A Reconfigurable Antenna Pattern Design Method and System Based on Zero-Forcing Precoding

Abstract

This invention discloses a method and system for designing reconfigurable antenna patterns based on zero-forcing precoding, belonging to the field of wireless communication technology. The method includes establishing a non-convex optimization model that maximizes the user signal-to-noise ratio (SNR) under single-mode conditions using zero-forcing precoding; solving the non-convex optimization model for maximizing the user SNR under single-mode conditions using a continuous convex approximation SCA algorithm; iteratively solving the non-convex optimization model by designing related convex problems using a first-order Taylor expansion; and simplifying the solution process of the non-convex optimization model by designing a new surrogate function using a principal-minimization MM algorithm, thereby calculating and obtaining a reconfigurable antenna pattern design scheme based on zero-forcing precoding. The model of this invention has a fast convergence speed and low time complexity in the solution process, improving the efficiency of solving the single-mode design problem of pattern reconfigurable systems under zero-forcing precoding, and promoting the application of pattern reconfigurable antennas in single-mode scenarios.

Description

Technical Field

[0001] This invention belongs to the field of wireless communication technology, specifically relating to a reconfigurable antenna pattern design method and system based on zero-forcing precoding. Background Technology

[0002] With the rapid development and evolution of wireless communication technology, the contradiction between the increasing number of wireless devices and communication rates and limited spectrum resources is intensifying. Faced with the unavoidable fading and interference in wireless propagation environments, how to make fuller use of spectrum resources has become a significant challenge in communication system design. Reconfigurable antenna technology, as an emerging technology in the physical layer of wireless communication, can reconfigure frequency, polarization, or antenna pattern by changing the current distribution on its antenna surface, thereby improving the spectrum utilization of the communication system with additional degrees of freedom. Among these, pattern-reconfigurable antennas have received more widespread attention and application research due to their ability to actively change the channel. In downlink multi-user multiple-output (MIMO) communication systems, considering the complexity of user processing at the receiver, pattern-reconfigurable systems tend to place the pattern-reconfigurable antenna at the transmitter, i.e., transmitter-reconfigurable.

[0003] Traditional pattern-reconfigurable systems select the optimal pattern from a discrete, finite number of patterns. This aims to choose the pattern that best performs the communication system among multiple channels implemented with different patterns; this is known as the mode selection scheme for pattern-reconfigurable systems. However, the mode selection scheme only realizes the performance gain by utilizing the degrees of freedom of pattern reconfigurability, without revealing the mechanism by which the reconfigurable pattern changes the channel to achieve performance gain. In recent years, the industry has proposed a mode design scheme for pattern-reconfigurable systems based on the assumption of continuous adjustability. This scheme sets the pattern of the reconfigurable antenna to be continuously variable, describes the mechanism by which the pattern-reconfigurable system changes the channel through a pattern sampling matrix, and explores the maximum theoretical performance gain of the pattern-reconfigurable system. Here, the pattern sampling matrix refers to the matrix composed of the antenna pattern gains corresponding to the multipath angles in the wireless propagation environment.

[0004] Currently, recent research has divided the modal design problem of pattern-reconfigurable systems into single-mode and multi-mode modal design problems. It utilizes the mechanism of reconfigurable patterns redistributing equivalent channel gain and the correlation between sub-channels to achieve optimal solutions to the modal design problem. Single-mode refers to all reconfigurable antennas using the same continuously adjustable pattern, while multi-mode refers to different reconfigurable antennas using different continuously adjustable patterns. Existing literature has proposed a joint design problem of symbol-level precoding and modal design, which decomposes the joint design problem into two sub-problems and then solves them using alternating optimization. Precoding refers to the design of antenna transmitted symbols so that the receiving user can effectively receive the target signal and achieve the system's preset performance indicators, thus completing an effective and reliable communication process. However, this approach is always an approximation of the original problem and cannot achieve optimal performance. Furthermore, this approach only considers symbol-level precoding, which requires updating the precoding matrix in each time slot, leading to high complexity. Summary of the Invention

[0005] The purpose of this invention is to address the problems in the prior art by providing a reconfigurable antenna pattern design method and system based on zero-forcing precoding, which effectively reduces the solution complexity, decreases the actual computational overhead, and promotes the application of pattern reconfigurable antennas in single-mode scenarios.

[0006] To achieve the above objectives, the present invention provides the following technical solution:

[0007] Firstly, a reconfigurable antenna pattern design method based on zero-forcing precoding is provided, including:

[0008] Using the user received signal-to-noise ratio under zero-forcing precoding, a non-convex optimization model that maximizes the user signal-to-noise ratio in a single mode is established.

[0009] The non-convex optimization model that maximizes the user signal-to-noise ratio under the single mode is solved by the continuous convex approximation SCA algorithm, and the non-convex optimization model is iteratively solved by designing related convex problems using the first-order Taylor expansion.

[0010] A new surrogate function is designed using the principal-minimize MM algorithm to simplify the solution process of the non-convex optimization model, and a reconfigurable antenna pattern design scheme based on zero-forcing precoding is obtained.

