A method and system for beam optimization based on smart reflectarray element grouping

Abstract

The application discloses a kind of based on intelligent reflecting surface array element grouping's beam optimization method and system, it is related to wireless communication technical field.The application includes: the initial optimization problem of signal transmission by intelligent reflecting surface IRS between base station and multiple users is constructed;Intelligent reflecting surface IRS reflecting element is grouped using Kronecker product method;Initial optimization problem is converted into the optimization problem P1 of grouped IRS reflecting element;Optimization problem P1 is converted into optimization problem P2 with reduced complexity using fractional programming theory;Auxiliary variable is introduced, and optimization problem P2 is decomposed into three decoupled sub-problems, and the optimal solution of each sub-problem is obtained using alternating iterative optimization method;According to optimal solution, the base station BS beamforming vector and IRS passive beamforming vector after joint optimization are obtained, and finally intelligent reflecting surface array element grouping's beam optimization method is obtained.The application can achieve the purpose of reducing the calculation complexity of IRS passive beamforming vector.

Description

Technical Field

[0001] This invention relates to the field of wireless communication technology, and in particular to a beam optimization method and system based on intelligent reflector array element grouping. Background Technology

[0002] Intelligent Reflecting Surface (IRS) is a new technology that reconfigures the wireless propagation environment through software-controlled reflection, significantly improving the performance of wireless communication networks. An IRS is a plane composed of numerous low-cost passive reflective elements. Each element can independently adjust the amplitude and phase of the incident signal, thereby controlling the signal transmission direction between the base station (BS) and users, achieving active control of the wireless channel.

[0003] However, IRS typically contains a large number of reflection elements, and this increase in the number of reflection elements presents several technical challenges to the communication system. The two most important challenges are as follows: First, a large number of IRS reflection elements significantly increases channel estimation time, leading to substantial latency and reduced communication quality for users. Second, the complexity of calculating the IRS passive beamforming vector increases with the number of reflection elements. Grouping adjacent reflection elements in the IRS and deriving the grouped IRS passive beamforming vector using relevant optimization algorithms can significantly reduce the computational complexity of the IRS passive beamforming vector while maintaining user communication quality.

[0004] IRS, as a novel wireless communication-assisted transmission technology, has been studied to some extent in the field of wireless communication coverage extension. Existing research mainly uses an alternating iterative optimization algorithm to optimize the beamforming vector at the BS end and the passive beamforming vector of the IRS.

[0005] Existing research shows that IRS-assisted wireless communication systems achieve higher user rates than traditional communication systems, especially when obstacles between users and base stations cause severe congestion, resulting in a more significant improvement. However, in practical applications, the IRS needs to adjust the reflection coefficients of the reflector elements in real time based on changes in CSI (Channel State Information). Since the IRS typically has multiple reflector elements, this introduces high computational complexity to the optimization of passive beamforming vectors, causing a certain amount of latency. Summary of the Invention

[0006] This invention addresses the problem of high computational complexity of IRS passive beamforming vector calculation in IRS-assisted multi-user communication scenarios in existing technologies.

[0007] To solve the above-mentioned technical problems, the present invention provides the following technical solution:

[0008] On one hand, this invention provides a beam optimization method based on intelligent reflector element grouping. This method is implemented by a beam optimization system based on intelligent reflector element grouping, which includes a base station, an intelligent reflector RIS, and multiple users. The method includes:

[0009] S1. Construct an initial optimization problem for signal transmission between the base station and multiple users via a smart reflector IRS; the initial optimization problem is to optimize the total rate of multiple users.

[0010] S2. The Kronecker product method is used to group the reflection elements of the intelligent reflector IRS; the initial optimization problem is transformed into the optimization problem P1 of the grouped IRS reflection elements.

[0011] S3. Using fractional programming theory, the optimization problem P1 is transformed into an optimization problem P2 with reduced complexity.

[0012] S4. Introduce auxiliary variables and decompose the optimization problem P2 into three decoupled subproblems. Use the alternating iterative optimization method to find the optimal solution of each subproblem in closed form.

[0013] S5. Based on the optimal solution, the jointly optimized base station (BS) beamforming vector and IRS passive beamforming vector are obtained, and finally, the beam optimization method for intelligent reflector array element grouping is obtained.

