Method, system and computer readable medium for determining phase shift coefficient of metasurface array

Abstract

The present disclosure provides a method for determining the phase shift coefficient of a metasurface array, comprising: step S1, constructing an interference suppression model of the metasurface array, wherein the goal in the interference suppression model is to maximize the communication quality index between the target base station and the metasurface array, and the decision variable in the interference suppression model is the phase shift coefficient of each metasurface unit included in the metasurface array, and is a positive integer; step S2, solving the interference suppression model based on a particle swarm optimization algorithm to obtain a corresponding current optimal solution; in the process of solving the problem based on the particle swarm optimization algorithm, each particle in the particle swarm includes dimensions, each dimension is used to represent the phase shift coefficient of each metasurface unit, and the fitness of the particle is the communication quality index between the target base station and the metasurface array when the phase shift coefficient scheme corresponding to the particle is deployed in the metasurface array, and the current optimal solution includes the current optimal value of the phase shift coefficient of each metasurface unit.

Description

Technical Field

[0001] The present disclosure relates to the field of wireless communication technology, and in particular to a method, system and computer-readable medium for determining a phase shift coefficient of a metasurface array. Background Art

[0002] In order to continuously improve the performance of wireless communications, it is necessary to study new information technologies with lower cost, lower power consumption and higher intelligence. In the current research on the frontier of communications, metasurface technology (DMA / Dynamic Metasurface Antennas / RIS / Reconfigurable Intelligent Surface / Large Intelligent Surface / Software DefinedSurface / Metasurface / IRS / Intelligent Reflecting Surface / Reconfigurable Meta-Surfaces / Holographic MIMO, etc.) meets the above requirements. Metamaterials are a type of artificial materials whose physical properties (especially dielectric constant and magnetic permeability) can be customized and regulated according to specific needs. When metamaterials are designed and deployed in the form of planar structures, namely metasurfaces, these planar structural units, namely metasurface units, are arranged according to certain rules to form a metasurface array. By adjusting its effective parameters, the transmission, reception, reflection or refraction of electromagnetic waves can be transformed and regulated.

[0003] Dynamic Metasurface Antennas (DMA) are a new type of array antenna developed in recent years based on metamaterials. They are a special type of metasurface array. Metasurface antennas contain a large number of tunable metamaterial antenna elements that can be packaged in a small physical space and support a wide range of operating frequencies. Metasurface antennas have the ability to customize beams and can flexibly and dynamically configurably control transmitted and received signals in the analog domain through simple, low-cost system hardware. In addition, compared with traditional antenna arrays (such as antenna arrays based on patch arrays and phase shifters), communication systems based on metasurface antennas require lower power and cost, and eliminate complex feeding networks and / or active phase shifters. Metasurface antennas have excellent characteristics such as low cost, low hardware complexity, and low power consumption, which makes metasurface antennas have good prospects in the application of ultra-large-scale MIMO transceiver in 6G wireless networks. Summary of the invention

[0004] The present disclosure aims to suppress interference, improve communication quality and transmission efficiency when a target base station communicates with a metasurface array.

[0005] In a first aspect, an embodiment of the present disclosure provides a method for determining a phase shift coefficient of a metasurface array, comprising:

[0006] Step S1: construct an interference suppression model of the metasurface array, wherein the goal of the interference suppression model is to maximize the communication quality index between the target base station and the metasurface array, and the decision variables in the interference suppression model are the parameters included in the metasurface array. The phase shift coefficient of each metasurface unit, and is a positive integer;

[0007] Step S2, solving the interference suppression model based on the particle swarm optimization algorithm to obtain the corresponding current optimal solution; in the process of solving based on the particle swarm optimization algorithm, each particle in the particle swarm includes dimensions, The dimensions are used to represent The phase shift coefficient of the metasurface unit, the fitness of the particle is the communication quality index between the target base station and the metasurface array when the phase shift coefficient scheme corresponding to the particle is deployed in the metasurface array, and the current optimal solution includes The current optimal value of the phase shift coefficient of each metasurface unit.

[0008] In some embodiments, step S2 includes:

[0009] Step S201, perform the initialization phase of the particle swarm optimization algorithm, the number of iterations The initial value is 1, and the generated The particle swarm of particles is initialized and used as the original particle swarm corresponding to the first iteration. is a preset positive integer;

[0010] Step S202: Determine Whether the original particle swarm corresponding to the iterative processing satisfies the preset iteration termination condition;

[0011] If it is determined that The original particle swarm corresponding to the first iterative processing does not meet the iteration termination condition, then execute step S203; if it is determined that the first If the original particle swarm corresponding to the iteration process satisfies the iteration termination condition, step S206 is executed;

[0012] Step S203: The velocity of each particle in the original particle group corresponding to the iterative processing is updated;

[0013] Step S204: update the speed of each particle according to the updated speed of each particle. The position of the particles in the original particle group corresponding to the first iterative processing is updated, and the The initial updated particle swarm corresponding to the iterative processing;

[0014] Step S205: The initial updated particle swarm corresponding to the first iterative process determines the The original particle group corresponding to the iterative processing;

[0015] After step S205 is completed, the number of iterations Perform an increment process to update, and execute step S202 again;

[0016] Step S206: output the current global historical optimal particle to obtain the current optimal solution.

[0017] In some embodiments, the step of generating an initialization particle group including a plurality of particles in step S201 includes: a step of generating a steering vector codebook, a step of generating a Fourier transform codebook, a step of generating a random codebook, and using each codeword in the steering vector codebook, the Fourier transform codebook, and the random codebook as a particle to form the initialization particle group;

[0018] The step of generating a steering vector codebook includes:

[0019] Step S2011a, divide the preset azimuth range into equal parts to obtain target azimuth, and divide the preset pitch angle range into equal parts to obtain The target pitch angle, The target azimuth and Target pitch angle formation Different combinations of target angle pairs, and is a positive integer, and is a positive integer;

[0020] Step S2012a, for each of the target angle pairs, determining an array steering vector for characterizing the response of the metasurface array to the target angle pair, and using the array steering vector as the codeword corresponding to the target angle pair, The target angles of different combinations correspond to The barcode words constitute the steering vector codebook;

[0021] The steps of generating a Fourier transform codebook include:

[0022] Step S2011b, according to the number of array rows of the metasurface array Build dimensional complex discrete Fourier transform matrix , and the number of array columns according to the metasurface array Build dimensional complex discrete Fourier transform matrix , ;

[0023] Step S2012b: Calculate dimensional complex discrete Fourier transform matrix and dimensional complex discrete Fourier transform matrix The Kronecker tensor product of Dimensional Phase Matrix ;

[0024] Step S2013b: Dimensional Phase Matrix Each line in is extracted as a codeword, and the extracted codewords constitute the Fourier transform codebook;

[0025] The steps of generating a random codebook include:

[0026] Step S2011c: Generate by randomly generating phase states random code words, each of which includes The corresponding metasurface unit The phase shift coefficient, .

