User selection and beam forming method in multi-IRSs auxiliary MIMO communication system

Abstract

The invention belongs to the technical field of wireless communication, and particularly relates to a user selection and beam forming method in a multi-IRSs auxiliary MIMO communication system. The method comprises the following steps: constructing a multi-IRSs auxiliary MIMO communication system model; the method comprises the following steps: based on a multi-IRSs auxiliary MIMO communication system model, jointly optimizing a beamforming vector transmitted at a base station, IRSs phase shift and a user selection variable, maximizing the weighting and rate of a user, and constructing an optimization problem P1; based on a weighted 1-norm, implicitly fusing user selection into a beamforming vector, introducing an auxiliary variable by utilizing fractional programming, and representing the optimization problem P1 as an optimization problem P2; solving the optimization problem P2 to obtain an optimal beamforming vector, an optimal IRSs phase shift and a user selection strategy at the base station; according to the method, the channel is reconfigured by properly selecting the user and controlling the phase of the IRS element, so that the sum rate of the selected user is maximized, and the system performance is improved.

Description

Technical Field

[0001] The present invention belongs to the technical field of wireless communications, and in particular relates to a user selection and beamforming method in a multi-IRSs assisted MIMO communication system. Background Art

[0002] With the rapid development of wireless communication technology, people's expectations for it are increasing. Although the existing 5G system has met the needs of high-speed and low-latency communication to a certain extent, it has problems such as high hardware cost, high power consumption and high complexity, which brings many inconveniences to operators and users. The goal of the sixth generation of mobile communications (6G) is to meet higher requirements than 5G, such as massive access of communication equipment, ultra-high traffic, ultra-low latency and ultra-high reliability. As one of the most representative new technologies, the Intelligent Reflecting Surface (IRS) has been widely concerned because it can change the wireless transmission environment in real time, expand coverage, eliminate interference and enhance energy efficiency. Compared with other technologies in 6G networks, IRS has significant advantages such as small size, low energy consumption and easy deployment.

[0003] IRS is a passive device that adds a design dimension to wireless communication systems. It can intervene in wireless channels and achieve a "certain degree" of controllable channels, depending on the algorithm and coverage density. IRS can reduce the energy consumption of electromagnetic wave propagation, expand communication coverage, resist interference noise and improve spectrum utilization.

[0004] With the rapid development of global mobile Internet, the number of mobile broadband users has shown explosive growth. Massive communication devices are connected to the network at the same time. The main challenge to be solved is how to effectively allocate and manage network resources. However, most of the current research on IRS-assisted communication systems does not consider the simultaneous access of massive devices to the network, ignoring the situation that the number of users in hot spots is far greater than the number of antennas at the base station. It still considers that the base station in the cell can serve massive communication devices under the same time-frequency resource block. Although many schemes have jointly designed beamforming and user selection at the base station in cellular networks without IRS, these algorithms cannot be applied to IRS-assisted networks because for selected users, the channel needs to be reconfigured by controlling the amplitude and phase of the IRS elements. Therefore, a new user selection algorithm is needed in the IRS-assisted communication system to optimize system performance. On the other hand, when the direct link of the communication system is severely blocked and large-scale centralized deployment is difficult, it is particularly necessary to use multiple intelligent reflecting surfaces (IRSs) to improve system performance. In multi-IRSs-assisted MIMO communication networks, there are massive communication devices. How to jointly optimize user selection, beamforming at the base station, and phase shifting of IRSs remains a challenge. Summary of the invention

[0005] In view of the shortcomings of the prior art, the present invention proposes a user selection and beamforming method in a multi-IRSs-assisted MIMO communication system, the method comprising:

[0006] S1: Construct a multi-IRSs-assisted MIMO communication system model;

[0007] S2: Based on the multi-IRSs assisted MIMO communication system model, the transmit beamforming vector, IRSs phase shift and user selection variables at the base station are jointly optimized to maximize the weighted sum rate of the user and construct the optimization problem P1;

[0008] S3: Based on the weighted 1-norm, the user selection is implicitly integrated into the beamforming vector, and the optimization problem P1 is expressed as the optimization problem P2 by using fractional programming and introducing auxiliary variables;

[0009] S4: Solve the optimization problem P2 and obtain the optimal beamforming vector, optimal IRSs phase shift and user selection strategy at the base station.

