Optimization method and device for maximizing sum rate in active RIS-assisted cell-free MIMO system, and medium

Abstract

The invention provides an optimization method and device for maximizing the sum rate in an active RIS-assisted cell-free MIMO system and a medium, and relates to the technical field of resource allocation of a mobile communication system. The method comprises the following steps: establishing an active RIS-assisted cell-free MIMO system model; based on the active RIS-assisted cell-free MIMO system model, jointly optimizing received beam forming, RIS reflection coefficient and power distribution construction and rate maximization problem solving model; decomposing the sum rate maximization problem solving model into a receiving beam forming sub-problem, an active RIS reflection coefficient design sub-problem and a user power allocation sub-problem; and iteratively optimizing the beam forming matrix, the RIS reflection coefficient matrix and the user power by adopting an alternating optimization method. The joint resource allocation scheme is obtained by using the iterative optimization algorithm. The method provided by the invention can effectively improve the sum rate of the active RIS assisted cell-free MIMO system.

Description

Technical Field

[0001] This invention relates to the field of resource allocation technology for mobile communication systems, and more specifically, to an optimization method, device, and medium for maximizing sum and rate in an active RIS-assisted cellless MIMO system. Background Technology

[0002] In practical applications, larger-scale antenna systems place higher demands on antenna integration within a limited space. Cellular-free MIMO (Multiple-Input Multiple-Output) systems, as one implementation approach, can significantly improve spectral efficiency and effectively expand coverage by deploying a large number of distributed radio frequency links and antennas over a wider geographical area. Meanwhile, reconfigurable intelligent surfaces (RIS), as low-cost and energy-efficient components, can replace some required base stations or construct new communication paths to further enhance system performance.

[0003] To overcome the inherent limitations of passive RIS, especially in harsh transmission environments with low signal-to-noise ratios or severe path loss in reflection channels, a novel active RIS architecture has been introduced. Active RIS integrates a reflective power amplifier in each RIS unit, achieving phase modulation and high-gain amplification of the reflected signal, thereby effectively compensating for path loss caused by multiplicative fading.

[0004] In active RIS-assisted cellless MIMO systems, the receiver beamforming design, RIS reflection coefficient design, and power allocation are generally interdependent and closely related in addressing the resource optimization problem of maximizing sum rate. However, existing work has only studied centralized MIMO scenarios with single active RIS assistance, and there is no relevant research on cellless MIMO systems with multiple active RIS assistance. Therefore, it is urgent to develop an optimization method for maximizing sum rate in active RIS-assisted cellless MIMO systems. Summary of the Invention

[0005] The present invention aims to solve at least one of the aforementioned technical problems existing in the prior art.

[0006] Therefore, the first aspect of the present invention provides a method for maximizing the sum and rate optimization in an active RIS-assisted cellless MIMO system.

[0007] A second aspect of the present invention provides a computer device.

[0008] A third aspect of the present invention provides a computer-readable storage medium.

[0009] This invention provides an optimization method for maximizing sum and rate in an active RIS-assisted cell-free MIMO system, comprising: Establish a model of an active RIS-assisted cellless MIMO system; Based on the aforementioned active RIS-assisted cellless MIMO system model, a model for solving the rate maximization problem is constructed by jointly optimizing receive beamforming, RIS reflection coefficient, and power allocation. The solution model for the sum rate maximization problem is decomposed into a receiver beamforming subproblem, an active RIS reflection coefficient design subproblem, and a user power allocation subproblem. Based on the subproblems obtained from the decomposition, an alternating optimization method is used to iteratively optimize the beamforming matrix, the RIS reflection coefficient matrix, and the user power.

[0010] The optimization method for maximizing sum and rate in an active RIS-assisted cellless MIMO system according to the above-described technical solution of the present invention may further have the following additional technical features: In the above technical solution, in the active RIS-assisted cellless MIMO system model, R active RIS-assisted B access points equipped with M reflective elements provide services to K single-antenna users. The channels between the access point, RIS, and single-antenna users follow a Rice distribution. Each access point performs beamforming processing on the received uplink signal and then transmits it back to the CPU via the forward backhaul link for merging and decoding.