[0027] In some embodiments, step S203 includes:

[0028] Step S2031, determine The initial inertia weight of each particle in the original particle group corresponding to the first iteration processing, The first iteration of the original particle group corresponds to The initial inertia weight of each particle :

[0029]

[0030] in, and are the preset maximum and minimum inertia weights, respectively. For the The first iteration of the original particle group corresponds to The similarity between a particle and the current global historical optimal particle, and is a positive integer;

[0031] Step S2032: Based on the hyperbolic tangent function, The initial inertia weight of each particle in the original particle group corresponding to the first iterative processing is adjusted to obtain the The final inertia weight of each particle in the original particle swarm corresponding to the first iteration process, The first iteration of the original particle group corresponds to The final inertia weight of the particles :

[0032]

[0033] in, is the hyperbolic tangent function, and are two pre-set positive constants and , is the preset maximum number of iterations;

[0034] Step S2033: According to The final inertia weight of each particle in the original particle group corresponding to the first iteration is The velocity of each particle in the original particle group corresponding to the first iteration is updated. The first iteration of the original particle group corresponds to The particle The final speed after the speed update is completed in each dimension :

[0035]

[0036] in, For the The first iteration is the process before the velocity of each particle in the original particle group corresponding to the first iteration is updated. The particle The original speed of the dimension, For the particle swarm The current individual best particle in history In the The value of the dimension, is the current global historical optimal particle In the The value of the dimension, For the The first iterative process corresponds to the first particle in the original particle swarm. The particle The value of the dimension, and In the Individual learning factors and social learning factors in the iterative process, and In the In the value range of the iteration The first random number and the second random number to be selected, And it is a positive integer.

[0037] In some embodiments, Individual learning factors in the iterative process and social learning factors Satisfy the following formula:

[0038]

[0039] in, and They are the preset maximum and minimum learning factor values ​​respectively.

[0040] In some embodiments, step S205 includes:

[0041] Step S2051, calculate the The particle aggregation degree of the initial updated particle swarm corresponding to the iterative processing ;

[0042]

[0043] in, is the total number of particles in the initialized particle swarm, For the The first iteration corresponds to the initial update of the particle swarm. The similarity between a particle and the current global historical optimal particle, and is a positive integer;

[0044] Step S2052: Determine The mutation probability of each particle in the initial updated particle swarm corresponding to the first iteration processing, The first iteration corresponds to the initial update of the particle swarm. The mutation probability of a particle :

[0045]

[0046] in, For the The mutation calculation parameters corresponding to the iteration processing are: It is a constant used for preset decreasing adjustment;

[0047] Step S2053: Each particle in the initial updated particle swarm corresponding to the first iteration process The first iteration corresponds to the initial update of the particle swarm. The mutation probability of a particle with the pre-generated one located in the range Random number Compare and compare the mutation probability Less than random number Time to The first iteration corresponds to the initial update of the particle swarm. Particles are mutated.

[0048] The particle swarm obtained after step S2053 is taken as the first The original particle swarm corresponding to the iterative processing.

[0049] In some embodiments, the first The similarity between a particle and the current global historical optimal particle Determined based on the following formula:

[0050]

[0051] in, For the particle swarm The Euclidean distance between a particle and the current global historical optimal particle, and are the preset upper and lower limits of the Euclidean distance, respectively. is a pre-configured proportional constant.

[0052] In some embodiments, the communication quality indicator between the target base station and the metasurface array includes: a signal to interference plus noise ratio at the metasurface array.

[0053] In a second aspect, an embodiment of the present disclosure provides a phase shift coefficient determination system for a metasurface array, configured to implement the phase shift coefficient determination method provided in the first aspect, the phase shift coefficient determination system comprising:

[0054] A modeling module is configured to construct an interference suppression model of the metasurface array, wherein the goal in the interference suppression model is to maximize the communication quality index between the target base station and the metasurface array, and the decision variables in the interference suppression model are the The phase shift coefficient of each metasurface unit, and is a positive integer;

[0055] The solution module is configured to solve the interference suppression model based on the particle swarm optimization algorithm to obtain the corresponding current optimal solution; in the process of solving the problem based on the particle swarm optimization algorithm, each particle in the particle swarm includes dimensions, The dimensions are used to represent The phase shift coefficient of the metasurface unit, the fitness of the particle is the communication quality index between the target base station and the metasurface array when the phase shift coefficient scheme corresponding to the particle is deployed in the metasurface array, and the current optimal solution includes The current optimal value of the phase shift coefficient of each metasurface unit.

[0056] In a third aspect, an embodiment of the present disclosure further provides a computer-readable medium having a computer program stored thereon, and when the computer program is executed by a processor, the steps in the phase shift coefficient determination method provided in the first aspect are implemented.

[0057] The technical solution disclosed in the present invention can combine the strategy of maximizing the communication quality index (for example, maximizing the signal to interference plus noise ratio) without obtaining the complete channel information, and use the particle swarm optimization algorithm to determine the phase shift coefficient matrix of the metasurface array, which can effectively improve the interference suppression effect of the system. At the same time, the technical solution disclosed in the present invention has a simple algorithm, can support the continuous adjustment of the phase angle of the metasurface unit, and can also adapt to complex and frequently changing channel environments. BRIEF DESCRIPTION OF THE DRAWINGS

[0058] Figure 1 A schematic diagram of a downlink system in a communication system based on a metasurface array;

[0059] Figure 2 A schematic diagram of establishing a three-dimensional Cartesian coordinate system using the center point of the metasurface array as the origin in the present disclosure;

[0060] Figure 3 A flow chart of a method for determining a phase shift coefficient of a metasurface array provided in an embodiment of the present disclosure;

[0061] Figure 4 This is a flow chart of an optional implementation method of step S2 in the embodiment of the present disclosure;

[0062] Figure 5 This is the fitness function change diagram when the number of iterations in the simulation is 100;

[0063] Figure 6 This is the fitness function change diagram when the number of iterations in the simulation is 1000;

[0064] Figure 7 A structural block diagram of a phase shift coefficient determination system for a metasurface array provided in an embodiment of the present disclosure;

[0065] Figure 8 The figure is a schematic diagram of the structure of an electronic device according to an embodiment of the present disclosure. DETAILED DESCRIPTION

[0066] In order to enable those skilled in the art to better understand the technical solution of the present disclosure, the present disclosure is further described in detail below in conjunction with the accompanying drawings and specific implementation methods.

[0067] Unless otherwise defined, the technical terms or scientific terms used in the present disclosure should be understood by people with ordinary skills in the field to which the present disclosure belongs. The "first", "second" and similar words used in the present disclosure do not indicate any order, quantity or importance, but are only used to distinguish different components. Similarly, similar words such as "one", "one" or "the" do not indicate quantitative restrictions, but indicate that there is at least one. "Including" or "comprising" and similar words mean that the elements or objects appearing before the word cover the elements or objects listed after the word and their equivalents, without excluding other elements or objects.

[0068] In each of the accompanying drawings, identical elements are represented by similar reference numerals. For the sake of clarity, the various parts in the accompanying drawings are not all drawn to scale. In addition, some well-known parts may not be shown in the drawings.

