Intelligent calculation method for radar three-dimensional target electromagnetic characteristics

Abstract

The invention relates to an intelligent calculation method for radar three-dimensional target electromagnetic characteristics. The method comprises the following steps: acquiring a radar three-dimensional target; discrete dipole array equivalent modeling is carried out on the radar three-dimensional target; constructing a moment information neural network model; predicting a current coefficient of the discrete dipole array obtained by modeling according to a moment information neural network model; according to an electromagnetic field superposition principle, radiation field calculation is carried out by combining a current coefficient on the basis of geometrical characteristic parameters of a discrete dipole array, and equivalent reconstruction of electromagnetic characteristics of a target space is realized. The method can improve the rapid prediction of the electromagnetic features of the specific radar target.

Description

Technical Field

[0001] The present application relates to the field of radar target imaging technology, and in particular to an intelligent calculation method for electromagnetic characteristics of three-dimensional radar targets. Background Art

[0002] In the field of radar target imaging, accurately obtaining the electromagnetic characteristics of the target is crucial to achieve high-resolution and high-precision imaging. As a computer simulation technology, electromagnetic full-wave numerical calculation plays an indispensable role in this field. It uses numerical discretization methods to solve the Maxwell equations satisfied by the target and the corresponding boundary conditions, and can accurately give the electromagnetic reflection and scattering characteristics of radar targets of different materials and shapes under electromagnetic wave irradiation. These characteristic data are the core basis for constructing radar target images, identifying and analyzing targets, and have extremely high engineering application value and academic research significance in many fields such as reconnaissance, aerospace, and civil security. However, the current radar target imaging exposes many shortcomings when using traditional electromagnetic numerical algorithms. Traditional algorithms solve the Maxwell equations based on computer discretization technology. This solution mode determines that it is inevitably limited by the scale of calculation and computing resources. When performing imaging calculations on long-distance and large-sized radar targets, due to the large number of target scatterers and the complex division of the computational grid, the amount of calculation increases exponentially, which not only consumes a lot of computing resources, but also greatly increases the computing time, seriously affecting the timeliness of imaging. When optimizing multi-band and multi-polarization imaging parameters in radar target imaging systems, and adjusting imaging algorithms for radar targets with complex structures and special materials, it is necessary to repeatedly seek the global optimal solution of imaging parameters in the multivariable domain. This requires that the electromagnetic response characteristics of the target must be recalculated every time the parameters change. When using traditional electromagnetic numerical calculation methods to participate in this optimization iteration process, the cumbersome calculation process and long calculation time have become bottlenecks that limit the overall design efficiency and imaging quality improvement of radar target imaging systems. In recent years, the rapid development of artificial intelligence (AI) technology has brought new solutions to the field of radar target imaging and promoted the rise of the emerging research direction of "intelligent electromagnetic computing". A large number of past practices have shown that AI technology has powerful data mining capabilities and can discover the intrinsic connections between massive data. In the field of radar target imaging, as long as enough electromagnetic simulation data is accumulated, AI technology can be used to mine and establish direct correlation mapping relationships between radar target geometric parameters, material properties and electromagnetic scattering characteristics, and imaging results, so as to achieve rapid prediction of specific radar target imaging characteristics. Intelligent electromagnetic computing is expected to break through the limitations of traditional electromagnetic computing that relies on numerical solutions to the Maxwell equations, get rid of the constraints of computing scale and resources on algorithms, decouple algorithms from specific imaging problems, and significantly improve the versatility, autonomy, and efficiency of imaging algorithms, providing a new path for the innovation of radar target imaging technology. However, at present, the comprehensive application of AI technology to radar target imaging still faces many difficulties. On the one hand, how to obtain high-quality, large-scale electromagnetic simulation data covering various scenarios to train AI models with excellent performance is the primary challenge. On the other hand, how to ensure the stability and accuracy of AI models in different environments and complex target conditions to avoid model overfitting still needs to be further studied. In addition, the compatibility of AI algorithms with existing radar target imaging systems also needs to be solved urgently. Summary of the invention

[0003] Based on this, it is necessary to provide an intelligent calculation method for the electromagnetic characteristics of radar three-dimensional targets to address the above technical problems and realize the rapid prediction of the imaging characteristics of specific radar targets.