[0014] Optionally, in the initial optimization problem, the received signal of the k-th user among multiple users is as shown in equation (1):

[0015]

[0016] in, Let M be the number of reflection elements in an IRS and the k-th user, and h be the channel between the IRS and the k-th user. M,k It is the channel between the Mth reflection element and the kth user, T is the transpose of the matrix, and H is the transpose conjugate of the matrix; Φ = ηdiag([θ1,...,θ) M ] T Let be the matrix representing the reflection coefficients of the IRS, where η∈[0,1] represents the on or off state of the M-th reflection element in the IRS. and This represents the reflection coefficient of the Mth array element; This is the channel between the BS and the IRS, where N is the number of antennas the base station is equipped with. This is the channel between BS and the Mth reflector element. All channels are assumed to be Saleh-Valenzuela millimeter-wave channels; P = [P1,...PK ] represents the beamforming matrix, where This represents the beamforming vector sent to the k-th user, where K is the number of users; s k This indicates that the signal sent by BS to the k-th user satisfies... k∈{1,2,...,K}; s represents the beamforming vector sent to the j-th user; j This indicates that the signal sent by BS to the j-th user satisfies... j∈{1,2,...,K};μ k This indicates that the mean is 0 and the variance is... Additive white Gaussian noise.

[0017] Alternatively, in the optimization problem of the grouped IRS reflection array elements, the received signal of the kth user among multiple users is as shown in equation (2):

[0018]

[0019] in, The matrix representing the phase shift factor after IRS grouping. This represents the phase shift factor of the Jth group, where J is the number of IRS element groups; This represents the channel between the IRS and the k-th user; This is the channel between the BS and the IRS.

[0020] Optionally, in the optimization problem of the grouped IRS reflection array elements, the signal-to-interference-plus-noise ratio (SINNR) for the k-th user among multiple users is shown in Equation (3):

[0021]

[0022] Optionally, the optimization problem P1 of the grouped IRS reflection array elements is shown in equations (4)-(6) below:

[0023]

[0024] st|φ j |=1,n=1,2,...,J (5)

[0025] tr(PP H )≤P max (6)

[0026] Where, |φ j |=1 indicates that the reflection amplitude of each group of reflection array elements in the IRS is set to 1; P max This indicates the maximum transmit power of the BS.

[0027] Alternatively, the optimization problem P2 with reduced complexity is shown in equations (7)-(10) below:

[0028]

[0029] st|φ j |=1,n=1,2,...,J (8)

[0030] tr(PP H )≤P max (9)

[0031] α k ≥0, k=1,2,...,K (10)

[0032] Where α=[α1,...,α K ] T These are auxiliary variables introduced.

[0033] Optionally, α k The methods for finding the optimal solution include:

[0034] Let variables P and For the variable α in the above formula (6) fixed, k Taking the partial derivative, we get α. k The optimal solution;

[0035] α k Substituting the optimal solution into equation (6) above, we can transform optimization problem P2 into optimization problem P3.

[0036] Alternatively, the methods for finding the optimal solution to P include:

[0037] make For the equivalent channel, the optimization problem P3 is re-expressed as the optimization problem P4.

[0038] Given α k and Introduce auxiliary variable β = [β1,...,β] K ] T By performing a second transformation on optimization problem P4, we obtain optimization problem P5.

[0039] Given P, β can be obtained by taking the partial derivative with respect to the optimization problem P5. k The optimal solution.

[0040] Fixed β k p is obtained through the Lagrange multiplier method k The optimal solution.

[0041] Optionally, The methods for finding the optimal solution include:

[0042] Given α and P, introduce auxiliary variables ρ = [ρ1,...,ρ K ] T The optimization problem P5 is transformed twice to obtain the optimization problem P6.

[0043] Given ρ, re-express the optimization problem P6 as the optimization problem P7.

[0044] The optimization problem P7 is re-expressed as optimization problem P8 using the constraint relaxation method.

[0045] According to the Lagrange dual decomposition method, the optimization problem P8 is re-expressed as the optimization problem P9.

[0046] Solving optimization problem P9, we finally obtain... The optimal solution.

[0047] On the other hand, the present invention provides a beam optimization system based on intelligent reflector array element grouping. This system is used to implement a beam optimization method based on intelligent reflector array element grouping. The system includes a base station, an intelligent reflector RIS, and multiple users; wherein:

[0048] A base station is used to transmit signals to multiple users.