[0069] Many specific details of the present disclosure are described below to provide a clearer understanding of the present disclosure. However, as those skilled in the art will appreciate, the present disclosure may be implemented without following these specific details.

[0070] In the related art, when adjusting the metasurface array to improve the communication quality index between the base station and the metasurface array, the adjustment based on discrete phase or the adjustment based on complete channel information is often adopted. Among them, the adjustment scheme based on discrete phase is often limited by the discreteness of the adjustable phase, which makes the accuracy insufficient and difficult to adapt to the complex and frequently changing channel environment, that is, the flexibility and accuracy steps; the adjustment scheme based on complete channel information is limited by the fact that the metasurface array itself has no information processing capability or weak information processing capability. Due to the sharp increase in the number of antennas, channel estimation becomes extremely complex and resource consumption is huge, which makes it very difficult to obtain complete channel information and difficult to implement, making it difficult to actually implement the technical solution. The related art also provides some other adjustment algorithms, but most of them have problems such as high computational complexity and insufficient adaptability.

[0071] In order to effectively improve at least one of the technical problems existing in the related art, the present disclosure provides a method for determining the phase shift coefficient of a metasurface array that does not require channel estimation and supports continuous phase adjustment.

[0072] Figure 1 Schematic diagram of the downlink system in a communication system based on a metasurface array. Figure 2 Schematic diagram of establishing a three-dimensional Cartesian coordinate system with the center point of the metasurface array as the origin in the present disclosure. Figure 1 and Figure 2 As shown, the downlink system includes: a metasurface array including metasurface units and Base stations (base stations composed of point antennas) are distributed at different locations around the metasurface array. As a specific application scenario, the base station acts as the signal transmitter and the metasurface array acts as the signal receiver. One of the base stations is used as the target base station, and the others A base station is used as an interference base station to maximize the communication quality index between the target base station and the metasurface array.

[0073] In the embodiments of the present disclosure, a communication quality indicator, the signal to interference plus noise ratio (SINR) at a metasurface array, is taken as an example for exemplary description.

[0074] Considering that the moving or rotating coordinate system has little effect on the channel model between the base station and the metasurface array, the origin of the three-dimensional Cartesian coordinate system can be set at the center point of the metasurface array without losing generality. At the same time, the channel model between the base station and the metasurface array established in the present disclosure is a simplified channel model; that is, it does not consider the multipath effect, shadow fading and other loss factors associated with the specific channel structure, does not consider the array response vector of the transmitting antenna, and assumes that there is only one unobstructed line of sight (LOS) between each base station and each metasurface unit in the metasurface array; in addition, it is assumed that the position change of the user end (also called the device end) is limited, so the channel changes slowly and the Doppler effect can be ignored.

[0075] In addition, considering the size of the metasurface array and the spacing between the metasurface units, the distance between the base station and the metasurface array can be ignored. Therefore, when considering the azimuth angles from the base station to each metasurface unit on the metasurface array, and pitch angle When , the position of the metasurface unit can be equated with the center point of the metasurface array (i.e., the origin of the three-dimensional Cartesian coordinate system), that is, the azimuth angle from the base station to each metasurface unit on the metasurface array is are equal and the pitch angle Are also equal.

[0076] Base stations The coordinates of the base stations in the three-dimensional Cartesian coordinate system are , No. The position vector of the base station is :

[0077]

[0078] Metasurface array metasurface unit (assumed to be located at the Line The position vector of the column is :

[0079]

[0080] and are the total number of rows and columns of the metasurface array, and are the horizontal spacing and vertical spacing of the super surface units, respectively. , and is a positive integer, And it is a positive integer.

[0081] Base stations Channel from base station (signal transmitter) to metasurface array (signal receiver) It can be modeled as:

[0082]

[0083] in, For the The channel coefficients of the channels from the base stations to the metasurface array, For the The complex amplitude of the transmission path of each base station, It is The steering vector from the base station to the metasurface array, For the The azimuth angle from the base station to the metasurface array (i.e., The azimuth angles from the base station to each metasurface unit in the metasurface array are ), For the The elevation angle from the base station to the metasurface array (i.e., The pitch angles from the base station to each metasurface unit in the metasurface array are ), is the frequency of the signal carrier, is the wavelength of the signal carrier, is the speed of light, is an imaginary unit, For the The delay from the base station to the metasurface array, is a natural constant, Represents the transpose of a matrix.

[0084] At this time, Base stations base stations are the target base stations, and the others For example, when the target base station is communicating with the metasurface array, the signal to interference plus noise ratio at the metasurface array is It can be expressed as:

[0085]

[0086] Reason The phase shift coefficients corresponding to the metasurface units constitute a dimensional matrix (also called the phase shift coefficient matrix of the metasurface array, The phase shift coefficient of each metasurface unit is , express The phase angle of each metasurface unit, Additive white Gaussian noise power at the metasurface array.

[0087] Based on the above, it can be seen that by adjusting the phase shift coefficient matrix of the metasurface array , which can realize the signal to interference plus noise ratio at the metasurface array when the target base station communicates with the metasurface array Therefore, maximizing The problem can be regarded as a process of determining an optimal phase shift coefficient matrix.

[0088] Figure 3 A flow chart of a method for determining a phase shift coefficient of a metasurface array provided in an embodiment of the present disclosure. Figure 3 The phase shift coefficient determination method includes:

[0089] Step S1, constructing an interference suppression model of a metasurface array.

[0090] The goal of the interference suppression model is to maximize the communication quality index between the target base station and the metasurface array, and the decision variables in the interference suppression model are the The phase shift coefficient of each metasurface unit, And it is a positive integer.

[0091] In some embodiments, the communication quality indicator between the target base station and the metasurface array includes: the signal to interference plus noise ratio at the metasurface array. Base stations The base station is the target base station, and the goal in the interference suppression model is:

[0092]

[0093] Step S2: Solve the interference suppression model based on the particle swarm optimization algorithm to obtain the corresponding current optimal solution.

[0094] In the process of solving the problem based on the particle swarm optimization algorithm (PSO), each particle in the particle swarm includes dimensions, The dimensions are used to represent The phase shift coefficient of each metasurface unit. The fitness of a particle is the communication quality index between the target base station and the metasurface array when the phase shift coefficient scheme corresponding to the particle is deployed on the metasurface array. The current optimal solutions include The current optimal value of the phase shift coefficient of each metasurface unit.

[0095] The technical solution disclosed in the present invention can combine the strategy of maximizing the communication quality index (for example, maximizing the signal to interference plus noise ratio) without obtaining the complete channel information, and use the particle swarm optimization algorithm to determine the phase shift coefficient matrix of the metasurface array, which can effectively improve the interference suppression effect of the system. At the same time, the technical solution disclosed in the present invention has a simple algorithm, can support the continuous adjustment of the phase angle of the metasurface unit, and can also adapt to complex and frequently changing channel environments.

[0096] It should be noted that after obtaining the current optimal phase shift coefficient matrix through step S2, the metasurface array can be adjusted accordingly based on the current optimal phase shift coefficient matrix; specifically, the phase angle corresponding to each metasurface unit can be determined according to the current optimal phase shift coefficient corresponding to each metasurface unit, and then each metasurface unit in the metasurface array can be adjusted to the corresponding phase angle.

