A method and apparatus for weighted sum rate optimization of a multi-user coexisting radio network

Abstract

The present application relates to the field of wireless communication, provide a kind of weighted sum and rate optimization method, device of multi-user coexistence radio network, the method comprises: constructing multi-user coexistence radio network model;Determine to maximize system weighted sum and rate as target to construct optimization problem according to multi-user coexistence radio network model, objective function of optimization problem is determined based on time slot access matrix, user association matrix and active beamforming, and satisfy the minimum transmission rate constraint of primary user and secondary user, base station transmit power constraint;Iterative solution is carried out to objective function until convergence, obtain the optimal time slot allocation result, user association result and active beamforming optimal solution.The present application solves the problem that multi-user coexistence radio network in prior art is difficult to maximize system weighted sum and rate under the constraint of primary and secondary user quality of service, realizes the technical effect of significantly improving system spectral efficiency and user fairness.

Description

Technical Field

[0001] This invention relates to the field of wireless communication technology, and in particular to a weighted sum rate optimization method and apparatus for multi-user coexisting radio networks. Background Technology

[0002] With the rapid development of 5G / 6G, the Internet of Things (IoT), and smart terminals, wireless communication networks face challenges such as scarce spectrum resources, low energy efficiency, and increased interference from multiple users. Traditional wireless communication technologies mainly rely on active beamforming and static resource allocation by base stations (BS), which is insufficient to meet the demands of massive device access and high throughput. In recent years, the rise of Smart Reflector (RIS) and Symbiotic Radio (SR) technologies has provided new solutions to these problems. RIS enhances signal coverage and suppresses interference by dynamically controlling the electromagnetic wave propagation environment through programmable metasurfaces; while SR technology allows primary users (PU) and secondary users (SU) to share spectrum and energy, improving resource utilization. However, how to deeply integrate RIS and SR to achieve efficient joint optimization (such as time-slot access, user association, and beamforming) in multi-user scenarios remains a research challenge in the field of wireless communication.

[0003] In existing technologies, most studies focus only on single technologies and lack a systematic design for joint optimization of time slots, users, and beamforming, resulting in low resource allocation efficiency. Due to the unit mode constraint of the RIS phase shift matrix, strong coupling interference between users, and the presence of mixed integer variables, the optimization problem exhibits high non-convexity, making it difficult to solve directly using traditional convex optimization methods. Summary of the Invention

[0004] This invention provides a weighted sum rate optimization method and apparatus for multi-user coexisting radio networks, which solves the problem in the prior art that multi-user coexisting radio networks are difficult to maximize the system weighted sum rate while satisfying the quality of service constraints of primary and secondary users, and achieves the technical effect of significantly improving system spectrum efficiency and user fairness.

[0005] This invention provides a weighted sum rate optimization method for multi-user coexisting radio networks, comprising the following steps: A multi-user coexisting radio network model is constructed, which includes a base station, a smart reflector node, a primary user, and secondary users; Based on the multi-user co-existing radio network model, an optimization problem is constructed with the goal of maximizing the system weighted sum rate. The objective function of the optimization problem is determined based on the time slot access matrix, the user correlation matrix, and active and passive beamforming, and satisfies the minimum transmission rate constraints for primary and secondary users and the base station transmit power constraints. The objective function is iteratively solved until convergence, yielding the optimal time slot allocation result, user association result, and optimal solution for active and passive beamforming.

[0006] According to the weighted sum rate optimization method for a multi-user coexisting radio network provided by the present invention, the step of iteratively solving the objective function until convergence to obtain the optimal time slot allocation result, user association result, and optimal solution for active and passive beamforming specifically includes: optimizing the time slot access matrix based on Stackelberg game to determine the optimal time slot allocation strategy; optimizing the user association matrix based on a many-to-one matching algorithm to determine the user association result; and, based on the optimal time slot allocation strategy and the user association result, jointly optimizing active and passive beamforming using a semidefinite relaxation and continuous convex approximation method to obtain the optimal solution for active and passive beamforming.

[0007] According to the weighted sum rate optimization method for a multi-user coexisting radio network provided by the present invention, the optimization of the time slot access matrix based on the Stackelberg game specifically includes: modeling the time slot access problem as a Stackelberg game problem, wherein the base station, as the leader, dynamically adjusts the time slot access price through a pricing mechanism, and the secondary users, as followers, select the optimal time slot; and using a particle swarm optimization algorithm to solve the Stackelberg game problem to find the Stackelberg equilibrium and determine the optimal time slot allocation strategy.

[0008] According to the weighted sum rate optimization method for a multi-user coexisting radio network provided by the present invention, the optimization of the user association matrix based on a many-to-one matching algorithm specifically includes: modeling the user association problem as a many-to-one matching problem; constructing preference lists for secondary users and intelligent reflector nodes respectively based on the many-to-one matching problem, and establishing a preliminary matching relationship between secondary users and intelligent reflector nodes; performing exchange matching based on the preliminary matching relationship, and determining the user association result according to the matching result.

[0009] According to the weighted sum rate optimization method for a multi-user coexisting radio network provided by the present invention, the method employs a semi-definite relaxation and continuous convex approximation method to jointly optimize active and passive beamforming to obtain the optimal solution for active and passive beamforming. Specifically, the method includes: transforming the non-convex beamforming optimization problem into a convex optimization problem using a semi-definite relaxation method; solving the convex optimization problem using a continuous convex approximation and Gaussian randomization method to obtain a local optimum solution for active beamforming; and optimizing the phase shift of each smart reflector based on the local optimum solution for active beamforming to obtain the optimal solution for passive beamforming.

