Practical intelligent metasurface-assisted RSMA system beam forming method

Abstract

The invention provides a beam forming method for a practical intelligent super-surface (PRIS) assisted rate segmentation multiple access (RSMA) system, so that multiple users and rate maximization are achieved, and an amplitude-phase coupling relation and 1-bit phase shift of the PRIS are considered. Due to the complex amplitude-phase coupling relation of the pratical-RIS, the optimization is difficult to directly solve. Therefore, the invention provides an effective variable separation algorithm developed on the basis of an alternating multiplication ADMM principle. It is worth mentioning that the algorithm provided by the invention is compatible with RIS reflection optimization under continuous phase shift or any amplitude-phase coupling relationship, and the problem of sum rate maximization of the passive-RIS assisted non-orthogonal multiple access NOMA and orthogonal multiple access OMA systems can be directly solved by using the algorithm provided by the invention. Compared with an ideal RIS and continuous phase shift 1-bit quantization, the method can obviously improve the system performance, and has important practical application significance.

Description

Technical Field

[0001] The invention belongs to the technical field of wireless communications, and mainly relates to a method for beamforming of a practical intelligent metasurface-assisted RSMA system. Background Art

[0002] The combination of Reconfigurable Intelligent Surface (RIS) and Rate-splitting Multiple Access (RSMA) can effectively improve communication quality and is the most needed technology for the sixth-generation (6G) communication system.

[0003] Reconfigurable intelligent metasurface RIS is a planar array composed of a large number of low-cost passive components, each of which can be independently controlled. RIS can adjust the phase shift of the incident signal according to the channel state information, so that the signal can be coherently superimposed in the user direction, thereby realizing the reconfiguration of the wireless propagation environment to significantly improve the communication quality and coverage of the wireless network. Compared with the traditional communication system of relay, RIS has no self-interference and complex RF links, and is a low-cost communication device. Based on the above advantages, RIS has received extensive research attention.

[0004] Rate Division Multiple Access (RSMA) is considered to be an indispensable, efficient and reliable multiple access strategy and interference management technology in 6G wireless networks. The core concept of RSMA is to divide user information into public information and private information at the transmitter, and encode them into public streams and private streams respectively. The public stream can be decoded by all users, while the private stream can only be decoded by the corresponding user. RSMA can not only decode the interference, but also treat the interference as noise, thereby achieving more flexible interference processing. Studies have shown that compared with traditional multiple access technologies such as Orthogonal Multiple Access (OMA) and Non-Orthogonal Multiple Access (NOMA), RSMA has higher spectrum efficiency.

[0005] Relevant studies have shown that the combination of RIS and RSMA can significantly improve system performance. However, most of the existing research on RIS is based on an ideal phase shift model, which usually assumes that the reflection amplitude of each reflective element is constant. In addition, due to hardware limitations, the phase shift of RIS is discrete in practice. In the research on RIS-assisted RSMA, the amplitude-phase coupling relationship and discrete phase shift are hardly considered, which will inevitably cause a large performance loss in practical applications. It is worth mentioning that recent studies have proposed an empirical formula for a practical intelligent metasurface (practical-RIS), which fully considers the coupling relationship between amplitude response and phase shift. However, there is currently no research combining practical-RIS with RSMA. In response to the above problems, the present invention proposes a beamforming method for a practical-RIS-assisted RSMA system. Summary of the invention

[0006] The present invention aims to solve the above technical problems and proposes a practical intelligent metasurface practical-RIS assisted rate division multiple access RSMA system beamforming method. Specifically, the technical solution of the method proposed by the present invention is as follows:

[0007] A practical intelligent metasurface assisted RSMA system beamforming method is provided. The method constructs a multi-user sum rate maximization problem in a practical intelligent metasurface practical-RIS assisted rate division multiple access RSMA system, and utilizes the coupled amplitude-phase relationship and 1-bit discrete phase shift of the practical intelligent metasurface to improve the spectrum efficiency. The specific steps include:

