A multi-RIS collaborative assistance user-centralized network rate optimization method

Abstract

The application belongs to the field of wireless communication, and particularly relates to a user-centered network rate optimization method assisted by multiple RISs, which comprises the following steps: constructing a user-centered dense base station deployment wireless communication system assisted by multiple RISs; taking the maximum user reachable rate as the target, mathematically modeling a non-convex, multivariable coupled wireless resource allocation problem; through the alternating iterative optimization thought, using the Lagrange dual theory, fractional programming method and other methods, the original complex and difficult to solve optimization problem is transformed into solving three sub-problems of user multi-association, base station transmit beamforming design and multiple RIS collaborative phase shift matrix design respectively; the application considers the maximum user multi-link reachable rate under the condition of associating different base stations and RISs, reduces the influence when the line-of-sight link between the base station and the user is blocked, and has high practicability.

Description

TECHNICAL FIELD

[0001] The application belongs to the field of wireless communication, and particularly relates to a user-centered network rate optimization method with multi-RIS cooperative assistance. BACKGROUND

[0002] High-frequency wireless communication, such as millimeter wave communication and terahertz communication, significantly improves system capacity with its wide spectrum resources, effectively alleviating the shortage of spectrum resources. However, compared with microwave communication, high-frequency signals face more serious path loss in the transmission process, and are more easily blocked by obstacles such as buildings, trees, etc. To solve this problem, a reconfigurable intelligent surface (RIS) can be introduced to reconstruct the wireless signal transmission environment. RIS is a two-dimensional surface composed of programmable electromagnetic metamaterials that can be adjusted in real time according to the environment and communication needs. By adjusting the amplitude or phase of each reflection unit, the outgoing signal can be adjusted, optimized or enhanced. Using the above characteristics, RIS can build a virtual line-of-sight link, thereby eliminating unfavorable channel factors such as blockage and fading, and improving the performance of the communication system. In addition, RIS has the advantages of green energy saving, easy deployment and low cost, and is widely used in wireless communication systems.

[0003] For RIS-assisted wireless communication systems, existing literature mostly focuses on single RIS or multiple distributed RIS scenarios, and only a few consider the potential gains from RIS cooperation. Research results show that compared with traditional single RIS or distributed RIS assistance, RIS cooperative assistance can further improve system performance. However, existing research on RIS cooperative assistance is mainly limited to single base station communication systems, and user-centered dense base station deployment scenarios remain to be further studied. The present application is dedicated to breaking this limitation. In the dense base station deployment scenario, the user multi-association problem is the key to guaranteeing the quality of service of users, involving power allocation, multi-cell interference problems; for the same serving base station, the effects of different RIS cooperative assistance are quite different; active-passive joint beamforming design is the basis for RIS-assisted high-frequency wireless communication systems to achieve higher transmission rates.

[0004] In summary, the existing technical problems are that the influence of RIS cooperation is less considered, the advantages of RIS are not fully utilized, and the single base station communication system is limited. Therefore, in order to maximize the achievable rate of users and fully utilize the effective gain of RIS cooperation, there is an urgent need for a resource scheduling method for a multi-RIS cooperative dense high-frequency wireless communication system. SUMMARY

[0005] To solve the above problems, the application provides a multi-RIS cooperative assistance user-centralized network rate optimization method, which aims to maximize the user multi-link reachability and rate under the strict constraint conditions of user maximum association number, base station maximum transmit power and RIS reflection unit phase through wireless resource allocation.

[0006] The specific scheme includes the following steps:

[0007] S1. Construct a user-centered multi-RIS cooperative assistance dense base station deployment high-frequency wireless communication network model;

[0008] S2. Construct a non-convex multivariate coupled wireless resource allocation optimization problem with the goal of maximizing user reachable rate; wherein the user reachable rate is the sum of the rates of all reachable links between the user and its associated multiple base stations;

[0009] S3. Decouple the wireless resource allocation optimization problem into three sub-problems: user multi-association, base station transmit beamforming design and multi-RIS cooperative phase shift matrix design;

[0010] S4. Jointly iteratively solve the three sub-problems to obtain the optimal user multi-association parameters, optimal base station transmit beamforming vector and optimal multi-RIS cooperative phase shift matrix.

[0011] Further, the wireless resource allocation optimization problem is decoupled into three sub-problems, including:

[0012] S31. In the wireless resource allocation optimization problem, fix the base station beamforming vector and the multi-RIS cooperative phase shift matrix, and relax the constraint condition C2 to 0 b ≤1, to obtain the user multi-association sub-problem;

[0013] S32. In the wireless resource allocation optimization problem, fix the user association parameters and introduce auxiliary variables Construct a new optimization problem; after processing the new optimization problem, a conversion problem is obtained, given the multi-RIS cooperative phase shift matrix, to obtain the base station transmit beamforming design sub-problem;

[0014] S33. In the conversion problem, given the base station beamforming vector, obtain the multi-RIS cooperative phase shift matrix sub-problem.

