Wireless communication network optimization method and system based on multi-intelligent reflector grouping

Abstract

The invention discloses a wireless communication network optimization method and system based on multi-intelligent-reflecting-surface grouping, and belongs to the field of intelligent-reflecting-surface-assisted wireless communication, and the method comprises the steps: building a multi-IRS-assisted wireless communication network according to the coverage characteristics of the equivalent cascade link information of a plurality of IRSs; according to the statistical property of the optimal beam domain channel state information, intelligent pre-grouping is carried out on reflection units of a plurality of IRSs, and each group of units share the same reflection coefficient; analyzing array performance gains of different pilot frequency overhead under a single-user single-antenna cascade channel; a system capacity of a multi-IRS intelligent packet assisted wireless communication network under a Rice channel is evaluated. According to the method provided by the invention, the number of pilots for estimating the channels related to the IRS can be greatly reduced while the multi-user communication service quality is improved.

Description

Technical Field

[0001] The present invention relates to the technical field of intelligent reflecting surface assisted wireless communication, and particularly to a method and system for optimizing a wireless communication network based on multi-intelligent reflecting surface grouping. Background Art

[0002] In the past few decades, the wireless communication field has been committed to proposing various wireless technologies to improve system capacity, but the wireless channel has traditionally been considered uncontrollable. In recent years, the intelligent reflecting surface (IRS) has emerged as a new technology that actively changes the wireless environment through reconfigurable and intelligent signal reflection. The intelligent reflecting surface usually consists of a planar array composed of a large number of reconfigurable passive elements, which can independently reflect electromagnetic signals to collaboratively reconstruct the signal propagation path. With advantages such as high array gain, low power consumption, and low hardware cost, it can improve system capacity, expand coverage, and increase spectral efficiency, and is expected to promote the development of wireless communication in future 6G networks. In previous studies, it was mostly assumed that the channel state information of all links in the intelligent reflecting surface assisted wireless communication system was completely known. In the single-user single-input single-output scenario, the number of intelligent reflecting surface elements has a square relationship with the asymptotic channel gain, which enables the reduction of the transmit power without sacrificing the signal-to-noise ratio, and then more elements can be used to improve the performance. However, the passive nature of the intelligent reflecting surface makes its elements lack radio frequency links, and accurate channel state information needs to be obtained through the base station or access point, which brings high signaling overhead and high-complexity beamforming problems. At the same time, the "multiplicative fading" effect introduced by the intelligent reflecting surface will greatly increase the equivalent path loss of the cascaded link. Therefore, a very large-scale intelligent reflecting surface is required to mitigate this effect. However, the existing methods for optimizing the reflection coefficient are difficult to adapt to a very large-scale intelligent reflecting surface, and although the active intelligent reflecting surface integrated with an amplifier can alleviate the demand for the number of elements, it has problems such as high power consumption and complex hardware architecture. Therefore, from the perspective of practical applications, there is an urgent need for an optimization scheme for a low-pilot-overhead wireless communication network based on multi-intelligent reflecting surface intelligent grouping that achieves an excellent compromise between pilot overhead and very large-scale intelligent reflecting surface optimization. Summary of the Invention

[0003] The present invention provides a method and system for optimizing a wireless communication network based on multi-intelligent reflecting surface grouping, which can not only simultaneously improve performance and reduce pilot overhead, but also has a theoretically supported convergence characteristic, and solves the problem of high pilot overhead in a multi-intelligent reflecting surface assisted wireless communication network.

[0004] In a first aspect embodiment of the present invention, a method for optimizing a wireless communication network based on multi-intelligent reflecting surface grouping is provided, including the following steps:

[0005] Step 1: Design the location deployment of multiple intelligent reflecting surfaces (IRSs) and construct a wireless communication system assisted by multiple IRSs.

[0006] Step 2: Based on the wireless communication system assisted by multiple IRSs, combine the location information of users, the base station (BS), and multiple IRSs, project the statistical channel state matrix onto the optimal beam domain, and use fractional programming, the alternating direction method of multipliers (ADMM), the max-min method, and the successive convex approximation (SCA) method to transform the non-convex objective function and non-convex constraint problems with unit modulus constraints and 0-1 grouping strategy integer constraints in the non-convex and multi-variable coupled resource allocation problem aiming at maximizing the system capacity into a joint optimization problem of the base station's active beamforming, the passive grouping precoding of multiple IRSs, and the grouping strategy. Solve the joint optimization problem to obtain the optimal grouping strategy based on the statistical cascaded channel state information.

[0007] Step 3: According to the optimal grouping strategy based on the statistical cascaded channel state information, pre-group the reflecting elements of the multiple IRSs, and use the fractional programming, ADMM, max-min, and SCA methods proposed in Step 2 to transform the system capacity maximization problem after grouping the reflecting elements of the multiple IRSs into a joint optimization problem of the base station's active beamforming and the grouped passive precoding of the multiple IRSs. Solve the joint optimization problem to obtain the maximum capacity of the base station and the multiple IRSs' active and passive precoding and the multi-user system in the real-time transmission stage.

[0008] Optionally, in an embodiment of the present invention, Step 1 specifically includes:

[0009] Design the location layout of multiple IRSs by analyzing the location distributions of the base station, IRSs, and users and combining the environmental information.