[0049] The Intelligent Reflector RIS (RIS) is used to group the reflector elements of the Intelligent Reflector RIS (IRS) using the Kronecker product method. The initial optimization problem is transformed into an optimization problem P1 of the grouped IRS reflector elements. Fractional programming theory is used to transform optimization problem P1 into a less complex optimization problem P2. Auxiliary variables are introduced, and optimization problem P2 is decomposed into three decoupled subproblems. An alternating iterative optimization method is used to find the optimal solution in closed form for each subproblem. Based on the optimal solution, the jointly optimized BS beamforming vector and IRS passive beamforming vector are obtained, finally yielding the beam optimization method for grouping the intelligent reflector elements.

[0050] Multiple users are used to receive signals transmitted from the base station.

[0051] Optionally, in the initial optimization problem, the received signal of the k-th user among multiple users is as shown in equation (1):

[0052]

[0053] in, Let M be the number of reflection elements in an IRS and the k-th user, and h be the channel between the IRS and the k-th user. M,k It is the channel between the Mth reflection element and the kth user, T is the transpose of the matrix, and H is the transpose conjugate of the matrix; Φ = ηdiag([θ1,...,θ) M ]T Let be the matrix representing the reflection coefficients of the IRS, where η∈[0,1] represents the on or off state of the M-th reflection element in the IRS. and This represents the reflection coefficient of the Mth array element; This is the channel between the BS and the IRS, where N is the number of antennas the base station is equipped with. This is the channel between BS and the Mth reflector element. All channels are assumed to be Saleh-Valenzuela millimeter-wave channels; P = [P1,...P K ] represents the beamforming matrix, where This represents the beamforming vector sent to the k-th user, where K is the number of users; s k This indicates that the signal sent by BS to the k-th user satisfies... k∈{1,2,...,K}; s represents the beamforming vector sent to the j-th user; j This indicates that the signal sent by BS to the j-th user satisfies... j∈{1,2,...,K};μ k This indicates that the mean is 0 and the variance is... Additive white Gaussian noise.

[0054] Alternatively, in the optimization problem of the grouped IRS reflection array elements, the received signal of the kth user among multiple users is as shown in equation (2):

[0055]

[0056] in, The matrix representing the phase shift factor after IRS grouping. This represents the phase shift factor of the Jth group, where J is the number of IRS element groups; This represents the channel between the IRS and the k-th user; This is the channel between the BS and the IRS.

[0057] Optionally, in the optimization problem of the grouped IRS reflection array elements, the signal-to-interference-plus-noise ratio (SINNR) for the k-th user among multiple users is shown in Equation (3):

[0058]

[0059] Optionally, the optimization problem P1 of the grouped IRS reflection array elements is shown in equations (4)-(6) below:

[0060]

[0061] st|φ j|=1,n=1,2,...,J (5)

[0062] tr(PP H )≤P max (6)

[0063] Where, |φ j |=1 indicates that the reflection amplitude of each group of reflection array elements in the IRS is set to 1; P max This indicates the maximum transmit power of the BS.

[0064] Alternatively, the optimization problem P2 with reduced complexity is shown in equations (7)-(10) below:

[0065]

[0066] st|φ j |=1,n=1,2,...,J (8)

[0067] tr(PP H )≤P max (9)

[0068] α k ≥0, k=1,2,...,K (10)

[0069] Where α=[α1,...,α K ] T These are auxiliary variables introduced.

[0070] Optionally, the smart reflective surface is further used for:

[0071] Let variables P and For the variable α in the above formula (6) fixed, k Taking the partial derivative, we get α. k The optimal solution;

[0072] α k Substituting the optimal solution into equation (6) above, we can transform optimization problem P2 into optimization problem P3.

[0073] Optionally, the smart reflective surface is further used for:

[0074] make For the equivalent channel, the optimization problem P3 is re-expressed as the optimization problem P4.

[0075] Given α k and Introduce auxiliary variable β = [β1,...,β] K ] T By performing a second transformation on optimization problem P4, we obtain optimization problem P5.

[0076] Given P, β can be obtained by taking the partial derivative with respect to the optimization problem P5. k The optimal solution.

[0077] Fixed β k p is obtained through the Lagrange multiplier method k The optimal solution.

[0078] Optionally, the smart reflective surface is further used for:

[0079] Given α and P, introduce auxiliary variables ρ = [ρ1,...,ρ K ] T The optimization problem P5 is transformed twice to obtain the optimization problem P6.