[0011] In the above technical solution, the decoded... The signal for each user is represented as:

[0012] in, Indicates the first [number] after decoding Signals for individual users; Indicates the first The first access point sends the data back to the CPU. Detection signals for each user; Indicates that for the first The first user's The beamforming matrix at the access point; H is located at the superscript to indicate the conjugate transpose; the first... The access point and the first The channels between RIS follow a Rice distribution, represented as follows: ;No. The RIS and the first i The channel between users follows a Ricean distribution, represented as follows: ;No. The access point and the first i The channel between users follows a Ricean distribution, represented as follows: ; It is the set of complex numbers; Indicates the first A reflection coefficient matrix of an active RIS; Indicates the first i The transmission power corresponding to each user; Indicates the first i Signals sent by each user; Indicates the first Additive white Gaussian noise vector at each RIS; Indicates the first Additive white Gaussian noise vectors at each access point; Indicates the user index. , Indicates except the first All other users of that user.

[0013] In the above technical solution, the joint optimization of receive beamforming, RIS reflection coefficient, and power allocation construction and rate maximization problem-solving model includes: Based on user transmit power constraints, RIS reflection power constraints, RIS reflection amplitude constraints, and receive beamforming normalization constraints, a model for solving the rate maximization problem is constructed with the goal of maximizing the total rate. The solution model for the sum-rate maximization problem is as follows:

[0014] in, Representation and rate objective function; Indicates the first The received signal-to-noise ratio for each user; st represents the constraint condition; Represents a set of users; Represents an active RIS set; Represents the set of access points; Indicates the first A set of reflective elements for a RIS; Indicates user transmit power constraints; Indicates the first The maximum transmit power for each user; F in subscript indicates the Frobenius norm; Indicates RIS reflection power constraint; Indicates the first The variance of additive white Gaussian noise at each RIS point; This indicates the maximum reflected power of the active RIS; Indicates RIS reflection amplitude constraint; This indicates the maximum reflection amplitude of the RIS active element; This indicates the receive beamforming normalization constraint.

[0015] In the above technical solution, the iterative optimization of the beamforming matrix, RIS reflection coefficient matrix, and user power using an alternating optimization method includes: Design of receiver beamforming based on the closed-loop solution of generalized Rayleigh entropy; Solving the reflection coefficient of active RIS using fractional programming; Solving user power allocation schemes based on the continuous convex approximation method.

[0016] In the above technical solution, the design of receiver beamforming based on the closed-form solution of generalized Rayleigh entropy includes: fixed Solve The optimization of the receive beamforming problem is as follows:

[0017] The optimal solution for the generalized Rayleigh entropy is:

[0018]

[0019]

[0020] in, The objective function for the receiver beamforming problem is expressed as follows: These are beamforming auxiliary variables obtained after mathematical transformation; and A constant related to prior channel state information; express The identity matrix; Indicates that the index is The constants related to prior channel state information, where ; This represents the variance of Gaussian white noise at the RIS. This represents the variance of the Gaussian white noise at the access point.

[0021] In the above technical solution, the step of solving the active RIS reflection coefficient based on the fractional programming method includes: fixed Solve The subproblem of optimizing the reflection coefficient design of active RIS is:

[0022] Introducing auxiliary variables , Using fractional programming, the design subproblem of the reflection coefficient of an active RIS is transformed into solving the following convex optimization problem:

[0023] in, The variable to be optimized; , , , and All of them are constants related to prior channel state information;

[0024]

[0025]

[0026]

[0027]

[0028]

[0029] in, Describe a cascaded channel and define the cascaded channel. ; express An identity matrix of order 1; Use a convex optimization problem solver to solve for the design scheme of active RIS reflection coefficients in the current iteration.

[0030] In the above technical solution, the step of solving the user power allocation scheme based on the continuous convex approximation method includes: fixed Solve The subproblem of optimizing user power allocation is as follows:

[0031] in, and To represent constants related to prior channel state information, auxiliary variables are introduced. and The user power allocation subproblem is transformed into solving the following convex optimization problem:

[0032] in:

[0033] Use a convex optimization problem solver to find the optimal power allocation scheme for the user in the current iteration.

[0034] The present invention also provides a computer device, including a processor and a memory, wherein the memory stores a computer program, and when the computer program is loaded and executed by the processor, it implements the optimization method for maximizing speed in an active RIS-assisted cellless MIMO system as described in any of the above technical solutions.

