Array antenna design model training method, design method, device and storage medium

Abstract

The present invention discloses an array antenna design model training method, a design method, a device and a storage medium, and relates to the field of antenna design technology. The present invention obtains focal length parameters and array element position distribution, determines the phase compensation of each array element according to the focal length parameters and the array element position distribution, performs phase matching on the array element according to the phase compensation to determine the array element size parameter, simulates based on the focal length parameters and all array element size parameters of the transmission array antenna to obtain the performance parameters of the transmission array antenna, forms sample data according to the focal length parameters, all array element size parameters and performance parameters of the transmission array antenna, and inputs multiple sample data into the array antenna design model so that the model learns the relationship between the performance parameters and the array element size parameters and the focal length, so that the model can predict the array element size and focal length based on the performance parameters. By directly obtaining the array element size and focal length of the antenna by inputting the performance parameters into the array antenna design model, the design efficiency of the transmission array antenna is improved.

Description

Technical Field

[0001] The present invention relates to the field of antenna design technology, and in particular to an array antenna design model training method, a design method, a device and a storage medium. Background Art

[0002] Transmittance array antenna is an antenna system composed of many identical single structures arranged in a certain pattern, also known as meta-lens antenna array. The independent units that make up the antenna array are called array elements or antenna units. Transmittance array (TA) antenna is a type of array antenna, which has attracted much attention in recent years because of its easy realization of high gain, low profile and beam steering. Transmittance array antenna consists of a feed source and an array plane divided into multiple unit structures. This structure can convert the incoming spherical electromagnetic wavefront into a plane wave, thereby generating a highly directional beam in the far field. To design the plane of the transmission array antenna, it is first necessary to generate the required transmission amplitude and phase distribution based on the focal length, and then design the unit structure to meet the required transmission characteristics of the location of the unit structure. In addition, the focal length of the transmission array antenna also plays an important role in the performance of the antenna. During the design, the focal length needs to be continuously adjusted to achieve the required antenna beam shape. However, once the focal length changes, the amplitude and phase distribution of the transmission array antenna plane needs to be recalculated, and each unit needs to be redesigned and replaced. Therefore, the design of array antenna is a complex and challenging project that requires a lot of numerical simulation and expertise in related fields.

[0003] In recent years, data-driven algorithms based on machine learning have become a very powerful antenna design algorithm in antenna design. In related technologies, artificial neural networks (ANNs) are used to design the unit structure of transmission array antennas to assist in the design of array antennas, but a complex matching process between the unit and the required amplitude and phase distribution and focal length adjustment is still required, resulting in low design efficiency of transmission array antennas. Summary of the invention

[0004] The present invention aims to solve at least one of the technical problems existing in the prior art. To this end, the present invention proposes an array antenna design model training method, a design method, a device and a storage medium, which can improve the design efficiency of a transmission array antenna.

[0005] On the one hand, an embodiment of the present invention provides an array antenna design model training method, comprising the following steps:

[0006] Obtaining a focal length parameter and array element position distribution, wherein the focal length parameter represents the distance between the feed source and the transmission array antenna plane;

[0007] Determining the phase compensation of each array element according to the focal length parameter and array element position distribution;

[0008] Performing phase matching on the array elements according to the phase compensation to determine the size parameters of the array elements;

[0009] Perform simulation based on the focal length parameter and all element size parameters of the transmission array antenna to obtain performance parameters of the transmission array antenna;

[0010] Forming sample data according to the focal length parameter, all array element size parameters of the transmission array antenna and the performance parameter;

[0011] Multiple sample data are input into an array antenna design model based on a neural network architecture for training to obtain a trained array antenna design model.

[0012] According to some embodiments of the present invention, the phase compensation It is expressed as:

[0013]

[0014] Where f represents the signal frequency, r ij represents the position of the ijth array element in the transmission array antenna, F represents the focal length parameter, represents the reference phase, and c represents the speed of light.