[0097] Figure 4 FIG. 4 is a flow chart of an optional implementation method of step S2 in the embodiment of the present disclosure. Figure 4 As shown, step S2 includes:

[0098] Step S201, perform the initialization phase of the particle swarm optimization algorithm, the number of iterations The initial value is 1, and the generated The particle swarm of particles is initialized and used as the original particle swarm corresponding to the first iteration.

[0099] In the initialization stage, not only the initialization particle swarm is generated, but also some parameters used in the entire particle swarm optimization algorithm are set, such as the maximum number of iterations, the maximum inertia weight, the minimum inertia weight, the initial velocity of the particles, etc.

[0100] Step S202: Determine Whether the original particle swarm corresponding to the iterative processing satisfies the preset iteration termination condition.

[0101] If it is determined that The original particle swarm corresponding to the first iterative processing does not meet the iteration termination condition, then execute step S203; if it is determined that the first If the original particle swarm corresponding to the iteration process satisfies the iteration termination condition, step S206 is executed.

[0102] In practical applications, the iteration termination condition can be preset according to actual needs; for example, the iteration process ends when the original population processed by the current iteration meets at least one of the following iteration termination conditions:

[0103] 1) Current number of iterations Greater than the preset maximum number of iterations .

[0104] 2) The fitness value of the current global historical optimal particle is greater than or equal to the preset fitness threshold.

[0105] Step S203: The velocity of each particle in the original particle group corresponding to the iterative processing is updated.

[0106] Step S204: update the speed of each particle according to the updated speed of each particle. The position of the particles in the original particle group corresponding to the first iterative processing is updated, and the The initial updated particle swarm corresponds to the iterative processing.

[0107] No. The first iterative process corresponds to the first particle in the original particle swarm. The particle The value of each dimension (i.e., position) is updated based on the following formula.

[0108]

[0109] in, For the The first iteration corresponds to the initial update of the particle swarm. The particle The value of the dimension, For the The first iterative process corresponds to the first particle in the original particle swarm. The particle The value of the dimension, For the The first iterative process corresponds to the first particle in the original particle swarm. The particle The final speed after the speed update is completed in all dimensions. In addition, It will also serve as the The first iterative process corresponds to the first particle in the original particle swarm. The particle The original speed of the dimension before the speed update .

[0110] It should be noted that according to the updated speed of each particle After the position of the particles in the original particle swarm corresponding to the iteration processing is updated, the particle swarm optimization algorithm will also perform "result sorting" processing, which generally includes: updating the individual historical optimal particle at each particle in the particle swarm The processing process and updating of the global historical optimal particle processing process.

[0111] In the particle swarm optimization algorithm, updating the individual best particle corresponding to each particle in the particle swarm is one of the key steps for each particle to gradually approach the optimal solution in the solution space. The individual best particle (Personal Best, also known as the individual optimal solution) refers to the position with the highest fitness value found by each particle in its search process so far.

[0112] In the traditional particle swarm optimization algorithm, the update of the individual historical best particle is done by comparing the current fitness value of the particle with the fitness value of the individual historical best particle found before the particle in each iteration. If the current fitness value of the particle is greater than the fitness value of the individual historical best particle found before the particle, the position of the current particle is updated to the new individual optimal position (that is, the current particle is used as the new individual historical best particle), and the new individual optimal fitness value corresponding to this position is recorded.

[0113] In the particle swarm optimization algorithm, updating the global best particle is one of the key steps for the entire particle swarm to gradually approach the global optimal solution in the solution space. The global best particle (Global Best, also known as the global optimal solution) refers to the position with the highest fitness value found by the entire particle swarm in the search process so far.

[0114] The global historical best particle is updated by comparing the current fitness values ​​of all particles in the population with the fitness value of the current global historical best particle in each iteration. If the current fitness value of a particle is greater than the fitness value of the current global historical best particle, the position of the particle is updated to the new global optimal position (the particle is used as the new global historical best particle), and the new global optimal fitness value corresponding to this position is recorded.

[0115] The specific steps of updating the global best particle in history and the individual best particle in history are the conventional steps after the position update step of the original particle swarm is completed in each iteration of the particle swarm optimization algorithm, and the specific process will not be repeated here.

[0116] Step S205: The initial updated particle swarm corresponding to the first iterative process determines the The original particle swarm corresponding to the iterative processing.

[0117] In some embodiments, the The initial updated particle swarm corresponding to the first iterative process is taken as the The original particle swarm corresponding to the iterative processing.

[0118] In other embodiments, the The initial updated particle swarm corresponding to the first iteration is further processed to be updated again, and the result of the update is used as the The original particle group corresponding to the iterative processing is described below with reference to an example.

[0119] After step S205 is completed, the number of iterations An increment process is performed to update, and step S202 is executed again.

[0120] Step S206: output the current global historical optimal particle to obtain the current optimal solution.

[0121] In some embodiments, the step of generating an initialized particle group including a plurality of particles in step S201 includes: a step of generating a steering vector codebook, a step of generating a Fourier transform codebook, a step of generating a random codebook, and using each codeword in the steering vector codebook, the Fourier transform codebook, and the random codebook as a particle to form an initialized particle group.

[0122] The step of generating a steering vector codebook includes:

[0123] Step S2011a, divide the preset azimuth range into equal parts to obtain target azimuth, and divide the preset pitch angle range into equal parts to obtain Target pitch angle.

[0124] The target azimuth and Target pitch angle formation Different combinations of target angle pairs, and is a positive integer, And it is a positive integer. For example, at this time The target angle pairs are: ,in Indicates Target angle pairs, and Respectively represent The target azimuth and target elevation angles of the target angle alignment are: And it is a positive integer.

[0125] The preset azimuth angle range and the preset elevation angle range are both preset manually according to actual needs. ,but The target azimuth and The target pitch angles are evenly distributed in Inside.

[0126] Step S2012a: for each target angle pair, determine an array steering vector for characterizing the response of the metasurface array to the target angle pair, and use it as the codeword corresponding to the target angle pair. The target angles of different combinations correspond to The barcode words constitute the steering vector codebook.

[0127] The array steering vectors include: Each of the metasurface units responds to the target angle with respect to the element steering vector. The metasurface unit is Target angle pair The element steering vector of the response :

[0128]

[0129] Wave Vector Represents the propagation direction of the wavefront, the dot product represents the wave vector With The position vector of the metasurface unit The phase difference between them.

[0130] Metasurface array to target angle alignment The array steering vector of the response can be expressed as:

[0131]

[0132] In the present disclosure, each codeword in the steering vector codebook represents a response state at a specific pitch angle and azimuth angle without considering interference. By generating a steering vector codebook and using it as a particle in an initialization particle group, there can be a sufficient number of response states to the incident wave at specific azimuth and pitch angles in the initialization particle group, thereby enhancing the superiority of the position of the initialization particle and laying the foundation for subsequent dynamic adjustment.