[0010] Preferably, the multi-IRSs assisted MIMO communication system model includes: 1 base station, L IRSs and K single-antenna users, the base station is equipped with M antenna arrays, using a uniform linear array structure; each IRS is configured with N = N x N y The number of reflection units is configured in a uniform planar array, where Nx and N y are the number of reflection units in the horizontal direction and the vertical direction respectively, and N is the total number of reflection units.

[0011] Preferably, the optimization problem P1 is expressed as:

[0012]

[0013] Among them, a k Indicates whether the kth user is selected by the base station, Θ l represents the diagonal reflection matrix of the lth IRS, w k represents the beamforming vector sent by the base station to the kth user, ω k represents the rate weight of k users, K represents the total number of users in the network, R k represents the achievable rate of the kth user, P represents the maximum transmit power of the base station, M represents the number of antennas at the base station, and θ l,n represents the reflection coefficient of the nth reflection unit of the lth IRS, represents the set of users in the network, represents the set of IRSs in the network, Represents the set of reflection units for each IRS.

[0014] Preferably, the optimization problem P2 is expressed as:

[0015]

[0016] Where η=[η1,…,η K ] T represents auxiliary variables, Θ l represents the diagonal reflection matrix of the lth IRS, w k represents the beamforming vector sent by the base station to the kth user, ω k represents the rate weight of k users, K represents the total number of users in the network, and γ k represents the signal-to-interference-to-noise ratio of the kth user downlink, P represents the maximum transmit power of the base station, μ k For ||w k || 2 The positive weight of M represents the number of antennas at the base station, θ l,n represents the reflection coefficient of the nth reflection unit of the lth IRS, represents the set of users in the network,

[0017] represents the set of IRSs in the network, Represents the set of reflection units for each IRS.

[0018] Preferably, the process of solving the optimization problem P2 includes:

[0019] S41: Given the beamforming vector at the base station and the IRSs phase shift matrix, solve the optimization problem P2 to obtain the value of the auxiliary variable;

[0020] S42: Given the IRSs phase shift, the optimization problem P2 is expressed as the optimization problem P3 according to the value of the auxiliary variable; solving the optimization problem P3, the beamforming vector at the base station and the user selection strategy are obtained;

[0021] S43: According to the values ​​of the beamforming vector and the auxiliary variables at the base station, the optimization problem P2 is expressed as the optimization problem P4; the optimization problem P4 is solved to obtain the phase shift of the IRSs;

[0022] S44: Repeat steps S41 to S43 until the optimization problem P2 converges, and the optimal beamforming vector, optimal IRSs phase shift and user selection strategy at the base station are obtained.

[0023] Furthermore, the value of the auxiliary variable is the signal to interference and noise ratio of the user downlink.

[0024] Furthermore, the optimization problem P3 is expressed as:

[0025]

[0026] Among them, w k represents the beamforming vector sent by the base station to the kth user, α k represents the intermediate parameter of the kth user, represents the combined channel of the kth user, i.e., the direct channel plus the reflected channel, w i represents the beamforming vector sent by the base station to the i-th user, represents the noise power at the user, K represents the total number of users in the network, P represents the maximum transmit power of the base station, M represents the number of antennas at the base station, and μ k For ||w k || 2 The positive weight of Represents a collection of users in the network.