[0010] The wireless communication system assisted by multiple IRSs includes: a base station with M antennas, D IRSs each consisting of N reflecting elements, and K single-antenna users; the uplink and downlink cascaded link transmissions follow the time-division duplex (TDD) protocol. Assume that the channel state information in the downlink cascaded link is estimated from the uplink cascaded link training. The signals received by user k from the BS-user k and BS-D-IRSs-user k channel transmissions are expressed as:

[0011]

[0012] where (·) H denotes the conjugate operation, and respectively represent the links from the BS to D IRSs, the links from D IRSs to user k, and the link from the BS to user k, and z kDenote the additive Gaussian white noise with zero mean and variance acting on user k, and use to represent the complex signal s transmitted by the BS, j denote the symbol with zero mean and unit variance sent by the BS to user j, represent the corresponding beamforming complex vector, and use to denote the equivalent reflection diagonal matrix after intelligent grouping, represent the adjustable equivalent reflection coefficient vector after intelligently dividing into Q groups, where denote the optimizable grouping strategy, where G q,n ∈{0,1} indicates whether the nth reflection unit is assigned to the qth group.

[0013] Optionally, in an embodiment of the present invention, step 2 specifically includes:

[0014] Step 201, transmission phase, the signal-to-noise ratio SINR at user k is:

[0015]

[0016] where, and respectively represent the reflection cascaded link from the base station to the multi-intelligent reflecting surface to user k and the superimposed channel of the link from the base station to user k;

[0017] Step 202, under the constraints of the maximum transmit power of the base station, the transmit beamforming constraint of the base station, the unit modulus constraint of the passive reflection precoding of the multi-intelligent reflecting surface, and the grouping strategy constraint, with the goal of maximizing the sum rate of the system, construct the optimization problem P0:

[0018]

[0019] s.t.G q,n ∈{0,1}

[0020]

[0021] Q≤Q0<<LN

[0022]

[0023] |v q |=1

[0024] where, represents the overall transmit beamforming vector of K users, Q0 represents the number of pilot frequencies allocated to the multi-IRSs side by the system, and is also the dimension of the equivalent cascaded channel related to the multi-intelligent reflecting surface or the maximum number of groups of the multi-intelligent reflecting surface reflection units, || || 2is the square of the vector two-norm, is the maximum transmit power of the base station;

[0025] Step 203: Combine the fractional programming method, the alternating direction method of multipliers, the maximization-minimization method, and the successive convex approximation method to transform the non-convex objective function and the non-convex constraint problem with unit modulus constraint and 0-1 grouping strategy integer constraint in the non-convex and multi-variable coupled resource allocation problem aiming at maximizing system capacity into a joint optimization problem of base station active beamforming, multi-intelligent reflecting surface passive grouping precoding optimization, and grouping strategy.

[0026] Let and represent the statistical channel state information of the cascaded channel and the direct channel of user k respectively. By fixing the base station transmit beamforming vector and the multi-intelligent reflecting surface reflection precoding vector, the original optimization problem P0 is transformed into an optimal grouping strategy optimization problem P1 based on statistical cascaded channel state information:

[0027]

[0028] s.t.G q,n ∈{0,1}

[0029]

[0030] where and represent the introduced auxiliary variables. The optimization problem P1 is a 0-1 integer programming problem. By relaxing the non-convex constraints in the L 21 regularized maximization problem, the optimization problem P1 is transformed into an optimizable quadratic programming problem. By introducing the auxiliary variable the optimization problem P1 is transformed into the optimization problem P 1.1 :

[0031]

[0032] where G n represents the nth column vector of the intelligent grouping strategy G to be optimized, is the adjoint matrix of ξ k is the real part of the parameter, is the adjustable equivalent reflection coefficient vector after the intelligent division into Q groups under the corresponding statistical cascaded channel, is the beamforming complex vector under the corresponding statistical cascaded channel. The optimization problem P 1.1 is a linearly constrained quadratic programming problem, and the optimal grouping strategy is obtained by using the Lagrange multiplier method.

[0033] ​Optionally, in an embodiment of the present invention, step 3 specifically includes:

[0034] Step 301, combining the proposed fractional programming method, alternating direction method of multipliers, maximize-minimize method, and successive convex approximation method, transforms the original optimization problem P0 into the following optimization problem P2:

[0035]

[0036] |v q | = 1

[0037] The optimization problem P2 contains four variables to be optimized. Borrowing the idea of alternating optimization, the optimization problem P2 is transformed into three optimization sub-problems to solve the auxiliary variables and ξ, the active beamforming vector w, and the intelligent grouped reflection precoding v;

[0038] Step 302, fix the active beamforming vector w and the intelligent grouped reflection precoding v, and then optimize the auxiliary variables and ξ:

[0039] First, by fixing the auxiliary variables the active beamforming vector w and the intelligent grouped reflection precoding v, the problem P2 to be optimized is transformed into the following optimization problem P 2.1 to solve the auxiliary variable ξ:

[0040]

[0041] Because the optimization problem P 2.1 is an unconstrained quadratic programming problem, by setting the derivative of the objective function equal to zero, the optimal auxiliary variable ξ is derived:

[0042]

[0043] where and

[0044] Step 303, fix the auxiliary variable ξ, the active beamforming vector w, and the intelligent grouped reflection precoding The optimization problem P2 can be transformed into the following optimization problem P 2.2 to solve the auxiliary variable

[0045]

[0046] Because the optimization problem P 2.2 is an unconstrained convex optimization problem, by setting the derivative of the objective function equal to zero, the optimal auxiliary variable is derived:

[0047]

[0048] Among them,

[0049] According to and 's formula, the auxiliary variables and the solution of ξ are further analyzed and decoupled. Then, the optimal solution containing only the channel state information and the active and passive precoding is expressed as:

[0050]

[0051] Among them, θ k = arg(a k ). and

[0052] Step 304, fix the auxiliary variables and ξ and the intelligent grouped reflection precoding v, and then optimize the active beamforming vector w:

[0053] First of all, by fixing the auxiliary variables and ξ and the intelligent grouped reflection precoding v, the optimization problem P2 is transformed into the optimization problem P 2.3 to solve the active beamforming vector w:

[0054]

[0055] Among them, and I K is a K×K dimensional identity matrix;

[0056] Because the optimization problem P 2.3 is a standard quadratic constrained quadratic programming problem, by using the Lagrange multiplier method, the optimal active beamforming vector is obtained:

[0057] w opt =(L + λI KM ) -1 ζ

[0058] Among them, λ represents the Lagrange multiplier, which is obtained by grid search, and I KM is a KM×KM dimensional identity matrix.