[0010] According to the present invention, a weighted sum rate optimization method for a multi-user co-occurring radio network is provided. The method involves transforming a non-convex beamforming optimization problem into a convex optimization problem using a semi-definite relaxation method; solving the convex optimization problem using continuous convex approximation and Gaussian randomization to obtain a local optimum for active beamforming. Specifically, the method includes: transforming the active beamforming vector into a positive semi-definite matrix; transforming the non-convex beamforming optimization problem using the positive semi-definite matrix to obtain an intermediate objective function; obtaining defined relaxation variables; relaxing the intermediate objective function using the relaxation variables and determining the constraints satisfied by the relaxation variables; processing the constraints using a continuous convex approximation method to obtain a convex optimization problem; solving the convex optimization problem to obtain the optimum solution of the positive semi-definite matrix; and obtaining a local optimum for active beamforming using Gaussian randomization based on the optimum solution of the positive semi-definite matrix.

[0011] According to the weighted sum rate optimization method for a multi-user coexisting radio network provided by the present invention, the optimization of the phase shift of each smart reflector to obtain the optimal solution for passive beamforming specifically includes: obtaining the defined phase shift matrix and auxiliary variables of the smart reflector; determining the transmission rates of the primary user and the secondary user as functions of the auxiliary variables to obtain a rate expression; introducing relaxation variables into the fractional terms in the rate expression, and determining the non-convex constraint as a linear inequality based on the relaxation variables; solving the linear inequality to obtain the optimal solution for the auxiliary variables; and obtaining the optimal solution for passive beamforming based on the optimal solution for the auxiliary variables using Gaussian randomization.

[0012] The present invention also provides a weighted sum rate optimization apparatus for a multi-user coexisting radio network, comprising the following modules: The model building module is used to build a multi-user coexisting radio network model, which includes a base station, a smart reflector node, a primary user, and secondary users. The optimization problem construction module is used to determine the optimization problem based on the multi-user co-existing radio network model, with the goal of maximizing the system weighted sum rate. The objective function of the optimization problem is determined based on the time slot access matrix, user association matrix, and active and passive beamforming, and satisfies the minimum transmission rate constraints for primary and secondary users and the base station transmit power constraints. The solution module is used to iteratively solve the objective function until convergence, and obtain the optimal time slot allocation result, user association result, and optimal solution for active and passive beamforming.

[0013] According to the present invention, a weighted sum rate optimization device for a multi-user coexisting radio network includes a solution module comprising: an optimal time slot allocation submodule for optimizing the time slot access matrix based on a Stackelberg game to determine the optimal time slot allocation strategy; an association submodule for optimizing the user association matrix based on a many-to-one matching algorithm to determine the user association result; and an optimal solution solving submodule for jointly optimizing active and passive beamforming using a semidefinite relaxation and continuous convex approximation method based on the optimal time slot allocation strategy and the user association result to obtain the optimal solution for active and passive beamforming.

[0014] According to the present invention, a weighted sum rate optimization device for a multi-user coexisting radio network includes an optimal time slot allocation submodule comprising: an optimal time slot selection unit, used to model the time slot access problem as a Stackelberg game problem, wherein the base station, as the leader, dynamically adjusts the time slot access price through a pricing mechanism, and the secondary user, as the follower, selects the optimal time slot; and an optimal time slot allocation unit, used to solve the Stackelberg equilibrium in the Stackelberg game problem using a particle swarm optimization algorithm to determine the optimal time slot allocation strategy.

[0015] According to the present invention, a weighted sum rate optimization device for a multi-user coexisting radio network includes an association submodule comprising: a problem modeling unit for modeling the user association problem as a many-to-one matching problem; a preliminary matching unit for constructing preference lists for secondary users and intelligent reflector nodes based on the many-to-one matching problem, and establishing a preliminary matching relationship between secondary users and intelligent reflector nodes; and a result association unit for performing exchange matching based on the preliminary matching relationship, and determining the user association result based on the matching result.

[0016] According to the present invention, a weighted sum rate optimization device for a multi-user coexisting radio network includes an optimal solution solving submodule comprising: a problem transformation solving unit, used to transform a non-convex beamforming optimization problem into a convex optimization problem using a semi-definite relaxation method; solving the convex optimization problem using a continuous convex approximation and Gaussian randomization method to obtain a local optimal solution for active beamforming; and an optimal solution solving unit, used to optimize the phase shift of each smart reflector based on the local optimal solution of active beamforming to obtain an optimal solution for passive beamforming.

[0017] According to the present invention, a weighted sum rate optimization device for a multi-user coexisting radio network includes a problem transformation and solution unit comprising: a vector transformation subunit for transforming an active beamforming vector into a positive semi-definite matrix; a problem transformation subunit for transforming a non-convex beamforming optimization problem based on the positive semi-definite matrix to obtain an intermediate objective function; a function relaxation subunit for obtaining defined relaxation variables, relaxing the intermediate objective function based on the relaxation variables, and determining the constraint conditions satisfied by the relaxation variables; a constraint condition processing subunit for processing the constraint conditions using a continuous convex approximation method to obtain a convex optimization problem; a problem solving subunit for solving the convex optimization problem to obtain the optimal solution of the positive semi-definite matrix; and a local optimal solution solving subunit for obtaining a local optimal solution of active beamforming based on the optimal solution of the positive semi-definite matrix using a Gaussian randomization method.

[0018] According to the present invention, a weighted sum rate optimization device for a multi-user coexisting radio network includes an optimal solution solving unit comprising: a variable acquisition subunit for acquiring the phase shift matrix and auxiliary variables of a defined smart reflector; an expression acquisition subunit for determining the transmission rates of the primary user and the secondary user as functions of the auxiliary variables to obtain a rate expression; an inequality acquisition subunit for introducing relaxation variables into the fractional terms in the rate expression and determining the non-convex constraints as linear inequalities based on the relaxation variables; an inequality solving subunit for solving the linear inequalities to obtain the optimal solution for the auxiliary variables; and an optimal solution solving subunit for obtaining the optimal solution for passive beamforming based on the optimal solution for the auxiliary variables using Gaussian randomization.