[0004] An intelligent calculation method for electromagnetic characteristics of radar three-dimensional targets, the method comprising: Acquire radar three-dimensional targets; perform discrete dipole array equivalent modeling on radar three-dimensional targets; Constructing a moment information neural network model; predicting the current coefficient of the discrete dipole array obtained by modeling according to the moment information neural network model; According to the principle of electromagnetic field superposition, the radiation field of the discrete dipole array is calculated based on the geometric characteristic parameters of the discrete dipole array and the current coefficient, so as to achieve the equivalent reconstruction of the electromagnetic characteristics of the target space.

[0005] The above-mentioned intelligent calculation method for electromagnetic characteristics of radar three-dimensional targets first performs discrete dipole array equivalent modeling on the radar three-dimensional target. By utilizing the equivalent characteristics of triangulated grids and dipoles, the three-dimensional target is equivalent to a discrete dipole array, and the geometric characteristics of the target are converted into two parameters: the position of the dipole array and the normalized dipole moment, to achieve the mapping of geometric characteristics to data space and provide input data for the subsequent moment information neural network model. Secondly, a moment information neural network model is constructed. This model innovatively adopts a moment feature extraction method based on quadratic calculation, incorporates high-order statistical features such as data autocoupling / mutual coupling, makes up for the shortcomings of traditional convolutional neural networks, and combines high-order features with traditional convolutional neural networks through additive neurons, so that the model can mine more data features, enhance the generalization ability of the model, and improve its adaptability to different complex targets. Finally, the moment information neural network model is used to predict the current coefficient of the discrete dipole array, and the radiation field of the discrete dipole array is calculated based on the principle of electromagnetic field superposition, so as to achieve equivalent reconstruction of the electromagnetic characteristics of the target space, quickly obtain the electromagnetic characteristics of specific radar targets, improve computing efficiency, and meet the timeliness requirements of radar target imaging. This application incorporates the grid division information and geometric characteristics of the target into the input data; introduces the analytical formula of the dipole radiation field in the post-processing of the output data, and uses the vector synthesis characteristics of the electromagnetic field to reduce the absolute accuracy requirements for the tiny amount of the dipole current coefficient. By creating a "physics-data" dual-driven intelligent electromagnetic computing method, the model's dependence on a large amount of training data is reduced, and the prediction accuracy and generalization ability are improved. BRIEF DESCRIPTION OF THE DRAWINGS

[0006] Figure 1 A schematic diagram of a flow chart of an intelligent calculation method for electromagnetic characteristics of radar three-dimensional targets in one embodiment; Figure 2 A schematic diagram of the construction of a moment information neural network model in one embodiment. DETAILED DESCRIPTION

[0007] In order to make the purpose, technical solution and advantages of the present application more clearly understood, the present application is further described in detail below in conjunction with the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are only used to explain the present application and are not used to limit the present application.

[0008] In one embodiment, Figure 1 As shown, an intelligent calculation method for electromagnetic characteristics of radar three-dimensional targets is provided, comprising the following steps: Step 102, obtaining a radar three-dimensional target; performing discrete dipole array equivalent modeling on the radar three-dimensional target.

[0009] By utilizing the good simulation characteristics of triangulated meshes for three-dimensional targets and the equivalent characteristics of triangular meshes and dipoles, any three-dimensional radar target can be converted into a discrete dipole array for analysis. This equivalent modeling method can simplify the processing of complex three-dimensional targets and realize the transformation from geometric space to data space. Compared with traditional electromagnetic numerical calculations directly on three-dimensional targets, the computational complexity based on the discrete dipole array model is lower, reducing the demand for computing resources.

[0010] Step 104, constructing a moment information neural network model; predicting the current coefficient of the discrete dipole array obtained by modeling according to the moment information neural network model.

[0011] A moment information neural network based on quadratic calculation and convolutional neural network is designed. The proposed moment feature extraction method can effectively integrate high-order statistical features such as data autocoupling / mutual coupling. For the electromagnetic characteristics analysis of array structures, there are autocoupling / mutual coupling effects between different array elements. Traditional convolutional neural networks cannot mine these high-order features. This model integrates high-order features into traditional convolutional neural networks through additive neurons, so that the network can not only mine the first-order linear features of the data, but also mine high-order cross features, thus providing support for building a more generalized intelligent electromagnetic computing architecture, improving the model's adaptability to different scenarios and complex targets, and helping to solve the limitations of traditional convolutional neural networks in the face of complex multi-target coupling scenarios.