[0035] The present invention also provides a computer-readable storage medium storing a program that, when loaded by a processor, implements an optimization method for maximizing sum and rate in an active RIS-assisted cellless MIMO system as described in any of the above technical solutions.

[0036] In summary, due to the adoption of the above-mentioned technical features, the beneficial effects of the present invention are: For active RIS-assisted cellless MIMO uplink communication systems, in order to improve the number of users and the data rate in the system, this invention proposes an effective method for jointly designing receive beamforming, active RIS reflection coefficient and user transmit power. The proposed joint optimization method can effectively realize the resource allocation of the system.

[0037] Specifically, this invention has the advantages of high system and rate performance, and can effectively utilize communication system resources. This method fully utilizes the inherent structure of the original optimization problem. First, it decomposes the complex joint optimization problem into a receive beamforming subproblem, an active RIS reflection coefficient design subproblem, and a power allocation subproblem. Then, it uses fractional programming to transform the active RIS reflection coefficient design subproblem into a convex optimization problem. Next, it introduces auxiliary variables to transform the power allocation subproblem into a convex optimization problem. Finally, it uses the aforementioned iterative optimization algorithm to obtain the joint resource allocation scheme. The method proposed in this invention can effectively improve the sum and rate of an active RIS-assisted cellless MIMO system.

[0038] Additional aspects and advantages of the invention will become apparent in the following description or may be learned by practice of the invention. Attached Figure Description

[0039] The above and / or additional aspects and advantages of the present invention will become apparent and readily understood from the description of the embodiments taken in conjunction with the following drawings, in which: Figure 1 This is a flowchart of an optimization method for maximizing sum and rate in an active RIS-assisted cellless MIMO system according to an embodiment of the present invention; Figure 2 This is a model diagram of an active RIS-assisted cellless MIMO system established in one embodiment of the present invention; Figure 3 This is a comparison chart of simulation results between the power allocation scheme proposed in one embodiment of the present invention and the benchmark scheme; Figure 4This is a comparison diagram of simulation results between the receiving beamforming scheme proposed in one embodiment of the present invention and the benchmark scheme. Detailed Implementation

[0040] To better understand the above-mentioned objectives, features, and advantages of the present invention, the present invention will be further described in detail below with reference to the accompanying drawings and specific embodiments. It should be noted that, unless otherwise specified, the embodiments and features described in these embodiments can be combined with each other.

[0041] Many specific details are set forth in the following description in order to provide a full understanding of the invention. However, the invention may also be practiced in other ways different from those described herein, and therefore the scope of protection of the invention is not limited to the specific embodiments disclosed below.

[0042] The following reference Figures 1 to 4 This invention describes optimization methods, apparatus, and media for maximizing sum and rate in an active RIS-assisted cellless MIMO system, according to some embodiments of the present invention.

[0043] Some embodiments of this application provide an optimization method for maximizing sum and rate in an active RIS-assisted cellless MIMO system.

[0044] like Figure 1 As shown, the first embodiment of the present invention proposes an optimization method for maximizing the sum and rate in an active RIS-assisted cellless MIMO system, including the following steps S1-S4.

[0045] S1. Establish a model of an active RIS-assisted cellless MIMO system.

[0046] like Figure 2 As shown, in the active RIS-assisted cellless MIMO system model, R active RIS equipped with M reflective elements assist B access points equipped with N antennas to provide services to K single-antenna users; wherein, the access point is schematically represented as a base station, and the channel between the access point, the RIS and the single-antenna users follows a Ricean distribution. Each access point performs beamforming processing on the received uplink signal and then transmits it back to the CPU through the forward backhaul link for merging and decoding.