[0019] The present invention also provides an electronic device, including a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein the processor executes the computer program to implement a weighted sum rate optimization method for a multi-user coexisting radio network as described above.

[0020] The present invention also provides a non-transitory computer-readable storage medium having a computer program stored thereon, which, when executed by a processor, implements a weighted sum rate optimization method for a multi-user coexisting radio network as described above.

[0021] The present invention also provides a computer program product, including a computer program that, when executed by a processor, implements a weighted sum rate optimization method for a multi-user coexisting radio network as described above.

[0022] This invention provides a weighted sum rate optimization method and apparatus for multi-user co-existing radio networks, which offers the following advantages: By jointly optimizing the time slot access matrix, user association matrix, and active / passive beamforming, and under the premise of satisfying the minimum transmission rate constraints for primary and secondary users and the base station transmit power constraints, iterative solutions are performed with the goal of maximizing the system's weighted sum rate, significantly improving the system's spectral efficiency; enhancing signal transmission quality through the collaborative optimization of intelligent reflectors; ensuring fairness among users, optimizing access opportunities for secondary users while guaranteeing the communication quality of primary users; achieving efficient resource allocation, achieving overall system performance optimization under complex constraints through joint optimization of time slot allocation, user association, and beamforming schemes; and improving network adaptability, enabling dynamic adjustment of resource allocation schemes according to channel conditions and user needs. These effects collectively solve key problems in traditional co-existing radio networks, such as uneven resource allocation, low spectrum utilization, and difficulty in guaranteeing user service quality. Attached Figure Description

[0023] To more clearly illustrate the technical solutions in this invention or the prior art, the drawings used in the description of the embodiments or the prior art will be briefly introduced below. Obviously, the drawings described below are some embodiments of this invention. For those skilled in the art, other drawings can be obtained from these drawings without creative effort.

[0024] Figure 1 This is a flowchart illustrating the weighted sum rate optimization method for multi-user coexisting radio networks provided by the present invention; Figure 2 This is a diagram of a multi-user co-existing radio network model with intelligent reflector-assisted operation provided by the present invention; Figure 3 This is a schematic diagram of the weighted sum rate optimization device for a multi-user coexisting radio network provided by the present invention; Figure 4 This is a schematic diagram of the structure of the electronic device provided by the present invention. Detailed Implementation

[0025] To make the objectives, technical solutions, and advantages of this invention clearer, the technical solutions of this invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some, not all, of the embodiments of this invention. All other embodiments obtained by those skilled in the art based on the embodiments of this invention without creative effort are within the scope of protection of this invention.

[0026] The following is an explanation of the English abbreviations and terms involved in this invention.

[0027] RIS (Reconfigurable Intelligent Surface) is a planar structure composed of programmable metamaterials that can dynamically control the reflection characteristics of electromagnetic waves to enhance the signal coverage and quality of wireless communication.

[0028] Symbiotic Radio (SR) is a wireless communication technology that combines the advantages of cognitive radio (CR) and ambient backscatter communication (AmBC) to enable spectrum and energy sharing between primary users (PU) and secondary users (SU).

[0029] In a coexisting radio network, a PU (Primary User) is a user who has priority access to the spectrum and is typically a high-priority communication device.

[0030] SU (Secondary User) is a low-priority user in a symbiotic radio network that communicates by sharing the spectrum and energy of PU.

[0031] BS (Base Station) is a core device in a wireless communication network, responsible for data transmission and control with user terminals.

[0032] TDMA (Time Division Multiple Access) is a multi-user access technology that divides time into multiple time slots, allowing different users to transmit data in different time slots.

[0033] CSI (Channel State Information) is a parameter that describes the characteristics of a wireless channel, including channel gain and phase, and is used to optimize signal transmission.

[0034] LoS (Line of Sight) is the path of a signal propagation directly from the transmitter to the receiver in wireless communication, and it typically has a high signal strength.

[0035] NLoS (Non-Line of Sight) is a phenomenon in wireless communication where signals propagate through indirect paths such as reflection and diffraction, and the signal strength is usually weak.

[0036] CSCG (Circularly Symmetric Complex Gaussian) is a probability distribution of complex random variables, often used to model noise and signals in wireless communication.

[0037] SINR (Signal-to-Interference-plus-Noise Ratio) is the ratio of signal power to interference plus noise power, used to measure the quality of a communication link.

[0038] SIC (Successive Interference Cancellation) is a signal processing technique that improves the quality of received signals by progressively decoding and eliminating interfering signals.

[0039] MINLP (Mixed-Integer Nonlinear Programming) is an optimization problem that involves integer variables and nonlinear objectives or constraints, and it typically has high solution complexity.

[0040] SE (Stackelberg Equilibrium) is an equilibrium state in game theory in which leaders (such as BS) and followers (such as SU) reach the optimal solution through strategic interactions.

[0041] PSO (Particle Swarm Optimization) is an optimization algorithm based on swarm intelligence that solves optimization problems by simulating the search behavior of particles in the solution space.

[0042] SDP (Semidefinite Programming) is a convex optimization problem in which the objective function and constraints involve positive semidefinite matrices. It is often used to solve beamforming problems in wireless communication.

[0043] CVX (Convex Optimization Toolbox) is a software tool for modeling and solving convex optimization problems, supporting a variety of optimization algorithms.

[0044] Alternating Optimization (AO) is an iterative optimization method that solves complex problems by alternately fixing some variables and optimizing others.

[0045] SDR (Semidefinite Relaxation) is a technique that transforms a non-convex optimization problem into a convex problem, and is often used to handle rank-constant optimization problems.