[0011] As a preferred embodiment, the signal-to-noise ratio (SNR) received by all users under the zero-forcing precoding method is the same without considering power allocation, and the calculation expression is as follows:

[0012]

[0013] Where P is the total power of the transmitter; σ2 It is the power of additive white Gaussian noise; N t The number of antennas equipped at the base station; H is the single-mode pattern reconfigurable channel, calculated as follows:

[0014] H = MH path

[0015] In the formula, M represents the matrix set of all user pattern sampling vectors; H path It is the set of all user physical sub-channels;

[0016]

[0017] H path =[H1,H2,…,H k ] T m k and H k These are the directional pattern sampling vector and physical sub-channel for the k-th user, respectively.

[0018] As a preferred embodiment, the expression for the non-convex optimization model that maximizes the user signal-to-noise ratio in a single-modal environment is as follows:

[0019]

[0020] st0≤M≤ε

[0021]

[0022] In the formula, It is a positive semi-definite matrix; ε is a real number representing the maximum gain of the reconfigurable pattern; Energy constraints representing all user subchannels.

[0023] As a preferred approach, in the step of solving the non-convex optimization model that maximizes the user signal-to-noise ratio under the single-mode condition using the continuous convex approximation SCA algorithm, the objective function is g(M) = tr((MRM) H ) -1 ), is a non-convex function under the sparse matrix independent variable M;

[0024] The objective function is approximated as a non-convex function using a first-order Taylor expansion, expressed as:

[0025]

[0026] In the formula, This represents the gradient of the function g(·).

[0027] The gradient of the objective function is:

[0028]

[0029] The specific expression for the design-related convex problem is as follows:

[0030]

[0031] st0≤M≤ε

[0032]

[0033] In the formula, M i Let M be the iteration point corresponding to the i-th iteration; the relevant convex problem is solved using the convex optimization toolbox; the obtained result M is used to solve the problem. * The direction of descent of the original objective function is M. * -M i The iteration point for the (i+1)th iteration is derived as follows:

[0034] M i+1 =M i +d(M * -M i )

[0035] In the formula, the iteration step size d∈(0,1) is calculated by the backtracking line search method.

[0036] As a preferred approach, under the continuous convex approximation SCA algorithm, the step size of the iteration is calculated using the backtracking line search method, including:

[0037] Calculate the maximum allowable step size s using the following formula max :

[0038] s max =min{1,min{-m i / △m i |△m i <0}}

[0039] Repeatedly give the maximum allowed step size s max Multiply by the parameter β until:

[0040] 0≤M i+1 ≤ε, and All of these hold true, where α∈[0.01,0.1], β∈[0.3,0.8], and M i+1 =M i +s·△M.

[0041] As a preferred approach, the method of using the principal-minimize MM algorithm to design a new surrogate function to simplify the solution process of the non-convex optimization model includes:

[0042] The original objective function can be rewritten using the properties of matrix inversion as follows:

[0043]

[0044] In the formula, adj(·) represents the adjoint matrix of the matrix;

[0045] According to the Hamilton-Cayley theorem, we obtain det(MRM) H ) / tr(MRM H ) and det(MRM H ) / tr(adj(MRM H They have the same monotonicity;

[0046] According to the constraints By combining the relationship between the total channel energy and the user sub-channel energy, we obtain:

[0047]

[0048] The simplified expression for the non-convex optimization model that maximizes the user signal-to-noise ratio in a single-modal environment is as follows:

[0049]

[0050] st0≤M≤ε

[0051]

[0052] The calculation results for reconfigurable antenna pattern design schemes based on zero-forcing precoding include:

[0053] Objective function f(M) = det(MRM) H Given a sparse matrix with independent variable M, the function is non-convex. A first-order Taylor expansion is used to approximate the non-convex function, and the expression is:

[0054]

[0055] The gradient of the objective function is:

[0056]

[0057] A relevant convex problem is constructed to solve the simplified non-convex optimization model that maximizes the user signal-to-noise ratio in a single-modal environment. The specific expression of the relevant convex problem is as follows:

[0058]

[0059] st0≤M≤ε

[0060]

[0061] In the formula, M i Let M be the iteration point corresponding to the i-th iteration; the relevant convex problem is solved using the convex optimization toolbox; the obtained result M is used to solve the problem. * The direction of descent of the original objective function is M. * -M i The iteration point for the (i+1)th iteration is derived as follows:

[0062] M i+1 =M i +d(M * -M i )

[0063] In the formula, the iteration step size d∈(0,1) is calculated by the backtracking line search method.

[0064] As a preferred approach, under the main-minimize MM algorithm, the step size of the iteration is calculated using the backtracking line search method, including:

[0065] Calculate the maximum allowable step size s using the following formula max :

[0066] s max =min{1,min{-m i / △m i |△m i <0}}

[0067] Repeatedly give the maximum allowed step size s mac Multiply by the parameter β until:

[0068] 0≤M i+1 ≤ε, and All of these hold true, where α∈[0.01,0.1], β∈[0.3,0.8], and M i+1 =M i +s·△M.

[0069] Secondly, a reconfigurable antenna pattern design system based on zero-forcing precoding is provided, including:

[0070] The non-convex optimization model building module is used to build a non-convex optimization model that maximizes the user signal-to-noise ratio in a single mode by utilizing the user received signal-to-noise ratio under zero-forcing precoding.