[0008] Step 1: Construct a practical-RIS-assisted RSMA system downlink transmission model and construct a multi-user and rate maximization optimization problem. The specific steps are as follows:

[0009] 1) In this system, the base station is equipped with M antennas and provides services for single-antenna users through linear precoding technology. In order to enhance the transmission effect, the base station is assisted by a reconfigurable smart surface containing N passive reflective elements. Since the direct link is blocked, the transmission signal of the base station can only be transmitted through the reflection link;

[0010] 2) Cascaded wireless channel uses PL = -30-10αlog 10 (d) dB represents large-scale fading, and the Rayleigh fading model is used for small-scale fading, where α is the path loss exponent and d is the distance between devices;

[0011] 3) The channel from the base station to practical-RIS is expressed as The practical-RIS channel to the kth user is expressed as The phase shift matrix of practical-RIS is expressed as According to the RSMA system design principle, the transmission signal at the base station is expressed as Among them, w c and Denotes the linear precoding vector of the public and private streams of user k, and defines s = [s c ,s 1 ,s 2 ,...,s K ] T is the data symbol at the base station, where (·) H ,(·) T are the conjugate transpose and transpose of the matrix respectively, is a circularly symmetric complex Gaussian distribution CSCG,

[0012] 4) The received signal of user k is expressed as in It is represented by the additive white Gaussian noise AWGN of user k;

[0013] 5) In the RSMA system, the signal at the base station can be divided into a public part and a private part. All public parts will use a unified code book to generate a public stream and allow all users to decode it. At the same time, the private part will be encoded as a private stream and only the corresponding user is allowed to decode it. Therefore, the signal-to-interference-to-noise ratio of the public stream of user k is The signal-to-interference-to-noise ratio of the private flow of user k is

[0014] 6) In the practical-RIS-assisted RSMA system scenario, the multi-user sum rate maximization problem is

[0015]

[0016] In the formula n=1,2,...,N,P max and γ th represents the maximum power of the BS and the QoS threshold of user k; constraint condition The phase shift for each RIS unit is θ n Scope; Constraints is the empirical formula of practical-RIS, where β min ,δ,α are constants related to the specific circuit implementation, and the formula guarantees its amplitude-phase coupling relationship;

[0017] Step 2: Introduce auxiliary variables ξ and ξ k , a variable separation algorithm is proposed, the specific steps are as follows:

[0018] 1) The objective function of the multi-user and rate maximization problem in the practical-RIS-assisted RSMA system can be equivalently expressed as

[0019]

[0020] Step 3: Adopt the alternating optimization algorithm to decompose the optimization problem represented by formula (2) into two sub-problems. The specific steps are as follows:

[0021] 1) The optimization problem expressed by formula (2) is converted into two sub-problems of alternating optimization of w and Θ, which are updated as follows:

[0022]

[0023] Step 4: When the phase shift matrix Θ is fixed, the precoding vector w is optimized. The specific steps are as follows:

[0024] 1) The optimization problem expressed by formula (3.a) is equivalent to

[0025]

[0026] 2) Due to constraints Non-convexity, its optimization can be restated as

[0027]

[0028] in, represents the real part of a complex number, (·) * is the conjugate of a complex number.

[0029] 3) μ in formulas (5.a) and (5.b) c,k and μ k The closed-form solution is

[0030] and

[0031] 4) The non-convex constraints in the optimization problem expressed in formula (4) After replacing it with formula (5), the problem is transformed into a convex optimization problem, which can be solved using general optimization tools;

[0032] Step 5: When the precoding vector w is fixed, optimize the phase shift matrix Θ. The specific steps are as follows:

[0033] 1) Introduce auxiliary variable θ = diag(θ H ), Θ is separated from the complex constraints, and an algorithm based on the principle of alternating multiplier method is proposed for decoupling. The optimization problem expressed by formula (3.b) can be restated as

[0034]

[0035] In the formula μ>0 is a given parameter, is the scaled dual variable;

[0036] 2) The optimization problem expressed in formula (6) is updated and iterated through the following steps:

[0037]

[0038]