[0015] Further, step S4 specifically includes:

[0016] S41. Solve the user multi-association sub-problem using the Lagrangian dual decomposition method to update the user association parameters;

[0017] S42. Selecting the auxiliary RIS group for the user and its associated base station based on the shortest distance criterion, solving the base station transmit beamforming matrix design subproblem according to the auxiliary RIS group and the updated user association parameter in step S41, and updating the base station transmit beamforming vector;

[0018] S43. Solving the multi-RIS cooperative phase shift matrix design subproblem based on the updated user association parameter in step S41 and the updated base station transmit beamforming vector in step S42, and updating the multi-RIS cooperative phase shift matrix;

[0019] S44. Judging whether the wireless resource allocation optimization problem converges (i.e., the user achievable rate obtained in the latest iteration does not increase compared to the user achievable rate obtained in the previous iteration, or the maximum number of iterations is reached), if yes, obtaining the optimal user multi-association parameter, the optimal base station transmit beamforming vector and the optimal multi-RIS cooperative phase shift matrix, if not, returning to step S41.

[0020] Further, step S41 solves the user multi-association subproblem using the Lagrange dual decomposition method, including:

[0021] S411. Introducing the Lagrange multiplier v to construct the Lagrange function and the Lagrange dual function g(v);

[0022] S412. Minimizing the Lagrange dual function to form the dual problem;

[0023] S413. Updating the user association parameter according to the Lagrange multiplier v;

[0024] S414. Updating the Lagrange multiplier v using the subgradient algorithm; then judging whether the Lagrange multiplier v converges, if yes, the dual problem reaches the global optimum, outputting the current user association parameter, if not, returning to step S413.

[0025] Further, step S42 specifically includes:

[0026] S421. Pairing all RISs two by two to obtain multiple RIS pairs, for each base station b associated with the user b = 1, 2, …, B, calculating the total path distance between all RIS pairs and the base station b, and selecting the RIS pair with the smallest total path distance as the auxiliary RIS group of the base station b; wherein the total path distance between any RIS pair (RIS i, RIS j) and the base station b is the sum of the distance from the base station b to RIS i, the distance from RIS i to RIS j, and the distance from RIS j to the user;

[0027] S422. Introduce a complex auxiliary vector ξ, and convert the base station transmit beamforming design sub-problem into a bi-convex optimization problem about the base station transmit beamforming vector and the complex auxiliary vector ξ;

[0028] S423. Fix the base station transmit beamforming vector in the bi-convex optimization problem, and solve to obtain the optimal complex auxiliary vector ξ opt ;

[0029] S424. Substitute the optimal complex auxiliary vector ξ opt into the bi-convex optimization problem to solve to obtain the optimal base station transmit beamforming vector.

[0030] Further, step S43 specifically comprises:

[0031] S431. Introduce an auxiliary parameter ε to reconstruct the multi-RIS cooperative phase shift matrix design sub-problem to obtain a reconstruction optimization problem;

[0032] S432. Fix the multi-RIS cooperative phase shift matrix in the reconstruction optimization problem, and solve to obtain the optimal auxiliary parameter ε opt ;

[0033] S433. Substitute the optimal auxiliary parameter ε opt into the reconstruction optimization problem to solve to obtain the optimal multi-RIS cooperative phase shift matrix.

[0034] Advantages of the present application:

[0035] Compared with the prior art, the present application not only considers the optimization of user multi-association, but also particularly proposes joint optimization and design of the base station transmit beamforming matrix and the multi-RIS cooperative phase shift matrix under the multi-association condition. In addition, the present application maximizes the use of the reflection characteristics of the multi-RIS by dynamically adjusting the cooperative strategy of the base station and the RIS, and significantly improves the multi-link reachable rate of the user. For the determination of the cooperative RIS, the present application uses the minimum distance criterion, i.e., determines the optimal auxiliary RIS combination by calculating the minimum value of the sum of distances between the base station, the RIS and the user.

