CN112784467B - Programmable super-surface array coding design method for radiation field generation - Google Patents

Abstract

The invention belongs to the technical field of microwave antenna array design, and discloses a metasurface array coding design method for radiation field generation. The main steps of the method include: (1) calculating the information matrix according to the expected radiation field; (2) initializing the metasurface array coding scheme; (3) fine-tuning the coding; (4) calculating the loss function; (5) updating the coding scheme; (6) ) iteratively updates the encoding. The method solves the problem existing in the prior art that the code cannot be effectively designed to generate a near-field radiation field. The error between the field and the expected radiation field is small; the real-time speed is fast, and the coding design result can be obtained quickly by setting the appropriate threshold value, which can adapt to the application of high-speed switching of the radiation field, and has wide application prospects in the fields of electromagnetic detection and intelligent communication.

Description

A Programmable Metasurface Array Code Design Method for Radiated Field Generation

technical field

The invention belongs to the technical field of array design, and relates to a method for designing a metasurface array code to generate a specified radiation field.

Background technique

Electromagnetic metamaterials can flexibly control electromagnetic beams, thereby designing various forms of radiation fields. At present, they have been used to generate vortex electromagnetic waves, control beam patterns, etc., and have great application prospects in electromagnetic detection, intelligent communication and other fields. Two-dimensional metasurface arrays composed of electromagnetic metamaterials have many advantages that break through the performance of traditional antennas, and their structures are simple and easy to implement. Flexible and diverse regulation provides a solution for intelligent perception using electromagnetic waves. However, how to design the encoding of metasurface arrays is an urgent problem to be solved. In the field of antenna design and manufacturing, only the beam shape is usually designed, and its far-field properties are expressed in the form of beam patterns. Therefore, the current coding optimization research of metasurface arrays mainly focuses on adjusting the far-field radiation, and the research on the near-field region of the metasurface Very few, and the conventional design can only roughly calculate the corresponding code according to the single beam irradiation angle, and cannot calculate the code according to the specific distance and focus position, which is invalid for multi-beam, complex beam, and near-field situations. However, in practical applications, the target may be located in the near-field region of the metasurface, so it is necessary to consider the dynamic control of near-field electromagnetic radiation by controlling the programmable metasurface.

SUMMARY OF THE INVENTION

The technical problem to be solved by the present invention is to solve the problem in the prior art that the code cannot be effectively designed to generate a near-field radiation field, and a metasurface array code design method for radiation field generation is proposed. The design method has the advantages of flexibility and controllability, High accuracy and low computational complexity.

In order to solve the above technical problems, the technical solutions adopted are as follows:

A metasurface array coding design method for radiation field generation, the method comprising:

(1) Calculate the information matrix according to the expected radiation field y: determine the calculation plane grid according to the expected radiation field, and then calculate the information matrix H, the calculation method is as follows

where h _qn is the matrix element of the information matrix, Q is the number of array elements, q is the count index of the array element, N is the number of plane grids of the radiation field, π is the pi, f is the frequency of the transmitted signal, and t _n is the nth The electromagnetic wave propagation delay corresponding to the radiation field grid, t _q is the electromagnetic wave propagation delay corresponding to the qth array element;

(2) Initialize the metasurface array coding scheme: calculate the pseudo-inverse of the information matrix, then obtain the inverse coding design, map it to 0 or 1 according to the minimum distance rule; select a loss function to calculate the current radiation field and the expected radiation field The difference between the two, the loss function is selected as mean square error or mutual information or cross entropy, calculate the initial loss function L, c is the vector formed by the rearrangement of the 0/1 matrix formed by the array element encoding, and the calculation formula of the array encoding is:

是信息矩阵的伪逆，g(x)是一个二分函数，当自变量x≤0.5时g(x)＝0，其它情况时为g(x)＝1；in

is the pseudo-inverse of the information matrix, g(x) is a dichotomous function, g(x)=0 when the independent variable x≤0.5, and g(x)=1 in other cases;

(3) fine-tuning coding: randomly select D array elements, change its coding to obtain a new coding c _new ;

(4) Calculate the loss function: calculate the loss L _new between the current radiation field and the desired radiation field according to the loss function determined in step (2);

(5) Update coding scheme: if the loss function calculated in step (4) decreases, that is, L _new < L, then accept and update the coding c=c _new , and make the loss L=L _new ; otherwise, if L _new ≥ L, then keep the encoding method unchanged, that is

(6) Determine whether to terminate the iteration: if the current loss is less than the preset threshold L _th , that is, L<L _th , or the number of iterations exceeds the preset value _Niter , then stop the iteration and output the current coding scheme; otherwise, repeat step (3) Go to step (6).