[0039] ν l+1 =v l +θ l+1 -θ l+1 (7.c)

[0040] In the formula, is the value of the optimization problem expressed by formula (6) at the l+1th iteration;

[0041] 2.1) The constraints in the optimization problem expressed by formula (7.a) are due to the constraints With non-convexity, its optimization can be restated as

[0042]

[0043] In the formula

[0044]

[0045] 2.2) μ in formulas (8.a) and (8.b) c,k and μ k The closed-form solution is

[0046]

[0047] 2.3) Non-convex constraints in the optimization problem represented by formula (7.a) Substituting into formula (8.a), the problem is transformed into a convex optimization problem, which can be solved using general optimization tools;

[0048] 2.4) A solution method is proposed to decompose the optimization problem represented by formula (7.b) into N independent sub-problems to form a decoupled structure, which is transformed into

[0049]

[0050] 2.5) Set the constraints Embedding the objective function, formula (9) is reformulated as

[0051]

[0052] 2.6) A one-dimensional search method is used for each sub-problem to determine the optimal phase one by one to maximize the objective function value.

[0053] 3) The method proposed in the present invention is compatible with 1-bit discrete phase shift RIS reflection optimization. It only needs to adjust the search range in step 2.5) to {-π, 0}, and the above steps can be used to effectively perform the optimization process.

[0054] The advantages of the present invention are:

[0055] 1) The present invention proposes and solves the optimization problem of multi-user and rate maximization in a practical-RIS-assisted RSMA system. Specifically, the present invention considers the coupling relationship between the phase shift and amplitude of RIS and proposes a performance enhancement method that is compatible with all coupling curves;

[0056] 2) The algorithm proposed in the present invention decouples the coupled variables in multiple complex constraints by introducing auxiliary variables, thereby allowing each constraint to be processed separately; at the same time, the algorithm is compatible with RIS reflection optimization under continuous phase shift or any amplitude-phase coupling relationship;

[0057] 3) The algorithm proposed in this paper can be directly used for multi-user and rate maximization problems in practical-RIS-assisted NOMA and OMA systems;

[0058] 4) The algorithm proposed in the present invention is superior to the ideal RIS and 1-bit quantization of continuous phase shift;

[0059] 5) The method proposed in this invention has better performance than the practical-RIS assisted OMA or NOMA system. BRIEF DESCRIPTION OF THE DRAWINGS

[0060] Figure 1 It is a model diagram of the downlink practical-RIS assisted RSMA communication system;

[0061] Figure 2 It is a flowchart of the optimization method for the optimization problem represented by formula (10);

[0062] Figure 3 The convergence of the algorithm proposed in this invention is demonstrated;

[0063] Figure 4 The effect of base station transmit power on different multiple access schemes is demonstrated;

[0064] Figure 5 Demonstrated the impact of base station transmit power on system performance;

[0065] Figure 6 Demonstrated the impact of the number of base station antennas on system performance;

[0066] Figure 7 The effect of the number of smart metasurface elements on system performance is demonstrated. DETAILED DESCRIPTION

[0067] The present invention proposes a beamforming method for a practical-RIS-assisted RSMA system. The technical solution of the embodiment of the present invention will be further described in detail below in conjunction with the accompanying drawings of the present invention.

[0068] Embodiment 1

[0069] Attached Figure 1 The figure shows the model diagram of the downlink practical-RIS-assisted RSMA communication system. The positions of the base station and RIS are set to (0,0,0) meters and (40,40,0) meters respectively; users are evenly distributed in a circle with a radius of 8 meters and a center of (50,50,0) meters.