[0036] In the face of multivariable coupled non-convex optimization problems, the present application decouples them into three sub-problems through decomposition technology, and realizes them by using the method of alternating iterative optimization, which presents closed-form updates in the optimization process and can reduce the computational complexity. First, in the case of fixing the base station transmit beamforming matrix and the RIS phase shift matrix, the user multi-association sub-problem is solved by the Lagrange dual decomposition method. Subsequently, by applying the fractional programming method, the optimization problems of the base station transmit beamforming matrix design and the RIS phase shift matrix design are simplified, and the closed-form optimal solution is obtained more easily. The algorithm proposed in the present application can greatly reduce the computational complexity while improving the system performance. BRIEF DESCRIPTION OF DRAWINGS

[0037] Figure 1 A flow chart of a multi-RIS collaborative assistance user-centralized network rate optimization method of the present application;

[0038] Figure 2 A flow chart for solving the joint optimization problem in the present application;

[0039] Figure 3 A multi-RIS collaborative assistance user-centralized wireless communication network model diagram of the present application;

[0040] Figure 4 The sum rate performance in the case of multi-user association and single-user association;

[0041] Figure 5 User and rate simulation diagram under different RIS selection methods. DETAILED DESCRIPTION

[0042] The technical solutions in the embodiments of the present application will be described clearly and completely below with reference to the drawings in the embodiments of the present application. Obviously, the described embodiments are only a part of the embodiments of the present application, rather than all the embodiments. Based on the embodiments in the present application, all other embodiments obtained by those skilled in the art without creative labor fall within the scope of protection of the present application.

[0043] The present application provides a multi-RIS collaborative assistance user-centralized network rate optimization method, as shown in Figures 1-3 , comprising the following steps:

[0044] S1. Construct a user-centered multi-RIS collaborative assistance dense base station deployment high-frequency wireless communication network model.

[0045] Specifically, the present application considers a user-centered multi-RIS assisted dense base station deployment high-frequency wireless communication network, and the network model is as shown in Figure 3 , which is composed of multiple base stations, multiple RISs and a user, wherein the base stations are all millimeter wave base stations, and the millimeter wave base stations are equipped with uniform linear arrays, and the RISs are equipped with uniform planar arrays. The RISs are randomly distributed between the base stations and the user, and it is assumed that there is a certain blockage probability for the line-of-sight link between the base stations and the user. By properly adjusting the phase of the RIS, a virtual line-of-sight link can be collaboratively constructed to ensure the communication quality of the user.

[0046] Specifically, in the network model constructed by the present application, the base station set is denoted as , the RIS set is denoted as B represents the number of base stations, and L represents the number of RISs; wherein the user is equipped with M antennas, and each base station is equipped with M tRoot antenna, assuming that each antenna of a user can be regarded as a virtual single-antenna user at the same location, each antenna, i.e., each virtual single-antenna user, is associated with different base stations and receives independent data streams from these base stations, realizing virtual spatial division multiplexing and improving user achievable rate; each RIS is equipped with N = N x × N y reflection units, N x denotes the number of horizontally arranged reflection units in the RIS, N y denotes the number of vertically arranged reflection units in the RIS; define Φ i = diag(φ i ) represents the phase shift matrix of the RIS, where φ i = [φ i,1 , φ i,2 ,..., φ i,N ] T denotes the phase shift vector of the RIS i, and each element in the phase shift vector can be further described as denotes the phase shift of the nth reflection unit of the RIS i, θ i,n ∈ [0, 2π) denotes the phase shift angle of the nth reflection unit of the RIS i; define denotes the channel between the base station and the user; denotes the channel between the base station b and the RIS i, denotes the channel between the base station b and ; denotes the channel between the RIS i and the RIS j; denotes the channel between the RIS i and the user, denotes the channel between the RIS j and the user; are all real matrices.

[0047] Most of the research on RIS-assisted millimeter wave communication assumes that all line-of-sight links between the base station and the user are blocked, but this assumption is too idealistic. In order to be more realistic, the present invention considers introducing an equivalent line-of-sight sphere model to approximate the actual line-of-sight area, where the equivalent line-of-sight sphere is a fixed-size spherical model with a radius r b ; the probability of a link with a length of L d being a line-of-sight link is Define the blockage coefficient e b to represent whether there is a line-of-sight link between the user and the base station b, when P LOS is greater than or equal to the threshold τ, define e b = 1, indicating that there is a line-of-sight link, otherwise e b = 0, indicating that the line-of-sight link is blocked.