[0027] Furthermore, the optimization problem P4 is expressed as:

[0028]

[0029] Among them, α k represents the intermediate parameter of the kth user, e k,k represents the product of the direct channel of the kth user and the beamforming vector of the kth user, e k,irepresents the product of the direct channel of the kth user and the beamforming vector of the ith user, u represents the reflection coefficient combination matrix of IRSs, d k,i represents the product of the reflection channel of the kth user and the beamforming vector of the i-th user, K represents the total number of users in the network, θ l,n represents the reflection coefficient of the nth reflection unit of the lth IRS, represents the noise power at the user, represents the set of IRSs in the network, Represents the set of reflection units for each IRS

[0030] The beneficial effects of the present invention are as follows: the present invention proposes a user selection and beamforming design method for a multi-IRSs-assisted MIMO communication system, implicitly integrating the user selection variable into the beamforming vector at the base station, which is used to solve the situation where there are a large number of users in the cell edge area and the number of users is much larger than the number of antennas at the base station. The present invention reconfigures the channel by appropriately selecting users and controlling the phase of the IRS elements, thereby maximizing the sum rate of the selected users, thereby improving the system performance. BRIEF DESCRIPTION OF THE DRAWINGS

[0031] Figure 1 It is a flow chart of the method for user selection and beamforming in a multi-IRSs assisted MIMO communication system of the present invention;

[0032] Figure 2 It is a schematic diagram of a MIMO communication system model based on multiple IRSs assistance in the present invention;

[0033] Figure 3 This is a graph showing how the total rate of a user changes with the number of iterations during the solution process of a preferred embodiment of the present invention. DETAILED DESCRIPTION

[0034] The following will be combined with the drawings in the embodiments of the present invention to clearly and completely describe the technical solutions in the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0035] The present invention proposes a user selection and beamforming method in a multi-IRSs assisted MIMO communication system. Figure 1 As shown, the method includes the following contents:

[0036] S1: Construct a multi-IRSs-assisted MIMO communication system model.

[0037] like Figure 2As shown in the figure, in a multiple IRSs assisted MIMO communication system, including a base station (BS), L IRSs and K single-antenna users, the base station is equipped with M antenna arrays, using a uniform linear array (ULA) structure, and each IRS is configured with N = N x N y The number of reflective units is configured in a uniform planar array (UPA), where N x and N y are the number of reflection units in the horizontal and vertical directions, respectively. Assuming a crowded hotspot scenario, there are many cell edge users, whose number is much larger than the number of antennas at the base station, that is, K>>M. Assuming that the direct link from the base station to the user is not blocked by obstacles, the user can receive the superimposed signal from the base station-user direct link and the base station-IRSs-user reflection link. Define the number of IRSs, the number of users, and the number of reflection units in each IRS as Definition Θ l =diag(θ l,1 ,…,θ l,N ) is the diagonal reflection matrix of the lth IRS, where φ l,n ∈[0,2π) represents the phase shift of the nth reflection unit of the lth IRS. Denote the channel gain from the base station to the lth IRS, the channel gain from the base station to user k, and the channel gain from the lth IRS to user k respectively. Define a binary selection vector a={a1,…,a K} T :

[0038]

[0039] The signal vector transmitted at the base station is x∈C N×1 It can be expressed as:

[0040]

[0041] where s k is the information sent by the base station to the kth user and satisfies E{|s k | 2}=1, is the beamforming vector sent by the BS to the kth user and satisfies P is the maximum transmission power of the base station. In each coherent time block, the signal y received by the kth user k It can be expressed as:

[0042]

[0043] in, represents the additive white Gaussian noise at the receiver of user k, represents the noise variance of the k-th user. Based on the above formula, the signal to interference noise ratio of the k-th user can be obtained as:

[0044]

[0045] Furthermore, the achievable rate R of the kth user is k It can be expressed as:

[0046] R k =log2(1+γ k )

[0047] S2: Based on the multi-IRSs-assisted MIMO communication system model, the transmit beamforming vector, IRSs phase shift and user selection variables at the base station are jointly optimized to maximize the weighted sum rate of the user and construct the optimization problem P1.