[0071] The convex optimization design module is used to solve the non-convex optimization model that maximizes the user signal-to-noise ratio under the single mode by using the continuous convex approximation SCA algorithm, and to achieve iterative solution of the non-convex optimization model by using the first-order Taylor expansion to design related convex problems.

[0072] The non-convex optimization model simplification solution module is used to design a new surrogate function using the principal-minimize MM algorithm to simplify the solution process of the non-convex optimization model and calculate a reconfigurable antenna pattern design scheme based on zero-forcing precoding.

[0073] Thirdly, an electronic device is provided, comprising:

[0074] A memory for storing at least one instruction; and a processor for executing the instructions stored in the memory to implement the reconfigurable antenna pattern design method based on zero-forcing precoding.

[0075] Fourthly, a computer-readable storage medium is provided, wherein at least one instruction is stored therein, the at least one instruction being executed by a processor in an electronic device to implement the reconfigurable antenna pattern design method based on zero-forcing precoding.

[0076] Compared with the prior art, the present invention has at least the following beneficial effects:

[0077] The reconfigurable antenna pattern design method based on zero-forcing precoding of this invention is applicable to base stations equipped with N... t The system employs a single antenna, with K single-antenna users at the receiver. At the transmitter, channel reconstruction is performed using a single-mode pattern-reconfigurable antenna array, and interference cancellation between multiple users is achieved through zero-forcing precoding. The scenario considered is a downlink in a multi-user MIMO system. This invention utilizes the user received signal-to-noise ratio (SNR) under zero-forcing precoding to establish a non-convex optimization model that maximizes the user SNR in a single-mode environment. The non-convex optimization model is solved using the Successive Convex Approximation (SCA) algorithm. Iterative solutions to the non-convex optimization model are achieved by designing related convex problems using first-order Taylor expansion. The original non-convex problem can be solved through a series of related convex problems. Considering the complexity of the function form in the above solution process, this invention uses the Majorization-Minimization (MM) algorithm to find a new surrogate function, and then designs a simpler non-convex surrogate function and the corresponding convex optimization problem. Under the same initial point conditions, it has similar complexity performance in a single iteration and fewer iterations to converge. Therefore, the algorithm has a fast convergence speed and low time complexity, which improves the efficiency of solving the single-mode design problem of the directional pattern reconfigurable system under zero-forcing precoding. Attached Figure Description

[0078] To more clearly illustrate the technical solutions of the embodiments of the present invention, the accompanying drawings used in the embodiments will be briefly introduced below. It should be understood that the following drawings only show some embodiments of the present invention. For those skilled in the art, other related drawings can be obtained based on these drawings without creative effort.

[0079] Figure 1 The graph shows the bit error rate versus signal-to-noise ratio of the method proposed in this embodiment of the invention, using QPSK modulation, with the transmitter equipped with N... t = 4 antennas, the number of users per antenna at the receiving end is K=4, and the number of multipaths is L=6.

[0080] Figure 2 The graph shows the average sum rate as a function of signal-to-noise ratio for the method proposed in this embodiment of the invention. It employs QPSK modulation and the transmitter is equipped with N... t = 4 antennas, the number of users per antenna at the receiving end is K=4, and the number of multipaths is L=6.

[0081] Figure 3 The graph shows the average sum rate as a function of the number of multipath paths in the method proposed in this embodiment of the invention. It employs QPSK modulation and the transmitter is equipped with N... t =8 antennas, the number of users per antenna at the receiving end is K=8, and the signal-to-noise ratio at the transmitting end is SNR=5dB.

[0082] Figure 4 The graph shows the average sum rate as a function of the number of users in the method proposed in this embodiment of the invention. It employs QPSK modulation and the transmitter is equipped with N... t =8 antennas, with a multipath number of L=12 and a transmission signal-to-noise ratio of SNR=5dB.

[0083] Figure 5 This is a graph showing the variation of iteration-related parameters of the method proposed in this embodiment of the invention with the number of iterations. Detailed Implementation

[0084] The technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. Based on the embodiments of the present invention, those skilled in the art can obtain other embodiments without creative effort.

[0085] Example 1

[0086] This invention proposes a reconfigurable antenna pattern design method based on zero-forcing precoding, comprising:

[0087] Using the user received signal-to-noise ratio under zero-forcing precoding, a non-convex optimization model that maximizes the user signal-to-noise ratio in a single mode is established.

[0088] The non-convex optimization model that maximizes the user signal-to-noise ratio under the single mode is solved by the continuous convex approximation SCA algorithm, and the non-convex optimization model is iteratively solved by designing related convex problems using the first-order Taylor expansion.

[0089] A new surrogate function is designed using the principal-minimize MM algorithm to simplify the solution process of the non-convex optimization model, and a reconfigurable antenna pattern design scheme based on zero-forcing precoding is obtained.

[0090] The reconfigurable antenna pattern design method based on zero-forcing precoding in this invention is applicable in the following situations:

[0091] The base station is equipped with N t The receiver has K single-antenna users, and the transmitter uses a single-mode pattern reconfigurable antenna array for channel reconstruction and zero-forcing precoding for interference cancellation between multiple users. The scenario considered is a downlink of multi-user MIMO.