[0133] The steps of generating a Discrete Fourier Transform (DFT) codebook include:

[0134] Step S2011b, according to the number of array rows of the metasurface array Build dimensional complex discrete Fourier transform matrix , and according to the number of array columns of the metasurface array Build dimensional complex discrete Fourier transform matrix , .

[0135] In this disclosure, a dimensional complex discrete Fourier transform matrix It can be expressed as:

[0136]

[0137] Referring to the above formula, we can construct a dimensional complex discrete Fourier transform matrix and a dimensional complex discrete Fourier transform matrix .

[0138] Step S2012b: Calculate dimensional complex discrete Fourier transform matrix and dimensional complex discrete Fourier transform matrix The Kronecker tensor product of Dimensional Phase Matrix .

[0139] in, dimensional complex discrete Fourier transform matrix and dimensional complex discrete Fourier transform matrix The Kronecker tensor product of can be expressed as follows:

[0140]

[0141] in, Represents the Kronecker tensor product operation.

[0142] Step S2013b: Dimensional Phase Matrix Each line in is extracted as a codeword, and the extracted The code words constitute the Fourier transform codebook.

[0143] The steps of generating a random codebook include:

[0144] Step S2011c: Generate by randomly generating phase states random code words, each of which includes The corresponding metasurface unit Phase shift coefficient.

[0145] In addition to the particles based on the steering vector codebook and the Fourier transform codebook, the remaining particles in the initialized particle swarm are still initialized by randomly generating phase states to increase the diversity of the population and ensure that the algorithm has the opportunity to explore different areas of the solution space, thereby increasing the possibility of finding the global optimal solution.

[0146] In the present disclosure, once all particles are initialized (including particles from a specific codebook and randomly generated particles), they will be sequentially merged into a size of The particle swarm matrix is ​​used for subsequent iterations of the algorithm.

[0147] The present disclosure enhances the diversity and superiority of the initialized particle group by introducing three particle initialization methods, including a generation method based on a steering vector codebook, a generation method based on a discrete Fourier transform (DFT) codebook, and a traditional random generation method. The steering vector codebook generates particle positions using a predefined set of azimuth and elevation angles, the DFT codebook generates a beamforming matrix based on Fourier transform, and the random generation method retains the exploration capability of the traditional algorithm.

[0148] Those skilled in the art should be aware that the above-mentioned scheme of forming an initialized particle group by combining three different particle initialization methods is only an optional implementation scheme in the present disclosure and does not limit the technical solution of the present disclosure.

[0149] In some embodiments, step S203 includes:

[0150] Step S2031, determine The initial inertia weight of each particle in the original particle group corresponding to the iterative processing.

[0151] No. The first iteration of the original particle group corresponds to The initial inertia weight of each particle .

[0152]

[0153] in, and are the preset maximum and minimum inertia weights, respectively. For the The first iteration of the original particle group corresponds to The similarity between a particle and the current global historical optimal particle, And it is a positive integer.

[0154] Step S2032: Based on the hyperbolic tangent function, The initial inertia weight of each particle in the original particle group corresponding to the first iterative processing is adjusted to obtain the The final inertia weight of each particle in the original particle swarm corresponding to the iterative processing.

[0155] No. The first iteration of the original particle group corresponds to The final inertia weight of the particles :

[0156]

[0157] in, is the hyperbolic tangent function, and are two pre-set positive constants and , is the preset maximum number of iterations. As an example, The value is 1.25. The value is 2.5.

[0158] In the disclosed embodiment, the setting of the inertia weight includes the consideration of the similarity between the particle and the current global historical optimal particle, and the correction of the hyperbolic tangent curve function. Specifically, the initial inertia weight is determined by the similarity, so that the particle can move more finely near the global optimal position, and is more inclined to move in the inertial direction at a position far away; then, it is further adjusted through the hyperbolic tangent function to adapt to different stages of the search process.

[0159] Step S2033: According to The final inertia weight of each particle in the original particle group corresponding to the first iteration is The velocity of each particle in the original particle group corresponding to the iterative processing is updated.

[0160] No. The first iteration of the original particle group corresponds to The particle The final speed after the speed update is completed in each dimension :

[0161]

[0162] in, For the The first iteration is the process before the velocity of each particle in the original particle group corresponding to the first iteration is updated. The particle The original speed of the dimension, For the particle swarm The current individual best particle in history In the The value of the dimension, is the current global historical optimal particle In the The value of the dimension, For the The first iterative process corresponds to the first particle in the original particle swarm. The particle The value of the dimension, and In the Individual learning factors and social learning factors in the iterative process, and In the In the value range of the iteration The first random number and the second random number to be selected, And it is a positive integer.

[0163] In some embodiments, Individual learning factors in the iterative process and social learning factors Satisfy the following formula:

[0164]

[0165] in, and They are the preset maximum and minimum learning factor values ​​respectively.

[0166] In this disclosure, individual learning factors and social learning factors Dynamically adjust according to the number of iterations to balance the ability of global exploration and local development. This adjustment ensures that in the early stages of iteration, particles pay more attention to individual historical experience (higher ), while in the later stages of the iteration, particles tend to learn from the group (higher ).

[0167] In the present disclosure, by introducing the hyperbolic tangent curve inertia weight adjustment strategy and dynamically adjusted individual and social learning factors, the search ability and convergence performance of the algorithm are significantly enhanced; the inertia weight is initially set according to the similarity between the particle and the current global historical optimal particle, and is adjusted at different search stages through the hyperbolic tangent function, so that the particle achieves a balance between global search and local optimization.

[0168] As the A specific example of further processing the initially updated particle group corresponding to the iterative processing to make it updated again. In some embodiments, step S205 includes:

[0169] Step S2051, calculate the The particle aggregation degree of the initial updated particle swarm corresponding to the iterative processing .

[0170]

[0171] in, is the total number of particles in the initialized particle swarm, For the The first iteration corresponds to the initial update of the particle swarm. The similarity between a particle and the current global historical optimal particle, And it is a positive integer.

[0172] Step S2052: Determine The mutation probability of each particle in the initial updated particle swarm corresponding to the iterative processing.

[0173] No. The first iteration corresponds to the initial update of the particle swarm. The mutation probability of a particle :

[0174]

[0175] in, For the The mutation calculation parameter corresponding to the iteration process is a preset decreasing adjustment constant.

[0176] Step S2053: Each particle in the initial updated particle swarm corresponding to the first iteration process The first iteration corresponds to the initial update of the particle swarm. The mutation probability of a particle with the pre-generated one located in the range Random number Compare and compare the mutation probability Less than random number Time to The first iteration corresponds to the initial update of the particle swarm. Particles are mutated.

[0177] In some embodiments, the mutation operation may be to perform random initialization processing on the particles, that is, to perform random initialization on the particles. The values ​​of each dimension in each particle are randomly generated. Of course, other "mutation" algorithms can also be used in the present disclosure to implement the update of particles, and the present disclosure does not limit this.

[0178] The particle swarm obtained after step S2053 is taken as the first The original particle swarm corresponding to the iterative processing.