[0048] The present invention constructs the optimization problem P1 by jointly optimizing the transmit beamforming vector, IRSs phase shift and user selection variable at the base station to maximize the weighted sum rate of the selected user. The optimization problem P1 is expressed as:

[0049]

[0050] Among them, a k Indicates whether the kth user is selected by the base station, Θ l represents the diagonal reflection matrix of the lth IRS, w k represents the beamforming vector sent by the base station to the kth user, ω k represents the rate weight of k users, K represents the total number of users in the network, R k represents the achievable rate of the kth user, P represents the maximum transmit power of the base station, M represents the number of antennas at the base station, and θ l,n represents the reflection coefficient of the nth reflection unit of the lth IRS, represents the set of users in the network, represents the set of IRSs in the network, Represents the set of reflection units for each IRS.

[0051] S3: Based on the weighted 1-norm, the user selection is implicitly integrated into the beamforming vector. Fractional programming is used to introduce auxiliary variables, and the optimization problem P1 is expressed as the optimization problem P2.

[0052] The base station only sends information to the selected users, while other users do not receive any information from the base station within the coherent time block. This shows that when a k =1, we have ||w k || 2 ≥0; when ak = 0, there is

[0053] ||w k || 2 = 0. Therefore, the above optimization problem P1 can be reformulated as P1.1:

[0054]

[0055] in, It means that when ||w k || 2 >0, there is In other cases, The 0-norm of a vector is the number of non-zero elements in the vector, so:

[0056]

[0057] Furthermore, you can use w k Implicitly represents a k , the signal-to-interference-noise ratio of the kth user can be restated as:

[0058]

[0059] Furthermore, the non-convex 0-norm can be approximately expressed as a weighted 1-norm:

[0060]

[0061] Among them, y m is the mth element of vector y, μ m is relative to y m In order to express the above formula more accurately, it is very important to choose an appropriate weight μ. The constraints can be further expressed as:

[0062]

[0063] Among them, μ k For ||w k || 2 The positive weight of , which satisfies the weight of the approximate representation It can be calculated at the i-th iteration:

[0064]

[0065] Among them, stability is guaranteed by ε>0, and needs to be set to be slightly smaller than By setting weights that are inversely proportional to the power required to send to each user, when the power is smaller, the resulting weight is larger, and users with larger weights will not be selected to satisfy the constraint.

[0066] Based on the above transformation, the optimization problem P1.1 can be transformed into P1.2:

[0067]

[0068] Through fractional programming, we introduce auxiliary variables η=[η1,…,η K ] T , the above optimization problem can be expressed as P2:

[0069]

[0070] The above optimization problem is still a problem with highly coupled optimization variables, which can be optimized alternately by solving each sub-problem.

[0071] S4: Solve the optimization problem P2 and obtain the optimal beamforming vector, optimal IRSs phase shift and user selection strategy at the base station.

[0072] S41: Given the beamforming vector at the base station and the IRSs phase shift matrix, solve the optimization problem P2 and obtain the value of the auxiliary variable.

[0073] When given w k and θ l When , the optimal solution η can be easily obtained by the first-order condition opt :

[0074]

[0075] Among them, η opt represents the optimal solution of η, γ k represents the signal to interference and noise ratio of the k-th user downlink.

[0076] S42: Given the IRSs phase shift, the optimization problem P2 is expressed as the optimization problem P3 according to the value of the auxiliary variable; the optimization problem P3 is solved to obtain the beamforming vector at the base station and the user selection strategy.

[0077] When η and Θ are given l When we need to consider only the last term of the objective function to optimize w k , so the subproblem that implicitly includes the user-selected beamforming can be equivalently expressed as P3:

[0078]

[0079] Among them, α k =ω k(1+η k ) is the intermediate parameter, ω k represents the weight value of the kth user, Represents the combined channel of the kth user, i.e., the direct channel plus the reflected channel. represents the channel gain from the base station to the lth IRS, represents the channel gain from the base station to user k, represents the channel gain from the lth IRS to user k, Θ l represents the diagonal reflection matrix of the lth IRS, Represents the noise power at the user.

[0080] The above problem is still a non-convex problem. Through further secondary transformation, the above optimization problem can be rewritten as P3.1:

[0081]

[0082] Where ξ=[ξ1,…,ξ K ] T As the auxiliary variable introduced, alternating optimization can be further used to solve the above problem.