[0012] Step 106 , according to the electromagnetic field superposition principle, based on the discrete dipole array geometric characteristic parameters combined with the current coefficient, the radiation field of the discrete dipole array is calculated to achieve equivalent reconstruction of the electromagnetic characteristics of the target space.

[0013] The current coefficient of the discrete dipole array is predicted according to the moment information neural network model, and then the radiation field is calculated on the basis of the discrete dipole array according to the principle of electromagnetic field superposition, and finally the equivalent reconstruction of the electromagnetic characteristics of the target space is realized. In this way, the electromagnetic characteristics of the radar three-dimensional target can be quickly obtained, avoiding the complex process of solving the Maxwell equations in traditional electromagnetic numerical calculations, greatly improving the calculation efficiency, meeting the timeliness requirements of radar target imaging, and also providing an effective way to solve the problem of long calculation time in the traditional algorithm for large-scale target imaging calculations. At the same time, intelligent electromagnetic computing is an emerging electromagnetic computing method developed in the past two or three years. At present, the main application objects are still limited to two-dimensional ideal problems or three-dimensional symmetric problems, and there are few studies on arbitrary three-dimensional targets. This application cleverly uses the good simulation characteristics of the triangulated grid for arbitrary three-dimensional targets and the equivalent characteristics of the triangular grid to the dipole, and innovatively establishes an intelligent electromagnetic computing model based on a three-dimensional randomly distributed dipole array, which is expected to expand the intelligent electromagnetic computing technology from two-dimensional problems and three-dimensional symmetric problems to arbitrary three-dimensional problems. On the other hand, this application proposes a "physics-data" dual-driven intelligent electromagnetic calculation method, in which the input data incorporates the grid division information and geometric characteristics of the target, and at the same time incorporates the analytical formula of the dipole radiation field in the subsequent calculations, implicitly constraining the physical characteristics of electromagnetic far-field radiation. From the perspective of vector field synthesis, the prediction error of the tiny current term has a weak effect on the calculation result of the synthetic total field, reducing the model's stringent requirements for the absolute accuracy of the tiny term of the current coefficient. This calculation method effectively reduces the model's dependence on a large amount of training data, improves the model's prediction accuracy and generalization ability, and solves the efficiency problems of traditional methods in the optimization iteration process due to complex calculations and long time, as well as the stability and accuracy of the model under different environments and complex target conditions.

[0014] The above-mentioned intelligent calculation method for electromagnetic characteristics of radar three-dimensional targets first performs discrete dipole array equivalent modeling on radar three-dimensional targets. By utilizing the good simulation characteristics of triangulated grids for any three-dimensional targets and the equivalent characteristics of dipoles, the complex three-dimensional targets are simplified and equivalent to discrete dipole arrays, and the geometric characteristics of the targets are converted into two parameters: the position of the dipole array and the normalized dipole moment, so as to realize the mapping of geometric characteristics to data space and provide input data for the moment information neural network. Secondly, a moment information neural network model is constructed. This model innovatively adopts a moment feature extraction method based on quadratic calculation, incorporates high-order statistical features such as data autocoupling / mutual coupling, makes up for the shortcomings of traditional convolutional neural networks, and combines high-order features with traditional convolutional neural networks through additive neurons, so that the model can mine more data features, enhance the generalization ability of the model, and improve its adaptability to different complex targets. Finally, the moment information neural network model is used to predict the current coefficient of the discrete dipole array, and the radiation field of the discrete dipole array is calculated by combining the electromagnetic field superposition principle to achieve equivalent reconstruction of the electromagnetic characteristics of the target space, quickly obtain the target electromagnetic characteristics, improve computing efficiency, and meet the timeliness requirements of radar target imaging. This application incorporates the grid division information and geometric characteristics of the target into the input data; introduces the dipole radiation field analytical formula in the post-processing of the output data, and uses the vector synthesis characteristics of the electromagnetic field to reduce the absolute accuracy requirements for the tiny amount of the dipole current coefficient. By creating a "physics-data" dual-driven intelligent electromagnetic computing method, the model's dependence on a large amount of training data is reduced, and the prediction accuracy and generalization ability are improved.