[0048] Therefore, the equivalent cascaded channel between the base station b and the user can be represented as

[0049]

[0050] The present application uses the Saleh-Valenzuela channel model to study, taking the channel from the base station b to the RIS i and the RIS i to the user as an example, respectively

[0051]

[0052] Where N BI represents the number of paths between the base station b and the RIS i, N IU represents the number of paths between the RIS i and the user; p BI,l represents the complex gain of the lth link between the base station b and the RIS i, p IU,l represents the complex gain of the lth link between the RIS i and the user; a r represents the array response vector of the receiving end, a t represents the array response vector of the sending end; And respectively represent the azimuth and elevation angles of arrival at the RIS; q l is the azimuth angle of departure from the base station; And respectively represent the azimuth and elevation angles of departure from the RIS; define the correlation coefficient c b between the user and the base station b, if c b = 1, it means that the user is associated with the base station b, if c b = 0, it means that the user is not associated with the base station b; define the set of base stations associated with the user (i.e. the base station that provides service to the user) as Define the signal sent by the base station b as:

[0053] x b = w b s b

[0054] Where s b represents the signal sent by the base station b to the user, represents the transmit beamforming vector of the base station b, is a real matrix; therefore, for the base station b, the signal received at the user is the useful signal received by the virtual single-antenna user from the associated base station b, and the signal sent by other base stations to other virtual users; it can also be understood that the signal received at the user is the sum of the useful signals received by M virtual single-antenna users from their associated base stations, which can be represented as:

[0055]

[0056] in The received signal-to-interference-plus-noise ratio (SIR) at the user's location represents complex additive white Gaussian noise; the received SIR at the user's location can be expressed as:

[0057]

[0058] Where σ 2 This represents the noise power. According to Shannon's formula, the data rate from base station b to the user is...

[0059] R b =log(1+γ) b ).

[0060] S2. To maximize the achievable rate for users, construct a non-convex, multivariable coupled wireless resource allocation optimization problem.

[0061] Specifically, in this invention, the reachable rate of a user is defined as the sum of the rates of all reachable links between the user and its associated base stations. It is primarily influenced by three key variables: user multi-association parameters, base station transmit beamforming matrix, and RIS phase shift matrix. By setting constraints on these key variables, a non-convex, multivariate coupled wireless resource allocation optimization problem is established with the goal of maximizing the reachable rate of the user, expressed as:

[0062]

[0063] Where C1 represents the constraint on the number of base stations a user can associate with, C2 is the association coefficient constraint, C3 is the maximum transmit power constraint of the base station, and C4 is the phase shift constraint of the RIS reflection unit; C={c1,c2,…,c B} represents the user-associated parameters, W = {w1, w2, ..., w B} represents the base station beamforming vector, Φ i Φ represents the phase shift matrix of RIS i. j express The phase shift matrix; the base station set is B represents the number of base stations; c is defined as follows: b =1, indicating the user and base station Related, otherwise c b =0, indicating that the user is not associated with base station b; R b This represents the link rate between the user and base station b; M represents the maximum number of associations, and P represents the link rate between the user and base station b. max This indicates the maximum transmit power of the base station, w b The transmit beamforming vector of base station b is represented by L; L represents the number of RIS, and N represents the total number of reflective elements in a single RIS; φ l,n This represents the phase shift of the nth reflection unit in RIS l.

[0064] S3. Decouple the wireless resource allocation optimization problem into three sub-problems of user multi-association, base station transmit beamforming design and multi-RIS collaborative phase shift matrix design.

[0065] Specifically, the wireless resource allocation optimization problem involves high coupling of multiple optimization variables, belongs to a non-convex optimization problem, and is difficult to solve directly; by the alternating iterative optimization idea, using the Lagrange dual theory, fractional programming and other methods, the complex and difficult to solve wireless resource allocation optimization problem in step 2 is decoupled into three sub-problems of user multi-association, base station transmit beamforming matrix design and multi-RIS collaborative phase shift matrix design, including:

[0066] S31. In the wireless resource allocation optimization problem, fix the base station beamforming vector W and the multi-RIS collaborative phase shift matrix Φ, and relax the constraint condition C2 in equation (1) to 0≤c b ≤1, obtain the user multi-association sub-problem, denoted as:

[0067]

[0068] S32. In the wireless resource allocation optimization problem, fix the user association parameter C according to the optimization result of the user multi-association sub-problem, and introduce an auxiliary variable Construct a new optimization problem, denoted as:

[0069]

[0070] Where When α is fixed, let Substitute the signal-to-noise ratio γ b into f2 to obtain the conversion problem, denoted as:

[0071]

[0072] Given the multi-RIS collaborative phase shift matrix Φ, obtain the base station transmit beamforming design sub-problem:

[0073]

[0074] S33. Similarly, based on the conversion problem of equation (4), given the base station beamforming vector W, obtain the multi-RIS collaborative phase shift matrix sub-problem:

[0075]

[0076] At this point, the original optimization problem (1) is successfully decoupled into the user multi-association sub-problem (2), the base station transmit beamforming matrix design sub-problem (5) and the multi-RIS collaborative phase shift matrix design sub-problem (6).

[0077] S4. Jointly solve the three sub-problems to obtain the optimal user multi-association parameter, the optimal base station transmit beamforming vector and the optimal multi-RIS cooperative phase shift matrix.