[0092] In one possible implementation, the user signal-to-noise ratio (SNR) under zero-forcing precoding is first considered. Without considering power allocation, the SNR of all receiving users is the same, and its corresponding expression is:

[0093]

[0094] In the formula, P is the total power of the transmitter, and σ 2 Here, H is the power of additive white Gaussian noise, and H is the channel with a reconfigurable single-mode radiation pattern, specifically expressed as:

[0095] H = MH path

[0096] In the formula, M represents the matrix set of all user pattern sampling vectors, and H path It is the set of all user physical sub-channels, represented as follows:

[0097]

[0098] and H path =[H1,H2,…,H K ] T m k and H k These are the directional pattern sampling vector and physical sub-channel for the k-th user, respectively.

[0099] The problem of maximizing the signal-to-noise ratio in a pattern-reconfigurable system based on zero-forcing precoding can be transformed into:

[0100]

[0101] st0≤M≤ε

[0102]

[0103] in, Let be a positive semi-definite matrix, and ε be a real number representing the maximum gain of the reconfigurable pattern. In the problem... In this context, 0 ≤ M ≤ ε represents the reconfigurable pattern gain to the transmitted signal being greater than or equal to 0 and less than or equal to the threshold, corresponding to the zeros of the reconfigurable pattern and the maximum gain of the main lobe, respectively. This means that after introducing reconfigurable patterns, the maximum channel gain for each user will not change. In other words, the performance gain of a pattern-reconfigurable system comes from the different antenna patterns that can be reconfigured, rather than from introducing additional energy to enhance the channel gain.

[0104] Furthermore, considering the objective function g(M) = tr((MRM) H ) -1 Given a sparse matrix with independent variable M, the objective function is non-convex. Therefore, the SCA algorithm can be used to solve it. A first-order Taylor expansion is applied to the objective function to approximate the original non-convex function, specifically in the form:

[0105]

[0106] in, This represents the gradient of the function g(·).

[0107] Specifically, the gradient of the original objective function is:

[0108]

[0109] Furthermore, the original nonconvex problem This can be solved through a series of related convex problems. The specific form of the related convex problems is as follows:

[0110]

[0111] st0≤M≤ε

[0112]

[0113] In the formula, M i This represents the iteration point corresponding to the i-th iteration. Problem Since this is a convex problem, it can be solved directly using the standard convex optimization toolbox, and the result M can be used... * The direction M of the descent of the original objective function can be indicated. * -M i .

[0114] Furthermore, we can deduce that the iteration point for the (i+1)th iteration is:

[0115] M i+1 =M i +d(M * -M i )

[0116] The iteration step size d∈(0,1) can be calculated using the backtracking line search method. Since the i-th iteration point and... Solution M of the problem * Since all points are within the feasible region, the iteration point of the (i+1)th iteration must also satisfy the feasible region.

[0117] In one possible implementation, considering the complexity of the above functional form, the tr((MRM) in the original optimization problem... H ) -1 The function can be solved by designing a new surrogate function using the MM algorithm.

[0118] Specifically, the original objective function can be further rewritten using the properties of matrix inversion:

[0119]

[0120] Where adj(·) denotes the adjoint matrix of the matrix.

[0121] Furthermore, according to the Hamilton-Cayley theorem, the adjoint matrix can be represented by the power of the original matrix and the trace of the original matrix. It has been proven that det(MRM) H ) / tr(MRM H ) and det(MRM H ) / tr(adj(MRM H Both exhibit the same monotonicity. Therefore, in the modal design scheme of a pattern reconfigurable system, det(MRM) H ) / tr(MRM H It can also be used as an objective function to maximize the user's signal-to-noise ratio.

[0122] Specifically, the new proxy function det(MRM) H ) / tr(MRM H The denominator consists of two parts: a numerator and a denominator. The numerator represents the channel capacity of the MIMO system, while the denominator represents the energy of the MIMO system channel.

[0123] Furthermore, considering the original optimization problem Constraints in Representing the energy constraints of all user sub-channels, and combining the relationship between the total channel energy and the user sub-channel energy, we can deduce:

[0124]

[0125] Within the feasible region of the optimization problem, the above equation provides a lower bound for the surrogate function, thus eliminating the adverse effects of the fractional form on the surrogate function. Introducing the above equation, the modal design problem based on this surrogate function can be further transformed into:

[0126]

[0127] st0≤M≤ε

[0128]

[0129] Further, the problem The objective function in the equation is f(M) = det(MRM). H The function is non-convex under the sparse matrix independent variable M. It can also be approximated as a non-convex function using a first-order Taylor expansion, specifically in the form:

[0130]

[0131] Specifically, the gradient of the original objective function is:

[0132]

[0133] Furthermore, non-convex problems This can be solved through a series of related convex problems. The specific form of the related convex problems is as follows:

[0134]

[0135] st0≤M≤ε

[0136]

[0137] Among them, M i This represents the iteration point corresponding to the i-th iteration. Problem Since this is a convex problem, it can be solved directly using the standard convex optimization toolbox, and the result M can be used... * The direction M of the descent of the original objective function can be indicated. * -M i .