[0080] Given ρ, re-express the optimization problem P6 as the optimization problem P7.

[0081] The optimization problem P7 is re-expressed as optimization problem P8 using the constraint relaxation method.

[0082] According to the Lagrange dual decomposition method, the optimization problem P8 is re-expressed as the optimization problem P9.

[0083] Solving optimization problem P9, we finally obtain... The optimal solution.

[0084] The beneficial effects of the technical solutions provided in the embodiments of the present invention include at least the following:

[0085] The proposed intelligent reflector grouping optimization model for wireless communication considers the impact of different numbers of IRS element groups and different numbers of users on the system downlink rate. It also compares the average optimization time per iteration between cases with different numbers of IRS element groups and cases with no IRS grouping. This significantly reduces the model optimization time while ensuring the total downlink user rate. Attached Figure Description

[0086] To more clearly illustrate the technical solutions in the embodiments of the present invention, the accompanying drawings used in the description of the embodiments will be briefly introduced below. Obviously, the accompanying drawings described below are only some embodiments of the present invention. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0087] Figure 1 This is a schematic diagram of the beam optimization method based on intelligent reflective array element grouping provided in an embodiment of the present invention;

[0088] Figure 2 This is a schematic diagram of the IRS reflection array element grouping provided in an embodiment of the present invention;

[0089] Figure 3 This is a comparison chart of the average optimization time per iteration for IRS array elements with different grouping numbers and IRS without grouping, provided by an embodiment of the present invention.

[0090] Figure 4 This is a schematic diagram of the total downlink user rate that can be achieved when the number of different groups of users in the IRS array element is different, as provided in the embodiment of the present invention.

[0091] Figure 5 This is a block diagram of a beam optimization system based on intelligent reflector array element grouping provided in an embodiment of the present invention. Detailed Implementation

[0092] To make the technical problems, technical solutions and advantages of the present invention clearer, a detailed description will be given below in conjunction with the accompanying drawings and specific embodiments.

[0093] like Figure 1 As shown, this embodiment of the invention provides a beam optimization method based on intelligent reflector array element grouping, which can be implemented by a beam optimization system based on intelligent reflector array element grouping. Figure 1 The flowchart shown is a beam optimization method based on intelligent reflector array element grouping. The processing flow of this method may include the following steps:

[0094] S1. Construct an initial optimization problem for signal transmission between the base station and multiple users via a smart reflector; the initial optimization problem is to optimize the total rate of multiple users.

[0095] In one feasible implementation, consider a downlink multi-user wireless communication system where the direct link between the user and the BS is blocked by an obstacle. The base station is equipped with N antennas, there are K users, each with one antenna, and an IRS has M reflector elements. The received signal of the k-th user among the multiple users is shown in equation (1) below:

[0096]

[0097] in, Let M be the number of reflection elements in an IRS and the k-th user, and h be the channel between the IRS and the k-th user. M,k It is the channel between the Mth reflection element and the kth user, T is the transpose of the matrix, and H is the transpose conjugate of the matrix; Φ = ηdiag([θ1,...,θ) M ] T Let be the matrix representing the reflection coefficients of the IRS, where η∈[0,1] represents the on or off state of the M-th reflection element in the IRS. and This represents the reflection coefficient of the Mth array element; This is the channel between the BS and the IRS, where N is the number of antennas the base station is equipped with. This is the channel between BS and the Mth reflector element. All channels are assumed to be Saleh-Valenzuela millimeter-wave channels; P = [P1,...P K ] represents the beamforming matrix, where This represents the beamforming vector sent to the k-th user, where K is the number of users; s k This indicates that the signal sent by BS to the k-th user satisfies... k∈{1,2,...,K}; s represents the beamforming vector sent to the j-th user; j This indicates that the signal sent by BS to the j-th user satisfies... j∈{1,2,...,K};μ k This indicates that the mean is 0 and the variance is... Additive white Gaussian noise.

[0098] like Figure 2-4 As shown, for the case of IRS element grouping, it is assumed that all elements are divided into J groups, and each group contains the same number of reflection elements B. Then the reflection array elements between (j-1)B+1 and jB belong to the j-th group. Assuming that all array elements in a group have the same phase shift factor, the IRS phase shift matrix can be re-expressed as the following equation (2):

[0099]

[0100] in, The matrix representing the phase shift factor after IRS grouping. Let J represent the phase shift factor of the Jth group. After grouping, the channel between the IRS and the kth user can be expressed as: The channel between the BS and the IRS can be represented as Therefore, in the optimization problem of the grouped IRS reflection array elements, the received signal of the kth user among multiple users is rewritten as the following equation (3):

[0101]

[0102] road; This is the channel between the BS and the IRS.