[0046] AmBC (Ambient Backscatter Communication) is a technology that uses ambient radio frequency signals (such as Wi-Fi and television signals) for low-power communication.

[0047] Cognitive Radio (CR) is an intelligent wireless communication technology that can sense and dynamically utilize idle spectrum resources.

[0048] RIS-BD (RIS with Backscatter Device) is a device that combines a smart reflector with backscatter communication technology to enhance signal transmission and energy harvesting.

[0049] NP-hard (Non-deterministic Polynomial-time hard) refers to problems with extremely high computational complexity, which are usually difficult to solve in polynomial time.

[0050] MIMO (Multiple-Input Multiple-Output) is a method that uses multiple antennas to improve the capacity and reliability of wireless communication.

[0051] RF (Radio Frequency) refers to the range of electromagnetic wave frequencies applicable to wireless communication.

[0052] With the explosive growth in the number of mobile devices and mobile data traffic, wireless networks face enormous challenges in terms of spectrum resources, power consumption, and capacity, and spectrum efficiency and energy efficiency urgently need optimization. Traditional mobile communication technologies focus on human-centered services, but have limitations in dealing with the surge in device connections and massive access brought about by the Internet of Things. The emerging Symbiotic Radio (SR) technology, by utilizing the advantages of Cognitive Radio (CR) and Ambient Backscatter Communication (AmBC), can promote coordination among various parts of the wireless communication system, enabling different users to cooperate and share spectrum resources, alleviate interference problems, and complete primary user (PU) and secondary user (SU) transmission in a resource-sharing manner. As a popular technology, Reconfigurable Intelligent Surface (RIS) can suppress interference, improve transmission, and enhance coverage through flexible and efficient passive modulation, thus compensating for some of the shortcomings of SR. However, research on the integration of the two is still in its early stages, and its potential in improving the overall system performance remains to be explored. To address the challenge that traditional technologies struggle to effectively guarantee the spectrum efficiency and user transmission needs of wireless networks amidst rapidly increasing device and traffic demands, this study organically combines symbiotic radio with RIS (Radio Relational Interconnection). The research focuses on methods for integrating these two technologies in multi-user, multi-access scenarios to provide high-quality transmission for a massive number of users with limited resources. While ensuring the transmission needs of users at all levels, the study also jointly optimizes time-slot access, user association, and active / passive beamforming to improve the overall transmission performance of the system.

[0053] This invention proposes a weighted sum rate optimization method for intelligent reflector-assisted multi-user co-existing radio networks. An intelligent reflector-assisted multi-user co-existing radio model is constructed, and system performance is improved by jointly optimizing time slot access, user association, and active / passive beamforming. Based on this model, considering transmit power limitations and minimum user transmission rate requirements, the optimization problem is constructed with the goal of maximizing the system's weighted sum rate, using the secondary user time slot access matrix, secondary user transmission association matrix, and active / passive beamforming as optimization variables. For this problem, an alternating optimization algorithm is proposed. The problem is decomposed into multiple sub-problems, and the relevant variables are solved repeatedly until convergence. This achieves a significant improvement in the overall transmission rate of the co-existing radio network while ensuring the transmission quality of users at all levels.

[0054] The following is combined with Figures 1-4 The embodiments of the present invention are described in detail.

[0055] Figure 1This is one of the flowcharts illustrating the weighted sum rate optimization method for multi-user coexisting radio networks provided by the present invention, such as... Figure 1 As shown, the method includes the following steps: S110. Construct a multi-user coexisting radio network model.

[0056] 1. Multi-user coexisting radio network model Specifically, the multi-user coexisting radio network model includes base stations, smart reflector nodes, primary users, and secondary users. For example... Figure 2 As shown, this subsection considers a smart reflector-assisted multi-user co-existing radio network model, which consists of a... Base station with root antenna One RIS-BD node, A single antenna PU and It consists of a single-antenna SU. In this cellular network, the BS is responsible for transmission to the PU, and the RIS-BD acts as a transmission relay, capable of harvesting energy from the environment, modulating information through backscatter communication, and providing transmission to the associated SU. Meanwhile, the RIS introduced in the BD contains... Each reflector unit can enhance the backscatter link through passive beamforming.

[0057] BS uses TDMA technology for Each PU provides the transmission. Assume the duration is... The frames are divided into There are three equal-length time slots, each allocated to a PU for downlink transmission. A SU can reside within a time slot, sharing the PU's spectrum and energy with its associated RIS-BD to achieve transmission. Let variables... User parasitism index This indicates that user SU Parasite on user PU Transmission occurs in the specified time slots, otherwise... Assume that a RIS-BD can only serve one SU, while a SU can be associated with multiple RIS-BDs. Let variables... For SU With RIS-BD The correlation index between them, when When, it indicates that the two are related; otherwise, they are related. .

[0058] 1.1 Channel Model All channels in the model follow a stable block fading channel model, meaning that the channel coefficients for each time slot remain constant during signal transmission. Since the signal suffers severe path loss after multiple reflections, resulting in a negligible reduction in signal strength, this paper only considers signals reflected once by the RIS. For example... Figure 1 As shown, from BS to PU BS to SU BS to RIS-BD RIS-BD To PU RIS-BD To SU The channel coefficients are respectively used as , , , , This means that it is assumed that the CSI of all channels can be perfectly acquired.

[0059] In this section, the present invention uses To represent the large-scale fading component of the channel, its magnitude is distance-dependent and is defined as follows: .in, The straight-line distance between two points. This represents the path loss per unit distance. This is the path loss factor. This section uses the widely used Ricean fading channel to model the channel, comparing BS and RIS-BD. Taking the transmission channel between them as an example, its channel coefficient can be expressed as: in, and These represent the line-of-sight (LoS) component and the non-line-of-sight (NLoS) component, respectively. yes Large-scale fading components; It means The Rice factor is a positive number. The larger the Rice factor, the more significant the influence of the line-of-sight component relative to the non-line-of-sight component. Each element follows a complex Gaussian distribution Line-of-sight component It can be defined as , in , The spacing between antennas, It's the wavelength. It is the angle of arrival at the RIS end. It is the transmission angle of the BS end.