[0015] According to some embodiments of the present invention, performing phase matching on the array element according to the phase compensation and determining the array element size parameter comprises the following steps:

[0016] Obtaining a mapping function between phase delay and array element size;

[0017] The optimization goal is to minimize the difference between the phase delay and the phase compensation of each array element, and the array element size parameter of each array element is determined in combination with the mapping function.

[0018] According to some embodiments of the present invention, the simulation based on the focal length parameter and all element size parameters of the transmission array antenna to obtain the performance parameters of the transmission array antenna includes the following steps:

[0019] The focal length parameters and all array element size parameters of the transmission array antenna are simulated using a full-wave simulation algorithm to obtain performance parameters of the transmission array antenna, wherein the performance parameters include gain, half-power beamwidth, sidelobe width and emission coefficient.

[0020] According to some embodiments of the present invention, inputting a plurality of sample data into an array antenna design model based on a neural network architecture for training to obtain a trained array antenna design model comprises the following steps:

[0021] Initializing the parameters of the array antenna design model, wherein the array antenna design model includes a forward network and a reverse network, the output of the forward network is cascaded with the input of the reverse network, the input parameters of the forward network include a focal length parameter and all array element size parameters of the transmission array antenna, and the output parameters of the forward network include performance parameters; the input parameters of the reverse network include performance parameters, and the output parameters of the reverse network include a focal length parameter and all array element size parameters of the transmission array antenna;

[0022] Inputting a plurality of the sample data into the array antenna design model for forward prediction to obtain output parameters of the forward network and output parameters of the reverse network;

[0023] Based on the model loss function, determining the model loss value according to the output parameters of the forward network and the output parameters of the reverse network;

[0024] The forward network parameters and the reverse network parameters of the array antenna design model are updated according to the model loss value to obtain a trained array antenna design model.

[0025] According to some embodiments of the present invention, the model loss function is expressed as:

[0026]

[0027] in, represents the output parameter of the forward network, represents the performance parameters in the sample data, represents the output parameter in the reverse network, Represents the focal length parameters in the sample data and all array element size parameters in the transmission array antenna.

[0028] On the other hand, an embodiment of the present invention further provides an array antenna design method, comprising the following steps:

[0029] Obtain expected performance parameters;

[0030] Inputting the expected performance parameters into an array antenna design model to obtain focal length parameters and all array element size parameters of the transmission array antenna;

[0031] The array antenna design model is obtained by training according to the array antenna design model training method described in the above embodiment.

[0032] On the other hand, an embodiment of the present invention further provides an array antenna design model training device, comprising:

[0033] The first module is used to obtain a focal length parameter and array element position distribution, wherein the focal length parameter represents the distance between the feed source and the transmission array antenna plane;

[0034] A second module is used to determine the phase compensation of each array element according to the focal length parameter and array element position distribution;

[0035] The third module is used to perform phase matching on the array element according to the phase compensation and determine the size parameters of the array element;

[0036] A fourth module is used to perform simulation based on the focal length parameter and all element size parameters of the transmission array antenna to obtain performance parameters of the transmission array antenna;

[0037] A fifth module, configured to form sample data according to the focal length parameter, all element size parameters of the transmission array antenna and the performance parameter;

[0038] The sixth module is used to input multiple sample data into the array antenna design model based on the neural network architecture for training to obtain a trained array antenna design model.

[0039] On the other hand, an embodiment of the present invention further provides an electronic device, including:

[0040] at least one processor;

[0041] at least one memory for storing at least one program;

[0042] When the at least one program is executed by the at least one processor, the at least one processor implements the array antenna design model training method or array antenna design method as described above.

[0043] On the other hand, an embodiment of the present invention further provides a computer-readable storage medium, which stores computer-executable instructions, and the computer-executable instructions are used to enable a computer to execute the array antenna design model training method or array antenna design method as described above.