[0103] Optionally, in the optimization problem of the grouped IRS reflection array elements, the SINR (Signal-to-interference-plus-noise Ratio) for the k-th user among multiple users can be obtained by the following equation (4):

[0104]

[0105] S2. The Kronecker product method is used to group the reflective elements of the smart reflective surface; the initial optimization problem is transformed into the optimization problem P1 of the grouped IRS reflective elements.

[0106] In one feasible implementation, mathematically, the corresponding optimization problem can be expressed as equations (5)-(7):

[0107]

[0108] st|φ j |=1,n=1,2,...,J (6)

[0109] tr(PP H )≤P max (7)

[0110] Where, |φ j |=1 indicates that the reflection amplitude of each group of reflection array elements in the IRS is set to 1, P max Let represent the maximum transmit power of BS. The above problem is a non-convex problem.

[0111] S3. Using fractional programming theory, the optimization problem P1 is transformed into an optimization problem P2 with reduced complexity.

[0112] S4. Introduce auxiliary variables and decompose the optimization problem P2 into three decoupled subproblems. Use the alternating iterative optimization method to find the optimal solution of each subproblem in closed form.

[0113] In one feasible implementation, in order to solve the above non-convex optimization problem, it can be transformed into a much less complex problem using fractional programming theory, as shown in equations (8)-(11):

[0114]

[0115] st|φ j |=1,n=1,2,...,J (9)

[0116] tr(PP H )≤P max (10)

[0117] α k ≥0, k=1,2,...,K (11)

[0118] Where α = [α1,...,α] K ] TThese are auxiliary variables introduced when problem (P1) is transformed into (P2). This application decomposes the problem into three decoupled subproblems and uses an alternating iterative optimization method to find the optimal solution in closed form for each subproblem.

[0119] Optionally, α k The methods for finding the optimal solution include:

[0120] First, when variables P and When fixed, apply formula (8) to variable α k Taking the partial derivative and setting the result to 0, we can obtain α. k The optimal solution is α k After finding the optimal solution, substituting it into formula (8) transforms problem (P2) into the following formulas (12)-(14):

[0121]

[0122] st|φ j |=1,n=1,2,...,J (13)

[0123] tr(PP H )≤P max (14)

[0124] Therefore, P can be optimized alternately. To solve the problem (P3).

[0125] Alternatively, the methods for finding the optimal solution to P include:

[0126] For ease of representation, Represented as an equivalent channel, problem (P3) can be reformulated as equations (15)-(16):

[0127]

[0128] sttr(pp H )≤P max (16)

[0129] The above optimization problem is given α k and The time can be expressed by a quadratic transformation as follows: (17)-(18):

[0130]

[0131] sttr(pp H )≤P max (18)

[0132] Where β = [β1,...,β] K ] T It utilizes the auxiliary variable introduced during the second transformation. When P is given, formula (17) applies to β. k By taking the partial derivative and setting the result to 0, we can obtain β. k The optimal solution is shown in equation (19):

[0133]

[0134] Next, fix β k p k The optimal solution can be obtained by the Lagrange multiplier method, as shown in equation (20):

[0135]

[0136] Where μ≥0 is the Lagrange multiplier of the power constraint (18), and the optimal solution of μ is shown in equation (21):

[0137] μ op ={μ≥0:tr(PP H ) = P max} (twenty one)

[0138] Optionally, The methods for finding the optimal solution include:

[0139] For ease of representation, Rewrite equation (22):

[0140]

[0141] Where θ=[θ1,...,θ J ] T .definition Given α and P, the original problem can be written as equation (23) using quadratic programming:

[0142]

[0143] in The above problem can be expressed by the constraint relaxation method as follows: (24)-(25):

[0144]

[0145]

[0146] The above problem can be solved using the Lagrange dual decomposition method, as shown in equations (26)-(27):

[0147]

[0148]

[0149] Differentiating equation (26) with respect to θ and setting the result to 0, we can obtain the optimal solution for θ as equation (28):

[0150]

[0151] in, Substituting equation (28) into equation (26) yields equation (29):

[0152]

[0153] The above problem can be rewritten as a semidefinite programming problem using the existing Shure complement method and solved using the CVX toolkit, resulting in the following equations (30)-(31):

[0154]

[0155]

[0156] S5. Based on the optimal solution, the jointly optimized base station (BS) beamforming vector and IRS passive beamforming vector are obtained, and finally, the beam optimization method for intelligent reflector array element grouping is obtained.