[0060] 1.2 Signal Model and Achievable Transmission Rate Considering the high transmission demands of users at all levels in real-world multi-user networks, this section adopts a "contention-based" transmission mode, where the same symbol transmission rate is set for PU and SU. This means that direct links and backscatter links interfere with each other. Definition For PU Transmission symbols, For SU The transmitted symbols all follow a standard circularly symmetric complex Gaussian distribution (CSCG). Then PU The received signal at that location can be represented as: SU The received signal at that location can be represented as: It is the beamforming vector at the transmitting end; Then, the reflection coefficient matrix of RIS-BDj can be expressed as: ,in . and The power is Additive white Gaussian noise.

[0061] because and The symbol transmission rates are the same, and The distribution is very complex, when decoding In this invention, the worst-case decoding scenario is considered, whereby all backscattered link signals within a time slot are treated as interference. Correspondingly, the decoding... The signal-to-interference-plus-noise ratio (SINR) is At this time, PU The transmission rate can be approximated as In SU End, complete the task After decoding, continuous interference cancellation (SIC) technology is required to complete the decoding. Decoding. Assuming the direct link signal is perfectly eliminated, decoding... The corresponding SINR is At this time, SU The achievable transmission rate is 1.3 Description of the Weighted Sum Rate Maximization Problem The goal of this paper is to maximize the system's weighted transmission rate by jointly optimizing active and passive beamforming, node correlation, and time slot allocation, while meeting the transmission needs of each user, using limited transmit power. The optimization problem can be expressed in the following form: (P1): st in, and It means and The weighting coefficients, and These represent the transmission rate thresholds for PU and SU, respectively. and This indicates the minimum transmission rate constraint for PU and SU; This indicates the phase shift constraint of RIS; Ensure that the BS's transmission power does not exceed its maximum transmission power. ; Represents a SU It can only transmit within a single time slot; This means that a RIS-BD can only be associated with one SU, and each SU At least one RIS-BD is required to provide transmission, SU The channel can accommodate a maximum of [number] simultaneous [occupants]. Transmission is performed using RIS-BD.

[0062] Clearly, this problem involves both binary and continuous variables, with strong coupling between them, making it an NP-hard mixed-integer nonlinear programming (MINLP) problem. To solve this difficult problem, in subsequent chapters, this invention first decomposes the problem (P1) to decouple the related variables. Then, using an alternating optimization approach, iteratively solves each subproblem to obtain the final optimized result.

[0063] S120. Based on the multi-user coexisting radio network model, determine the optimization problem with the goal of maximizing the system weighted sum rate.

[0064] S130. Iterate the objective function until convergence, and obtain the optimal time slot allocation result, user association result, and optimal solution for active and passive beamforming.

[0065] 2. Weighted sum rate maximization algorithm based on alternating optimization According to the weighted sum rate optimization method for a multi-user coexisting radio network provided by the present invention, the objective function is iteratively solved until convergence to obtain the optimal time slot allocation result, user association result, and optimal solution for active and passive beamforming. Specifically, the method includes: optimizing the time slot access matrix based on Stackelberg game to determine the optimal time slot allocation strategy; optimizing the user association matrix based on a many-to-one matching algorithm to determine the user association result; and, based on the optimal time slot allocation strategy and user association result, jointly optimizing active and passive beamforming using semidefinite relaxation and continuous convex approximation methods to obtain the optimal solution for active and passive beamforming.

[0066] Specifically, the objective function of the optimization problem is determined based on the time slot access matrix, the user association matrix, and active / passive beamforming, and satisfies the minimum transmission rate constraints for primary and secondary users, as well as the base station transmit power constraints. This section first decomposes the original problem into three corresponding sub-problems: the time slot access problem, the node association problem, and the active / passive beamforming optimization problem. The problems are solved iteratively using an alternating optimization algorithm. Finally, the complete algorithm processing procedure is presented.

[0067] 2.1 Time-slot access scheme based on Stackelberg game framework According to the present invention, a weighted sum rate optimization method for a multi-user coexisting radio network is provided, which optimizes the time slot access matrix based on a Stackelberg game. Specifically, it includes: modeling the time slot access problem as a Stackelberg game problem, in which the base station, as the leader, dynamically adjusts the time slot access price through a pricing mechanism, and the secondary users, as followers, select the optimal time slot; and using a particle swarm optimization algorithm to solve the Stackelberg game problem to find the Stackelberg equilibrium and determine the optimal time slot allocation strategy.

[0068] Specifically, the SU (Supply Provider) obviously prefers to access the time slot that maximizes its own benefits. However, in reality, transmission resources within each time slot are limited, and the SU's access will interfere with the PU (Power Provider). Therefore, the BS (Browser / Server) needs to fully consider system load and overall performance when deciding on the SU's access time slot. This means that both parties need to constantly "bargain" on this issue to achieve a "win-win" situation.

[0069] Based on the above characteristics, in this section, the present invention models the time slot access problem as a Stackelberg game problem involving multiple followers. In this game, the BS acts as the leader, employing a pricing mechanism to balance the system load within each time slot and maximize the overall system transmission performance; each SU acts as a follower, making decisions in a non-cooperative competitive manner to ensure the maximization of its own benefits.

[0070] It needs to rationally select access time slots based on access prices to ensure maximum revenue. SU has two objectives: maximizing its own benefits and minimizing the cost of access time slots. Therefore, SU... The payoff function consists of two conflicting objectives: First item For SU The benefit function is about A monotonically increasing function is usually modeled as a logarithmic function or a sigmoid function. The second term... It is SU The cost of accessing the time slot. For time slots The access price.