The programmable metasurface array coding design method for radiation field generation proposed by the present invention solves the metasurface array coding through the information matrix and the expected radiation field, and then uses the loss function to optimize the coding result, which has the following beneficial effects:

The first effect is flexible and controllable, and any form of radiation field distribution can be designed;

The second effect is high accuracy, and the error between the radiation field generated according to the designed code and the expected radiation field is small;

The third effect is real-time and fast. By setting an appropriate threshold, the coding design result can be quickly obtained, which can adapt to the application of high-speed switching of radiation fields.

Description of drawings

Fig. 1 is the flow chart of the metasurface array coding design of the present invention;

Fig. 2 is the expected radiation field of embodiment 1;

Fig. 3 is the radiation field corresponding to the optimized design code of Embodiment 1;

Fig. 4 is the array element coding of the optimized design of embodiment 1;

Fig. 5 is the expected radiation field of embodiment 2;

Fig. 6 is the radiation field corresponding to the optimized design code of Embodiment 2;

FIG. 7 shows the array element coding optimized in Embodiment 2. FIG.

Detailed ways

In order to make the objectives, technical solutions and advantages of the present invention clearer, the present invention will be further described in detail below with reference to the embodiments. It should be understood that the specific embodiments described herein are only used to explain the present invention, but not to limit the present invention.

Example 1

This embodiment provides a metasurface array coding design method for radiation field generation. The metasurface antenna used in the embodiment is a 1-bit programmable metasurface array, that is, the phase change corresponding to the 0/1 state of each array element is 0 or π, the number of array elements is Q=12×12=144, the array elements are evenly distributed in the two-dimensional plane, the distance between the array elements is 0.03 meters, the frequency of the transmitted single-frequency continuous wave signal is f=5GHz, and the transmitting antenna is located at On one side of the radiation field, a reflective structure is formed. Figure 1 shows the flow of the array coding design method. The specific design steps are as follows:

(1) The desired radiation field y is shown in Fig. 2, the calculation plane grid is determined according to y, and the grid is divided into N=50×50=2500, t _n is the electromagnetic wave propagation delay corresponding to the nth radiation field grid, t _q is the electromagnetic wave propagation delay corresponding to the qth array element, calculate the information matrix

然后得到反推的编码设计，根据最小距离法则将其映射为0或者1。选择某一损失函数计算当前辐射场与期望辐射场之间的差异，损失函数选择为均方误差，即

其中H_n为H的第n行，y_n为y的第n个元素，计算初始损失函数

计算阵列编码(2) Initializing the metasurface array encoding scheme: computing the pseudo-inverse of the information matrix

Then the inverse coding design is obtained, which is mapped to 0 or 1 according to the minimum distance rule. Select a loss function to calculate the difference between the current radiation field and the expected radiation field. The loss function is selected as the mean square error, that is

where H _n is the nth row of H, y _n is the nth element of y, and calculates the initial loss function

Computational array encoding

(3) fine-tuning coding: randomly select D=1 array element, change its coding to obtain new coding c _new ;

(4) Calculate the loss function: Calculate the loss between the current radiation field and the desired radiation field according to the loss function determined in step (2).

(5) Update coding scheme: if the loss function calculated in step (4) decreases, that is, L _new < L, then accept and update the coding c=c _new , and make the loss L=L _new ; otherwise, if L _new ≥ L, then keep the encoding method unchanged;

(6) Determine whether to terminate the iteration: if the current loss is less than the preset threshold, that is, L<L _th , or the iteration exceeds a certain number of times, the iteration is stopped and the current coding scheme is output. Repeat steps (3) to (6). In this embodiment, L _th =0.05 is set, and the maximum number of iterations _Niter =200. The final optimized radiation field is shown in Figure 3, and the corresponding 0/1 code is shown in Figure 4.