[0157] In this embodiment of the invention, the proposed intelligent reflector grouping optimization model for wireless communication considers the impact of different numbers of IRS element groups and different numbers of users on the system downlink rate, and compares the average optimization time for different numbers of IRS element groups and the case of no IRS grouping. This can significantly reduce the model optimization time while ensuring the total downlink user rate.

[0158] like Figure 5 As shown, this embodiment of the invention provides a beam optimization system based on intelligent reflector array element grouping. This system is used to implement a beam optimization method based on intelligent reflector array element grouping. The system includes a base station, an intelligent reflector RIS, and multiple users. Figure 5 The diagram shown is a block diagram of a beam optimization system based on intelligent reflector array element grouping, wherein:

[0159] A base station is used to transmit signals to multiple users.

[0160] The Intelligent Reflector RIS (RIS) is used to group the reflector elements of the Intelligent Reflector RIS (IRS) using the Kronecker product method. The initial optimization problem is transformed into an optimization problem P1 of the grouped IRS reflector elements. Fractional programming theory is used to transform optimization problem P1 into a less complex optimization problem P2. Auxiliary variables are introduced, and optimization problem P2 is decomposed into three decoupled subproblems. An alternating iterative optimization method is used to find the optimal solution in closed form for each subproblem. Based on the optimal solution, the jointly optimized BS beamforming vector and IRS passive beamforming vector are obtained, finally yielding the beam optimization method for grouping the intelligent reflector elements.

[0161] Multiple users are used to receive signals transmitted from the base station.

[0162] Optionally, in the initial optimization problem, the received signal of the k-th user among multiple users is as shown in equation (1):

[0163]

[0164] in, Let M be the number of reflection elements in an IRS and the k-th user, and h be the channel between the IRS and the k-th user. M,k It is the channel between the Mth reflection element and the kth user, T is the transpose of the matrix, and H is the transpose conjugate of the matrix; Φ = ηdiag([θ1,...,θ) M ] T Let be the matrix representing the reflection coefficients of the IRS, where η∈[0,1] represents the on or off state of the M-th reflection element in the IRS. and This represents the reflection coefficient of the Mth array element; This is the channel between the BS and the IRS, where N is the number of antennas the base station is equipped with. This is the channel between BS and the Mth reflector element. All channels are assumed to be Saleh-Valenzuela millimeter-wave channels; P = [P1,...P K ] represents the beamforming matrix, where This represents the beamforming vector sent to the k-th user, where K is the number of users; s k This indicates that the signal sent by BS to the k-th user satisfies... k∈{1,2,...,K}; s represents the beamforming vector sent to the j-th user; j This indicates that the signal sent by BS to the j-th user satisfies... j∈{1,2,...,K};μ k This indicates that the mean is 0 and the variance is... Additive white Gaussian noise.

[0165] Alternatively, in the optimization problem of the grouped IRS reflection array elements, the received signal of the kth user among multiple users is as shown in equation (2):

[0166]

[0167] in, The matrix representing the phase shift factor after IRS grouping. This represents the phase shift factor of the Jth group, where J is the number of IRS element groups; This represents the channel between the IRS and the k-th user; This is the channel between the BS and the IRS.

[0168] Optionally, in the optimization problem of the grouped IRS reflection array elements, the signal-to-interference-plus-noise ratio (SINNR) for the k-th user among multiple users is shown in Equation (3):

[0169]

[0170] Optionally, the optimization problem P1 of the grouped IRS reflection array elements is shown in equations (4)-(6) below:

[0171]

[0172] st|φ j |=1,n=1,2,...,J (5)

[0173] tr(PP H )≤P max (6)

[0174] Where, |φ j |=1 indicates that the reflection amplitude of each group of reflection array elements in the IRS is set to 1; P max This indicates the maximum transmit power of the BS.

[0175] Alternatively, the optimization problem P2 with reduced complexity is shown in equations (7)-(10) below:

[0176]

[0177] st|φ j |=1,n=1,2,...,J (8)

[0178] tr(PP H )≤P max (9)

[0179] α k ≥0, k=1,2,...,K (10)

[0180] Where α=[α1,...,αK ] T These are auxiliary variables introduced.