[0179] In the present disclosure, a reinitialization strategy based on the average similarity of particles is introduced to avoid the particle swarm from falling into the local optimum and enhance the global search capability. In each iteration, the system calculates the average similarity of the current population and determines whether to reinitialize some particles in combination with the probability function. Particles with higher similarity are more inclined to re-initialize their positions randomly (the speed can also be initialized). The reinitialized particles can jump out of the current local optimal solution space and maintain the diversity and activity of the population, thereby ensuring the balance between the convergence speed and search accuracy of the algorithm. At the same time, this strategy effectively avoids the problem of premature convergence of the population and improves the global exploration capability of the particle swarm in complex search spaces.

[0180] In some embodiments, the first The similarity between a particle and the current global historical optimal particle Determined based on the following formula:

[0181]

[0182] in, For the particle swarm The Euclidean distance between a particle and the current global historical optimal particle, and are the preset upper and lower limits of the Euclidean distance, respectively. is a pre-configured proportional constant.

[0183] Of course, the algorithm for determining the similarity between two particles shown above is only an optional implementation in the present disclosure, and it does not limit the technical solution of the present disclosure. In the present disclosure, other algorithms can also be used to determine the similarity between two particles; for example, the cosine similarity of the vectors corresponding to the two particles is used as the similarity between the two particles, etc.

[0184] In order to verify the effect of the technical solution of the present disclosure, the technical solution of the present disclosure was simulated. The simulation parameters and corresponding values ​​are shown in Table 1 below:

[0185] Table 1. Simulation parameters and values

[0186]

[0187] Figure 5 This is the change graph of the fitness function when the number of iterations in the simulation is 100. Figure 6 This is the fitness function change diagram when the number of iterations in the simulation is 1000. Figure 5 and Figure 6 As shown, Figure 5 and Figure 6 The simulation results of fitness function changes of six different algorithms during the optimization process are shown, including basic PSO (including discrete PSO and continuous PSO), improved PSO (including PSO after initialization speed, PSO after initialization + speed parameter improvement), genetic algorithm (GA) and row and column scanning algorithm. The following will analyze from the two perspectives of convergence speed and fitness size in conjunction with the attached figure.

[0188] See also Figure 5 As shown, from the perspective of convergence speed:

[0189] Row-column scanning algorithm (fitness 6): This method converges quickly in the first few generations, and the fitness rises rapidly and reaches a stable state, indicating that it is very effective in early exploration, but it may also fall into a local optimal solution due to too fast convergence.

[0190] PSO after initialization speed (fitness 3) and PSO after initialization + speed parameter improvement (fitness 4): Both show faster convergence speed, but PSO after initialization + speed parameter improvement (fitness 4) maintains a strong exploration ability in the later stage, and the fitness value tends to be stable, which better avoids the problem of local optimal solution.

[0191] GA (fitness 5): The convergence speed of the genetic algorithm is fast in the early stage, but gradually slows down in the later stage, which may be due to the lack of local search ability due to the reduction of individual diversity.

[0192] Continuous PSO (fitness 2) and discrete PSO (fitness 1): These two methods converge slowly, especially discrete PSO, whose fitness increases slowly and stays at a low value.

[0193] See also Figure 5 As shown, from the perspective of fitness:

[0194] PSO after initialization + speed parameter improvement (fitness 4): The final fitness is 30.1714, which is the highest among all methods, indicating that this method performs best in exploring the solution space and finding high-quality solutions.

[0195] PSO (fitness 3) and GA (fitness 5) after initialization speed: The final fitness is 28.0467 and 26.0094 respectively, which also performs well.

[0196] Row and column scanning (fitness 6): The final fitness is 23.3442. Although it is better than continuous PSO and discrete PSO, it is slightly inferior to PSO after initialization + speed parameter improvement (fitness 4) and PSO after initialization speed (fitness 3).

[0197] Continuous PSO (fitness 2) and discrete PSO (fitness 1): The final fitnesses are 18.873 and 14.5817 respectively, indicating that these methods have limitations in finding the optimal solution.

[0198] It can be seen that when the number of iterations is 100, the PSO algorithm achieves a better balance between global search and local search by improving the initialization and speed parameters, so that the algorithm can quickly find high-quality solutions in the early stage and continue to optimize in the later stage to finally obtain the optimal solution.

[0199] See also Figure 6 As shown, from the perspective of convergence speed:

[0200] Row-column scanning (fitness 6): This method converges quickly in the early stage, and the fitness rises rapidly and reaches a stable state after the early generations. This shows that the row-column scanning method is very effective in the initial exploration and can quickly find a solution, but its later improvement space is limited, and the fitness tends to stabilize after reaching a certain value.

[0201] Initialization + speed parameter improved PSO (fitness 4) and initialization improved PSO (fitness 3): Both improved PSOs showed rapid fitness improvement in the early stage, among which the initialization + speed parameter improved PSO (fitness 4) was better than the initialization improved PSO (fitness 3), indicating that it has a better balance between exploration and development and can effectively avoid falling into the local optimum.

[0202] GA (fitness 5): The genetic algorithm shows fast convergence in the early stage, and the fitness increases rapidly, but its convergence speed is fast and slows down in the later stage, and finally stabilizes, which is related to the relatively weak local development ability of GA in the later stage.

[0203] Continuous PSO (fitness 2) and discrete PSO (fitness 1): These two methods converge slowly, especially discrete PSO, whose fitness improves slowly and fails to reach a higher fitness level.

[0204] See also Figure 6 As shown, from the perspective of fitness:

[0205] Initialization + PSO after speed parameter improvement (fitness 4): The final fitness is 30.6439, the highest among all methods, indicating that this method is the best in exploring the solution space and avoiding local optimality. The improved speed parameter enables the algorithm to better balance between global search and local development, thereby improving fitness.

[0206] PSO (fitness 3) and GA (fitness 5) after initialization improvement: The final fitness values ​​are 28.7438 and 28.5923 respectively. These two methods also show good performance, but not as good as the effect of PSO (fitness 4) after initialization + speed parameter improvement.

[0207] Row and column scanning (fitness 6): The final fitness is 23.3813. Although it is not as good as the improved version of PSO and GA, it converges faster in the early stage.

[0208] Continuous PSO (fitness 2) and discrete PSO (fitness 1): The final fitnesses are 21.75 and 16.8206, respectively, which are the lowest among all the methods, indicating that they have great limitations in finding the optimal solution.

[0209] It can be found that when the number of iterations is 1000, PSO can better balance the global search and local search processes by improving the initialization and speed parameters. In the early stage, the optimization of its speed parameters makes the fitness rise rapidly, avoiding falling into the local optimum too early. At the same time, the fitness is kept stable by adjusting the speed in the later stage, so that the final fitness reaches the highest. This improvement enables the algorithm to have a faster convergence speed and can be developed stably in the later stage, so as to obtain the best fitness value.

[0210] Figure 7 The structure block diagram of a phase shift coefficient determination system of a metasurface array provided in an embodiment of the present disclosure. Figure 7As shown, the phase shift coefficient determination system can be used to implement the phase shift coefficient determination method provided in any of the previous embodiments. The phase shift coefficient determination system includes: a modeling module and a solving module.