[0044] The above technical solution of the present invention has at least one of the following advantages or beneficial effects: by obtaining the focal length parameter and the array element position distribution, determining the phase compensation of each array element according to the focal length parameter and the array element position distribution, performing phase matching on the array element according to the phase compensation to determine the array element size parameter, and then simulating based on the focal length parameter and all the array element size parameters of the transmission array antenna to obtain the performance parameters of the transmission array antenna, and then forming sample data based on the focal length parameter, all the array element size parameters of the transmission array antenna and the performance parameters, and inputting the sample data into the array antenna design model based on the neural network architecture so that the model learns the relationship between the performance parameters and the array element size parameters and the focal length, so that the model can predict the array element size and focal length based on the performance parameters. In the process of designing the transmission array antenna, the present application directly obtains the array element size and focal length of the transmission array antenna by inputting the performance parameters into the array antenna design model, thereby improving the design efficiency of the transmission array antenna. BRIEF DESCRIPTION OF THE DRAWINGS

[0045] Figure 1 is a flow chart of an array antenna design model training method provided by an embodiment of the present invention;

[0046] Figure 2 is a schematic diagram of a series neural network architecture provided by an embodiment of the present invention;

[0047] Figure 3 is a schematic diagram of the array antenna design model training process provided by an embodiment of the present invention;

[0048] Figure 4 is a flow chart of an array antenna design method provided by an embodiment of the present invention;

[0049] Figure 5 is a schematic diagram of the structure of an electronic device provided by an embodiment of the present invention;

[0050] Figure 6 It is a schematic diagram of the structure of a transmission array antenna according to an embodiment of the present invention. DETAILED DESCRIPTION

[0051] Embodiments of the present invention are described in detail below, examples of which are shown in the accompanying drawings, wherein the same or similar reference numerals throughout represent the same or similar components or elements having the same or similar functions. The embodiments described below with reference to the accompanying drawings are exemplary and are only used to explain the present invention, and cannot be understood as limiting the present invention.

[0052] In the description of the present invention, it should be understood that descriptions involving orientation, such as up, down, left, right, etc., the orientations or positional relationships indicated are based on the orientations or positional relationships shown in the drawings, and are only for the convenience of describing the present invention and simplifying the description, rather than indicating or implying that the device or element referred to must have a specific orientation, be constructed and operated in a specific orientation, and therefore should not be understood as a limitation on the present invention.

[0053] In the description of the present invention, if there is a description of first, second, etc., it is only for the purpose of distinguishing technical features, and cannot be understood as indicating or implying relative importance or implicitly indicating the number of the indicated technical features or implicitly indicating the sequence of the indicated technical features.

[0054] The array antenna design model training method or array antenna design method in the embodiment of the present application can be applied to a terminal or a server, or can be software running in a terminal or a server. The terminal can be a tablet computer, a laptop computer, a desktop computer, etc., but is not limited to this. The server can be an independent physical server, or a server cluster or distributed system composed of multiple physical servers, or a cloud server that provides basic cloud computing services such as cloud services, cloud databases, cloud computing, cloud functions, cloud storage, network services, cloud communications, middleware services, domain name services, security services, CDN, and big data and artificial intelligence platforms.

[0055] Reference Figure 1 The array antenna design model training method of the embodiment of the present invention includes but is not limited to step S110, step S120, step S130, step S140, step S150 and step S160.

[0056] Step S110, obtaining a focal length parameter and array element position distribution, wherein the focal length parameter represents the distance between the feed source and the transmission array antenna plane;

[0057] Step S120, determining the phase compensation of each array element according to the focal length parameter and the array element position distribution;

[0058] Step S130, performing phase matching on the array element according to the phase compensation to determine the array element size parameters;

[0059] Step S140, performing simulation based on the focal length parameter and all element size parameters of the transmission array antenna to obtain performance parameters of the transmission array antenna;

[0060] Step S150, forming sample data according to the focal length parameter, all array element size parameters and performance parameters of the transmission array antenna;

[0061] Step S160: input multiple sample data into an array antenna design model based on a neural network architecture for training to obtain a trained array antenna design model.