[0181] Optionally, the smart reflective surface is further used for:

[0182] Let variables P and For the variable α in the above formula (6) fixed, k Taking the partial derivative, we get α. k The optimal solution;

[0183] α k Substituting the optimal solution into equation (6) above, we can transform optimization problem P2 into optimization problem P3.

[0184] Optionally, the smart reflective surface is further used for:

[0185] make For the equivalent channel, the optimization problem P3 is re-expressed as the optimization problem P4.

[0186] Given α k and Introduce auxiliary variable β = [β1,...,β] K ] T By performing a second transformation on optimization problem P4, we obtain optimization problem P5.

[0187] Given P, β can be obtained by taking the partial derivative with respect to the optimization problem P5. k The optimal solution.

[0188] Fixed β k p is obtained through the Lagrange multiplier method k The optimal solution.

[0189] Optionally, the smart reflective surface is further used for:

[0190] Given α and P, introduce auxiliary variables ρ = [ρ1,...,ρ K ] T The optimization problem P5 is transformed twice to obtain the optimization problem P6.

[0191] Given ρ, re-express the optimization problem P6 as the optimization problem P7.

[0192] The optimization problem P7 is re-expressed as optimization problem P8 using the constraint relaxation method.

[0193] According to the Lagrange dual decomposition method, the optimization problem P8 is re-expressed as the optimization problem P9.

[0194] Solving optimization problem P9, we finally obtain... The optimal solution.

[0195] In this embodiment of the invention, the proposed intelligent reflector grouping optimization model for wireless communication considers the impact of different numbers of IRS element groups and different numbers of users on the system downlink rate, and compares the average optimization time for different numbers of IRS element groups and the case of no IRS grouping. This can significantly reduce the model optimization time while ensuring the total downlink user rate.

[0196] Those skilled in the art will understand that all or part of the steps of the above embodiments can be implemented by hardware or by a program instructing related hardware. The program can be stored in a computer-readable storage medium, such as a read-only memory, a disk, or an optical disk.

[0197] The above description is only a preferred embodiment of the present invention and is not intended to limit the present invention. Any modifications, equivalent substitutions, improvements, etc., made within the spirit and principles of the present invention should be included within the protection scope of the present invention.

Claims

1. A beam optimization method based on intelligent reflector array element grouping, characterized in that, The method is implemented by a beam optimization system based on intelligent reflector array element grouping. The system includes a base station, an intelligent reflector RIS, and multiple users. The method includes: S1. Construct an initial optimization problem for signal transmission between the base station and multiple users via the intelligent reflective surface IRS; the initial optimization problem is to optimize the total rate of multiple users; In the initial optimization problem, the received signal of the kth user among multiple users is shown in equation (1) below: in, Let M be the channel between the IRS and the k-th user, and h be the number of reflection elements in the IRS. M,k It is the channel between the Mth reflection element and the kth user, T is the transpose of the matrix, and H is the transpose conjugate of the matrix; Φ = ηdiag([θ1,...,θ) M ] T Let be the matrix representing the reflection coefficients of the IRS, where η∈[0,1] represents the on or off state of the M-th reflection element in the IRS. and ) represents the reflection coefficient of the Mth array element; This is the channel between the BS and the IRS, where N is the number of antennas the base station is equipped with. This is the channel between BS and the Mth reflector element. All channels are assumed to be Saleh-Valenzuela millimeter-wave channels; P = [P1,...P K ] represents the beamforming matrix, where This represents the beamforming vector sent to the k-th user, where K is the number of users; s k This indicates that the signal sent by BS to the k-th user satisfies... s represents the beamforming vector sent to the j-th user; j This indicates that the signal sent by BS to the j-th user satisfies... μ k This indicates that the mean is 0 and the variance is... Additive white Gaussian noise; S2. The Kronecker product method is used to group the reflective elements of the intelligent reflective surface IRS; the initial optimization problem is transformed into an optimization problem P1 of the grouped IRS reflective elements; The optimization problem P1 of the grouped IRS reflection array elements is shown in equations (5)-(7) below: s.t.|φ j |=1,n=1,2,…,J (6) tr(PP H )≤P max (7) in, The matrix representing the phase shift factor after IRS grouping. γ represents the phase shift factor of the Jth group, where J is the number of IRS element groups, and γ k Let |φ| represent the signal-to-interference-plus-noise ratio (SIR) for the k-th user among multiple users. j |=1 indicates that the reflection amplitude of each group of reflection array elements in the IRS is set to 1; P max This indicates the maximum transmit power of the BS; S3. Using fractional programming theory, the optimization problem P1 is transformed into an optimization problem P2 with reduced complexity; The optimization problem P2 with reduced complexity is shown in equations (8)-(11) below: s.t.|φ j |=1,n=1,2,…,J (9) tr(PP H )≤P max (10) a k ≥0,k=1,2,...,K (11) Where α=[α1,...,α K ] T Auxiliary variables introduced; S4. Introduce auxiliary variables and decompose the optimization problem P2 into three decoupled subproblems. Use the alternating iterative optimization method to find the optimal solution of each subproblem in closed form. The decomposition of the optimization problem P2 into three decoupled subproblems includes: When variables P and When fixed, apply formula (8) to variable α k Taking the partial derivative and setting the result to 0, we obtain α. k The optimal solution is α k After finding the optimal solution, substitute it into formula (8) to transform problem (P2) into the following formulas (12)-(14): s.t.|φ j |=1,n=1,2,…,J (13) tr(PP H )≤P max (14) S5. Based on the optimal solution, the jointly optimized base station (BS) beamforming vector and IRS passive beamforming vector are obtained, and finally, the beam optimization method for intelligent reflector array element grouping is obtained.