[0078] Specifically, as shown in FIG. 4, step S4 includes: Figure 2

[0079] S41. Parameter initialization: initialize the user association parameter C, the base station transmit beamforming vector W, and the phase shift matrix of each RIS to provide initial conditions for subsequent iterative optimization;

[0080] S42. Solve the user multi-association sub-problem by using the Lagrangian dual decomposition method, i.e., determine which base stations the user is associated with to provide services for the user, and update the user association parameter C, which specifically includes:

[0081] The user multi-association sub-problem is a convex optimization problem, which can be analyzed by using the Lagrangian duality theory; introduce the Lagrangian multiplier v, and construct the Lagrangian function

[0082]

[0083] to realize the relaxation of the coupling constraint, and further construct the Lagrangian dual function Minimize the Lagrangian dual function on the feasible region of the dual multiplier to form the dual problem as

[0084]

[0085] In each iteration process, the user association parameter (the association coefficient of the user and each base station) can be updated by the following way

[0086]

[0087] denotes the association coefficient of the user and the base station at the t+1th iteration; denotes the index value b of the base station b is equal to

[0088] Update the Lagrangian multiplier v by the sub-gradient method as follows,

[0089]

[0090] is the step size of the tth iteration

[0091] According to equations (8) and (9), the parameters at the user end and the base station end are updated iteratively, and finally the optimal user multi-association scheme is obtained.

[0092] ​S43. Selecting multiple RISs for the user and its associated base stations for assisting communication based on the shortest distance criterion; then solving the base station transmit beamforming matrix design sub-problem, updating the base station transmit beamforming vector.

[0093] Specifically, after successfully solving the user multi-association sub-problem, it is crucial to select the RIS combination serving the target user and the associated base station. In order to fully exploit the advantages of RIS, the present application proposes an RIS selection scheme based on the shortest distance. Mainly includes: first, all RISs are numbered, such as RIS1, RIS 2, …, RIS L; then, any two RISs are paired, such as (RIS1, RIS2), (RIS2, RIS1), (RIS1, RIS 3), …, (RIS i, RIS j); finally, for each base station associated with the user, an auxiliary RIS group is selected, which specifically includes calculating the total path distance between each RIS pair at the base station and the user until all the above RIS combinations are traversed; let d b,i , d i,j , and d j be the distances between the base station b to RIS i, RIS i to RIS j, and RIS j to the user, respectively, according to the formula The RIS pair with the smallest total path distance is selected for cooperation to assist the communication between the user and the base station. This selection method can reduce transmission loss and improve the overall transmission rate and reliability of the system.

[0094] Specifically, the base station transmit beamforming design sub-problem of formula (5) is a multi-partition programming problem, therefore, the present application considers further transforming it into a more easily handled function form such as

[0095]

[0096] where is a complex auxiliary vector. Therefore, the optimization problem of the base station transmit beamforming vector W can be equivalent to a bi-convex optimization problem about W and ξ, which is specifically expressed as

[0097]

[0098] For the solution of the optimization problem (10), it can be divided into two steps. First, optimize the complex auxiliary vector ξ in the case of fixing the base station transmit beamforming vector W, to get the optimal complex auxiliary vector ξ opt ; then, using the complex auxiliary vector ξ opt to solve the optimal base station beamforming vector W opt .

[0099] When the base station transmits the beamforming vector W, let f 2,2 Take the partial derivative of ξ b , we can get

[0100]

[0101] By setting , we can get

[0102]

[0103] After that, the optimization objective function can be re-expressed as f 2.2 (W,ξ opt ). Regarding the optimization problem of the base station transmit beamforming vector W, the Lagrange multiplier method can be used to solve it. Specifically, introduce the Lagrange multiplier λ = [λ1, λ2,..., λ B ] T Construct the Lagrange function, which is specifically described as

[0104]

[0105] The optimal base station transmit beamforming vector W opt can be obtained by setting , which is specifically The optimal Lagrange multiplier λ should satisfy and can be obtained by bisection method.

[0106] In the whole process, the complex auxiliary vector, the base station transmit beamforming vector and the Lagrange multiplier are iteratively updated until the base station transmit beamforming problem converges.

[0107] S44. Based on the user association parameter updated in step S41 and the base station transmit beamforming vector updated in step S42, solve the multi-RIS cooperative phase shift matrix design sub-problem to update the multi-RIS cooperative phase shift matrix.