[0062] In step S110 of some embodiments, referring to Figure 6 , the transmission array antenna mainly includes a phase conversion plane and a feed source. The phase conversion plane is composed of multiple dielectric unit structures (i.e., array elements). The phase conversion plane can be divided into a linear array and a planar array according to the arrangement of the units. The linear array is a linear array in which the centers of each unit are arranged in a straight line in sequence with equal or unequal distances. Multiple linear arrays are arranged at a certain interval on a certain plane to form a plane array. If the centers of each unit are arranged on a sphere, a spherical array is formed. The feed source is used to radiate electromagnetic waves to the phase conversion plane. The feed source is generally located at the center of the phase conversion plane, and the distance between the feed source and the phase conversion plane is the focal length F. The array element position distribution characterizes the position of each array element in the transmission array antenna on the plane of the transmission array antenna. Exemplarily, the phase conversion plane of the transmission array antenna is formed by array elements of a 10×10 structure. The structure of each array element is a square three-dimensional column with a square hole in the middle of the square three-dimensional column. The array element can form a beam, and its performance parameters include gain, board power beam width of the radiation lobe pattern, sidelobe level and other parameters.

[0063] In step S120 of some embodiments, in order to achieve gain enhancement, it is necessary to calculate the phase compensation of each array element on the antenna surface. After the reference phase of each array element is phase compensated, the phase distribution on the array surface can be obtained. Specifically, the array element phase compensation It is expressed as:

[0064]

[0065] Where f represents the signal frequency, r ij represents the position of the ijth array element in the transmission array antenna, F represents the focal length parameter, represents the reference phase, and c represents the speed of light.

[0066] It is understandable that, in order to construct different sample data, the focal length parameter and the reference phase can be any value. For example, the focal length parameter can be randomly selected within the range of 20 mm to 40 mm, and the reference phase can be randomly selected within the range of 0 degrees to 360 degrees.

[0067] In step S130 of some embodiments, the step of performing phase matching on the array element according to the phase compensation and determining the array element size parameter includes but is not limited to the following steps:

[0068] Step 210, obtaining a mapping function between phase delay and array element size;

[0069] Step 220 , taking the minimum difference between the phase delay and the phase compensation of each array element as the optimization goal, and determining the array element size parameter of each array element in combination with the mapping function.

[0070] In step 210 of some embodiments, the size variation of the array element can provide different phase delays to match the calculated phase compensation. Specifically, the unit structure is composed of square columns with square holes, and each square column has a square hole at the center. To obtain different phase delays, the period P (i.e., the width of the square column) of the unit structure is scanned, and the scanning period ranges from 1.4 mm to 1.8 mm, with a step length of 0.5 mm; the height h of the column varies within a certain range, ranging from 3 mm to 7 mm; the width of the hole varies between 0.6 mm and P-0.4 mm, and the scanning step length is 0.05 mm; by scanning the array element sizes of the above height, column width, and hole width, a series of phase delays are obtained. Then, based on a series of array element sizes and corresponding phase delays, a mapping function of phase delay and array element size is obtained through regression analysis.

[0071] In step 220 of some embodiments, after determining the phase compensation according to the focal length parameter and the reference phase, the spacing of all units of each specific transmission array antenna of the plurality of sample data used for training the model is kept fixed, and the period of the unit structure is randomly selected. The phase matching process needs to meet the following optimization objectives:

[0072]

[0073] in, is the calculated phase compensation, is the actual phase delay from the array element.

[0074] Through the above optimization objectives, combined with the mapping function of phase delay and array element size, the array element size parameters (including array element height, hole width and column width) when the phase delay and phase compensation error of each array element are minimized can be determined. By performing phase matching on each array element of the transmission array antenna, the focal length parameters and all array element size parameters when the reference phase is constant can be obtained.