2. The method according to claim 1, characterized in that, For the case of IRS array element grouping, assuming all array elements are divided into J groups, each group contains the same number of reflection array elements B, then the reflection array elements between (j-1)B+1 and jB belong to the j-th group. Assuming that all array elements in a group have the same phase shift factor, the IRS phase shift matrix can be re-expressed as the following equation (2): In the optimization problem of the grouped IRS reflection array elements, the received signal of the kth user among multiple users is as shown in equation (3): in, This represents the channel between the IRS and the k-th user; This is the channel between the BS and the IRS.

3. The method according to claim 2, characterized in that, In the optimization problem of the grouped IRS reflection array elements, the signal-to-interference-plus-noise ratio (SINNR) of the k-th user among multiple users is shown in the following equation (4):

4. The method according to claim 1, characterized in that, The methods for finding the optimal solution to P include: make To represent the equivalent channel, optimization problem P3 is reformulated as optimization problem P4; Given α k and Introduce auxiliary variable β = [β1,...,β] K ] T By performing a second transformation on optimization problem P4, we obtain optimization problem P5; Given P, β is obtained by taking the partial derivative with respect to the optimization problem P5. k The optimal solution; Fixed β k p is obtained through the Lagrange multiplier method k The optimal solution.

5. The method according to claim 4, characterized in that, The The methods for finding the optimal solution include: Given α and P, introduce auxiliary variables ρ = [ρ1,...,ρ K ] T By performing a second transformation on optimization problem P5, we obtain optimization problem P6; Given ρ, re-express the optimization problem P6 as the optimization problem P7; The optimization problem P7 is re-expressed as optimization problem P8 using the constraint relaxation method; According to the Lagrange dual decomposition method, the optimization problem P8 is re-expressed as the optimization problem P9; Solving the optimization problem P9, we finally obtain... The optimal solution.

6. A beam optimization system based on intelligent reflector array element grouping, the system being used to implement the beam optimization method based on intelligent reflector array element grouping as described in any one of claims 1-5, wherein the beam optimization system based on intelligent reflector array element grouping includes a base station, an intelligent reflector RIS, and multiple users; wherein: The base station is used to transmit signals to the plurality of users; The intelligent reflector RIS is used to group the reflector elements of the intelligent reflector IRS using the Kronecker product method. The initial optimization problem is transformed into an optimization problem P1 of the grouped IRS reflector elements. Fractional programming theory is used to transform the optimization problem P1 into a less complex optimization problem P2. Auxiliary variables are introduced, and the optimization problem P2 is decomposed into three decoupled subproblems. An alternating iterative optimization method is used to find the optimal solution in closed form for each subproblem. Based on the optimal solution, the jointly optimized BS beamforming vector and IRS passive beamforming vector are obtained, and finally, the beam optimization method for grouping the intelligent reflector elements is obtained. The multiple users are used to receive signals transmitted by the base station.

Images