[0108] Specifically, after updating the base station transmit beamforming matrix, the multi-RIS cooperative phase shift matrix design sub-problem (6) can be obtained according to the decomposition of the wireless resource allocation optimization problem in step S3. The multi-RIS cooperative phase shift matrix design sub-problem is a non-convex optimization problem, and the present application converts this sub-problem into a problem with lower complexity by fractional programming method. First, auxiliary parameters are introduced to reconstruct the multi-RIS cooperative phase shift matrix design sub-problem, and the auxiliary parameters are solved by setting the partial derivative to zero; then, the multi-RIS cooperative phase shift matrix design sub-problem is converted into a reconstructed optimization problem, and the closed-form approximate solution of the RIS phase shift problem can be directly obtained; finally, the phase shift matrix of other auxiliary RIS is optimized according to the same process. It can be noted that Φ i and Φ jIn the channel is symmetric. Therefore, the optimization method proposed for Φ i is also applicable to Φ j . The present application is illustrated by taking the optimization of Φ j when Φ i is fixed as an example. First, the channel is reconstructed as follows

[0109]

[0110] where A i,j = g i + g j Φ j F i,j , B b,j = e b H b + g j Φ j G b,j are the reconstructed channel parameters.

[0111] Since Φ i = diag(φ i ), φ i = [φ i,1 , φ i,2 ,..., φ i,N ] T , the channel can be further expressed as Substituting the reconstructed channel into the objective function of equation (6), the new optimization objective function is obtained as

[0112]

[0113] Define auxiliary variables x b,i,j = diag(A i,j )G b,i w b , z b,j = B b,j w b , x b′,i,j = diag(A i,j )G b′,i w b′ , z b′,j = B b′,j w b′ . By a quadratic transformation, f 3.1 (φ i ) is converted to

[0114]

[0115] where is introduced. Thus, the corresponding reconstruction optimization problem can be described as

[0116]

[0117] The solution of this optimization problem can be divided into two main steps. First, fix the variable φ i to simplify the original problem and solve the optimal auxiliary variable ε opt . Second, based on the optimal auxiliary variable ε opt , further optimize the remaining variable φ i . Specifically, let Thus, the optimization objective function can be obtained as

[0118]

[0119] where

[0120]

[0121] The second term in equation (17) can be expanded as Since P is a Hermitian matrix, φ i H Pφ i can be rewritten as

[0122]

[0123] Subsequently, equation (18) is substituted into equation (17) and the constant term q2, f 3.3 (φ i ,ε opt ) can be re-expressed as

[0124]

[0125] where (a) holds because η n and the constant term q3 is as follows

[0126]

[0127] Discarding the terms irrelevant to the phase shift, maximizing f 3.4 (φ i,n ,ε opt ) is equivalent to maximizing Thus, the optimal φ can be updated by In addition, since Φ i and Φ j are symmetric in the channel , i.e., only the optimization of Φ​i and Φ j One of them, the other can be solved using similar methods.

[0128] S45. Determine whether the wireless resource allocation optimization problem converges, if yes, get the optimal user multi-association parameter, the optimal base station transmit beamforming vector and the optimal multi-RIS cooperative phase shift matrix, if not, return to step S42.

[0129] In an embodiment, the present application compares the relationship between the sum rate (i.e., the user achievable rate) and the base station maximum transmit power P max , under different numbers of user antennas (i.e., different numbers of associations), as shown in Figure 4 From the figure, it can be seen that when M = 1, i.e., the user can only be associated with one base station at most, the system sum rate is significantly lower than that of multi-association. This is because multi-association can better utilize multi-base station resources. In addition, as P max increases, the sum rate of all schemes increases, and the algorithm proposed in the present application is superior to other schemes, which shows that multi-RIS cooperation can effectively improve the system sum rate.

[0130] Figure 5 The figure shows the change of the sum rate of different schemes under the base station maximum transmit power P max . As P max increases, the sum rate of all algorithms is increasing. Since the scheme proposed in the present application can select the RIS pair that provides excellent assistance effect, the performance is superior to the fixed cooperation RIS pair scheme. In addition, it can also be found from the figure that the performance of the single RIS scheme is poorer than that of other multi-RIS cooperation schemes.

[0131] In the present application, unless otherwise explicitly specified and limited, the terms "mounting", "setting", "connecting", "fixing", "rotating" and the like should be understood broadly, for example, it can be fixedly connected, or it can be detachably connected, or it can be integrated; it can be mechanically connected, or it can be electrically connected; it can be directly connected, or it can be indirectly connected through an intermediate medium; it can be the internal communication of two elements or the interaction relationship between two elements, unless otherwise explicitly limited, the above-mentioned terms in the present application can be understood according to the specific meaning according to the specific circumstances by those skilled in the art.

[0132] Although embodiments of the present application have been shown and described, those skilled in the art can understand that various changes, modifications, replacements and variations can be made to these embodiments without departing from the principles and spirits of the present application, the scope of the present application is defined by the appended claims and their equivalents.