[0047] Specifically, after decoding, the first The signal for each user is represented as:

[0048] in, Indicates the first [number] after decoding Signals for individual users; Indicates the first The first access point sends the data back to the CPU. Detection signals for each user; Indicates that for the first The first user's The beamforming matrix at the access point; H is located at the superscript to indicate the conjugate transpose; the first... The access point and the first The channels between RIS follow a Rice distribution, represented as follows: ;No. The channel between the i-th RIS and the i-th user follows a Rice distribution. ;No. The channel between each access point and the i-th user follows a Ricean distribution, represented as follows: ; It is the set of complex numbers; Indicates the first A reflection coefficient matrix of an active RIS; This represents the transmission power corresponding to the i-th user; This represents the signal sent by the i-th user; This represents the transmission power corresponding to the k-th user; Indicates the first Signals sent by each user; Indicates the first Additive white Gaussian noise vector at each RIS; Indicates the first Additive white Gaussian noise vectors at each access point Indicates the user index. , Indicates except the first All other users of that user.

[0049] It should be noted that in this disclosure, i and k are both user indices. When used as subscripts for parameters, they are only used to distinguish between different users and do not affect the illustration of the parameters themselves.

[0050] S2. Based on the active RIS-assisted cellless MIMO system model, jointly optimize the receiving beamforming, RIS reflection coefficient and power allocation to construct and solve the rate maximization problem model.

[0051] In some embodiments, step S2 includes: Based on user transmit power constraints, RIS reflection power constraints, RIS reflection amplitude constraints, and receive beamforming normalization constraints, a model is constructed to solve the rate maximization problem with the goal of maximizing the total rate.

[0052] Specifically, the solution model for the sum-rate maximization problem is as follows:

[0053] in, Representation and rate objective function; Indicates the first The received signal-to-noise ratio for each user; st represents the constraint condition; Represents a set of users; Represents an active RIS set; Represents the set of access points; Indicates the first A set of reflective elements for a RIS; Indicates user transmit power constraints; Indicates the first The maximum transmit power for each user; F in subscript indicates the Frobenius norm; Indicates RIS reflection power constraint; Indicates the first The variance of additive white Gaussian noise at each RIS point; This indicates the maximum reflected power of the active RIS; Indicates RIS reflection amplitude constraint; This indicates the maximum reflection amplitude of the RIS active element; This indicates the receive beamforming normalization constraint.

[0054] S3. The solution model for the sum rate maximization problem is decomposed into a receiver beamforming sub-problem, an active RIS reflection coefficient design sub-problem, and a user power allocation sub-problem.

[0055] S4. Based on the sub-problems obtained from the decomposition, the beamforming matrix, RIS reflection coefficient matrix, and user power are iteratively optimized using an alternating optimization method.

[0056] In some embodiments, step S4 includes steps S41-S43.

[0057] S41. Design of receiver beamforming based on the closed-form solution of generalized Rayleigh entropy. Includes: fixed Solve The optimization of the receive beamforming problem is as follows:

[0058] The optimal solution for the generalized Rayleigh entropy is:

[0059]

[0060]

[0061] in, The objective function for the receiver beamforming problem is expressed as follows: These are beamforming auxiliary variables obtained after mathematical transformation; and A constant related to prior channel state information; express The identity matrix; Indicates that the index is The constants related to prior channel state information, where ; This represents the variance of Gaussian white noise at the RIS. This represents the variance of the Gaussian white noise at the access point.

[0062] S42. Solving for the reflection coefficient of an active RIS using fractional programming. This includes: fixed Solve The subproblem of optimizing the reflection coefficient design of active RIS is:

[0063] To facilitate the decoupling of variables, auxiliary variables are introduced. , ,and , , , and These are all constants related to prior channel state information, and are defined as follows:

[0064]

[0065]

[0066]

[0067]

[0068]

[0069] in, Describe a cascaded channel and define the cascaded channel. ; express An identity matrix of order 1; In the original objective function It can be transformed into:

[0070] Then, auxiliary variables, i.e., the variables to be optimized, are introduced. Using fractional programming, the design subproblem of the reflection coefficient of an active RIS is transformed into solving the following convex optimization problem:

[0071] Use a convex optimization problem solver to solve for the design scheme of active RIS reflection coefficients in the current iteration.

[0072] S43. Solving user power allocation schemes based on the continuous convex approximation method. This includes: fixed Solve The subproblem of optimizing user power allocation is as follows:

[0073] in, and To represent constants related to prior channel state information, auxiliary variables are introduced. and Transformed through the following variables:

[0074] The user power allocation subproblem is transformed into solving the following convex optimization problem:

[0075] Use a convex optimization problem solver to find the optimal power allocation scheme for the user in the current iteration.