[0015] In one embodiment, discrete dipole array equivalent modeling is performed on a radar three-dimensional target, including: The radar 3D target is meshed with unstructured triangular meshes using a meshing algorithm. All the obtained triangular meshes are clustered. Every two triangular meshes with common edges define a dipole. The established dipoles are numbered and recorded as , N is the sum of the numbers of all discrete dipoles; Define the position of each dipole as the center of the common edge of the pair of triangle meshes. n The position of the dipole is r n , define the normalized dipole moment of each dipole as the product of the three-dimensional vector pointing from the center of gravity of the negative film to the center of gravity of the positive film and the length of the common side, record n The normalized dipole moment of a dipole is ,in, l n Indicates n The length of the common side of the dipoles; Indicates nThe barycentric coordinates of the dipole positive / negative triangle patches.

[0016] In a specific embodiment, for a specific target, a mature meshing algorithm is used to perform unstructured triangular meshing. This process can be implemented with the help of existing software, such as Feko, Ansys ICEM CFD, etc. Then, all discrete nodes, that is, the three-dimensional space coordinates of all triangular mesh vertices are extracted and recorded. All the obtained triangular meshes are clustered, and each pair of triangular meshes with common edges can define a dipole model, and one of the triangular facets is randomly designated as a positive facet and the other triangular facet is randomly designated as a negative facet. The established dipole models are numbered in sequence and recorded as ( N is the sum of the number of all discrete dipoles). Define the position of each dipole as the center of the common edge of this pair of triangle meshes. n The position of the dipole is r n The normalized dipole moment of each dipole is defined as the product of the three-dimensional vector from the center of gravity of the negative film to the center of gravity of the positive film and the length of the common side. n The normalized dipole moment of a dipole is .in, l n Indicates n The length of the common side of the dipoles; Indicates n The centroid coordinates of the positive / negative triangle patches of the dipoles are obtained. Then, the physical structure information such as the position and normalized dipole moment of each dipole is extracted according to the node coordinates of the triangular mesh.

[0017] In one of the embodiments, constructing a moment information neural network model includes a moment feature extraction module, a convolutional feature extraction module and a fully connected layer.

[0018] In one of the embodiments, the moment feature extraction module is used to extract statistical features of the input sample data, and output them after combining through additive neurons; the convolution feature extraction module is used to extract abstract correlation features from the statistical features of the sample data through the traditional convolutional neural network architecture; the fully connected layer is used to directly correlate and map the abstract correlation features and the current coefficients for output.

[0019] In one embodiment, the statistical features include first-order moment features and second-order moment features; the moment feature extraction module is used to extract statistical features from the input sample data, including: The statistical features of the input sample data are extracted as follows: ; ; in, is the parameter to be estimated, represents the input sample data, and T represents the transposition operation.

[0020] In one embodiment, the calculation process of the parameter to be estimated includes: Establish a gradient calculation formula, solve the gradient calculation formula according to the gradient descent algorithm, and obtain the value of the parameter to be estimated; the process of establishing the gradient calculation formula includes: The loss function is recorded as L ,but L Relative to The gradient of is: ; Depend on have to, ,and , so: ; In the above formula, Obtained by the back-propagation algorithm of traditional CNN; Similarly, we can get L Relative to The gradient of is: ; In the above formula, Represents the vectorization operation rules of sample data.

[0021] In one embodiment, the current coefficient of the discrete dipole array obtained by predicting the modeling based on the moment information neural network model includes: The parameter information of the discrete dipole array obtained by modeling and the external irradiation wave information are encoded in data format and input into the moment information neural network model as input data for current coefficient prediction; the data format is: ; in, Indicates external irradiation wave information.

[0022] In a specific embodiment, the model needs to use a neural network to intelligently predict the induced current coefficient of each dipole array under external electromagnetic wave irradiation conditions based on the physical characteristic parameters of the discrete dipole array. First, the input / output data set of the model is set. The input set contains two parts of information, one is the external irradiation wave information, and the other is the discrete dipole array parameter information. The data format of the input set is: .

[0023] The above input data set contains not only the information of the incident wave’s polarization, frequency, propagation direction, etc., but also the information of the dipole array’s spatial distribution, length, polarization, etc.

[0024] For the output data set, its form is relatively simple, that is, the current coefficient of each discrete dipole, so its form is an N-dimensional vector: .