[0070] The cascaded wireless channel uses PL = 37.3 + 22.0log (d) dB to represent large-scale fading and Rayleigh fading model to represent small-scale fading, where α is the path loss exponent and d is the distance between devices; noise power

[0071] 1) The channel from the base station to practical-RIS is expressed as The practical-RIS channel to the kth user is expressed as According to the RSMA system design principle, the transmission signal at the base station is expressed as The received signal of user k is expressed as

[0072] 2) In the RSMA system, the signal at the base station can be divided into a public part and a private part, and encoded into a public stream and a private stream respectively. The signal-to-interference-to-noise ratio of the public stream of user k is The signal-to-interference-to-noise ratio of the private flow of user k is

[0073] 3) S101, in the practical-RIS-assisted RSMA system scenario, the multi-user sum rate maximization problem is

[0074]

[0075] In the formula n=1,2,...,N,P max and γ threpresents the maximum power of the BS and the QoS threshold of user k; constraint condition is the phase shift range of each RIS unit; constraints It is the empirical formula of practical-RIS, which guarantees its amplitude-phase coupling relationship.

[0076] 4) S102, randomly initialize the precoding vector w, the RIS phase shift matrix Θ and the rate decision threshold.

[0077] 5) Introduce auxiliary variables ξ and ξ k , an effective variable separation algorithm is proposed. The specific steps are as follows:

[0078] 5.1) The objective function of the multi-user and rate maximization problem in the practical-RIS-assisted RSMA system can be equivalently expressed as

[0079]

[0080] 6) Using the alternating optimization algorithm, the optimization problem represented by formula (2) is decomposed into two sub-problems. The specific steps are as follows:

[0081] 6.1) The optimization problem expressed by formula (2) is converted into two sub-problems of alternating optimization of w and Θ, which are updated as follows:

[0082]

[0083] 7) S103, when the phase shift matrix Θ is fixed, the precoding vector w is optimized. The specific steps are as follows:

[0084] 7.1) The optimization problem expressed by formula (3.a) is equivalent to

[0085]

[0086] 7.2) Due to the constraints Non-convexity, its optimization can be restated as

[0087]

[0088]

[0089] 7.3) μ in formulas (5.a) and (5.b) c,k and μ k The closed-form solution is

[0090] and

[0091] 7.4) The non-convex constraints in the optimization problem expressed in formula (4) are After replacing it with formula (5), the problem is transformed into a convex optimization problem and can be solved using general optimization tools.

[0092] 8) S104, when the precoding vector w is fixed, optimize the phase shift matrix Θ. The specific steps are as follows:

[0093] 8.1) Introduce auxiliary variable θ = diag(θ H ), separating Θ from the complex constraints. An algorithm based on the principle of alternating multiplier method is proposed for decoupling. The optimization problem expressed by formula (3.b) is reformulated as

[0094]

[0095] In the formula μ>0 is a given parameter, is the scaled dual variable.

[0096] 8.2) Attachment Figure 2 The optimization process of the problem represented by formula (6) is expressed. The optimization problem is updated and iterated through the following steps:

[0097]

[0098] ν l+1 =v l +θ l+1 -θ l+1 (7.c)

[0099] In the formula, is the value of the optimization problem expressed by formula (6) at the l+1th iteration;

[0100] 8.2.1) The constraints in the optimization problem represented by formula (7.a) are due to the constraints With non-convexity, its optimization can be restated as

[0101]

[0102] In the formula

[0103]

[0104] 8.2.2) In formulas (8.a) and (8.b), μ c,k and μ k The closed-form solution is

[0105]

[0106] 8.2.3) The non-convex constraints in the optimization problem represented by formula (7.a) Substituting into formula (8.a), the problem is transformed into a convex optimization problem, which can be solved using general optimization tools.

[0107] 8.2.4) A concise and efficient solution method is proposed to decompose the optimization problem represented by formula (7.b) into N independent sub-problems to form a decoupled structure, which is transformed into

[0108]

[0109] 8.2.5) Constraints Embedding objective function, formula (9) is expressed as

[0110]

[0111] 8.2.6) A one-dimensional search method is used for each sub-problem to determine the optimal phase one by one to maximize the objective function value.

[0112] 9) S105, based on the optimized w m+1 and θ m+1 , calculate the multi-user sum rate for the nth iteration.