[0076] In one specific embodiment, the effectiveness of the proposed maximization and rate optimization method is verified through simulation on the Matlab platform. The simulation parameters are set as follows: number of APs (Access Points). Number of active RIS Number of users Number of AP antennas Number of RIS components Thermal noise power RIS power gain Maximum reflected power Dimensions of the residential area .

[0077] Figure 3 The sum-rate performance of distributed MIMO systems under different power allocation schemes was compared. Considering that the power allocation subproblem is a multi-constraint optimization problem, this embodiment uses genetic algorithms and fractional programming (FP) algorithms as benchmark algorithms for comparison. Figure 3It can be observed that under the power allocation scheme based on continuous convex approximation proposed in this embodiment, the system's sum rate approximates the performance of the power allocation scheme based on genetic algorithm, and is superior to the power allocation scheme based on fractional programming algorithm. This demonstrates the effectiveness of the algorithm proposed in this embodiment.

[0078] Figure 4 The sum and rate performance of distributed MIMO systems under different beamforming schemes were compared. Figure 4 It can be observed that the system and rate under the beamforming scheme based on zero forcing (ZF) are poor. This is because the scheme only eliminates signal interference between users, while the scheme based on generalized Rayleigh entropy proposed in this embodiment further considers the interference caused by thermal noise at the base station and RIS, so as to achieve the optimal receiving beamforming design.

[0079] Other embodiments of the present invention also provide a computer device including a processor and a memory, wherein the memory stores a computer program that, when loaded and executed by the processor, implements the optimization method for maximizing sum and rate in an active RIS-assisted cellless MIMO system as described in any of the above embodiments.

[0080] Some embodiments of the present invention provide a computer-readable storage medium storing a program that, when loaded by a processor, implements the optimization method for maximizing sum and rate in an active RIS-assisted cellless MIMO system as described in any of the above embodiments.

[0081] In this specification, the illustrative expressions of the terms used do not necessarily refer to the same embodiments or examples. Moreover, the specific features, structures, materials, or characteristics described may be combined in any suitable manner in one or more embodiments or examples.

[0082] Any modifications, equivalent substitutions, improvements, etc., made within the spirit and principles of this invention shall be included within the scope of protection of this invention.

Claims

1. A method for maximizing the sum and rate in an active RIS-assisted cellless MIMO system, characterized in that, include: Establish a model of an active RIS-assisted cellless MIMO system; Based on the aforementioned active RIS-assisted cellless MIMO system model, a model for solving the rate maximization problem is constructed by jointly optimizing receive beamforming, RIS reflection coefficient, and power allocation. The solution model for the sum rate maximization problem is decomposed into a receiver beamforming sub-problem, an active RIS reflection coefficient design sub-problem, and a user power allocation sub-problem. Based on the subproblems obtained from the decomposition, an alternating optimization method is used to iteratively optimize the beamforming matrix, the RIS reflection coefficient matrix, and the user power.

2. The optimization method for maximizing sum and rate in an active RIS-assisted cellless MIMO system according to claim 1, characterized in that, In the active RIS-assisted cellless MIMO system model, R active RIS-assisted access points equipped with M reflective elements and B access points equipped with N antennas provide services to K single-antenna users. The channels between the access point, RIS, and single-antenna users follow a Rice distribution. Each access point performs beamforming processing on the received uplink signal and then transmits it back to the CPU via the forward backhaul link for merging and decoding.

3. The optimization method for maximizing sum and rate in an active RIS-assisted cellless MIMO system according to claim 2, characterized in that, After decoding The signal for each user is represented as: in, Indicates the first [number] after decoding Signals for individual users; Indicates the first The first access point sends the data back to the CPU. Detection signals for each user; Indicates that for the first The first user's The beamforming matrix at the access point; H is located at the superscript to indicate the conjugate transpose; the first... The access point and the first The channels between RIS follow a Rice distribution, represented as follows: ;No. The channel between the i-th RIS and the i-th user follows a Rice distribution. ;No. The channel between each access point and the i-th user follows a Ricean distribution, represented as follows: ; It is the set of complex numbers; Indicates the first A reflection coefficient matrix of an active RIS; This represents the transmission power corresponding to the i-th user; This represents the signal sent by the i-th user; Indicates the first Additive white Gaussian noise vector at each RIS; Indicates the first Additive white Gaussian noise vectors at each access point; Indicates the user index. , Indicates except the first All other users of that user.

