wireless fidelity (WIreleS-FIdelity, WI-FI) interface). The memory 1005 may be a high-speed random access memory (RandomAcceS Memory, RAM) memory, or a stable non-volatile memory (Non-Volatile Memory, NVM), such as a disk memory. The memory 1005 may also be a storage device independent of the aforementioned processor 1001.

Those skilled in the art will appreciate that the structure shown in FIG. 5 does not constitute a limitation on the optimized design device for the microstrip patch antenna, and may include more or fewer components than shown, or a combination of certain components, or a different arrangement of components.

Optionally, the memory 1005 is electrically connected to the processor 1001 , and the processor 1001 can be used to control the operation of the memory 1005 , and can also read data in the memory 1005 to achieve an optimized design of the microstrip patch antenna.

Optionally, as shown in FIG. 5 , the memory 1005 as a storage medium may include an operating system, a data storage module, a network communication module, a user interface module, and an optimization design program for a microstrip patch antenna.

Optionally, in the optimization design device for the microstrip patch antenna shown in Figure 5, the network interface 1004 is mainly used for data communication with other devices; the user interface 1003 is mainly used for data interaction with the user; the processor 1001 and the memory 1005 in the optimization design device for the microstrip patch antenna of the present application can be set in the optimization design device for the microstrip patch antenna.

As shown in FIG5 , the microstrip patch antenna optimization design device calls the microstrip patch antenna optimization design program stored in the memory 1005 through the processor 1001, and performs the relevant steps of the microstrip patch antenna optimization design method provided in the embodiment of the present application:

Upon receiving the antenna parameter, determining a parameter type corresponding to the antenna parameter;

Determining a forward prediction model or a reverse inversion model based on the parameter type, wherein the forward prediction model is a hybrid model;

The antenna parameters are processed based on the forward prediction model or the reverse inversion model to determine a target S parameter spectrum curve or a target structural parameter of the antenna.

Optionally, the processor 1001 may call an optimization design program for a microstrip patch antenna stored in the memory 1005, and further perform the following operations:

Upon receiving an input value corresponding to an antenna parameter, determining the input value;

According to the judgment result, it is determined that the parameter type of the antenna parameter is a structural parameter or an S-parameter spectrum curve.

Optionally, the processor 1001 may call an optimization design program for a microstrip patch antenna stored in the memory 1005, and further perform the following operations:

When the parameter type is a structural parameter, determining that the forward prediction model is a long short-term memory model hybrid attention mechanism model;

When the parameter type is an S-parameter spectrum curve, the reverse inversion model is determined to be a converter model.

Optionally, the processor 1001 may call an optimization design program for a microstrip patch antenna stored in the memory 1005, and further perform the following operations:

Determine the target S-parameter spectrum curve of the antenna based on the forward prediction model processing structure parameters; or,

The target structural parameters of the antenna are determined by processing the S-parameter spectrum curve based on the inverse inversion model.

Optionally, the processor 1001 may call an optimization design program for a microstrip patch antenna stored in the memory 1005, and further perform the following operations:

The gating unit inside the long short-term memory model captures the long-term dependencies in the structural parameters;

Determining the time dependency and parameter information corresponding to the structural parameters based on the long-term dependency;

Based on the attention mechanism, weighted summation is performed on the hidden states of the long-term dependency to extract the attention features;

The target S-parameter spectrum curve of the antenna is predicted based on the time dependency, the parameter information, and the feature of interest.

Optionally, the processor 1001 may call an optimization design program for a microstrip patch antenna stored in the memory 1005, and further perform the following operations:

Based on the S-parameter spectrum curve as an input sequence of an encoder in a converter model;

Based on the self-attention mechanism, the input sequence is modeled for correlation to determine parameter features;

The parameter features are mapped to the structural parameter space of the antenna based on the output layer of the converter model to determine the target structural parameters.

Optionally, the processor 1001 may call an optimization design program for a microstrip patch antenna stored in the memory 1005, and further perform the following operations:

Performing data dimension reduction on the antenna parameters and extracting key features;

Learning the nonlinear relationship between the structural parameters and the antenna performance based on the key features and the back propagation algorithm;

The reverse inversion model is trained based on the nonlinear relationship.

In addition, the present application also proposes an optimization design device for a microstrip patch antenna, the optimization design device for a microstrip patch antenna comprising:

A selection module, configured to determine, upon receiving antenna parameters, a parameter type corresponding to the antenna parameters; and determine a forward prediction model or a reverse inversion model based on the parameter type, wherein the forward prediction model is a hybrid model;

A forward spectrum curve prediction module, used to process structural parameters based on a forward prediction model to determine the target S-parameter spectrum curve of the antenna;

The inverse structural parameter inversion module is used to process the S parameter spectrum curve based on the inverse inversion model to determine the target structural parameters of the antenna.