[0113] 10) S106, determine whether the convergence requirement is met. If the absolute value of the difference between the updated sum rate and the previous iteration sum rate is not greater than the decision threshold, the maximum value of the multi-user sum rate is given and the algorithm ends; otherwise, the multi-user sum rate value is saved and the algorithm goes to step S103 to continue updating the optimization parameters until the convergence condition is met.

[0114] Embodiment 2

[0115] 1) The channel from the base station to practical-RIS is expressed as The practical-RIS channel to the kth user is expressed as According to the design principle of NOMA system, the transmission signal at the base station is expressed as The received signal of user k is expressed as

[0116] 2) In the NOMA system, the signal-to-interference-noise ratio of user k’s signal decoded by user l is The signal-to-interference-noise ratio of user K is

[0117] 3) In the practical-RIS-assisted NOMA system scenario, the multi-user sum rate maximization problem is

[0118]

[0119] In the formula n=1,2,...,N,Pmax and γ th represents the maximum power of the BS and the QoS threshold of user k; constraint condition is the phase shift range of each RIS unit; constraints It is the empirical formula of practical-RIS, which guarantees its amplitude-phase coupling relationship.

[0120] 4) The following steps are solved using the method proposed by the present invention.

[0121] Embodiment 3

[0122] 1) In the ideal RIS-assisted RSMA system scenario, the multi-user sum rate maximization problem is

[0123]

[0124] In the formula n=1,2,...,N,P max and γ th represents the maximum power of the BS and the QoS threshold of user k; constraint condition is the phase shift range for each RIS unit; The constraints of the ideal RIS are that β min =1.

[0125] 2) The following steps are solved using the method proposed by the present invention.

[0126] To verify the effectiveness of the present invention, the present invention uses computer simulation experiments to prove the effectiveness of the algorithm proposed in this article. All simulation results in the figure are the average values ​​of 200 independent experiments, and the optimal phase shift and random phase shift are represented by "wo" and "wr" respectively.

[0127] Attached Figure 3 The convergence of the algorithm proposed in this paper is verified. It can be clearly seen from the figure that with the increase of the number of iterations, the multi-user and rate of the NOMA system and RSMA system both show an upward trend and eventually reach convergence.

[0128] Attached Figure 4 The performance comparison analysis of RSMA system with traditional OMA and NOMA schemes under different transmission powers at the BS is shown considering practical RIS. As the BS transmit power increases, the multi-user and rate also increase. This result shows that our proposed algorithm shows better robustness under different circumstances. Importantly, the practical-RIS-assisted RSMA scheme in this invention has better system performance compared to the equivalent NOMA and OMA schemes.

[0129] Attached Figure 5 The relationship between multi-user and rate and BS transmission power is shown. As the power increases, the system performance of all schemes improves significantly. min As the value of increases, the system performance also improves. Importantly, under the same configuration, the optimal phase shift (wo) has better system performance than the random phase shift (wr).

[0130] Attached Figure 6 , Attachment Figure 7 It is shown that the system performance improves with the increase of the number of base station antennas and the number of RIS reflector elements. Compared with the ideal RIS, our proposed scheme can achieve more than 190% performance improvement and more than 50% performance improvement compared with 1-bit quantization of continuous phase shift.