[0025] The moment information neural network model relies on the classic network model in the field of computer vision: convolutional neural network (CNN), which is obtained through improved design. Traditional convolution operations can only extract the first-order moment information of the input sample data. In the analysis of electromagnetic problems, different dipole models involve a large number of electromagnetic self-coupling and mutual coupling effects. The essence of this electromagnetic coupling effect is the quadratic term and cross-term operation between the sample data. This application integrates the second-order moment information of the sample data into the traditional CNN network model by introducing quadratic calculations, which will be more conducive to the learning of high-order statistical features of the input data. More importantly, the introduction of quadratic terms and cross terms fully analogizes the calculation rules of self-impedance elements and mutual impedance elements in the moment method, and indirectly incorporates the constraint information of electromagnetic self-coupling / mutual coupling, thereby improving the generalization ability of the network model and the prediction accuracy of the current coefficient of the discrete dipole array to a certain extent.

[0026] The architecture of the moment information neural network model is as follows Figure 2 As shown, it contains three modules: Moment feature extraction module: responsible for extracting the statistical features of the input samples, extracting statistical features including first-order moments, second-order moments, and cross terms through quadratic calculations and convolution calculations, and combining them through additive neurons.

[0027] Assume that the input sample data is a vector , then the first-order moment operation rule is: ; The second-order moment operation rule is .in, is the parameter to be estimated.

[0028] The output of the moment feature extraction module is . Let the loss function be L ,but L Relative to The gradient of is: ; Depend on have to, ,and , so: (Formula 1) In the above formula, It can be obtained by the traditional CN;N back-propagation algorithm.

[0029] Similarly, we can get L Relative to The gradient of is: (Formula 2) In the above formula, for the convenience of writing, Represents the vectorization operation rules of sample data.

[0030] After obtaining the gradient calculation formulas such as (Equation 1) and (Equation 2), the gradient descent algorithm can be used to calculate the weight parameters of the moment feature extraction module. training and learning.

[0031] Convolutional feature extraction module: extracts abstract correlation features from the statistical features of sample data through the traditional convolutional neural network architecture. Among them, the convolution layer is responsible for data feature extraction and contains a multi-layer structure; the nonlinear activation layer is implemented using the ReLU function and is responsible for the sparseness of data features; the pooling layer is responsible for data dimensionality reduction and improving feature invariance.

[0032] Fully connected layer: responsible for mapping the output nodes.

[0033] This application innovatively designs a moment information neural network based on quadratic calculation and convolutional neural network, effectively integrating high-order statistical features such as data autocoupling / mutual coupling, and providing support for building a more generalized intelligent electromagnetic computing architecture. For the electromagnetic characteristics analysis of array structures, there are a large number of autocoupling / mutual coupling effects between different array elements, which are reflected in mathematical form as operations such as second-order moments and cross terms of sample data, while traditional convolutional neural networks generally do not involve the extraction of these high-order statistical features, and are unable to mine the mutual coupling features between data. The present invention innovatively proposes a moment feature extraction method based on quadratic calculation, and integrates it into the traditional convolutional neural network through additive neurons, so that the network model can not only mine the first-order linear features of the data, but also mine the higher-order cross features, providing support for building a more generalized intelligent electromagnetic computing architecture.

[0034] In one embodiment, the radiation field calculation is performed based on the discrete dipole array according to the electromagnetic field superposition principle, including: According to electromagnetic theory, the time-harmonic dipole at the origin of the coordinate system is calculated in space. r The radiation field at the point and the single time-harmonic dipole located at the origin of the coordinate system in space r The radiation field at the point is replaced by coordinate variables to obtain the radiation field for a dipole whose spatial position is not at the origin; Based on the superposition principle of electromagnetic fields, the radiation field of the discrete dipole array is obtained by vector superposition of the radiation fields of all dipoles.

[0035] In one embodiment, the time-harmonic dipole at the origin of a coordinate system is calculated based on electromagnetic theory. r The radiation field at a point, including: According to electromagnetic theory, the time-harmonic dipole at the origin of the coordinate system is calculated in space. r The radiation field at the point is: ; in, m represents the dipole moment, represents the wave impedance of the medium, Represents the spatial wave number.