In this embodiment, the forward spectrum curve prediction module is used to predict the S parameter spectrum curve of the microstrip patch antenna with high precision and high efficiency. The reverse structural parameter inversion module is used to invert the corresponding structural parameters according to the input S parameter spectrum curve, so as to design the microstrip patch antenna structure, and compare the predicted S parameter spectrum curve with the real S parameter spectrum curve to verify whether the inverted structural parameters meet the design standards. The performance optimization module is used to optimize the performance of the microstrip patch antenna, that is, to optimize the S parameter spectrum curve of the microstrip patch. Generally, the S parameter of the microstrip patch antenna is below -10dB at the resonance point, which can be regarded as a good antenna performance. Further performance optimization is performed to make the return loss as small as possible. It is set that when the return loss of the microstrip patch antenna at the resonance point is below -15dB, it is judged to meet the conditions for the completion of optimization.

Furthermore, it also includes an accuracy assessment module for S-parameter spectrum curve prediction and structural parameter inversion, which determines whether it meets the requirements by comparing the mean square error (MSE).

In this embodiment, the LSTM+Attention model is used to quickly predict the structural parameters input by the user, and the S parameter spectrum curve obtained by HFS simulation of the group of structural parameters is compared, and the mean square error is used to judge the quality of the prediction. When the mean square error is within a certain acceptable range, it indicates that the prediction is successful. The reverse structural parameter inversion module uses the Transformer model to quickly predict the S parameter spectrum curve input by the user, and puts the predicted structural parameters into HFS for simulation, compares the S parameter spectrum curves of the two, and uses the mean square error to judge the quality of the prediction. When the mean square error is within a certain acceptable range, it indicates that the prediction is successful. The performance optimization module uses the VAE+BPNN+SWO model to optimize the performance of the structural parameters or S parameter spectrum curves that need to be optimized. For microstrip patch antennas, return loss is a major factor affecting performance evaluation. In this experiment, the return loss at the resonance point is mainly optimized to make it as small as possible. Generally, a return loss below -10dB can be regarded as good performance. When optimization is selected, the optimization target is achieved when the return loss is below -15dB, that is, the optimization is completed.

In addition, the present application also proposes a computer-readable storage medium, which stores an optimization design program for a microstrip patch antenna. When the optimization design program for the microstrip patch antenna is executed by a processor, the steps of the optimization design method for the microstrip patch antenna described in the above embodiments are implemented.

Those skilled in the art will appreciate that the embodiments of the present application may be provided as methods, systems or computer program products. Therefore, the present application may take the form of a complete hardware embodiment, a complete software embodiment, or an embodiment combining software and hardware. Moreover, the present application may take the form of a computer program product implemented on one or more computer-usable storage media (including but not limited to disk storage, CD-ROM, optical storage, etc.) containing computer-usable program codes.

The present application is described with reference to the flowcharts and/or block diagrams of the methods, devices (systems) and computer program products according to the embodiments of the present application. It should be understood that each process and/or box in the flowchart and/or block diagram, as well as the combination of the processes and/or boxes in the flowchart and/or block diagram, can be implemented by computer program instructions. These computer program instructions can be provided to a processor of a general-purpose computer, a special-purpose computer, an embedded processor or other programmable data processing device to generate a machine, so that the instructions executed by the processor of the computer or other programmable data processing device generate a device for implementing the functions specified in one or more processes in the flowchart and/or one or more boxes in the block diagram.

These computer program instructions may also be stored in a computer-readable memory that can direct a computer or other programmable data processing device to operate in a specific manner, so that the instructions stored in the computer-readable memory produce a manufactured product including an instruction device that implements the functions specified in one or more processes in the flowchart and/or one or more boxes in the block diagram.

These computer program instructions may also be loaded onto a computer or other programmable data processing device so that a series of operational steps are executed on the computer or other programmable device to produce a computer-implemented process, whereby the instructions executed on the computer or other programmable device provide steps for implementing the functions specified in one or more processes in the flowchart and/or one or more boxes in the block diagram.

It should be noted that in the claims, any reference signs placed between brackets shall not be construed as limiting the claims. The word "comprising" does not exclude the presence of components or steps not listed in the claims. The word "a" or "an" preceding a component does not exclude the presence of a plurality of such components. The present application may be implemented by means of hardware comprising several different components and by means of a suitably programmed computer. In a unit claim enumerating several means, several of these means may be embodied by the same item of hardware. The use of the words first, second, and third etc. does not indicate any order. These words may be interpreted as names.