[0211] The modeling module is configured to construct an interference suppression model of the metasurface array. The goal of the interference suppression model is to maximize the communication quality index between the target base station and the metasurface array. The decision variables in the interference suppression model are the The phase shift coefficient of each metasurface unit, And it is a positive integer.

[0212] The solution module is configured to solve the interference suppression model based on the particle swarm optimization algorithm to obtain the corresponding current optimal solution; in the process of solving the problem based on the particle swarm optimization algorithm, each particle in the particle swarm includes dimensions, The dimensions are used to represent The phase shift coefficient of each metasurface unit. The fitness of a particle is the communication quality index between the target base station and the metasurface array when the phase shift coefficient scheme corresponding to the particle is deployed on the metasurface array. The current optimal solutions include The current optimal value of the phase shift coefficient of each metasurface unit.

[0213] Of course, the phase shift coefficient determination system may also include a deployment module, and the deployment module is configured to deploy the metasurface array according to the current optimal solution.

[0214] For the detailed description of the above functional modules, please refer to the corresponding contents in the previous embodiments, which will not be repeated here.

[0215] Based on the same inventive concept, an embodiment of the present disclosure also provides an electronic device. Figure 8 FIG. 1 is a schematic diagram of the structure of an electronic device according to an embodiment of the present disclosure. Figure 8 As shown, an embodiment of the present disclosure provides an electronic device including: one or more processors 101, a memory 102, and one or more I / O interfaces 103. The memory 102 stores one or more programs, and when the one or more programs are executed by the one or more processors, the one or more processors implement any phase shift coefficient determination method in the above embodiments; the one or more I / O interfaces 103 are connected between the processor and the memory, and are configured to implement information interaction between the processor and the memory.

[0216] Among them, the processor 101 is a device with data processing capabilities, including but not limited to a central processing unit (CPU), etc.; the memory 102 is a device with data storage capabilities, including but not limited to random access memory (RAM, more specifically SDRAM, DDR, etc.), read-only memory (ROM), electrically erasable programmable read-only memory (EEPROM), flash memory (FLASH); the I / O interface (read-write interface) 103 is connected between the processor 101 and the memory 102, and can realize information interaction between the processor 101 and the memory 102, including but not limited to a data bus (Bus), etc.

[0217] In some embodiments, the processor 101 , the memory 102 , and the I / O interface 103 are connected to each other via a bus 104 , and further connected to other components of the computing device.

[0218] In some embodiments, the one or more processors 101 include a field programmable gate array.

[0219] According to an embodiment of the present disclosure, a computer readable medium is further provided. The computer readable medium stores a computer program, wherein when the program is executed by a processor, the steps in determining the phase shift coefficient in any of the above embodiments are implemented.

[0220] In particular, according to an embodiment of the present disclosure, the process described above with reference to the flowchart can be implemented as a computer software program. For example, an embodiment of the present disclosure includes a computer program product, including a computer program carried on a machine-readable medium, the computer program containing a program code for executing the method shown in the flowchart. In such an embodiment, the computer program can be downloaded and installed from a network through a communication part, and / or installed from a removable medium. When the computer program is executed by a central processing unit (CPU), the above-mentioned functions defined in the system of the present disclosure are executed.

[0221] It should be noted that the computer-readable medium shown in the present disclosure may be a computer-readable signal medium or a computer-readable storage medium or any combination of the above two. The computer-readable storage medium may be, for example, but not limited to, an electrical, magnetic, optical, electromagnetic, infrared, or semiconductor system, device or device, or any combination of the above. More specific examples of computer-readable storage media may include, but are not limited to: an electrical connection with one or more wires, a portable computer disk, a hard disk, a random access memory (RAM), a read-only memory (ROM), an erasable programmable read-only memory (EPROM or flash memory), an optical fiber, a portable compact disk read-only memory (CD-ROM), an optical storage device, a magnetic storage device, or any suitable combination of the above. In the present disclosure, a computer-readable storage medium may be any tangible medium containing or storing a program that can be used by or in combination with an instruction execution system, device or device. In the present disclosure, a computer-readable signal medium may include a data signal propagated in a baseband or as part of a carrier wave, in which a computer-readable program code is carried. This propagated data signal may take a variety of forms, including but not limited to electromagnetic signals, optical signals, or any suitable combination of the above. Computer-readable signal media may also be any computer-readable medium other than computer-readable storage media, which may send, propagate or transmit a program for use by or in conjunction with an instruction execution system, apparatus or device. The program code contained on the computer-readable medium may be transmitted using any appropriate medium, including but not limited to: wireless, wire, optical cable, RF, etc., or any suitable combination of the above.

[0222] The flow chart and block diagram in the accompanying drawings illustrate the possible architecture, function and operation of the system, method and computer program product according to various embodiments of the present disclosure. In this regard, each box in the flow chart or block diagram can represent a module, a program segment or a part of the code, and the aforementioned module, program segment or a part of the code contains one or more executable instructions for realizing the specified logical function. It should also be noted that in some alternative implementations, the functions marked in the box can also occur in a different order from the order marked in the accompanying drawings. For example, two boxes represented in succession can actually be executed substantially in parallel, and they can sometimes be executed in the opposite order, depending on the functions involved. It should also be noted that each box in the block diagram and / or flow chart, and the combination of the boxes in the block diagram and / or flow chart can be implemented with a dedicated hardware-based system that performs the specified function or operation, or can be implemented with a combination of dedicated hardware and computer instructions.

[0223] It is to be understood that the above embodiments are merely exemplary embodiments used to illustrate the principles of the present disclosure, but the present disclosure is not limited thereto. For those of ordinary skill in the art, various modifications and improvements can be made without departing from the spirit and substance of the present disclosure, and these modifications and improvements are also considered to be within the scope of protection of the present disclosure.