[0083] For a given w k , ξ can be solved by solving the first-order conditions to find the optimal solution:

[0084]

[0085] When ξ is given, in order to solve w k , first define The problem can then be reformulated as P3.2:

[0086]

[0087] in,

[0088] Obviously, the matrix A and the matrix M are semi-positive definite matrices. The above optimization problem is a convex quadratic constrained quadratic programming problem, which can be solved by the Lagrangian dual method. The Lagrangian function can be expressed as:

[0089]

[0090] Among them, λ≥0 and v≥0 are introduced dual variables, so the Lagrangian dual problem can be expressed as P3.3:

[0091]

[0092] The solution of the Lagrangian dual problem can be divided into two stages. First, when the dual variables λ and υ are given, solve the problem about Second, when the variable When , λ and υ are updated by solving an external problem. These variables are updated iteratively until convergence conditions are met.

[0093] For about To solve the internal problem of , we can solve the first-order condition to get the optimal solution:

[0094]

[0095] For the external problem, in each iteration, the subgradient method can be used to update λ and υ:

[0096]

[0097] in, Indicates the update step size.

[0098] After the beamforming vector at the base station is obtained, since the beamforming vector at the base station implicitly includes the user selection strategy, the user selection strategy can be obtained from the beamforming vector at the base station.

[0099] S43: According to the values ​​of the beamforming vector and the auxiliary variables at the base station, the optimization problem P2 is expressed as the optimization problem P4; the optimization problem P4 is solved to obtain the phase shift of the IRSs.

[0100] For a given η and w k , we only need to consider the last term of the objective function to optimize Θ l , for the convenience of processing, define

[0101] and The IRSs phase shift optimization subproblem can be expressed as P4:

[0102]

[0103] Similarly, through further secondary transformation, the above optimization problem can be further rewritten as P4.1:

[0104]

[0105] Where ρ=[ρ1,…,ρ K ] T As the auxiliary variable introduced, alternating optimization is further used to solve the above problem. For a given Θ l ,ρ can be optimally solved by solving the first-order condition:

[0106]

[0107] For a given ρ, the optimization problem can be reformulated as P4.2:

[0108]

[0109] in, Obviously, Q is a semi-positive definite matrix. Obviously, the above optimization problem is about a concave function of u. The non-convexity of the optimization problem is caused by the constraints. In order to solve the above problem, the present invention proposes a low-complexity method based on the alternating direction method of multipliers (ADMM) to solve u. According to the standard form of ADMM, an auxiliary variable q is introduced and converted to P4.3:

[0110]

[0111] Among them, g(θ l,n ) is the characteristic function, that is, when |θ l,n |=1, g(θ l,n )=0, otherwise, g(θ l,n )=∞. Furthermore, its corresponding Lagrangian augmented function is:

[0112]

[0113] Among them, χ R =[χ R1,1 ,…,χ R1,N ,…,χ RL,1 ,…,χ RL,N ] T and χ I =[χ I1,1 ,…,χ I1,N ,…,χ IL,1 ,…,χ IL,N ] T The Lagrangian coefficients corresponding to the two restrictions of Re{qu}=0 and Im{qu}=0 respectively, τ>0 indicates a penalty parameter, is the characteristic function, that is, when |θ l,n When |=1, In other cases, Furthermore, its dual problem is P4.4:

[0114]

[0115] Solve the dual problem using the ADMM algorithm:

[0116]

[0117] in, t is the number of iterations. It can be seen that it is essentially an alternating optimization method that optimizes a single variable by fixing the other variables.

[0118] For the problem that needs to be solved in the first step, extract the items related to u in G. The problem can be written as:

[0119]

[0120] By adding an irrelevant constant term, the objective function can be rewritten as:

[0121]

[0122] Observing the objective function above, we find that in order to maximize the objective function, due to the existence of the indicative function, u must satisfy |θ l,n |=1, otherwise the value of the objective function will be -∞. Looking at the first item again, the physical meaning it represents is that we need to find the vector The nearest vector. So the solution to u is the definition of the projection operation and satisfies |θ l,n |=1, let So we have:

[0123]

[0124] For the problem that needs to be solved in the second step, this is a convex function about q, and the closed-form solution can be obtained by differentiation:

[0125]

[0126] At this point, ADMM can be used to solve the problem.