[0075] In some embodiments of step S140, the step of performing simulation based on the focal length parameter and all element size parameters of the transmission array antenna to obtain the performance parameters of the transmission array antenna includes but is not limited to the following steps:

[0076] Step S310, using a full-wave simulation algorithm to simulate the focal length parameters and all element size parameters of the transmission array antenna to obtain performance parameters of the transmission array antenna, wherein the performance parameters include gain, half-power beamwidth, sidelobe width and reflection coefficient.

[0077] In this embodiment, a full-wave simulation algorithm (FWS) is used to simulate and calculate a specific transmission array antenna structure (focal length and array element size parameters) to obtain performance parameters, and the performance parameters include but are not limited to gain (gain Gain_e in the E plane, gain Gain_h in the H plane), half-power beam width (half-power beam width HPBW_e in the E plane, half-power beam width HPBW_h in the H plane) and side lobe width (side lobe SL_e in the E plane, side lobe SL_h in the H plane) and transmission coefficient (transmission coefficient S 11 ). Sample data is obtained based on the focal length, array element size parameters and corresponding performance parameters.

[0078] It should be noted that the embodiment of the present invention needs to form multiple sets of sample data of focal length parameters and array element size parameters under the same array element position distribution, array element dielectric material and operating frequency, and train the array antenna design model through multiple sets of sample data, so that the array antenna design model can be used for the design of focal length and array element size of transmission array antenna under certain array element position distribution, array element dielectric material and operating frequency. Exemplarily, the embodiment of the present invention can obtain a series of sample data through a series of focal length parameters and corresponding array element size parameters and performance parameters when the specification is 10×10 array elements, the phase conversion plane with array element spacing D, the operating frequency is 80GHz, the dielectric constant ε is 2.66, and the loss tangent (tanδ) is 0.04.

[0079] In some embodiments of step S160, the step of inputting a plurality of sample data into an array antenna design model based on a neural network architecture for training to obtain a trained array antenna design model includes but is not limited to the following steps:

[0080] Step S410, initializing parameters of an array antenna design model, wherein the array antenna design model includes a forward network and a reverse network, the output of the forward network is cascaded with the input of the reverse network, the input parameters of the forward network include a focal length parameter and all array element size parameters of the transmission array antenna, and the output parameters of the forward network include performance parameters; the input parameters of the reverse network include performance parameters, and the output parameters of the reverse network include focal length parameters and all array element size parameters of the transmission array antenna;

[0081] Step S420, inputting a plurality of sample data into the array antenna design model for forward prediction, and obtaining output parameters of the forward network and output parameters of the reverse network;

[0082] Step S430, based on the model loss function, determining the model loss value according to the output parameters of the forward network and the output parameters of the reverse network;

[0083] Step S440, updating the forward network parameters and the reverse network parameters of the array antenna design model according to the model loss value to obtain a trained array antenna design model.

[0084] In step S410 of some embodiments, the array antenna design model may adopt a series neural network architecture, referring to Figure 2 , the tandem neural network architecture consists of two parts: the forward network (FN) and the inverse network (IN). Both the forward network and the inverse network contain 4 hidden layers (Hidden Layer), and the number of neural units in each layer ranges from dozens to hundreds, which can be adjusted according to the training data. FN predicts the electromagnetic response by taking the antenna geometry vector S as input. In the antenna geometry vector, S1 and S2 represent the period of the unit structure and the focal length of the transmission array antenna, respectively, and S3~Sn correspond to the height (h) and hole width (L) of all array elements of the array antenna. FN outputs a series of parameters, including the gain Gain_e of the E plane, the gain Gain_h of the H plane, the half-power beam width HPBW_e of the E plane, the half-power beam width HPBW_h of the H plane, the side lobe SL_e of the E plane, the side lobe SL_h of the H plane, and the emission coefficient S11. The output of FN is directly used as the input of IN to form a closely connected tandem network structure. IN outputs the corresponding unit structure period, focal length, array element height, and hole width. By exploiting the interaction between FN and IN, an automated design process for transmissive array antennas is achieved.