Although the preferred embodiments of the present application have been described, those skilled in the art may make other changes and modifications to these embodiments once they have learned the basic creative concept. Therefore, the appended claims are intended to be interpreted as including the preferred embodiments and all changes and modifications falling within the scope of the present application.

Obviously, those skilled in the art can make various changes and modifications to the present application without departing from the spirit and scope of the present application. Thus, if these modifications and variations of the present application fall within the scope of the claims of the present application and their equivalents, the present application is also intended to include these modifications and variations.

Claims (7)

1. The method for optimally designing the microstrip patch antenna is characterized by comprising the following steps of:

when antenna parameters are received, determining the parameter types corresponding to the antenna parameters;

Determining a forward prediction model or a reverse inversion model based on the parameter type, wherein the forward prediction model is a hybrid model;

processing the antenna parameters based on the forward prediction model or the inverse inversion model, and determining a target S parameter spectrum curve or a target structure parameter of the antenna;

The step of determining a forward prediction model or a reverse inversion model based on the parameter type, wherein the forward prediction model is a hybrid model, comprises the following steps:

when the parameter type is a structural parameter, determining that the forward prediction model is a long-period memory model and short-period memory model mixed attention mechanism model;

When the parameter type is an S-parameter spectrum curve, determining the inverse inversion model as a converter model;

The step of processing the antenna parameters based on the forward prediction model or the inverse inversion model to determine a target S parameter spectrum curve or a target structure parameter of the antenna comprises the following steps:

Determining the target S parameter spectrum curve of the antenna based on the forward prediction model processing structure parameters; or alternatively, the first and second heat exchangers may be,

Processing an S-parameter spectrum curve based on the inverse inversion model, and determining a target structure parameter of the antenna;

The step of determining the target S parameter spectral curve or the target structural parameter of the antenna includes:

Performing data dimension reduction on the antenna parameters and extracting key features;

Learning a nonlinear relationship between the structural parameters and antenna performance based on the key features and a back propagation algorithm;

Training the inverse inversion model based on the nonlinear relationship.

2. The method for optimizing design of microstrip patch antenna according to claim 1, wherein said step of determining a type of parameter corresponding to an antenna parameter when receiving said antenna parameter comprises:

When receiving an input value corresponding to an antenna parameter, judging the input value;

And determining the parameter type of the antenna parameter as a structural parameter or an S-parameter spectrum curve according to the judging result.

3. The method of optimizing design of a microstrip patch antenna according to claim 1, wherein said step of determining said target S-parameter spectral curve of said antenna based on said forward prediction model processing structure parameters comprises:

Capturing a long-term dependency relationship in the structural parameters based on a gating unit in the long-term and short-term memory model;

determining the time dependence and parameter information corresponding to the structural parameters based on the long-term dependence;

Weighting and summing the hidden states of the long-term dependency relationship based on an attention mechanism, and extracting attention features;

Predicting the target S-parameter spectral curve of the antenna based on the time dependence, the parameter information and the feature of interest.

4. The method of optimizing design of microstrip patch antenna according to claim 1, wherein said step of determining a target structural parameter of said antenna based on said inverse inversion model processing S-parameter spectral curves comprises:

Taking the S parameter spectrum curve as an input sequence of an encoder in a converter model;

Performing relevance modeling on the input sequence based on a self-attention mechanism, and determining parameter characteristics;

the target structural parameters are determined based on mapping the parametric features to a structural parameter space of the antenna by an output layer of the converter model.

5. An apparatus for implementing the method for optimally designing a microstrip patch antenna according to claim 1, said apparatus comprising:

The selection module is used for determining the parameter type corresponding to the antenna parameter when the antenna parameter is received; determining a forward prediction model or a reverse inversion model based on the parameter type, wherein the forward prediction model is a hybrid model;

The forward spectrum curve prediction module is used for processing structural parameters based on a forward prediction model and determining the target S parameter spectrum curve of the antenna;

and the inverse structure parameter inversion module is used for processing the S-parameter spectrum curve based on the inverse inversion model and determining the target structure parameter of the antenna.

6. An apparatus for optimizing a microstrip patch antenna, comprising a memory, a processor, and an optimizing program for the microstrip patch antenna stored in the memory and operable on the processor, wherein the processor performs the steps of the optimizing method for the microstrip patch antenna according to any one of claims 1 to 4 when executing the optimizing program for the microstrip patch antenna.

7. A computer-readable storage medium, wherein the computer-readable storage medium has stored thereon an optimization design program for a microstrip patch antenna, which when executed by a processor, implements the steps of the optimization design method for a microstrip patch antenna according to any one of claims 1 to 4.