[0127] S44: Repeat steps S41 to S43 until the optimization problem P2 converges, and the optimal beamforming vector, optimal IRSs phase shift and user selection strategy at the base station are obtained.

[0128] The system executes the optimal beamforming vector, optimal IRSs phase shift and user selection strategy at the base station to obtain the maximum user sum rate, thereby improving system performance.

[0129] Evaluation of the present invention:

[0130] The present invention is simulated, and the simulation scenario considers a three-dimensional coordinate single-cell network system, where the BS is located at (0m, 0m, 25m), and the lth IRS is located at (d1(cos(2πl / L)), d1(sin(2πl / L)), 10m). K users are evenly distributed in a circular area with a radius of [R1, R2], and the height of all users is set to 0m. Considering small-scale fading, all channels use the following channel model:

[0131]

[0132] Among them, β is the Rice factor, which controls the size relationship between the LOS component (mirror component) and the NLOS component (scattering component), G LoS and G NLoS They are the deterministic LOS component and NLOS component respectively.

[0133] Considering the path loss, the path loss model is assumed to be:

[0134]

[0135] Among them, C0 is the path loss when the reference distance D0 = 1m, d is the link distance, and α is the path loss factor. The number of IRS reflection units N and the number of BS transmitting antennas M are set according to the simulation needs. Assuming that the noise power at each user is the same, Table 1 gives the parameter values ​​for system initialization.

[0136] Table 1 System initialization parameters

[0137]

[0138]

[0139] Set the number of reflection units of each IRSs N = 30, Nx = 5, the number of BS antennas M = 8, d1 = 80m, R1 = 60m, R2 = 100m, set the maximum number of iterations to 30, and initialize the beamforming of the base station to Maximum Ratio Transmission (MRT). The IRSs phase shift is initialized as a random phase shift, and the rate weight of each user is set to ω k =1, simulation as Figure 3 As shown in the figure, the total rate selected by users in the proposed algorithm gradually increases and tends to be stable. Compared with the case without IRS, the performance is improved by about 34%.

[0140] In summary, the present invention implicitly combines the user selection strategy into the beamforming vector of the base station, thereby simplifying the problem, reducing the complexity of the problem, and maximizing the user sum rate under the premise of a given base station transmission power, thereby improving system performance.

[0141] The above embodiments further illustrate the purpose, technical solutions and advantages of the present invention in detail. It should be understood that the above embodiments are only preferred implementation modes of the present invention and are not intended to limit the present invention. Any modifications, equivalent substitutions, improvements, etc. made to the present invention within the spirit and principles of the present invention should be included in the protection scope of the present invention.

Claims

1. A method for user selection and beamforming in a multi-IRSs assisted MIMO communication system, characterized in that: include: S1: Construct a multi-IRSs-assisted MIMO communication system model; S2: Based on the multi-IRSs assisted MIMO communication system model, the transmit beamforming vector, IRSs phase shift and user selection variables at the base station are jointly optimized to maximize the weighted sum rate of the user and construct the optimization problem P1; S3: Based on the weighted 1-norm, the user selection is implicitly integrated into the beamforming vector, and the optimization problem P1 is expressed as the optimization problem P2 by using fractional programming and introducing auxiliary variables; S4: Solve the optimization problem P2 and obtain the optimal beamforming vector, optimal IRSs phase shift and user selection strategy at the base station.

2. The method for user selection and beamforming in a multi-IRSs assisted MIMO communication system according to claim 1, characterized in that: The multi-IRSs assisted MIMO communication system model includes: 1 base station, L IRSs and K single-antenna users, the base station is equipped with M antenna arrays, using a uniform linear array structure; each IRS is configured with N = N x N y The number of reflection units is configured in a uniform planar array, where N x and N y are the number of reflection units in the horizontal direction and the vertical direction respectively, and N is the total number of reflection units.