[0085] In some embodiments of step S430, the model loss function is expressed as:

[0086]

[0087] in, represents the output parameter of the forward network, represents the performance parameters in the sample data, represents the output parameter in the reverse network, Represents the focal length parameters in the sample data and all array element size parameters in the transmission array antenna.

[0088] Reference Figure 3, during the training of the array antenna design model, the corresponding antenna geometry vector in the sample data obtained after FWS is input into FN. FN generates the performance parameters of the antenna, which are then input into IN. The output of IN is compared with the focal length, array element size and other data in the sample data obtained after FWS to update the model parameters. This training strategy ensures mutual constraints between IN and FN during the training process, and makes full use of the training data from FWS, making the IN training process more efficient. The trained model can automatically give all the geometric parameters of the transmission array antenna, thereby quickly completing the design of the transmission array antenna, reducing the unit structure matching process in the traditional design process, and making the design of the transmission array antenna more intelligent and efficient.

[0089] Reference Figure 4 The present invention also provides an array antenna design method, comprising the following steps:

[0090] Step S510, obtaining expected performance parameters;

[0091] Step S520, inputting the expected performance parameters into the array antenna design model to obtain the focal length parameters and all array element size parameters of the transmission array antenna;

[0092] The array antenna design model is trained according to the array antenna design model training method of the above embodiment.

[0093] In this embodiment, after the array antenna design model is trained, the required performance parameters of the transmission array antenna (including the gain Gain_e of the E plane, the gain Gain_h of the H plane, the half-power beam width HPBW_e of the E plane, the half-power beam width HPBW_h of the H plane, the side lobe SL_e of the E plane, the side lobe SL_h of the H plane, and the transmission coefficient S 11 ) is used as the input of the inverse network. The inverse network can quickly give the focal length of the transmission array antenna and the parameters of all unit structures that constitute the transmission array antenna. The output parameters are used to directly design the transmission array antenna, and the transmission array antenna can exhibit the performance corresponding to the input parameters.

[0094] On the other hand, an embodiment of the present invention further provides an array antenna design model training device, comprising:

[0095] The first module is used to obtain a focal length parameter and array element position distribution, wherein the focal length parameter represents the distance between the feed source and the transmission array antenna plane;

[0096] A second module is used to determine the phase compensation of each array element according to the focal length parameter and array element position distribution;

[0097] A third module is used to perform phase matching according to the phase compensation and array element position distribution, and determine array element size parameters;

[0098] A fourth module is used to perform simulation based on the focal length parameter and all element size parameters of the transmission array antenna to obtain performance parameters of the transmission array antenna;

[0099] A fifth module, configured to form sample data according to the focal length parameter, all element size parameters of the transmission array antenna and the performance parameter;

[0100] The sixth module is used to input multiple sample data into the array antenna design model based on the neural network architecture for training to obtain a trained array antenna design model.

[0101] It can be understood that the contents of the above-mentioned array antenna design model training method embodiment are all applicable to the present system embodiment, the functions specifically implemented by the present system embodiment are the same as those in the above-mentioned array antenna design model training method embodiment, and the beneficial effects achieved are also the same as those achieved by the above-mentioned array antenna design model training method embodiment.

[0102] Reference Figure 5 , Figure 5 is a schematic diagram of an electronic device provided by an embodiment of the present invention. The electronic device of the embodiment of the present invention includes one or more control processors and a memory, Figure 5 A control processor and a memory are taken as an example.

[0103] The control processor and the memory can be connected via a bus or other means. Figure 5 The example of connecting through bus is taken in the following.

[0104] The memory, as a non-transitory computer-readable storage medium, can be used to store non-transitory software programs and non-transitory computer executable programs. In addition, the memory may include a high-speed random access memory, and may also include a non-transitory memory, such as at least one disk storage device, a flash memory device, or other non-transitory solid-state storage device. In some embodiments, the memory may optionally include a memory remotely arranged relative to the control processor, and these remote memories may be connected to the electronic device via a network. Examples of the above-mentioned network include, but are not limited to, the Internet, an intranet, a local area network, a mobile communication network, and combinations thereof.