4. The optimization method for maximizing sum and rate in an active RIS-assisted cellless MIMO system according to claim 3, characterized in that, The joint optimization model for receiving beamforming, RIS reflection coefficient, and power allocation, along with the solution model for the rate maximization problem, includes: Based on user transmit power constraints, RIS reflection power constraints, RIS reflection amplitude constraints, and receive beamforming normalization constraints, a model for solving the rate maximization problem is constructed with the goal of maximizing the total rate. The solution model for the sum-rate maximization problem is as follows: in, Representation and rate objective function; Indicates the first The received signal-to-noise ratio for each user; st represents the constraint condition; Represents a set of users; Represents the set of active RIS; Represents the set of access points; Indicates the first A set of reflective elements for a RIS; Indicates user transmit power constraints; Indicates the first The maximum transmit power for each user; F in subscript indicates the Frobenius norm; Indicates RIS reflection power constraint; Indicates the first The variance of additive white Gaussian noise at each RIS point; This indicates the maximum reflected power of the active RIS; Indicates RIS reflection amplitude constraint; This indicates the maximum reflection amplitude of the RIS active element; This indicates the receive beamforming normalization constraint.

5. The optimization method for maximizing sum and rate in an active RIS-assisted cellless MIMO system according to claim 4, characterized in that, The iterative optimization of the beamforming matrix, RIS reflection coefficient matrix, and user power using an alternating optimization method includes: Design of receiver beamforming based on the closed-loop solution of generalized Rayleigh entropy; Solving the reflection coefficient of active RIS using fractional programming; Solving user power allocation schemes based on the continuous convex approximation method.

6. The optimization method for maximizing sum and rate in an active RIS-assisted cellless MIMO system according to claim 5, characterized in that, The design of receiver beamforming based on the closed-form solution of generalized Rayleigh entropy includes: fixed Solve The optimization of the receive beamforming problem is as follows: The optimal solution for the generalized Rayleigh entropy is: in, The objective function for the receiver beamforming problem is expressed as follows: These are beamforming auxiliary variables obtained after mathematical transformation; and All of them are constants with index k that are related to prior channel state information; express The identity matrix; Indicates that the index is The constants related to prior channel state information, where ; This represents the variance of Gaussian white noise at the RIS. This represents the variance of the Gaussian white noise at the access point.

7. The optimization method for maximizing sum and rate in an active RIS-assisted cellless MIMO system according to claim 6, characterized in that, The method for solving the active RIS reflection coefficient based on fractional programming includes: fixed Solve The subproblem of optimizing the reflection coefficient design of active RIS is: Introducing auxiliary variables , Using fractional programming, the design subproblem of the reflection coefficient of an active RIS is transformed into solving the following convex optimization problem: in, The variable to be optimized; , , , and All of them are constants related to prior channel state information; in, Describe a cascaded channel and define the cascaded channel. ; express An identity matrix of order 1; Use a convex optimization problem solver to solve for the design scheme of active RIS reflection coefficients in the current iteration.

8. The optimization method for maximizing sum and rate in an active RIS-assisted cellless MIMO system according to claim 7, characterized in that, The method for solving user power allocation schemes based on continuous convex approximation includes: fixed Solve The subproblem of optimizing user power allocation is as follows: in, and Represents constants related to prior channel state information. Indicates except the first For each user, introduce auxiliary variables for all other users. and The user power allocation subproblem is transformed into solving the following convex optimization problem: in: Use a convex optimization problem solver to find the optimal power allocation scheme for the user in the current iteration.

9. A computer device, characterized in that, It includes a processor and a memory, wherein the memory stores a computer program that, when loaded and executed by the processor, implements an optimization method for maximizing speed in an active RIS-assisted cellless MIMO system as described in any one of claims 1 to 8.

10. A computer-readable storage medium, characterized in that, The system stores a program that, when loaded by a processor, implements an optimization method for maximizing sum and rate in an active RIS-assisted cellless MIMO system as described in any one of claims 1 to 8.

Images