[0059] Step 305, fix the auxiliary variables and ξ and the active beamforming vector w, and then optimize the intelligent grouped reflection precoding v:

[0060] First of all, by fixing the auxiliary variables With ξ and the active beamforming vector w, the optimization problem P2 is transformed into the following optimization problem P 2.4 to solve the intelligent grouped reflection precoding v :

[0061]

[0062] s.t. |v q | = 1

[0063] where

[0064] By borrowing the max - min algorithm, the optimization problem P is solved by constructing an approximate sub - problem 2.4 , let v t represent the solution of the sub - problem at the t - th iteration. Therefore, the optimization problem P 2.4 is transformed as follows:

[0065]

[0066] s.t. |v q | = 1

[0067] where λ max represents the maximum eigenvalue of the matrix U, and I Q is the Q×Q - dimensional identity matrix. Therefore, the optimal solution of the optimization problem P 2.4.1 is obtained as follows:

[0068]

[0069] Substitute the optimized active beamforming vector w and the intelligent grouped reflection precoding v into the objective function of the optimization problem P0 to obtain the maximum capacity of the multi - user system.

[0070] The second - aspect embodiment of the present invention provides a wireless communication network optimization system based on multi - intelligent reflecting surface grouping for the wireless communication network optimization method based on multi - intelligent reflecting surface grouping described in the above - mentioned embodiment, including:

[0071] A wireless channel transmission module, configured to build a multi - intelligent reflecting surface - assisted wireless communication system according to the coverage characteristics of the equivalent cascaded link information of multiple intelligent reflecting surfaces;

[0072] A multi - intelligent reflecting surface intelligent grouping module, configured to perform intelligent pre - grouping on the reflection units of multiple intelligent reflecting surfaces according to the statistical characteristics of the optimal beam - domain channel state information, and each group of units shares the same reflection coefficient;

[0073] An asymptotic performance analysis module for analyzing the array performance gain with different pilot overheads in a single-user single-antenna cascaded channel;

[0074] A system capacity evaluation module for evaluating the system capacity of a multi-intelligent reflecting surface intelligent grouping assisted wireless communication system in a Rice channel.

[0075] The wireless communication network optimization method and system based on multi-intelligent reflecting surface grouping proposed by the present invention can, while improving the multi-user communication service quality, significantly reduce the number of pilots used to estimate the channels related to the IRS.

[0076] The additional aspects and advantages of the present invention will be partly given in the following description, partly will become obvious from the following description, or will be learned through the practice of the present invention. Brief Description of the Drawings

[0077] The above-mentioned and / or additional aspects and advantages of the present invention will become obvious and easy to understand from the following description of the embodiments in conjunction with the drawings, wherein:

[0078] Figure 1 Is a flowchart of a wireless communication network optimization method based on multi-intelligent reflecting surface grouping according to an embodiment of the present invention;

[0079] Figure 2 Is a schematic diagram of the scenario of the intelligent grouping architecture method of a multi-IRS assisted wireless communication network according to an embodiment of the present invention;

[0080] Figure 3 Is a system structure diagram of the intelligent grouping architecture method of a multi-IRS assisted wireless communication network according to an embodiment of the present invention;

[0081] Figure 4 Is a schematic diagram of the intelligent grouping module of a multi-IRS assisted wireless communication network according to an embodiment of the present invention;

[0082] Figure 5 Is a specific step diagram of the asymptotic performance analysis module of a multi-intelligent grouping reflecting surface according to an embodiment of the present invention;

[0083] Figure 6 Is a comparison diagram of the asymptotic performance analysis of a multi-intelligent grouping reflecting surface according to an embodiment of the present invention;

[0084] Figure 7 Is a comparison diagram of the coverage performance of different IRS optimization schemes in the system capacity evaluation module of the intelligent grouping method of a multi-IRS assisted wireless communication network according to an embodiment of the present invention;

[0085] Figure 8 Is a comparison diagram of the cell number performance of different IRS optimization schemes in the system capacity evaluation module of the intelligent grouping method of a multi-IRS assisted wireless communication network according to an embodiment of the present invention;

[0086] Figure 9 This is a power consumption performance comparison chart of different IRS optimization schemes in the system capacity evaluation module of the multi-IRS-assisted wireless communication network intelligent grouping method according to the embodiments of the present invention;

[0087] Figure 10 It is a schematic structural diagram of a wireless communication network optimization system based on multi-intelligent reflecting surface grouping provided according to the embodiments of the present invention.

[0088] Note: IEG-IRS represents intelligent grouping IRS, AG-IRS represents adjacent grouping IRS, U-IRS represents uniform grouping IRS, Random RCV represents randomly initialized parameter IRS, and Without IRS represents without using IRS. Detailed implementation manners

[0089] The embodiments of the present invention will be described in detail below. Examples of the embodiments are shown in the drawings, where the same or similar reference numerals denote the same or similar elements or elements having the same or similar functions throughout. The embodiments described below with reference to the drawings are exemplary and are intended to explain the present invention and should not be construed as limiting the present invention.

[0090] Figure 1 It is a flowchart of a wireless communication network optimization method based on multi-intelligent reflecting surface grouping provided according to the embodiments of the present invention.