[0105] Those skilled in the art will understand that Figure 5 The device structure shown in the figure does not constitute a limitation on the electronic device, and may include more or less components than shown in the figure, or combine certain components, or arrange the components differently.

[0106] The non-transient software programs and instructions required to implement the array antenna design model training method or array antenna design method applied to electronic devices in the above-mentioned embodiments are stored in the memory. When executed by the control processor, the array antenna design model training method or array antenna design method applied to electronic devices in the above-mentioned embodiments is executed.

[0107] In addition, an embodiment of the present invention also provides a computer-readable storage medium, which stores computer-executable instructions, and the computer-executable instructions are executed by one or more control processors, so that the one or more control processors can execute the array antenna design model training method or the array antenna design method in the above method embodiment.

[0108] It will be appreciated by those skilled in the art that all or some of the steps and systems in the disclosed method above may be implemented as software, firmware, hardware and appropriate combinations thereof. Some physical components or all physical components may be implemented as software executed by a processor, such as a central processing unit, a digital signal processor or a microprocessor, or may be implemented as hardware, or may be implemented as an integrated circuit, such as an application specific integrated circuit. Such software may be distributed on a computer-readable medium, which may include a computer storage medium (or a non-transitory medium) and a communication medium (or a temporary medium). As known to those skilled in the art, the term computer storage medium includes volatile and non-volatile, removable and non-removable media implemented in any method or technology for storing information (such as computer-readable instructions, data structures, program modules or other data). Computer storage media include, but are not limited to, RAM, ROM, EEPROM, flash memory or other memory technologies, CD-ROM, digital versatile disks (DVD) or other optical disk storage, magnetic cassettes, magnetic tapes, disk storage or other magnetic storage devices, or any other medium that may be used to store desired information and may be accessed by a computer. Furthermore, it is well known to those skilled in the art that communication media typically embodies computer readable instructions, data structures, program modules, or other data in a modulated data signal such as a carrier wave or other transport mechanism, and may include any information delivery media.

[0109] The embodiments of the present invention are described in detail above in conjunction with the accompanying drawings, but the present invention is not limited to the above embodiments, and various changes can be made within the knowledge scope of ordinary technicians in the relevant technical field without departing from the purpose of the present invention.

Claims

1. A method for training an array antenna design model, characterized in that: The following steps are involved: Obtaining a focal length parameter and array element position distribution, wherein the focal length parameter represents the distance between the feed source and the transmission array antenna plane; Determining the phase compensation of each array element according to the focal length parameter and array element position distribution; Performing phase matching on the array elements according to the phase compensation to determine the size parameters of the array elements; Perform simulation based on the focal length parameter and all element size parameters of the transmission array antenna to obtain performance parameters of the transmission array antenna; Forming sample data according to the focal length parameter, all array element size parameters of the transmission array antenna and the performance parameter; Inputting a plurality of sample data into an array antenna design model based on a neural network architecture for training, thereby obtaining a trained array antenna design model; Among them, phase compensation It is expressed as: Where f represents the signal frequency, r ij represents the position of the ijth array element in the transmission array antenna, F represents the focal length parameter, represents the reference phase, c represents the speed of light; The step of inputting a plurality of sample data into an array antenna design model based on a neural network architecture for training to obtain a trained array antenna design model comprises the following steps: Initializing the parameters of the array antenna design model, wherein the array antenna design model includes a forward network and a reverse network, the output of the forward network is cascaded with the input of the reverse network, the input parameters of the forward network include a focal length parameter and all array element size parameters of the transmission array antenna, and the output parameters of the forward network include performance parameters; the input parameters of the reverse network include performance parameters, and the output parameters of the reverse network include a focal length parameter and all array element size parameters of the transmission array antenna; Inputting a plurality of the sample data into the array antenna design model for forward prediction to obtain output parameters of the forward network and output parameters of the reverse network; Based on the model loss function, determining the model loss value according to the output parameters of the forward network and the output parameters of the reverse network; Updating the forward network parameters and the reverse network parameters of the array antenna design model according to the model loss value to obtain a trained array antenna design model; The model loss function is expressed as: in, represents the output parameter of the forward network, represents the performance parameters in the sample data, represents the output parameter in the reverse network, Represents the focal length parameters in the sample data and all array element size parameters in the transmission array antenna.