[0036] In one embodiment, based on the superposition principle of electromagnetic fields, vector superposition is performed on the radiation fields of all dipoles to obtain the radiation field of the discrete dipole array, including: Based on the superposition principle of electromagnetic fields, the radiation field of the discrete dipole array is obtained by vector superposition of the radiation fields of all dipoles: ; In the above formula, N represents the total number of dipoles, r n Indicates n The spatial position of the dipoles, E n , H n Indicates n The radiation field of a dipole in space.

[0037] In a specific embodiment, for the problem of equivalent modeling of the electromagnetic scattering / radiation characteristics of a three-dimensional target, when a discrete dipole reference array model is established based on a triangular mesh and the corresponding physical structure characteristic parameters and current coefficients are obtained, the equivalent reconstruction of the electromagnetic characteristics of the target space can be achieved.

[0038] To achieve the above problem, we must first solve the problem of accurately calculating the radiation characteristics of arbitrarily distributed dipole arrays. According to electromagnetic theory, a single time-harmonic dipole located at the origin of the coordinate system has a r The radiation field at the point is: ; In the above formula, m represents the dipole moment, represents the wave impedance of the medium, Represents the spatial wave number.

[0039] For a dipole whose spatial position is not at the origin, its radiation field can be obtained by replacing the coordinate variables in the above equation. In addition, according to the superposition principle of electromagnetic fields, the radiation field of an array composed of multiple dipoles with different polarization characteristics can be obtained by vector superposition of the radiation field of a single dipole. Now assume that the number of discrete dipole arrays is N ,Location , the dipole moment of each dipole m If all parameters are known, the radiation field in space can be calculated as follows: ; In the above formula, N represents the total number of dipoles; r n Indicates n The spatial position of the dipoles; E n , H n Indicates n The radiation field of a dipole in space.

[0040] It is not difficult to find that the above calculation process only involves simple operations of vectors and matrices (sum, product, etc.), and does not involve any time-consuming calculation processes such as iteration and inversion. Therefore, its calculation efficiency is extremely high, laying the foundation for the rapid estimation of the electromagnetic properties of any three-dimensional target.

[0041] This application is mainly used for rapid estimation of electromagnetic characteristics of radar three-dimensional targets. The grid division information of the target is integrated into the input data, and the structural physical characteristics of the calculation object are effectively integrated. On the other hand, the intelligent computing model is only used to predict the current coefficient of the dipole array. In the subsequent vector field synthesis process, the analytical formula of the dipole radiation field is integrated, which implies the physical characteristic constraints of electromagnetic far-field radiation. Moreover, from the perspective of vector field synthesis, the prediction error of the tiny current term has a weak effect on the calculation results of the synthetic total field, which effectively reduces the model's stringent requirements for the absolute accuracy of the current coefficient predicted by the network. Through the integration of the above-mentioned physical information, it is expected to reduce the model's dependence on data, establish a "physics-data" dual-driven intelligent electromagnetic computing model, and improve the accuracy and generalization ability of the model.

[0042] It should be understood that although Figure 1 The steps in the flowchart are shown in sequence as indicated by the arrows, but these steps are not necessarily executed in the order indicated by the arrows. Unless otherwise specified in this document, there is no strict order restriction for the execution of these steps, and these steps can be executed in other orders. Moreover, Figure 1At least part of the steps may include multiple sub-steps or multiple stages. These sub-steps or stages are not necessarily executed at the same time, but can be executed at different times. The execution order of these sub-steps or stages is not necessarily sequential, but can be executed in turn or alternately with other steps or at least part of the sub-steps or stages of other steps.

[0043] The technical features of the above embodiments may be combined arbitrarily. To make the description concise, not all possible combinations of the technical features in the above embodiments are described. However, as long as there is no contradiction in the combination of these technical features, they should be considered to be within the scope of this specification.

[0044] The above-described embodiments only express several implementation methods of the present application, and the descriptions thereof are relatively specific and detailed, but they cannot be understood as limiting the scope of the present application. It should be pointed out that, for a person of ordinary skill in the art, several variations and improvements can be made without departing from the concept of the present application, and these all belong to the protection scope of the present application. Therefore, the protection scope of the present application shall be subject to the attached claims.