Claims

1. A multi-RIS co-assisted user-centralized network rate optimization method, characterized in that, The method comprises the following steps: S1. Constructing a user-centered multi-RIS cooperative assisted dense base station deployment high-frequency wireless communication network model; S2. Constructing a non-convex multi-variable coupled wireless resource allocation optimization problem with the maximum user reachable rate as the target, wherein the user reachable rate is the sum of the rates of all reachable links between the user and the multiple base stations associated with the user; The non-convex multi-variable coupled wireless resource allocation optimization problem is expressed as: C4: |φ l,n | = 1, l = 1, 2,..., L, n = 1, 2,..., N wherein C1 represents the number of user-associable base stations constraint, C2 is the association coefficient constraint, C3 is the maximum base station transmit power constraint, C4 is the RIS reflection unit phase shift constraint; C = {c1, c2, …, c B} represents the user association parameter, c b ∈ C represents the association coefficient of the user and the base station b; W = {w1, w2, …, w B} represents the base station beamforming vector, w b represents the transmit beamforming vector of the base station b; Φ i represents the phase shift matrix of the RIS i, Φ j represents the phase shift matrix of the RIS ; the base station set is B represents the number of base stations; c b = 1 represents that the user is associated with the base station , otherwise c b = 0 represents that the user is not associated with the base station b; R b represents the rate of the link between the user and the base station b; M represents the maximum association number, P max represents the maximum base station transmit power, w b represents the transmit beamforming vector of the base station b; L represents the number of RISs; φ l,n represents the phase shift of the nth reflection unit of the RIS l, and N represents the total number of reflection units of a single RIS. S3. Decoupling the wireless resource allocation optimization problem into three sub-problems of user multi-association, base station transmit beamforming design and multi-RIS cooperative phase shift matrix design; S4. Jointly and iteratively solving the three sub-problems to obtain the optimal user multi-association parameter, the optimal base station transmit beamforming vector and the optimal multi-RIS cooperative phase shift matrix.

2. The multi-RIS-collaborative-assisted user-centralized network rate optimization method of claim 1, wherein, The dense base station deployment high frequency wireless communication network model includes a plurality of base stations, a plurality of RISs and a user, and the base station set is denoted as The RIS set is denoted as B represents the number of base stations, and L represents the number of RISs; wherein the user is equipped with M antennas, each base station is equipped with M t antennas, and it is assumed that the M antennas of the user are regarded as M virtual single-antenna users at the same position, each antenna is respectively associated with a different base station, and receives independent data streams from the base stations to realize virtual spatial division multiplexing; each RIS is equipped with N = N x × N y reflective units, N x represents the number of reflective units arranged horizontally in the RIS, and N y represents the number of reflective units arranged vertically in the RIS; Definition of Φ i = diag(φ i ) denotes the phase shift matrix of RIS , where φ i = [φ i,1 , φ i,2 ,..., φ i,N ] T denotes the phase shift vector of RIS i, denotes the phase shift of the n-th reflecting unit of RIS i, θ i,n ∈ [0, 2π) denotes the phase shift angle of the n-th reflecting unit of RIS i; Definition of H b denotes the channel between base station and user; G b,i denotes the channel between base station b and RIS i; F i,j denotes the channel between RIS i and RIS ; g i denotes the channel between RIS i and user; Introducing an equivalent line-of-sight sphere model with a radius r b , the probability of a link with length L d being a line-of-sight link is Define a blockage coefficient e b to represent whether a line-of-sight link exists between a user and a base station b, when P LOS is greater than or equal to a threshold τ, define e b = 1, indicating that a line-of-sight link exists, otherwise e b = 0, indicating that the line-of-sight link is blocked; Thus, the equivalent cascaded channel between the base station b and the user is expressed as: Φ j denotes the phase shift matrix of the RIS denotes the phase shift matrix of the RIS j denotes the channel between the RIS j and the user b,j denotes the channel between the base station b and the RIS j; wherein: where N BI denotes the number of paths between the base station b and the RIS i, N IU denotes the number of paths between the RIS i and the user; p BI,l denotes the complex gain of the lth link between the base station b and the RIS i, p IU,l denotes the complex gain of the lth link between the RIS i and the user; a r denotes the array response vector of the receiving end, a t denotes the array response vector of the transmitting end; and denote the azimuth and elevation angles of arrival at the RIS, respectively; is the azimuth angle of departure from the base station; and denote the azimuth and elevation angles of departure from the RIS, respectively; define the association coefficient c b between the user and the base station b, if c b = 1, it means that the user is associated with the base station b, if c b = 0, it means that the user is not associated with the base station b; define the set of base stations associated with the user as define the signal transmitted by the base station b as: x b = w b s b where s b represents the signal transmitted by the base station b to the user, w b represents the transmit beamforming vector of the base station b; thus, for the base station b, the signal received at the user is the useful signal received by the virtual user from the relevant base station b, and the signals transmitted by the other base stations to other virtual users, represented as: wherein denotes the complex additive white Gaussian noise at the user; the received signal-to-interference-and-noise ratio at the user is denoted by where σ 2 is the noise power.