Example 2

This embodiment provides a metasurface array coding design method for radiation field generation. The metasurface antenna used in the embodiment is a 1-bit programmable metasurface array, that is, the phase change corresponding to the 0/1 state of each array element is 0 or π, the number of array elements is Q=20×20=400, the array elements are evenly distributed in a two-dimensional plane, the distance between the array elements is 0.03 meters, the frequency of the transmitted single-frequency continuous wave signal is f=5GHz, and the transmitting antenna is located at On one side of the radiation field, a reflective structure is formed. The specific design steps are as follows:

(1) The desired radiation field y is shown in Fig. 5, the calculation plane grid is determined according to y, and the grid is divided into N=28×28=784, t _n is the electromagnetic wave propagation delay corresponding to the nth radiation field grid, t _q is the electromagnetic wave propagation delay corresponding to the qth array element, calculate the information matrix

然后得到反推的编码设计，根据最小距离法则将其映射为0或者1。选择某一损失函数计算当前辐射场与期望辐射场之间的差异，损失函数选择为均方误差，即

其中H_n为H的第n行，y_n为y的第n个元素，计算初始损失函数

计算阵列编码(2) Initializing the metasurface array encoding scheme: computing the pseudo-inverse of the information matrix

Then the inverse coding design is obtained, which is mapped to 0 or 1 according to the minimum distance rule. Select a loss function to calculate the difference between the current radiation field and the expected radiation field. The loss function is selected as the mean square error, that is

where H _n is the nth row of H, y _n is the nth element of y, and calculates the initial loss function

Computational array encoding

(3) fine-tuning coding: randomly select D=5 array elements, change its coding to obtain new coding c _new ;

(4) Calculate the loss function: Calculate the loss between the current radiation field and the desired radiation field according to the loss function determined in step (2).

(5) Update coding scheme: if the loss function calculated in step (4) decreases, that is, L _new < L, then accept and update the coding c=c _new , and make the loss L=L _new ; otherwise, if L _new ≥ L, then keep the encoding method unchanged;

(6) Determine whether to terminate the iteration: if the current loss is less than the preset threshold, that is, L<L _th , or the iteration exceeds a certain number of times, the iteration is stopped and the current coding scheme is output. Repeat steps (3) to (6). In this embodiment, L _th =0.05 is set, and the maximum number of iterations _Niter =600. The final optimized radiation field is shown in Figure 6, and the corresponding 0/1 code is shown in Figure 7.

Although the illustrative specific embodiments of the present invention are described above so that those skilled in the art can understand the present invention, the present invention is not limited to the scope of the specific embodiments. Such changes are within the scope defined by the appended claims, and all inventions and creations utilizing the inventive concept are included in the protection list.

Claims (1)

1. a programmable metasurface array coding design method for radiation field generation, is characterized in that, described method comprises the following steps: (1) Calculate the information matrix according to the expected radiation field y: determine the calculation plane grid according to the expected radiation field, and then calculate the information matrix H, the calculation method is as follows

where h _qn is the matrix element of the information matrix, Q is the number of array elements, q is the count index of the array element, N is the number of plane grids of the radiation field, π is the pi, f is the frequency of the transmitted signal, and t _n is the nth The electromagnetic wave propagation delay corresponding to the radiation field grid, t _q is the electromagnetic wave propagation delay corresponding to the qth array element; (2) Initialize the metasurface array coding scheme: calculate the pseudo-inverse of the information matrix, then obtain the inverse coding design, map it to 0 or 1 according to the minimum distance rule; select a loss function to calculate the current radiation field and the expected radiation field The difference between the two, the loss function is selected as mean square error or mutual information or cross entropy, calculate the initial loss function L, c is the vector formed by the rearrangement of the 0/1 matrix formed by the array element encoding, and the calculation formula of the array encoding is:

in

is the pseudo-inverse of the information matrix, g(x) is a dichotomous function, g(x)=0 when the independent variable x≤0.5, and g(x)=1 in other cases; (3) fine-tuning coding: randomly select D array elements, change its coding to obtain a new coding c _new ; (4) Calculate the loss function: calculate the loss L _new between the current radiation field and the desired radiation field according to the loss function determined in step (2); (5) Update coding scheme: if the loss function calculated in step (4) decreases, that is, L _new < L, then accept and update the coding c=c _new , and make the loss L=L _new ; otherwise, if L _new ≥ L, then keep the encoding method unchanged, that is

(6) Determine whether to terminate the iteration: if the current loss is less than the preset threshold L _th , that is, L<L _th , or the number of iterations exceeds the preset value _Niter , then stop the iteration and output the current coding scheme; otherwise, repeat step (3) Go to step (6).

Images