[0091] As Figure 1 shown, the wireless communication network optimization method based on multi-intelligent reflecting surface grouping includes the following steps:

[0092] Step 1, design the position deployment of multiple intelligent reflecting surfaces and construct a multi-intelligent reflecting surface-assisted wireless communication system.

[0093] Further, in Step 1, by analyzing the position distributions of the base station, IRS, and users, and combining environmental information (such as building blockage, scattering, etc.), design the position layout of multiple IRSs to eliminate the problem of poor communication quality in non-line-of-sight areas, thereby improving the overall service quality of the BS coverage area.

[0094] The multi-IRS-assisted wireless communication system includes a BS with M antennas, D IRS reflecting surfaces composed of N reflecting units, and K single-antenna users. Since IRS is a passive reflecting device, considering that the uplink and downlink cascaded link transmissions follow the time-division duplex protocol, and according to the channel reciprocity, it is assumed that the channel state information in the downlink cascaded link is estimated by the uplink cascaded link training.

[0095] Denote the signal received at user \(k\) from the BS - user \(k\) and BS - D - IRSs - user \(k\) channels as:

[0096]

[0097] where \((\cdot)^H\) H denotes the conjugate operation, and denote the BS - to - D block IRSs link, the D block IRSs - to - user \(k\) link, and the BS - to - user \(k\) link respectively. \(z\) k denotes the additive Gaussian white noise with zero mean and variance \(\sigma^2\) acting on user \(k\). Let \(x\) denote the complex signal transmitted by the BS, where \(s\) j denotes the symbol with zero mean and unit variance sent from the BS to user \(j\), and \(\mathbf{w}\) denotes the corresponding beamforming complex vector. Let \(\mathbf{\Theta}\) denote the equivalent reflection diagonal matrix after intelligent grouping, where \(\mathbf{\theta}\) denotes the adjustable equivalent reflection coefficient vector after intelligent grouping into \(Q\) groups, q,n and \(G_{n,q}\in\{0,1\}\) indicates whether the \(n\) - th reflection element is assigned to the \(q\) - th group.

[0098] Step 2: According to the multi - intelligent reflecting surface - assisted wireless communication system, combining the position information of the user, the base station, and the multi - intelligent reflecting surface, project the statistical channel state matrix onto the optimal beam domain. Using the fractional programming method, the alternating direction method of multipliers, the maximization - minimization method, and the successive convex approximation method, transform the non - convex objective function and the non - convex constraint problem with unit - modulus constraints and 0 - 1 grouping strategy integer constraints in the non - convex, multi - variable coupled resource allocation problem aiming at maximizing the system capacity into a joint optimization problem of base - station active beamforming, multi - intelligent reflecting surface passive grouping precoding, and grouping strategy, and solve the joint optimization problem to obtain the optimal grouping strategy based on the statistical cascaded channel state information.

[0099] Optionally, in an embodiment of the present invention, step 2 specifically includes:

[0100] Step 201: Transmission stage, the signal - to - noise ratio SINR at user \(k\) is:

[0101]

[0102] where and denote the (reflected) cascaded link from the BS to the multi - IRSs to user \(k\) and the superimposed channel from the BS to user \(k\) respectively.

[0103] Step 202: Under the constraints of the maximum transmit power of the base station, the transmit beamforming of the base station, the unit modulus constraint of the multi-IRS passive reflection precoding, and the grouping strategy constraint, with the goal of maximizing the sum rate of the system, construct the optimization problem P0:

[0104]

[0105] s.t. G q,n ∈ {0, 1}

[0106]

[0107] Q ≤ Q0 << LN

[0108]

[0109] |v q | = 1

[0110] where, represents the overall transmit beamforming vector of K users; Q0 represents the number of pilot frequencies allocated by the system to the multi-IRS side, and is also the dimension of the equivalent cascaded channel related to the multi-IRS or the number of groups of the maximum multi-IRS reflection units, || || 2 is the square of the vector two-norm, is the maximum transmit power of the base station.

[0111] Step 203: Combine the fractional programming method, the alternating direction multiplier method, the max-min method, and the successive convex approximation method to transform the non-convex objective function and the non-convex constraint problem with unit modulus constraint and 0-1 grouping strategy integer constraint in the non-convex and multi-variable coupled resource allocation problem aiming at maximizing the system capacity into a joint optimization problem of base station active beamforming, multi-intelligent reflecting surface passive grouping precoding optimization, and grouping strategy.

[0112] Let and represent the statistical channel state information of the cascaded channel and the direct channel of user k respectively. By fixing the BS transmit beamforming vector and the multi-IRS reflection precoding vector, transform the original optimization problem P0 into an optimal grouping strategy optimization problem P1 based on the statistical cascaded channel state information:

[0113]

[0114] s.t. G q,n ∈ {0, 1}

[0115]

[0116] where, and Denote the introduced auxiliary variable. The optimization problem P1 is a 0-1 integer programming problem. By utilizing the non-convex constraints in the relaxation constraints of the regular maximization problem, the optimization problem P1 is transformed into an optimizable quadratic programming problem by introducing an auxiliary variable 21 The optimization problem P1 is transformed as follows. The optimization problem P is transformed as follows 1.1 :

[0117]

[0118] where G n denotes the n-th column vector of the intelligent grouping strategy G to be optimized. is the adjoint matrix of ξ k . is the real part of the parameter. is the statistical adjustable equivalent reflection coefficient vector corresponding to the intelligent division into Q groups. is the corresponding statistical beamforming complex vector. The optimization problem P 1.1 is a linearly constrained quadratic programming problem. The optimal grouping strategy is obtained by using the Lagrange multiplier method.