3. The method for user selection and beamforming in a multi-IRSs assisted MIMO communication system according to claim 1, characterized in that: The optimization problem P1 is expressed as: Among them, a k Indicates whether the kth user is selected by the base station, Θ l represents the diagonal reflection matrix of the lth IRS, w k represents the beamforming vector sent by the base station to the kth user, ω k represents the rate weight of k users, K represents the total number of users in the network, R k represents the achievable rate of the kth user, P represents the maximum transmit power of the base station, M represents the number of antennas at the base station, and θ l，n represents the reflection coefficient of the nth reflection unit of the lth IRS, represents the set of users in the network, represents the set of IRSs in the network, Represents the set of reflection units for each IRS.

4. The method for user selection and beamforming in a multi-IRSs assisted MIMO communication system according to claim 1, characterized in that: The optimization problem P2 is expressed as: Where η=[η1,…,η K ] T represents auxiliary variables, Θ l represents the diagonal reflection matrix of the lth IRS, w k represents the beamforming vector sent by the base station to the kth user, ω k represents the rate weight of k users, K represents the total number of users in the network, and γ k represents the signal-to-interference-to-noise ratio of the kth user downlink, P represents the maximum transmit power of the base station, μ k For ||w k || 2 The positive weight of M represents the number of antennas at the base station, θ l，n represents the reflection coefficient of the nth reflection unit of the lth IRS, represents the set of users in the network, represents the set of IRSs in the network, Represents the set of reflection units for each IRS.

5. The method for user selection and beamforming in a multi-IRSs assisted MIMO communication system according to claim 1, characterized in that: The process of solving the optimization problem P2 includes: S41: Given the beamforming vector at the base station and the IRSs phase shift matrix, solve the optimization problem P2 to obtain the value of the auxiliary variable; S42: Given the IRSs phase shift, the optimization problem P2 is expressed as the optimization problem P3 according to the value of the auxiliary variable; solving the optimization problem P3, the beamforming vector at the base station and the user selection strategy are obtained; S43: According to the values ​​of the beamforming vector and the auxiliary variables at the base station, the optimization problem P2 is expressed as the optimization problem P4; the optimization problem P4 is solved to obtain the phase shift of the IRSs; S44: Repeat steps S41 to S43 until the optimization problem P2 converges, and the optimal beamforming vector, optimal IRSs phase shift and user selection strategy at the base station are obtained.

6. The method for user selection and beamforming in a multi-IRSs assisted MIMO communication system according to claim 5, characterized in that: The value of the auxiliary variable is the signal to interference and noise ratio of the user downlink.

7. The method for user selection and beamforming in a multi-IRSs assisted MIMO communication system according to claim 5, characterized in that: The optimization problem P3 is expressed as: Among them, w k represents the beamforming vector sent by the base station to the kth user, α k represents the intermediate parameter of the kth user, represents the combined channel of the kth user, i.e., the direct channel plus the reflected channel, w i represents the beamforming vector sent by the base station to the i-th user, represents the noise power at the user, K represents the total number of users in the network, P represents the maximum transmit power of the base station, M represents the number of antennas at the base station, and μ k For ||w k || 2 The positive weight of Represents a collection of users in the network.

8. The method for user selection and beamforming in a multi-IRSs assisted MIMO communication system according to claim 5, characterized in that: The optimization problem P4 is expressed as: Among them, α k represents the intermediate parameter of the kth user, e k，k represents the product of the direct channel of the kth user and the beamforming vector of the kth user, e k，i represents the product of the direct channel of the kth user and the beamforming vector of the ith user, u represents the reflection coefficient combination matrix of IRSs, d k，i represents the product of the reflection channel of the kth user and the beamforming vector of the i-th user, K represents the total number of users in the network, θ l，n represents the reflection coefficient of the nth reflection unit of the lth IRS, represents the noise power at the user, represents the set of IRSs in the network, Represents the set of reflection units for each IRS.

Images