Claims

1. An intelligent calculation method for electromagnetic characteristics of radar three-dimensional targets, characterized in that: The method comprises: Acquire a radar three-dimensional target; perform discrete dipole array equivalent modeling on the radar three-dimensional target; Constructing a moment information neural network model; predicting the current coefficient of the discrete dipole array obtained by modeling according to the moment information neural network model; According to the electromagnetic field superposition principle, the radiation field of the discrete dipole array is calculated based on the geometric characteristic parameters of the discrete dipole array and combined with the current coefficient to achieve equivalent reconstruction of the electromagnetic characteristics of the target space.

2. The method according to claim 1, characterized in that The discrete dipole array equivalent modeling is performed on the radar three-dimensional target, including: The radar three-dimensional target is meshed unstructuredly using a meshing algorithm, and all the obtained triangular meshes are clustered. Every two triangular meshes with common edges define a dipole, and the established dipoles are numbered and recorded as , N is the sum of the numbers of all discrete dipoles; Define the position of each dipole as the center of the common edge of the pair of triangle meshes. n The position of the dipole is r n , define the normalized dipole moment of each dipole as the product of the three-dimensional vector pointing from the center of gravity of the negative film to the center of gravity of the positive film and the length of the common side, record n The normalized dipole moment of a dipole is ,in, l n Indicates n The length of the common side of the dipoles; Indicates n The barycentric coordinates of the dipole positive / negative triangle patches.

3. The method according to claim 1, characterized in that The moment information neural network model comprises a moment feature extraction module, a convolution feature extraction module and a full link layer.

4. The method according to claim 3, characterized in that The moment feature extraction module is used to extract statistical features of the input sample data, and output them after combining through additive neurons; the convolution feature extraction module is used to extract abstract correlation features from the statistical features of the sample data through the traditional convolutional neural network architecture; the full connection layer is used to directly correlate the abstract correlation features and the current coefficients for mapping and output.

5. The method according to claim 4, characterized in that The statistical features include first-order moment features and second-order moment features; the moment feature extraction module is used to extract statistical features from the input sample data, including: The statistical features of the input sample data are extracted as follows: in, is the parameter to be estimated, represents the input sample data, and T represents the transposition operation.

6. The method according to claim 5, characterized in that The calculation process of the parameters to be estimated includes: A gradient calculation formula is established, and the gradient calculation formula is solved according to a gradient descent algorithm to obtain a value of the parameter to be estimated; the process of establishing the gradient calculation formula includes: The loss function is recorded as L ,but L Relative to The gradient of is: Depend on have to, ,and , so: In the above formula, Obtained by the back-propagation algorithm of traditional CNN; Similarly, we can get L Relative to The gradient of is: In the above formula, Represents the vectorization operation rules of sample data.

7. The method according to claim 2, characterized in that The current coefficient of the discrete dipole array obtained by predicting the moment information neural network model includes: The parameter information of the discrete dipole array obtained by modeling and the external irradiation wave information are encoded in data format and input into the moment information neural network model as input data to predict the current coefficient; the data format is: in, Indicates external irradiation wave information.

8. The method according to claim 1, characterized in that The radiation field calculation is performed based on the discrete dipole array according to the electromagnetic field superposition principle, including: According to electromagnetic theory, the time-harmonic dipole at the origin of the coordinate system is calculated in space. r The radiation field at the point and the single time-harmonic dipole located at the origin of the coordinate system in space r The radiation field at the point is replaced by coordinate variables to obtain the radiation field for a dipole whose spatial position is not at the origin; Based on the superposition principle of electromagnetic fields, the radiation field of the discrete dipole array is obtained by vector superposition of the radiation fields of all dipoles.

9. The method according to claim 8, characterized in that According to electromagnetic theory, the time-harmonic dipole at the origin of the coordinate system is calculated in space. r The radiation field at a point, including: According to electromagnetic theory, the time-harmonic dipole at the origin of the coordinate system is calculated in space. r The radiation field at the point is: in, m represents the dipole moment, represents the wave impedance of the medium, Represents the spatial wave number.

10. The method according to claim 8, characterized in that Based on the superposition principle of electromagnetic fields, the radiation field of the discrete dipole array is obtained by vector superposition of the radiation fields of all dipoles, including: Based on the superposition principle of electromagnetic fields, the radiation field of the discrete dipole array is obtained by vector superposition of the radiation fields of all dipoles: In the above formula, N represents the total number of dipoles, r n Indicates n The spatial position of the dipoles, E n , H n Indicates n The radiation field of a dipole in space.

Images