[0119] Step 3: According to the optimal grouping strategy based on the statistical cascaded channel state information, pre-group the reflecting units of the multi-intelligent reflecting surface. By using the fractional programming method, the alternating direction multiplier method, the maximization-minimization method, and the successive convex approximation method proposed in Step 2, the problem of maximizing the system capacity after grouping the reflecting units of the multi-intelligent reflecting surface is transformed into a joint optimization problem of the base station active beamforming and the multi-intelligent reflecting surface grouped passive precoding. The joint optimization problem is solved to obtain the active and passive precoding of the base station and the multi-intelligent reflecting surface and the maximum capacity of the multi-user system in the real-time transmission stage.

[0120] Optionally, in an embodiment of the present invention, Step 3 specifically includes:

[0121] Step 301: Combine the proposed fractional programming method, alternating direction multiplier method, maximization-minimization method, and successive convex approximation method to transform the original optimization problem P0 into the following optimization problem P2:

[0122]

[0123] |v q | = 1

[0124] The optimization problem P2 contains four variables to be optimized. Borrowing the idea of alternating optimization, the optimization problem P2 is transformed into three optimization sub-problems to solve the auxiliary variables and ξ, the active beamforming vector w, and the intelligent grouping reflection precoding v;

[0125] Step 302: Fix the active beamforming vector w and the intelligent grouped reflection precoding v, and then optimize the auxiliary variables and ξ:

[0126] First, by fixing the auxiliary variables the active beamforming vector w and the intelligent grouped reflection precoding v, the problem P2 to be optimized is transformed into the following optimization problem P 2.1 to solve for the auxiliary variable ξ:

[0127]

[0128] Since the optimization problem P 2.1 is an unconstrained quadratic programming problem, by setting the derivative of the objective function equal to zero, the optimal auxiliary variable ξ is derived:

[0129]

[0130] where, and

[0131] Step 303: Fix the auxiliary variable ξ, the active beamforming vector w, and the intelligent grouped reflection precoding The optimization problem P2 can be transformed into the following optimization problem P 2.2 to solve for the auxiliary variable

[0132]

[0133] Since the optimization problem P 2.2 is an unconstrained convex optimization problem, similarly, by setting the derivative of the objective function equal to zero, the optimal auxiliary variable is derived:

[0134]

[0135] where,

[0136] According to and the formulas of, the solutions of the auxiliary variables and ξ are further analyzed and decoupled, and then, the optimal solution containing only the channel state information and the active and passive precoding is expressed as:

[0137]

[0138] where, θ k = arg(a k ), and

[0139] Step 304, fix the auxiliary variables and ξ, as well as the intelligent grouped reflection precoding v, and then optimize the active beamforming vector w:

[0140] First, by fixing the auxiliary variables and ξ, as well as the intelligent grouped reflection precoding v, the optimization problem P2 is transformed into the optimization problem P 2.3 to solve for the active beamforming vector w:

[0141]

[0142] where, and I K is a K×K dimensional identity matrix;

[0143] Since the optimization problem P 2.3 is a standard quadratic constrained quadratic programming problem, by using the Lagrange multiplier method, the optimal active beamforming vector is obtained:

[0144] w opt =(L + λI KM ) -1 ζ

[0145] where, λ represents the Lagrange multiplier, obtained by grid search, and I KM is a KM×KM dimensional identity matrix.

[0146] Step 305, fix the auxiliary variables and ξ, as well as the active beamforming vector w, and then optimize the intelligent grouped reflection precoding v:

[0147] First, by fixing the auxiliary variables and ξ, as well as the active beamforming vector w, the problem to be optimized P2 is transformed into the following optimization problem P 2.4 to solve for the intelligent grouped reflection precoding v:

[0148]

[0149] s.t. |v q | = 1

[0150] where,

[0151] By borrowing the max-min algorithm, the optimization problem P 2.4 is solved by constructing an approximate subproblem, letting v t represent the solution of the subproblem at the t-th iteration. Therefore, the optimization problem P 2.4 is transformed into the following:

[0152]

[0153] such that |v q | = 1

[0154] wherein, λ max represents the maximum eigenvalue of matrix U, and I Q is the Q×Q dimensional identity matrix. Therefore, the optimal solution of the optimization problem P 2.4.1 is obtained as follows:

[0155]

[0156] Substitute the active beamforming vector w and the intelligent grouping reflection precoding v after optimization into the objective function of the optimization problem P0 to obtain the maximum capacity of the multi-user system.

[0157] Figure 2 , Figure 6 , Figure 7 , Figure 8 and Figure 9 The (a) mentioned in and represents the scenario with a line-of-sight direct link, and (b) represents the scenario with a non-line-of-sight direct link.

[0158] Figure 2 shows the intelligent grouping architecture method scenario of the multi-IRS-assisted wireless communication network according to the embodiments of the present invention. Among them, Scenario 1 means that the line of sight between the base station and the user is blocked by a building, and the signal transmitted by the base station is reflected by multiple IRSs to multiple users; while in Scenario 2, there is a line-of-sight direct link, and the signal transmitted by the base station can not only be reflected by multiple IRSs, but also be transmitted to multiple users through the direct link.

[0159] Figure 3 shows the system structure of the intelligent grouping architecture method of the multi-IRS-assisted wireless communication network according to the embodiments of the present invention. The base station communicates with multiple users through the reflection channel and the direct channel. Figure 4 shows the intelligent grouping module of the multi-IRS-assisted wireless communication network according to the embodiments of the present invention.