[0138] Furthermore, we can deduce that the iteration point for the (i+1)th iteration is:

[0139] M i+1 =M i +d(M * -M i )

[0140] The iteration step size d∈(0,1) can be calculated using the backtracking line search method. Since the i-th iteration point and... Solution M of the problem * Since all points are within the feasible region, the iteration point of the (i+1)th iteration must also satisfy the feasible region.

[0141] This invention proposes a low-complexity single-mode design scheme for pattern reconfigurable systems based on zero-forcing precoding. Compared with existing symbol-level precoding joint design schemes, the proposed scheme has lower complexity and faster convergence speed. By utilizing the signal-to-noise ratio under zero-forcing precoding and the channel model of pattern reconfigurability, this invention addresses the optimization problem of the single-mode design scheme for pattern reconfigurable systems, clarifying the constraints and sources of performance gain.

[0142] This invention employs the SCA algorithm to solve the simplified non-convex optimization problem, and constructs the relevant convex problem through a first-order Taylor expansion to obtain the iterative direction. This iterative algorithm can achieve convergence, verifying the feasibility of single-mode design of pattern reconfigurable systems, and promoting the application of pattern reconfigurable antennas in more practical MIMO scenarios.

[0143] The invention further proposes a low-complexity MM algorithm with a simpler surrogate function form. Under the same initial conditions, it has similar complexity performance in a single iteration and fewer iterations to converge.

[0144] Example 2

[0145] Taking the SCA algorithm as an example, the reconfigurable antenna pattern design method based on zero-forcing precoding of this invention will be explained:

[0146] S1. The transmitter acquires channel state information and constructs a set H of physical sub-channels for all users based on each user's physical sub-channel. path =[H1,H2,…,H K ] T ,

[0147] S2. Randomly determine the initial point M0 for iteration. In the SCA algorithm, the initial point must be within the feasible region, i.e., satisfying the relations: 0 ≤ M0 ≤ ε.

[0148] S3. Initialize the current iteration count to i = 0, set the maximum residual norm at iteration exit to τ, and the maximum iteration count to i. max The linear backtracking parameters are α and β.

[0149] S4. Proceed to the iteration and solve the corresponding convex optimization problem:

[0150]

[0151] st0≤M≤ε

[0152]

[0153] Get the corresponding result Furthermore, calculate the search direction.

[0154] S5. Based on the search direction in the previous step, calculate the step size using the linear backtracking method.

[0155] First calculate the maximum allowable step size s = s max :

[0156] s max =min{1,min{-m i / △m i |△m i <0}}

[0157] Then keep giving s max Multiply by the parameter β until 0 ≤ M i+1 ≤ε, and All of these hold true, where α∈[0.01,0.1], β∈[0.3,0.8], and M i+1 =M i +s·△M.

[0158] S6, Update Iteration Point: M i+1 =M i +s·△M, i=i+1. If ||M i -M i-1 || F <τ or i>i max If the iteration fails, the iteration ends; otherwise, repeat steps S4, S5, and S6.

[0159] S7. After the iteration exits, based on the final result M * Calculate the corresponding channel H.

[0160] Example 3

[0161] If the MM algorithm is considered, the reconfigurable antenna pattern design method based on zero-forcing precoding in this invention includes the following steps:

[0162] S1. The transmitter acquires channel state information and constructs a set H of physical sub-channels for all users based on each user's physical sub-channel. path =[H1,H2,…,H K ] T ,

[0163] S2. Randomly determine the initial point M0 for iteration. In the MM algorithm, the initial point must be within the feasible region, i.e., satisfying the relations: 0 ≤ M0 ≤ ε.

[0164] S3. Initialize the current iteration count to i = 0, set the maximum residual norm at iteration exit to τ, and the maximum iteration count to i. max The linear backtracking parameters are α and β.

[0165] S4. Proceed to the iteration and solve the corresponding convex optimization problem:

[0166]

[0167] st0≤M≤ε

[0168]

[0169] Get the corresponding result Furthermore, calculate the search direction.

[0170] S5. Based on the search direction in the previous step, calculate the step size using the linear backtracking method.

[0171] First calculate the maximum allowable step size s = s max :

[0172] s max =min{1,min{-m i / △m i |△m i <0}}

[0173] Then keep giving s max Multiply by the parameter β until 0 ≤ M i+1 ≤ε, and All of these hold true, where α∈[0.01,0.1], β∈[0.3,0.8], and M i+1 =M i +s·△M.

[0174] S6, Update Iteration Point: M i+1 =M i +s·△M, i=i+1. If ||M i -M i-1 || F <τ or i>i max If the iteration fails, the iteration ends; otherwise, repeat steps S4, S5, and S6.

[0175] S7. After the iteration exits, based on the final result M * Calculate the corresponding channel H.

[0176] The following simulation experiments further illustrate the reconfigurable antenna pattern design method based on zero-forcing precoding in this invention:

[0177] The proposed scheme was simulated using Monte Carlo simulation.