[0071] From the follower's perspective, when the access price of a time slot is determined, each follower needs to solve the following optimization problem in order to choose their own association strategy.

[0072] (P2): st Specifically, each follower calculates its own revenue function when accessing different time slots and selects the time slot that maximizes its own revenue. Furthermore, it's easy to see that the optimal solution in problem (P2) is related to the pricing of each time slot. For the BS, it needs to set a uniform price for all SUs in each time slot, aiming to maximize the overall system revenue through cooperation. The BS's revenue can be defined as the sum of SU access fees minus resource costs. The payoff function in the equation can be expressed as: in, It is about The monotonically increasing function is used to measure the resource cost of BS transmission, time slots. The more SUs (Suits) that are internally connected, the higher the transmission resource cost.

[0073] The pricing problem of BS, as a leader problem, aims to maximize total revenue through cooperation and can be expressed in the following form: (P3): st Since pricing is often dynamic in the leader problem, this section uses the PSO algorithm to obtain the optimal pricing strategy to maximize the overall system revenue.

[0074] According to the principles of Stackelberg games, the followers' decisions depend on the leader's decisions, and the leader's optimal decisions also change as the equilibrium of the followers' decisions changes. In this problem, the Stackelberg equilibrium (SE) is defined as follows.

[0075] If a solution exists For any set If both of the following two equations are satisfied, then it is called... For this game, the SE (Stakeholder) is...

[0076] When SE appears, it means that the leader and followers have completed the "bargaining" and the algorithm has converged.

[0077] 2.2 User Association Strategy Based on Many-to-One Matching Mechanism According to the present invention, a weighted sum rate optimization method for a multi-user coexisting radio network optimizes the user association matrix based on a many-to-one matching algorithm. Specifically, the method includes: modeling the user association problem as a many-to-one matching problem; constructing preference lists for secondary users and intelligent reflector nodes based on the many-to-one matching problem, and establishing a preliminary matching relationship between secondary users and intelligent reflector nodes; performing exchange matching based on the preliminary matching relationship, and determining the user association result based on the matching result.

[0078] Specifically, due to constraints The existence of node association subproblems can be viewed as a many-to-one matching problem between SU ​​and RIS-BD.

[0079] This invention defines SU For RIS-BD Preference value , that is, SU With RIS-BD The achievable transmission rate during matching. For any two RIS-BDs , ,when At that time, it means SU The former is preferred for matching. This invention ranks each SU in descending order of preference value. Build a preference list This preference relationship means that each SU Prefers RIS-BD, which can provide greater reverse transfer capacity. Matching. In this design, SU This is undoubtedly "selfish." On the one hand, it aligns with the principle of ensuring SU transmission quality in parasitic transmission modes; on the other hand, it also conforms to SU... Limited computing power and difficulty in conveying the actual situation of cooperation strategies.

[0080] Similarly, this invention also applies to RIS-BD The side introduces a preference list, represented as Define RIS-BD For SU Preference value , indicating RIS-BD with SU During matching, the PU in its time slot and SU The weighted transmission rate sum. Same as above, They are also sorted in descending order of preference value. The reason for this design is that RIS-BD, as a node, can collect transmission information, thereby effectively balancing the transmission quality of the two.

[0081] This invention notes that when a set of RIS-BDs is associated with a SU, other SUs in the same transmission time slot are interfered with. This means that the preference value of a SU depends not only on its associated RIS-BD but also on the matching pairs of other SUs and RIS-BDs in the same time slot. This interdependent relationship is called an externality. The existence of externalities causes the preference order to change continuously, making it difficult to achieve a stable matching state. To address the impact of externalities, this invention proposes an optimization strategy based on exchange matching to achieve stable matching.

[0082] The matching scheme proposed in this invention is shown in Algorithm 1. This algorithm consists of two parts: initialization and exchange matching. In the initialization phase, this invention constructs a preference list for all RIS-BD and SU, and establishes preliminary matching relationships according to the following operations.

[0083] Step 1: For each RIS-BD that has not yet been matched, add it to the preference list. The preferred SU sends a pairing request. Upon receiving the pairing request, the SU then... Select the most preferred RIS-BD for matching. RIS-BDs and SUs that have already been matched are no longer eligible for matching. Repeat this process until each SU has a RIS-BD to match.

[0084] Step 2: For unmatched RIS-BD, re-install according to... Send a pairing request to the most preferred SU. For SU In other words, when the number of matched RIS-BDs is less than At that time, it will be in accordance with The system sequentially receives pairing requests from RIS-BDs. Otherwise, all pairing requests are rejected. This process is repeated until all RIS-BDs have been paired or there are no more matchable SUs in the remaining RIS-BDs' preference lists.

[0085] After initialization, the exchange and matching phase begins, which mainly includes the following two steps: Step 1: Check if each RIS-BD can form a swap-blocking pair with other RIS-BDs. If a swap-blocking pair exists, perform a swap operation to update the match; otherwise, maintain the current match.

[0086] Step 2: Repeat Step 1 until there are no more blocked pairs.

[0087] After the swap matching is completed, based on the matching results To adjust the correlation index This yields the optimized user association results.