[0160] Figure 5 shows the specific steps of the asymptotic performance analysis module of the multi-intelligent grouping reflecting surface according to the embodiments of the present invention. In a single-IRS-assisted single-input single-output (SISO) system, first analyze the cascaded steering vector from the base station to the IRS to the user: α N (θ) = [1,…,e j(N-1)Δ , and then analyze the statistical cascaded steering vector phase similarity, and obtain the equivalent grouped cascaded steering vector by grouping the IRS units: Finally, the cascaded channel distribution from the base station to the intelligent reflecting surface to the single user is obtained based on the Rice channel modeling:

[0161] Figure 6 The asymptotic performance analysis of the multi-intelligent grouped reflecting surface in the embodiments of the present invention and the comparison with other grouped, ungrouped, and IRS-assisted methods without grouping are shown. The simulation conditions are as follows: carrier frequency 5 GHz, M = K = 1, N = 10000, L = 1, P max = 10 dBm, σ k = -100 dBm, the number of groups Q ranges from 1 to 20, the coordinates of the base station are (0, 0), the coordinates of the intelligent reflecting surface are (6 m, 8 m), the coordinates of the user are (300 m, 0), and the path loss is where d represents the distance. In the figure, IEG-IRS represents intelligent grouped IRS, AG-IRS represents adjacent grouped IRS, U-IRS represents uniformly grouped IRS, Random RCV represents randomly initialized parameter IRS, and Without IRS represents without using IRS.

[0162] Figure 7 The comparison results of different IRS optimization schemes in the system capacity evaluation module of the multi-IRS-assisted wireless communication network intelligent grouping method in the embodiments of the present invention are shown. The simulation conditions are as follows: carrier frequency 5 GHz, M = K = 4, N = 10000, L = 2, Q = 4, P max = 10 dBm, σ k = -100 dBm, the coordinates of the base station are (-1 m, 0), the coordinates of 2 intelligent reflecting surfaces are (6 m, 8 m) and (6 m, -8 m), the coordinates of the multi-user distribution center are (L, 0), where the range of L is from 0 to 300 m with a step of 30 m, and the path loss is where d represents the distance.

[0163] Figure 8 The comparison results of the unit number performance of different IRS optimization schemes in the system capacity evaluation module of the multi-IRS-assisted wireless communication network intelligent grouping method in the embodiments of the present invention are shown. The simulation conditions are as follows: carrier frequency 5 GHz, M = K = 4, L = 2, Q = 4, P max = 10 dBm, σ k = -100 dBm, the coordinates of the base station are (0, 0), the coordinates of 2 intelligent reflecting surfaces are (6 m, 8 m) and (6 m, -8 m), the coordinates of the multi-user distribution center are (300 m, 0), the number of reflecting units of each IRS ranges from 400 to 10000, and the path loss is where d represents the distance.

[0164] Figure 9Shows the comparison results of the power consumption performance of different IRS optimization schemes in the system capacity evaluation module of the multi-IRS-assisted wireless communication network intelligent grouping method according to the embodiments of the present invention. The simulation conditions are as follows: the carrier frequency is 5 GHz, M = K = 4, N = 10,000, L = 2, Q = 4, σ k = -100 dBm, the coordinates of the base station are (0, 0), the coordinates of the two intelligent reflecting surfaces are (6 m, 8 m) and (6 m, -8 m), the coordinates of the multi-user distribution center are (300 m, 0), and the maximum transmit power P max range is from -10 dBm to 40 dBm, with a step size of 5 dBm, and the path loss is where d represents the distance.

[0165] Secondly, the embodiments of the present invention also propose a wireless communication network optimization system based on multi-intelligent reflecting surface grouping.

[0166] As Figure 10 shown, the wireless communication network optimization system based on multi-intelligent reflecting surface grouping includes: a wireless channel transmission module, a multi-intelligent reflecting surface intelligent grouping module, an asymptotic performance analysis module, and a system capacity evaluation module.

[0167] Among them, the wireless channel transmission module is used to build a multi-intelligent reflecting surface-assisted wireless communication system according to the coverage characteristics of the equivalent cascaded link information of multiple intelligent reflecting surfaces;

[0168] The multi-intelligent reflecting surface intelligent grouping module is used to perform intelligent pre-grouping on the reflection units of multiple intelligent reflecting surfaces according to the statistical characteristics of the optimal beam domain channel state information, and each group of units shares the same reflection coefficient;

[0169] The asymptotic performance analysis module is used to analyze the array performance gain with different pilot overheads in a single-user single-antenna cascaded channel;

[0170] The system capacity evaluation module is used to evaluate the system capacity of the multi-intelligent reflecting surface intelligent grouping-assisted wireless communication system in a Rice channel.

[0171] It should be noted that the foregoing explanations of the embodiments of the wireless communication network optimization method based on multi-intelligent reflecting surface grouping also apply to the wireless communication network optimization system based on multi-intelligent reflecting surface grouping of this embodiment, and will not be repeated here.

[0172] The wireless communication network optimization method and system based on multi-intelligent reflecting surface (IRS) grouping proposed according to the embodiments of the present invention, a wireless channel transmission module based on multi-IRS, which builds a multi-IRS-assisted wireless communication network according to the coverage characteristics of the equivalent cascaded link information of multiple IRSs; a multi-IRS intelligent grouping module based on statistical cascaded channel state information, which intelligently pre-groups the reflecting units of multiple IRSs according to the statistical characteristics of the optimal beam domain channel state information, and each group of units shares the same reflection coefficient; an asymptotic performance analysis module based on multi-IRS intelligent grouping, which analyzes the array performance gain with different pilot overheads in a single-user single-antenna cascaded channel; a system capacity evaluation module based on multi-IRS intelligent grouping, which evaluates the system capacity of the multi-IRS intelligent grouping-assisted wireless communication network in a Rice channel. The intelligent grouping optimization architecture and method of the multi-IRS-assisted wireless communication network proposed by the present invention can improve the quality of multi-user communication services while significantly reducing the number of pilots used to estimate the channels related to IRSs.