[0178] The test conditions are as follows:

[0179] In a multi-user MIMO downlink, the transmitter is equipped with N t With one transmitting antenna and K single-antenna receiving users, assuming the transmitter can obtain perfect channel state information, the receiving vector r can be expressed as:

[0180] r = HWs + n

[0181] In the formula, s is the modulated transmit vector, W is the precoding matrix for zero-forcing precoding, H is the reconfigurable channel matrix, and n is the noise vector. Specifically, it is assumed that the noise is additive white Gaussian noise, and each element of n follows a mean of 0 and a variance of σ. 2 The Gaussian distribution.

[0182] Furthermore, the parameters of the linear backtracking method in the proposed iterative algorithm are initialized as follows: α = 0.01, β = 0.5, τ = 10. -3 i max =200. The above parameters apply to both the SCA and MM algorithms.

[0183] Simulation results are as follows Figures 1 to 5 As shown in the figure. The simulation includes a physical channel, i.e., a traditional zero-forcing precoding MIMO system without considering pattern-reconfigurable antennas, which can be considered the baseline scheme of this invention. Additionally, the simulation includes a theoretical upper bound, i.e., an interference-free MIMO system among multiple users, which can be considered the theoretically optimal result achievable by the MIMO system channel.

[0184] Please see Figure 1 and Figure 2 , Figure 1 The bit error rate (BER) curves of different modal design schemes were simulated. First, two schemes with reconfigurable radiation patterns were introduced. Compared with the physical channel without reconfigurable radiation patterns, the BER of the communication system was reduced, effectively improving the reliability of the communication system. In the radiation pattern design scheme, since the surrogate function form of the MM algorithm is simpler, the error corresponding to the first-order Taylor expansion in finding the corresponding convex problem is smaller. Therefore, the performance gain of the MM algorithm proposed in this invention is slightly better than that of the SCA algorithm. Figure 2 The average and rate curves of different modal design schemes were simulated. Figure 4The results show that the reconfigurable design schemes of both the MM algorithm and the SCA algorithm are feasible, and both can achieve performance gains that approximate the theoretical upper limit.

[0185] Please see Figures 3 to 5 , Figure 3 The average sum and rate curves of the modal design schemes under different numbers of multipaths were simulated. It can be seen that the performance gain of the modal design schemes gradually approaches the theoretical upper limit with the increase of the number of multipaths, further verifying the performance advantage of the pattern reconfigurable system compared with the traditional MIMO system. Figure 4 The average sum and rate curves of the modal design schemes under different numbers of users were simulated. First, the traditional physical channel experiences increased interference and a severe performance degradation as the number of users increases. However, the communication system that introduces a reconfigurable radiation pattern can change the channel correlation between users, reduce interference, and thus improve the effectiveness of the communication system. Figure 5 The changes of some important parameters, including the objective function value and the iteration residual, during the iteration process of the two single-modal design schemes proposed in this invention were simulated. It can be seen that, under the same random initial point, as the iteration progresses, the objective function value and the iteration residual of both algorithms gradually decrease until convergence. However, at the same accuracy, the MM algorithm has significantly fewer convergence times than the SCA algorithm, indicating that the complexity of the MM algorithm is much lower than that of the SCA algorithm, further verifying the practical applicability of the proposed design scheme.

[0186] Another embodiment of the present invention also proposes a reconfigurable antenna pattern design system based on zero-forcing precoding, comprising:

[0187] The non-convex optimization model building module is used to build a non-convex optimization model that maximizes the user signal-to-noise ratio in a single mode by utilizing the user received signal-to-noise ratio under zero-forcing precoding.

[0188] The convex optimization design module is used to solve the non-convex optimization model that maximizes the user signal-to-noise ratio under the single mode by using the continuous convex approximation SCA algorithm, and to achieve iterative solution of the non-convex optimization model by using the first-order Taylor expansion to design related convex problems.

[0189] The non-convex optimization model simplification solution module is used to design a new surrogate function using the principal-minimize MM algorithm to simplify the solution process of the non-convex optimization model and calculate a reconfigurable antenna pattern design scheme based on zero-forcing precoding.

[0190] Another embodiment of the present invention provides an electronic device, characterized in that it comprises:

[0191] A memory for storing at least one instruction; and a processor for executing the instructions stored in the memory to implement the reconfigurable antenna pattern design method based on zero-forcing precoding.

[0192] Another embodiment of the present invention also proposes a computer-readable storage medium storing at least one instruction, which is executed by a processor in an electronic device to implement the reconfigurable antenna pattern design method based on zero-forcing precoding.

[0193] For example, the instructions stored in the memory can be divided into one or more modules / units. These modules / units are stored in a computer-readable storage medium and executed by the processor to complete the reconfigurable antenna pattern design method based on zero-forcing precoding of the present invention. The one or more modules / units can be a series of computer-readable instruction segments capable of performing specific functions, which describe the execution process of the computer program in the server.

[0194] The electronic device may be a smartphone, laptop, PDA, or cloud server, among other computing devices. It may include, but is not limited to, a processor and memory. Those skilled in the art will understand that the electronic device may also include more or fewer components, or combinations of certain components, or different components; for example, it may also include input / output devices, network access devices, buses, etc.

[0195] The processor can be a Central Processing Unit (CPU), or other general-purpose processors, digital signal processors (DSPs), application-specific integrated circuits (ASICs), field-programmable gate arrays (FPGAs), or other programmable logic devices, discrete gate or transistor logic devices, discrete hardware components, etc. A general-purpose processor can be a microprocessor or any conventional processor.