3. The multi-RIS-collaborative-assisted user-centralized network rate optimization method of claim 1, wherein, Decoupling the wireless resource allocation optimization problem into three sub-problems includes: S31. In the wireless resource allocation optimization problem, the fixed base station beamforming vector and the multi-RIS cooperative phase shift matrix, the constraint condition C2 is relaxed to 0≤c b ≤1, to obtain a user multi-association sub-problem; S32. In the wireless resource allocation optimization problem, the fixed user association parameter is introduced and an auxiliary variable is introduced A new optimization problem is constructed; after processing the new optimization problem, a conversion problem is obtained, given a multi-RIS cooperative phase shift matrix, a base station transmit beamforming design sub-problem is obtained; S33. In the conversion problem, given the base station beamforming vector, the multi-RIS cooperative phase shift matrix sub-problem is obtained.

4. The multi-RIS-collaborative-assisted user-centralized network rate optimization method of claim 1, wherein, Step S4 specifically comprises: S41. Solving the user multi-association sub-problem by using the Lagrange dual decomposition method to update the user association parameter; S42. Selecting the auxiliary RIS group for the user and the base station associated with the user based on the shortest distance criterion, solving the base station transmit beamforming matrix design sub-problem based on the auxiliary RIS group and the user association parameter updated in step S41, and updating the base station transmit beamforming vector; S43. Solving the multi-RIS cooperative phase shift matrix design sub-problem based on the user association parameter updated in step S41 and the base station transmit beamforming vector updated in step S42, and updating the multi-RIS cooperative phase shift matrix; S44. Judging whether the wireless resource allocation optimization problem converges, if yes, the optimal user multi-association parameter, the optimal base station transmit beamforming vector and the optimal multi-RIS cooperative phase shift matrix are obtained, and if not, returning to step S41.

5. The multi-RIS-collaborative-assisted user-centralized network rate optimization method of claim 4, wherein, Step S41 solves the user multi-association sub-problem by using the Lagrange dual decomposition method, comprising: S411. Introduce a Lagrange multiplier v to construct a Lagrange function and a Lagrange dual function g(v); S412. Minimizing the Lagrange dual function to form a dual problem; S413. Updating the user association parameter according to the Lagrange multiplier v; S414. Updating the Lagrange multiplier v by using the sub-gradient algorithm; then judging whether the Lagrange multiplier v converges, if yes, the dual problem reaches the global optimum, and the current user association parameter is output, and if not, returning to step S413.

6. The multi-RIS-collaborative-assisted user-centralized network rate optimization method of claim 4, wherein, Step S42 specifically comprises: S421. Pairing all RISs two by two to obtain a plurality of RIS pairs, for each base station associated with the user Calculate the total path distance between all RIS pairs and the base station b, and select the RIS pair with the smallest total path distance as the auxiliary RIS group of the base station b; wherein the total path distance of any RIS pair (RISi, RISj) and the base station b is the sum of the distance from the base station b to RISi, the distance from RISi to RISj, and the distance from RISj to the user; denotes a set of base stations, denotes a set of RISs;​​ S422. Introducing a complex auxiliary vector ξ to convert the base station transmit beamforming design sub-problem into a bi-convex optimization problem about the base station transmit beamforming vector and the complex auxiliary vector ξ; S423. Fixing the base station transmit beamforming vector in the bi-convex optimization problem, and solving to obtain the optimal complex auxiliary vector ξ opt ; S424. obtaining the optimal complex auxiliary vector ξ opt The optimal base station transmit beamforming vector is obtained by substituting the double convex optimization problem.

7. The multi-RIS-collaborative-assisted user-centralized network rate optimization method of claim 4, wherein, Step S43 specifically comprises: S431. Introducing an auxiliary parameter ε to reconstruct the multi-RIS cooperative phase shift matrix design sub-problem to obtain a reconstructed optimization problem; S432. Fix the multi-RIS cooperative phase shift matrix in the reconstruction optimization problem, and solve to obtain the optimal auxiliary parameter ε opt ; S433. Substitute the optimal auxiliary parameter ε opt into the reconstruction optimization problem to obtain the optimal multi-RIS cooperative phase shift matrix.

Images