[0173] In the description of this specification, the descriptions referring to terms such as "one embodiment", "some embodiments", "example", "specific example", or "some examples" etc. mean that the specific features, structures, materials, or characteristics described in connection with the embodiment or example are included in at least one embodiment or example of the present invention. In this specification, the schematic representations of the above terms do not necessarily refer to the same embodiment or example. Moreover, the specific features, structures, materials, or characteristics described can be combined in a suitable manner in any one or N embodiments or examples. In addition, without contradiction, those skilled in the art can combine and combine the different embodiments or examples described in this specification and the features of different embodiments or examples.

[0174] In addition, the terms "first" and "second" are only used for descriptive purposes and cannot be understood as indicating or implying relative importance or implicitly specifying the quantity of the indicated technical features. Thus, the features defined with "first" and "second" may explicitly or implicitly include at least one of such features. In the description of the present invention, the meaning of "N" is at least two, such as two, three, etc., unless otherwise specifically defined.

[0175] Any process or method description in the flowchart or described in other ways herein can be understood as representing a module, segment, or part of code including one or N executable instructions for implementing a customized logical function or process, and the scope of the preferred embodiments of the present invention includes additional implementations, where the functions can be executed in a substantially simultaneous manner or in a reverse order according to the involved functions, rather than in the order shown or discussed, which should be understood by those skilled in the art of the embodiments of the present invention.

Claims

1. A method for optimizing a wireless communication network based on multi-intelligent reflecting surface grouping, characterized in that It includes the following steps: Step 1: Design the position deployment of multiple intelligent reflecting surfaces and construct a wireless communication system assisted by multiple intelligent reflecting surfaces; Step 2: Based on the wireless communication system assisted by multiple intelligent reflecting surfaces, combined with the position information of users, base stations, and multiple intelligent reflecting surfaces, project the statistical channel state matrix onto the optimal beam domain. Using the fractional programming method, alternating direction multiplier method, maximization-minimization method, and successive convex approximation method, transform the non-convex objective function and non-convex constraint problems with unit modulus constraints and 0-1 grouping strategy integer constraints in the non-convex, multi-variable coupled resource allocation problem aiming at maximizing system capacity into a joint optimization problem of base station active beamforming, multi-intelligent reflecting surface passive grouping precoding, and grouping strategy. Solve the joint optimization problem to obtain the optimal grouping strategy based on statistical cascaded channel state information; Step 3: According to the optimal grouping strategy based on statistical cascaded channel state information, pre-group the reflecting units of multiple intelligent reflecting surfaces. Using the fractional programming method, alternating direction multiplier method, maximization-minimization method, and successive convex approximation method proposed in Step 2, transform the system capacity maximization problem after grouping the reflecting units of multiple intelligent reflecting surfaces into a joint optimization problem of base station active beamforming and multi-intelligent reflecting surface grouped passive precoding. Solve the joint optimization problem to obtain the maximum capacity of the base station and multi-intelligent reflecting surface active and passive precoding and multi-user system in the real-time transmission stage.

2. The method according to claim 1, characterized in that Specifically, Step 1 includes: By analyzing the position distribution of base stations, intelligent reflecting surfaces, and users, combined with environmental information, design the position layout of multiple intelligent reflecting surfaces; The wireless communication system assisted by multiple intelligent reflecting surfaces includes: a base station with M antennas, D intelligent reflecting surfaces composed of N reflecting units, and K single-antenna users; the uplink and downlink cascaded link transmissions follow the time-division duplex protocol. Assuming that the channel state information in the downlink cascaded link is estimated by the uplink cascaded link training, the signals received by user k from the BS-user k and BS-D-IRSs-user k channel transmissions are expressed as: Among them, (·) H represents the conjugate operation, and respectively represent the BS-to-D-block IRSs link, the D-block IRSs-to-user-k link, and the BS-to-user-k link, z k represents the additive Gaussian white noise with zero mean and variance acting on user k, and represents the complex signal transmitted by the BS, s j represents the symbol with zero mean and unit variance sent by the BS to user j, represents the corresponding beamforming complex vector, and represents the equivalent reflection diagonal matrix after intelligent grouping, represents the adjustable equivalent reflection coefficient vector after intelligent division into Q groups, where represents the optimizable grouping strategy, where G q,n ∈{0,1} indicates whether the nth reflection unit is assigned to the qth group.