[0196] The memory can be an internal storage unit of the server, such as a hard drive or RAM. Alternatively, it can be an external storage device, such as a plug-in hard drive, Smart Media Card (SMC), Secure Digital (SD) card, or Flash Card. Furthermore, the memory can include both internal and external storage units. The memory is used to store computer-readable instructions and other programs and data required by the server. It can also be used to temporarily store data that has been output or will be output.

[0197] It should be noted that the information interaction and execution process between the above-mentioned module units are based on the same concept as the method embodiment. For details on their specific functions and technical effects, please refer to the method embodiment section. They will not be repeated here.

[0198] Those skilled in the art will clearly understand that, for the sake of convenience and brevity, the above-described division of functional units and modules is merely an example. In practical applications, the above functions can be assigned to different functional units and modules as needed, that is, the internal structure of the device can be divided into different functional units or modules to complete all or part of the functions described above. The functional units and modules in the embodiments can be integrated into one processing unit, or each unit can exist physically separately, or two or more units can be integrated into one unit. The integrated unit can be implemented in hardware or as a software functional unit. Furthermore, the specific names of the functional units and modules are only for easy differentiation and are not intended to limit the scope of protection of this application. The specific working process of the units and modules in the above system can be referred to the corresponding process in the foregoing method embodiments, and will not be repeated here.

[0199] If the integrated unit is implemented as a software functional unit and sold or used as an independent product, it can be stored in a computer-readable storage medium. Based on this understanding, all or part of the processes in the methods of the above embodiments of this application can be implemented by a computer program instructing related hardware. The computer program can be stored in a computer-readable storage medium, and when executed by a processor, it can implement the steps of the various method embodiments described above. The computer program includes computer program code, which can be in the form of source code, object code, executable files, or certain intermediate forms. The computer-readable medium can include at least: any entity or device capable of carrying the computer program code to a photographing device / terminal device, a recording medium, a computer memory, a read-only memory (ROM), a random access memory (RAM), an electrical carrier signal, a telecommunication signal, and a software distribution medium. Examples include USB flash drives, portable hard drives, magnetic disks, or optical disks.

[0200] In the above embodiments, the descriptions of each embodiment have different focuses. For parts that are not described in detail or recorded in a certain embodiment, please refer to the relevant descriptions of other embodiments.

[0201] The above-described embodiments are only used to illustrate the technical solutions of this application, and are not intended to limit them. Although this application has been described in detail with reference to the foregoing embodiments, those skilled in the art should understand that modifications can still be made to the technical solutions described in the foregoing embodiments, or equivalent substitutions can be made to some of the technical features. Such modifications or substitutions do not cause the essence of the corresponding technical solutions to deviate from the spirit and scope of the technical solutions of the embodiments of this application, and should all be included within the protection scope of this application.

Claims

1. A method for designing reconfigurable antenna pattern based on zero-forcing precoding, characterized in that, include: Using the user received signal-to-noise ratio under zero-forcing precoding, a non-convex optimization model that maximizes the user signal-to-noise ratio in a single mode is established. The non-convex optimization model for maximizing the user signal-to-noise ratio in the single-modal case can be solved using any of the following methods: The solution is obtained by using the continuous convex approximation SCA algorithm, and the non-convex optimization model is iteratively solved by designing related convex problems using first-order Taylor expansion. Alternatively, a new surrogate function can be designed using the principal-minimize MM algorithm to simplify the solution process of the non-convex optimization model and obtain a reconfigurable antenna pattern design scheme based on zero-forcing precoding. The signal-to-noise ratio (SNR) received by all users under the zero-forcing precoding method is the same without considering power allocation, and the calculation expression is as follows: in, It is the total power of the transmitting end; It is the power of additive white Gaussian noise; The number of antennas equipped at the base station; It is a single-mode pattern reconfigurable channel, and the calculation expression is: In the formula, A set of matrices representing all user pattern sampling vectors; It is the set of all user physical sub-channels; , and The first Individual user's radiation pattern sampling vector and physical sub-channel; The expression for the non-convex optimization model that maximizes the user signal-to-noise ratio in a single-modal environment is as follows: s.t. In the formula, It is a positive semi-definite matrix; It is a real number representing the maximum gain of the reconfigurable pattern; Energy constraints representing all user subchannels; The process of designing a new surrogate function using the principal-minimize MM algorithm to simplify the solution of the nonconvex optimization model includes: The original objective function can be rewritten using the properties of matrix inversion as follows: In the formula, The adjoint matrix of a matrix; According to the Hamilton-Cayley theorem, we get and They have the same monotonicity; According to the constraints By combining the relationship between the total channel energy and the user sub-channel energy, we obtain: The simplified expression for the non-convex optimization model that maximizes the user signal-to-noise ratio in a single-modal environment is as follows: s.t. The calculation results for reconfigurable antenna pattern design schemes based on zero-forcing precoding include: objective function In sparse matrix independent variables The following is a non-convex function. We approximate the non-convex function using a first-order Taylor expansion, and the expression is: The gradient of the objective function is: A relevant convex problem is constructed to solve the simplified non-convex optimization model that maximizes the user signal-to-noise ratio in a single-modal environment. The specific expression of the relevant convex problem is as follows: s.t. In the formula, For the first The iteration point corresponding to the next iteration; related convex problems are solved using the convex optimization toolbox; the results are then used... The direction of descent of the original objective function is obtained as follows derive the first The iteration point for the next iteration is: In the formula, the step size of the iteration is... It was calculated using the backtracking line search method.