[0088] The overall process of Algorithm 1 (user association algorithm based on exchange matching) is as follows: Phase 1: Initialization 1. Construct a preference list for each SU. ; 2. Construct a preference list for each RIS-BD. ; 3. Define the set of unmatched SUs as ; 4. Define the set of unmatched RIS-BDs as ; 5. Define SU The number of RIS-BDs already associated is ; 6. Copy Preference List ; 7. While : 8.For : 9. From Choose your favorite SU; 10. Move the selected SU from Remove from; 11.For : 12. Based on the preference list Select the most preferred RIS-BD match from all matching requests; 13. Transfer the matched RIS-BD from Remove from; 14. Remove the matched SU from Remove from; 15. Update the matching pairs; 16. While and : 17.For : 18. From Choose your favorite SU; 19. Move the selected SU from Remove from; 20.For : 21. If : 22. Based on the preference list Select your preferred match from all matching requests. Matching 1 RIS-BD; 23. Transfer the matched RIS-BD from Remove from; 24. Else: 25. Reject all pairing requests; Phase Two: Exchange Matching 26. Define indicator variables ; 27. While : 28. Settings ; 29.For or or : 30.For or or : 31. If Forming a blocking pair: 32. Perform a swap operation to generate a new matching combination. and ; 33. Settings ; Return the final matching result .

[0089] 2.3 Joint Optimization Algorithm for Active and Passive Beamforming According to the weighted sum rate optimization method for a multi-user coexisting radio network provided by the present invention, active and passive beamforming are jointly optimized using semidefinite relaxation and continuous convex approximation methods to obtain the optimal solutions for active and passive beamforming. Specifically, the method includes: transforming the non-convex beamforming optimization problem into a convex optimization problem using a semidefinite relaxation method; solving the convex optimization problem using continuous convex approximation and Gaussian randomization methods to obtain a local optimum solution for active beamforming; and optimizing the phase shift of each smart reflector based on the local optimum solution for active beamforming to obtain the optimal solution for passive beamforming.

[0090] Specifically, in this subsection, given the optimization results for time slot access and user association, the AO technique is used to decouple the joint optimization of beamforming and the time allocation subproblem. Specifically, the transmit beamforming of the BS and the passive beamforming of the RIS are optimized sequentially within each time slot, repeating the above steps until convergence.

[0091] First, this invention uses the Monte Carlo method to replace the sample mean with... The expected value of the mathematical expression: in, It is a symbol The total number of transmissions.

[0092] 1) Active beamforming optimization According to the weighted sum rate optimization method for a multi-user co-occurring radio network provided by the present invention, a semi-definite relaxation method is used to transform a non-convex beamforming optimization problem into a convex optimization problem. The convex optimization problem is then solved using continuous convex approximation and Gaussian randomization to obtain a local optimum for active beamforming. Specifically, this includes: transforming the active beamforming vector into a positive semi-definite matrix; transforming the non-convex beamforming optimization problem based on the positive semi-definite matrix to obtain an intermediate objective function; obtaining defined relaxation variables; relaxing the intermediate objective function based on the relaxation variables and determining the constraints satisfied by the relaxation variables; processing the constraints using a continuous convex approximation method to obtain a convex optimization problem; solving the convex optimization problem to obtain the optimum solution of the positive semi-definite matrix; and obtaining a local optimum for active beamforming based on the optimum solution of the positive semi-definite matrix using Gaussian randomization.

[0093] Specifically, at this stage, given In this case, optimize BS in time slots Transmit beamforming in .

[0094] Problem P1 can now be simplified to: st This invention uses SDR technology to solve this non-convex problem. Specifically, this invention defines a completely new variable. Clearly, this variable is a positive semidefinite variable with rank 1. According to... The present invention can transform the problem into the following form: (P4): st Tr( ) rank( )=1 This invention is designed , , , , At this point, Rs,m and Rc,n can be expressed in the following form It is obvious that and All are about Since the objective function is a fractional function, it remains non-convex. To address this issue, this invention introduces slack variables and employs a continuous convex approximation to relax the optimization problem.

[0095] First, define slack variables. , , , , , Slack variables satisfy the following condition: At this time, for and However, the problem remains non-convex. Therefore, to address this issue, this invention employs a continuous convex approximation method and Taylor series expansion to transform the two constraints into: in, , , , , , These are slack variables, and , , , This corresponds to the value of the last iteration of the relaxation variable. At this point, the present invention ignores the constraint rank( When )=1, the problem becomes a standard semidefinite programming (SDP) problem. In this case, the present invention uses CVX to solve the problem, obtaining... The optimal solution is obtained by Gaussian randomization. The local optimal solution.

[0096] 2) Passive beamforming optimization According to the weighted sum rate optimization method for a multi-user coexisting radio network provided by the present invention, the phase shift of each smart reflector is optimized to obtain the optimal solution for passive beamforming. Specifically, the method includes: obtaining the defined phase shift matrix and auxiliary variables of the smart reflector; determining the transmission rates of the primary user and the secondary user as functions of the auxiliary variables to obtain a rate expression; introducing relaxation variables into the fractional terms in the rate expression, and determining the non-convex constraint as a linear inequality based on the relaxation variables; solving the linear inequality to obtain the optimal solution for the auxiliary variables; and obtaining the optimal solution for passive beamforming based on the optimal solution for the auxiliary variables using Gaussian randomization.

[0097] Specifically, at this stage, given In this case, the phase shift of each RIS-BD is optimized.

[0098] First, for an already associated SU RIS-BD This invention defines , , .

[0099] At this time RIS-BD Time slot The transmission rate of the primary user can be expressed as: RIS-BD The associated SU The transmission rate at this point can be written as: This time slot The transmission rate of other SUs can be written as The optimization problem at this point can be expressed in the following form: (P5): st rank( ) = 1 It is not difficult to see at this point, , It is about Similar to the fractional function described above, this invention also introduces relaxation variables to relax the non-convex portion. Let... , At this point, the slack variable , , , satisfy: This invention also employs continuous convex approximation and Taylor series expansion to introduce relaxation variables to relax the constraints: At this point, problem (P5) is transformed into a standard SDP problem. The present invention then uses "CVX + Gaussian randomization" to obtain the optimal solution.