3. The method according to claim 2, wherein Specifically, Step 2 includes: Step 201: In the transmission stage, the signal-to-noise ratio SINR at user k is: Among them, and respectively represent the cascaded reflection link from the base station to the RIS to user k and the superimposed channel of the link from the base station to user k; Step 202: Under the constraints of the maximum transmit power of the base station, the base station transmit beamforming constraint, the unit modulus constraint of the multi-intelligent reflecting surface passive reflection precoding, and the grouping strategy constraint, aiming at maximizing the sum rate of the system, construct the optimization problem P0: s.t.G q,n ∈{0,1} Q≤Q0<<LN |v q |＝1 Among them, represents the overall transmit beamforming vector of K users. Q0 represents the number of pilot frequencies allocated by the system to the multi-IRS side, which is also the dimension of the equivalent cascaded channel related to the multi-intelligent reflecting surface or the number of groups of the maximum multi-intelligent reflecting surface reflection units. || || 2 is the square of the vector two-norm, is the maximum transmit power of the base station; Step 203: Combine the fractional programming method, alternating direction multiplier method, maximization-minimization method, and successive convex approximation method to transform the non-convex objective function and non-convex constraint problems with unit modulus constraints and 0-1 grouping strategy integer constraints in the non-convex, multi-variable coupled resource allocation problem aiming at maximizing system capacity into a joint optimization problem of base station active beamforming, multi-intelligent reflecting surface passive grouping precoding optimization, and grouping strategy; Let and respectively represent the statistical channel state information of the cascaded channel and the direct channel of user k. By fixing the transmit beamforming vector of the base station and the reflection precoding vector of the IRS, the original optimization problem P0 is transformed into an optimal grouping strategy optimization problem P1 based on the statistical cascaded channel state information: s.t.G q,n ∈{0,1} Among them, and represent the introduced auxiliary variables. The optimization problem P1 is a 0-1 integer programming problem. By utilizing the non-convex constraints in the relaxation constraints of the L 21 regular maximization problem, the optimization problem P1 is transformed into an optimizable quadratic programming problem. By introducing the auxiliary variable the optimization problem P1 is transformed into the optimization problem P 1.1 : Among them, G n represents the n-th column vector of the intelligent grouping strategy G to be optimized, is the adjoint matrix of ξ k , is the real part of the parameter, is the statistically adjustable equivalent reflection coefficient vector after the corresponding intelligent division into Q groups, is the corresponding statistically beamforming complex vector, and the optimization problem P 1.1 is a linearly constrained quadratic programming problem, and the optimal grouping strategy is obtained by using the Lagrange multiplier method.

4. The method according to claim 3, wherein Specifically, Step 3 includes: Step 301: Combine the proposed fractional programming method, alternating direction multiplier method, maximize-minimize method, and successive convex approximation method to transform the original optimization problem P0 into the following optimization problem P2: |v q |＝1 The optimization problem P2 contains four variables to be optimized. Borrowing the idea of alternating optimization, the optimization problem P2 is transformed into three sub-optimization problems to solve for the auxiliary variables and ξ, the active beamforming vector w, and the intelligent grouped reflection precoding v; Step 302, fix the active beamforming vector w and the intelligent grouped reflection precoding v, and then optimize the auxiliary variables and ξ: First, by fixing the auxiliary variables the active beamforming vector w and the intelligent grouped reflection precoding v, the problem P2 to be optimized is transformed into the following optimization problem P 2.1 to solve the auxiliary variable ξ: Because of the optimization problem P 2.1 is an unconstrained quadratic programming problem. By setting the derivative of the objective function equal to zero, the optimal auxiliary variable ξ is derived as follows: Among them, and Step 303, fix the auxiliary variable ξ and the active beamforming vector w and the intelligent grouped reflection precoding The optimization problem P2 can be transformed into the following optimization problem P 2.2 to solve the auxiliary variable Because of the optimization problem P 2.2 is an unconstrained convex optimization problem. By setting the derivative of the objective function equal to zero, the optimal auxiliary variable is derived as follows: Among them, According to and 's formula, the auxiliary variables and the solution of ξ are further analyzed and decoupled. Then, the optimal solution containing only the channel state information and the active and passive precoding is expressed as: where θ k = arg(a k ), and Step 304, fix the auxiliary variables and ξ as well as the intelligent grouped reflection precoding v, and then optimize the active beamforming vector w: First, by fixing the auxiliary variables and ξ, and intelligent grouped reflection precoding v , the optimization problem P2 is transformed into the optimization problem P 2.3 to solve for the active beamforming vector w: Among them, and I K is a K×K dimensional identity matrix; Because of the optimization problem P 2.3 is a standard quadratic constrained quadratic programming problem. By using the Lagrange multiplier method, the optimal active beamforming vector is obtained: w opt =(L + λI KM ) -1 ζ where λ represents the Lagrange multiplier, which is obtained by searching through the grid method, and I KM is an identity matrix of dimension KM×KM. Step 305, fix the auxiliary variables and ξ and the active beamforming vector w, and then optimize the intelligent grouped reflection precoding v: First, by fixing the auxiliary variables and ξ and the active beamforming vector w, the optimization problem P2 is transformed into the following optimization problem P 2.4 to solve for the intelligent grouped reflection precoding v: s.t. |v q | = 1 Among them, By borrowing the maximize-minimize algorithm, the optimization problem P is solved by constructing an approximate sub-problem 2.4 , let v t represent the solution of the sub-problem at the t-th iteration. Therefore, the optimization problem P 2.4 is transformed as follows: s.t. |v q | = 1 wherein, λ max represents the largest eigenvalue of matrix U, and I Q is the Q×Q dimensional identity matrix. Therefore, the optimal solution of the optimization problem P 2.4.1 is obtained as follows: The active beamforming vector after optimization is completed w and the intelligent grouped reflection precoding v are substituted into the objective function of the optimization problem P0 to obtain the maximum capacity of the multi-user system.

5. A wireless communication network optimization system based on multi-intelligent reflecting surface grouping, for the method for optimizing a wireless communication network based on multi-intelligent reflecting surface grouping according to any one of claims 1-4, characterized in that, It includes: A wireless channel transmission module, which is used to build a multi-intelligent reflecting surface assisted wireless communication system according to the coverage characteristics of the equivalent cascaded link information of multiple intelligent reflecting surfaces; A multi-intelligent reflecting surface intelligent grouping module, which is used to perform intelligent pre-grouping on the reflecting units of multiple intelligent reflecting surfaces according to the statistical characteristics of the optimal beam domain channel state information, and each group of units shares the same reflection coefficient; An asymptotic performance analysis module, which is used to analyze the array performance gain with different pilot overheads in a single-user single-antenna cascaded channel; A system capacity evaluation module, which is used to evaluate the system capacity of a multi-intelligent reflecting surface intelligent grouping assisted wireless communication system in a Rice channel.

Images