Claims

1. A practical intelligent metasurface assisted RSMA system beamforming method, which constructs the multi-user and rate maximization problem in the practical intelligent metasurface practical-RIS assisted rate division multiple access RSMA system, and uses the coupled amplitude-phase relationship and 1-bit discrete phase shift of the practical intelligent metasurface to improve the spectrum efficiency. The specific steps include: Step 1: Construct a practical-RIS-assisted RSMA system downlink transmission model and construct a multi-user and rate maximization optimization problem. The specific steps are as follows: 1) In this system, the base station is equipped with M antennas and provides services for single-antenna users through linear precoding technology. In order to enhance the transmission effect, the base station is assisted by a reconfigurable smart surface containing N passive reflective elements. Since the direct link is blocked, the transmission signal of the base station can only be transmitted through the reflection link; 2) Cascaded wireless channel uses PL = -30-10αlog 10 (d) dB represents large-scale fading, and the Rayleigh fading model is used for small-scale fading, where α is the path loss exponent and d is the distance between devices; 3) The channel from the base station to practical-RIS is expressed as The practical-RIS channel to the kth user is expressed as The phase shift matrix of practical-RIS is expressed as According to the RSMA system design principle, the transmission signal at the base station is expressed as Among them, w c and Denotes the linear precoding vector of the public and private streams of user k, and defines s = [s c ,s1,s2,...,s K ] T is the data symbol at the base station, where (·) H ,(·) T are the conjugate transpose and transpose of the matrix respectively, is a circularly symmetric complex Gaussian distribution CSCG, 4) The received signal of user k is expressed as in It is represented by the additive white Gaussian noise AWGN of user k; 5) In the RSMA system, the signal at the base station can be divided into a public part and a private part. All public parts will use a unified code book to generate a public stream and allow all users to decode it. At the same time, the private part will be encoded as a private stream and only the corresponding user is allowed to decode it. Therefore, the signal-to-interference-to-noise ratio of the public stream of user k is The signal-to-interference-to-noise ratio of the private flow of user k is 6) In the practical-RIS-assisted RSMA system scenario, the multi-user sum rate maximization problem is In the formula n=1,2,...,N,P max and γ th represents the maximum power of the BS and the QoS threshold of user k; constraint condition The phase shift for each RIS unit is θ n Scope; Constraints is the empirical formula of practical-RIS, where β min ,δ,α are constants related to the specific circuit implementation, and the formula guarantees its amplitude-phase coupling relationship; Step 2: Introduce auxiliary variables ξ and ξ k , a variable separation algorithm is proposed, the specific steps are as follows: 1) The objective function of the multi-user and rate maximization problem in the practical-RIS-assisted RSMA system can be equivalently expressed as Step 3: Adopt the alternating optimization algorithm to decompose the optimization problem represented by formula (2) into two sub-problems. The specific steps are as follows: 1) The optimization problem expressed by formula (2) is converted into two sub-problems of alternating optimization of w and Θ, which are updated as follows: Step 4: When the phase shift matrix Θ is fixed, the precoding vector w is optimized. The specific steps are as follows: 1) The optimization problem expressed by formula (3.a) is equivalent to 2) Due to constraints Non-convexity, its optimization can be restated as in, represents the real part of a complex number, and (·)* is the conjugate of the complex number. 3) μ in formulas (5.a) and (5.b) c,k and μ k The closed-form solution is 4) The non-convex constraints in the optimization problem expressed in formula (4) After replacing it with formula (5), the problem is transformed into a convex optimization problem, which can be solved using general optimization tools; Step 5: When the precoding vector w is fixed, optimize the phase shift matrix Θ. The specific steps are as follows: 1) Introduce auxiliary variable θ = diag(θ H ), Θ is separated from the complex constraints, and an algorithm based on the principle of alternating multiplier method is proposed for decoupling. The optimization problem expressed by formula (3.b) can be restated as In the formula μ>0 is a given parameter, is the scaled dual variable; 2) The optimization problem expressed in formula (6) is updated and iterated through the following steps: n l+1 =n l +θ l+1 -θ l+1 (7.c) In the formula, is the value of the optimization problem expressed by formula (6) at the l+1th iteration; 2.1) The constraints in the optimization problem expressed by formula (7.a) are due to the constraints With non-convexity, its optimization can be restated as In the formula 2.2) μ in formulas (8.a) and (8.b) c,k and μ k The closed-form solution is 2.3) Non-convex constraints in the optimization problem represented by formula (7.a) Substituting into formula (8.a), the problem is transformed into a convex optimization problem, which can be solved using general optimization tools; 2.4) A solution method is proposed to decompose the optimization problem represented by formula (7.b) into N independent sub-problems to form a decoupled structure, which is transformed into 2.5) Set the constraints Embedding the objective function, formula (9) is reformulated as 2.6) A one-dimensional search method is used for each sub-problem to determine the optimal phase one by one to maximize the objective function value.

Images