2. The array antenna design model training method according to claim 1, characterized in that The performing phase matching of the array element according to the phase compensation and determining the array element size parameters comprises the following steps: Obtaining a mapping function between phase delay and array element size; The optimization goal is to minimize the difference between the phase delay and the phase compensation of each array element, and the array element size parameter of each array element is determined in combination with the mapping function.

3. The array antenna design model training method according to claim 1, characterized in that: The simulation based on the focal length parameter and all element size parameters of the transmission array antenna to obtain the performance parameters of the transmission array antenna comprises the following steps: The focal length parameters and all array element size parameters of the transmission array antenna are simulated using a full-wave simulation algorithm to obtain performance parameters of the transmission array antenna, wherein the performance parameters include gain, half-power beamwidth, sidelobe width and emission coefficient.

4. A method for designing an array antenna, characterized in that: The following steps are involved: Obtain expected performance parameters; Inputting the expected performance parameters into an array antenna design model to obtain focal length parameters and all array element size parameters of the transmission array antenna; The array antenna design model is trained according to the array antenna design model training method according to any one of claims 1 to 3.

5. An array antenna design model training device, characterized in that: include: The first module is used to obtain a focal length parameter and array element position distribution, wherein the focal length parameter represents the distance between the feed source and the transmission array antenna plane; A second module is used to determine the phase compensation of each array element according to the focal length parameter and array element position distribution; The third module performs phase matching on the array element according to the phase compensation to determine the size parameters of the array element; A fourth module is used to perform simulation based on the focal length parameter and all element size parameters of the transmission array antenna to obtain performance parameters of the transmission array antenna; A fifth module, configured to form sample data according to the focal length parameter, all element size parameters of the transmission array antenna and the performance parameter; The sixth module is used to input multiple sample data into the array antenna design model based on the neural network architecture for training, so as to obtain a trained array antenna design model; Among them, phase compensation It is expressed as: Where f represents the signal frequency, r ij represents the position of the ijth array element in the transmission array antenna, F represents the focal length parameter, represents the reference phase, c represents the speed of light; The sixth module is specifically used to perform the following steps: Initializing the parameters of the array antenna design model, wherein the array antenna design model includes a forward network and a reverse network, the output of the forward network is cascaded with the input of the reverse network, the input parameters of the forward network include a focal length parameter and all array element size parameters of the transmission array antenna, and the output parameters of the forward network include performance parameters; the input parameters of the reverse network include performance parameters, and the output parameters of the reverse network include a focal length parameter and all array element size parameters of the transmission array antenna; Inputting a plurality of the sample data into the array antenna design model for forward prediction to obtain output parameters of the forward network and output parameters of the reverse network; Based on the model loss function, determining the model loss value according to the output parameters of the forward network and the output parameters of the reverse network; Updating the forward network parameters and the reverse network parameters of the array antenna design model according to the model loss value to obtain a trained array antenna design model; The model loss function is expressed as: in, represents the output parameter of the forward network, represents the performance parameters in the sample data, represents the output parameter in the reverse network, Represents the focal length parameters in the sample data and all array element size parameters in the transmission array antenna.

6. An electronic device, characterized in that: include: at least one processor; at least one memory for storing at least one program; When the at least one program is executed by the at least one processor, the at least one processor implements the array antenna design model training method as described in any one of claims 1 to 3 or the array antenna design method as described in claim 4.

7. A computer-readable storage medium storing a program executable by a processor, characterized in that: The program executable by the processor is used to implement the array antenna design model training method as described in any one of claims 1 to 3 or the array antenna design method as described in claim 4 when executed by the processor.

Images