2. The reconfigurable antenna pattern design method based on zero-forcing precoding according to claim 1, characterized in that, In the step of solving the non-convex optimization model that maximizes the user signal-to-noise ratio in the single-mode using the continuous convex approximation SCA algorithm, the objective function is: In sparse matrix independent variables The following are non-convex functions; The objective function is approximated as a non-convex function using a first-order Taylor expansion, expressed as: In the formula, Representation function The gradient; The gradient of the objective function is: The specific expression for the design-related convex problem is as follows: s.t. In the formula, For the first The iteration point corresponding to the next iteration; related convex problems are solved using the convex optimization toolbox; the results are then used... The direction of descent of the original objective function is obtained as follows derive the first The iteration point for the next iteration is: In the formula, the step size of the iteration is... It was calculated using the backtracking line search method.

3. The reconfigurable antenna pattern design method based on zero-forcing precoding according to claim 2, characterized in that, In the continuous convex approximation SCA algorithm, the step size of the iteration is calculated using the backtracking line search method, including: Calculate the maximum allowable step size using the following formula : Repeatedly give the maximum allowed step size Multiply by parameter , until: and All are true, among which , , .

4. The reconfigurable antenna pattern design method based on zero-forcing precoding according to claim 1, characterized in that, In the main-minimize MM algorithm, the step size of the iteration is calculated using the backtracking line search method, including: Calculate the maximum allowable step size using the following formula : Repeatedly give the maximum allowed step size Multiply by parameter , until: and All are true, among which , , .

5. A reconfigurable antenna pattern design system based on zero-forcing precoding, characterized in that, include: The non-convex optimization model building module is used to build a non-convex optimization model that maximizes the user signal-to-noise ratio in a single mode by utilizing the user received signal-to-noise ratio under zero-forcing precoding. The convex optimization design module is used to solve the non-convex optimization model that maximizes the user signal-to-noise ratio under the single mode by using the continuous convex approximation SCA algorithm, and to achieve iterative solution of the non-convex optimization model by using the first-order Taylor expansion to design related convex problems. The non-convex optimization model simplification solution module is used to design a new surrogate function using the principal-minimize MM algorithm to simplify the solution process of the non-convex optimization model and calculate the reconfigurable antenna pattern design scheme based on zero-forcing precoding. The signal-to-noise ratio (SNR) received by all users under the zero-forcing precoding method is the same without considering power allocation, and the calculation expression is as follows: in, It is the total power of the transmitting end; It is the power of additive white Gaussian noise; The number of antennas equipped at the base station; It is a single-mode pattern reconfigurable channel, and the calculation expression is: In the formula, A set of matrices representing all user pattern sampling vectors; It is the set of all user physical sub-channels; , and The first Individual user's radiation pattern sampling vector and physical sub-channel; The expression for the non-convex optimization model that maximizes the user signal-to-noise ratio in a single-modal environment is as follows: s.t. In the formula, It is a positive semi-definite matrix; It is a real number representing the maximum gain of the reconfigurable pattern; Energy constraints representing all user subchannels; The process of designing a new surrogate function using the principal-minimize MM algorithm to simplify the solution of the nonconvex optimization model includes: The original objective function can be rewritten using the properties of matrix inversion as follows: In the formula, The adjoint matrix of a matrix; According to the Hamilton-Cayley theorem, we get and They have the same monotonicity; According to the constraints By combining the relationship between the total channel energy and the user sub-channel energy, we obtain: The simplified expression for the non-convex optimization model that maximizes the user signal-to-noise ratio in a single-modal environment is as follows: s.t. The calculation results for reconfigurable antenna pattern design schemes based on zero-forcing precoding include: objective function In sparse matrix independent variables The following is a non-convex function. We approximate the non-convex function using a first-order Taylor expansion, and the expression is: The gradient of the objective function is: A relevant convex problem is constructed to solve the simplified non-convex optimization model that maximizes the user signal-to-noise ratio in a single-modal environment. The specific expression of the relevant convex problem is as follows: s.t. In the formula, For the first The iteration point corresponding to the next iteration; related convex problems are solved using the convex optimization toolbox; the results are then used... The direction of descent of the original objective function is obtained as follows derive the first The iteration point for the next iteration is: In the formula, the step size of the iteration is... It was calculated using the backtracking line search method.

6. An electronic device, characterized in that, include: Memory, storing at least one instruction; and The processor executes instructions stored in the memory to implement the reconfigurable antenna pattern design method based on zero-forcing precoding as described in any one of claims 1 to 4.

7. A computer-readable storage medium, characterized in that: The computer-readable storage medium stores at least one instruction, which is executed by a processor in an electronic device to implement the reconfigurable antenna pattern design method based on zero-forcing precoding as described in any one of claims 1 to 4.

Images