Claims

1. A method for determining a phase shift coefficient of a metasurface array, characterized in that: include: Step S1: construct an interference suppression model of the metasurface array, wherein the goal of the interference suppression model is to maximize the communication quality index between the target base station and the metasurface array, and the decision variables in the interference suppression model are the parameters included in the metasurface array. The phase shift coefficient of each metasurface unit, and is a positive integer; Step S2, solving the interference suppression model based on the particle swarm optimization algorithm to obtain the corresponding current optimal solution; in the process of solving based on the particle swarm optimization algorithm, each particle in the particle swarm includes dimensions, The dimensions are used to represent The phase shift coefficient of the metasurface unit, the fitness of the particle is the communication quality index between the target base station and the metasurface array when the phase shift coefficient scheme corresponding to the particle is deployed in the metasurface array, and the current optimal solution includes The current optimal value of the phase shift coefficient of each metasurface unit; Step S2 includes: Step S201, perform the initialization phase of the particle swarm optimization algorithm, the number of iterations The initial value is 1, and the generated The particle swarm of particles is initialized and used as the original particle swarm corresponding to the first iteration. is a preset positive integer; Step S202: Determine Whether the original particle swarm corresponding to the iterative processing satisfies the preset iteration termination condition; If it is determined that The original particle swarm corresponding to the first iterative processing does not meet the iteration termination condition, then execute step S203; if it is determined that the first If the original particle swarm corresponding to the iteration process satisfies the iteration termination condition, step S206 is executed; Step S203: The velocity of each particle in the original particle group corresponding to the iterative processing is updated; Step S204: update the speed of each particle according to the updated speed of each particle. The position of the particles in the original particle group corresponding to the first iterative processing is updated, and the The initial updated particle swarm corresponding to the iterative processing; Step S205: The initial updated particle swarm corresponding to the first iterative process determines the The original particle group corresponding to the iterative processing; After step S205 is completed, the number of iterations Perform an increment process to update, and execute step S202 again; Step S206: output the current global historical optimal particle to obtain the current optimal solution; Wherein, step S205 includes: Step S2051, calculate the The particle aggregation degree of the initial updated particle swarm corresponding to the iterative processing ; ； in, is the total number of particles in the initialized particle swarm, For the The first iteration corresponds to the initial update of the particle swarm. The similarity between a particle and the current global historical optimal particle, and is a positive integer; Step S2052: Determine The mutation probability of each particle in the initial updated particle swarm corresponding to the first iteration processing, The first iteration corresponds to the initial update of the particle swarm. The mutation probability of a particle : ； in, For the The mutation calculation parameters corresponding to the iteration processing are: It is a constant used for preset decreasing adjustment; Step S2053: Each particle in the initial updated particle swarm corresponding to the first iteration process The first iteration corresponds to the initial update of the particle swarm. The mutation probability of a particle with the pre-generated one located in the range Random number Compare and compare the mutation probability Less than random number Time to The first iteration corresponds to the initial update of the particle swarm. Particles undergo mutation operations; The particle swarm obtained after step S2053 is taken as the first The original particle swarm corresponding to the iterative processing.

2. The method for determining the phase shift coefficient according to claim 1, characterized in that: The step of generating an initialization particle group including a plurality of particles in step S201 includes: generating a steering vector codebook, generating a Fourier transform codebook, generating a random codebook, and using each codeword in the steering vector codebook, the Fourier transform codebook, and the random codebook as a particle to form the initialization particle group; The step of generating a steering vector codebook includes: Step S2011a, divide the preset azimuth range into equal parts to obtain target azimuth, and divide the preset pitch angle range into equal parts to obtain The target pitch angle, The target azimuth and Target pitch angle formation Different combinations of target angle pairs, and is a positive integer, and is a positive integer; Step S2012a, for each of the target angle pairs, determining an array steering vector for characterizing the response of the metasurface array to the target angle pair, and using the array steering vector as the codeword corresponding to the target angle pair, The target angles of different combinations correspond to The barcode words constitute the steering vector codebook; The steps of generating a Fourier transform codebook include: Step S2011b, according to the number of array rows of the metasurface array Build dimensional complex discrete Fourier transform matrix , and the number of array columns according to the metasurface array Build dimensional complex discrete Fourier transform matrix , ; Step S2012b: Calculate dimensional complex discrete Fourier transform matrix and dimensional complex discrete Fourier transform matrix The Kronecker tensor product of Dimensional Phase Matrix ; Step S2013b: Dimensional Phase Matrix Each line in is extracted as a codeword, and the extracted codewords constitute the Fourier transform codebook; The steps of generating a random codebook include: Step S2011c: Generate by randomly generating phase states random code words, each of which includes The corresponding metasurface unit The phase shift coefficient, .

3. The method for determining a phase shift coefficient according to claim 1, characterized in that: Step S203 includes: Step S2031, determine The initial inertia weight of each particle in the original particle group corresponding to the first iteration processing, The first iteration corresponds to the first particle swarm in the original particle swarm. The initial inertia weight of each particle : ； in, and are the preset maximum and minimum inertia weights, respectively. For the The first iteration corresponds to the first particle swarm in the original particle swarm. The similarity between a particle and the current global historical optimal particle, and is a positive integer; Step S2032: Based on the hyperbolic tangent function, The initial inertia weight of each particle in the original particle group corresponding to the first iterative processing is adjusted to obtain the The final inertia weight of each particle in the original particle swarm corresponding to the first iteration process, The first iteration corresponds to the first particle swarm in the original particle swarm. The final inertia weight of the particles : ； in, is the hyperbolic tangent function, and are two pre-set positive constants and , is the preset maximum number of iterations; Step S2033: According to The final inertia weight of each particle in the original particle group corresponding to the first iteration is The velocity of each particle in the original particle group corresponding to the first iteration is updated. The first iteration corresponds to the first particle swarm in the original particle swarm. The particle The final speed after the speed update is completed in each dimension : ； in, For the The first iteration is the process before the velocity of each particle in the original particle group corresponding to the first iteration is updated. The particle The original speed of the dimension, For the particle swarm The current individual best particle in history In the The value of the dimension, is the current global historical optimal particle In the The value of the dimension, For the The first iterative process corresponds to the first particle in the original particle swarm. The particle The value of the dimension, and In the Individual learning factors and social learning factors in the iterative process, and In the In the value range of the iteration The first random number and the second random number to be selected, And it is a positive integer.

4. The method for determining the phase shift coefficient according to claim 3, characterized in that: In the Individual learning factors in the iterative process and social learning factors Satisfy the following formula: ； in, and They are the preset maximum and minimum learning factor values ​​respectively.

5. The method for determining the phase shift coefficient according to claim 3 or 4, characterized in that: Particle swarm The similarity between a particle and the current global historical optimal particle Determined based on the following formula: ； in, For the particle swarm The Euclidean distance between a particle and the current global historical optimal particle, and are the preset upper and lower limits of the Euclidean distance, respectively. is a pre-configured proportional constant.

6. The method for determining a phase shift coefficient according to claim 1, characterized in that: The communication quality indicator between the target base station and the metasurface array includes: a signal to interference plus noise ratio at the metasurface array.

7. A phase shift coefficient determination system for a metasurface array, characterized in that: The system is configured to implement the phase shift coefficient determination method according to any one of claims 1 to 6, wherein the phase shift coefficient determination system comprises: A modeling module is configured to construct an interference suppression model of the metasurface array, wherein the goal in the interference suppression model is to maximize the communication quality index between the target base station and the metasurface array, and the decision variables in the interference suppression model are the parameters included in the metasurface array. The phase shift coefficient of each metasurface unit, and is a positive integer; The solution module is configured to solve the interference suppression model based on the particle swarm optimization algorithm to obtain the corresponding current optimal solution; in the process of solving the problem based on the particle swarm optimization algorithm, each particle in the particle swarm includes dimensions, The dimensions are used to represent The phase shift coefficient of the metasurface unit, the fitness of the particle is the communication quality index between the target base station and the metasurface array when the phase shift coefficient scheme corresponding to the particle is deployed in the metasurface array, and the current optimal solution includes The current optimal value of the phase shift coefficient of each metasurface unit.

8. A computer-readable medium having a computer program stored thereon, characterized in that: When the computer program is executed by a processor, the steps in the method for determining a phase shift coefficient according to any one of claims 1 to 6 are implemented.

Images