[0100] The present invention has the following beneficial effects: 1. This invention improves system performance by constructing a smart reflector-assisted multi-user co-existing radio model and jointly optimizing time slot access, user association, and active and passive beamforming.

[0101] 2. Based on the multi-user coexisting radio model, this invention considers the transmit power budget and the minimum transmission rate requirements of users at all levels. With the goal of maximizing the system weighted sum rate, it constructs a system weighted sum rate maximization optimization problem using the secondary user time slot access matrix, the secondary user transmission correlation matrix, and active and passive beamforming as optimization variables.

[0102] 3. This invention derives closed-form expressions for the signal-to-interference-plus-noise ratio (SIR) and transmission rate of the received signals of users at each level in a multi-user coexisting radio network under "contention" mode.

[0103] 4. This invention addresses the system weighted sum rate maximization problem by designing an iterative algorithm based on alternating optimization, which decomposes the problem into multiple subproblems and solves each optimization variable separately.

[0104] The weighted sum rate optimization apparatus for a multi-user coexisting radio network provided by the present invention will be described below. The weighted sum rate optimization apparatus for a multi-user coexisting radio network described below can be referred to in correspondence with the weighted sum rate optimization method for a multi-user coexisting radio network described above.

[0105] like Figure 3 The image shows a weighted sum rate optimization apparatus for a multi-user coexisting radio network provided by the present invention, comprising: The model building module 310 is used to build a multi-user coexisting radio network model, which includes base stations, smart reflector nodes, primary users, and secondary users. The optimization problem construction module 320 is used to determine the optimization problem based on the multi-user coexisting radio network model, with the goal of maximizing the system weighted sum rate. The objective function of the optimization problem is determined based on the time slot access matrix, user correlation matrix and active and passive beamforming, and satisfies the minimum transmission rate constraints of primary users and secondary users, and the base station transmit power constraints. The solver module 330 is used to iteratively solve the objective function until convergence, and obtain the optimal time slot allocation result, user association result and optimal solution for active and passive beamforming.

[0106] According to the present invention, a weighted sum rate optimization device for a multi-user coexisting radio network includes a solution module 330 comprising: an optimal time slot allocation submodule for optimizing the time slot access matrix based on a Stackelberg game to determine the optimal time slot allocation strategy; an association submodule for optimizing the user association matrix based on a many-to-one matching algorithm to determine the user association result; and an optimal solution solving submodule for jointly optimizing active and passive beamforming using a semidefinite relaxation and continuous convex approximation method based on the optimal time slot allocation strategy and the user association result to obtain the optimal solution for active and passive beamforming.

[0107] According to the present invention, a weighted sum rate optimization device for a multi-user coexisting radio network includes an optimal time slot allocation submodule comprising: an optimal time slot selection unit, used to model the time slot access problem as a Stackelberg game problem, wherein the base station, as the leader, dynamically adjusts the time slot access price through a pricing mechanism, and the secondary user, as the follower, selects the optimal time slot; and an optimal time slot allocation unit, used to solve the Stackelberg equilibrium in the Stackelberg game problem using a particle swarm optimization algorithm to determine the optimal time slot allocation strategy.

[0108] According to the present invention, a weighted sum rate optimization device for a multi-user coexisting radio network includes an association submodule comprising: a problem modeling unit for modeling the user association problem as a many-to-one matching problem; a preliminary matching unit for constructing preference lists for secondary users and intelligent reflector nodes based on the many-to-one matching problem, and establishing a preliminary matching relationship between secondary users and intelligent reflector nodes; and a result association unit for performing exchange matching based on the preliminary matching relationship, and determining the user association result based on the matching result.

[0109] According to the present invention, a weighted sum rate optimization device for a multi-user coexisting radio network includes an optimal solution solving submodule comprising: a problem transformation solving unit, used to transform a non-convex beamforming optimization problem into a convex optimization problem using a semi-definite relaxation method; solving the convex optimization problem using a continuous convex approximation and Gaussian randomization method to obtain a local optimal solution for active beamforming; and an optimal solution solving unit, used to optimize the phase shift of each smart reflector based on the local optimal solution of active beamforming to obtain an optimal solution for passive beamforming.

[0110] According to the present invention, a weighted sum rate optimization device for a multi-user coexisting radio network includes a problem transformation and solution unit comprising: a vector transformation subunit for transforming an active beamforming vector into a positive semi-definite matrix; a problem transformation subunit for transforming a non-convex beamforming optimization problem based on the positive semi-definite matrix to obtain an intermediate objective function; a function relaxation subunit for obtaining defined relaxation variables, relaxing the intermediate objective function based on the relaxation variables, and determining the constraint conditions satisfied by the relaxation variables; a constraint condition processing subunit for processing the constraint conditions using a continuous convex approximation method to obtain a convex optimization problem; a problem solving subunit for solving the convex optimization problem to obtain the optimal solution of the positive semi-definite matrix; and a local optimal solution solving subunit for obtaining a local optimal solution of active beamforming based on the optimal solution of the positive semi-definite matrix using a Gaussian randomization method.

[0111] According to the present invention, a weighted sum rate optimization device for a multi-user coexisting radio network includes an optimal solution solving unit comprising: a variable acquisition subunit for acquiring the phase shift matrix and auxiliary variables of a defined smart reflector; an expression acquisition subunit for determining the transmission rates of the primary user and the secondary user as functions of the auxiliary variables to obtain a rate expression; an inequality acquisition subunit for introducing relaxation variables into the fractional terms in the rate expression and determining the non-convex constraints as linear inequalities based on the relaxation variables; an inequality solving subunit for solving the linear inequalities to obtain the optimal solution for the auxiliary variables; and an optimal solution solving subunit for obtaining the optimal solution for passive beamforming based on the optimal solution for the auxiliary variables using Gaussian randomization.

[0112] Figure 4 An example is a schematic diagram of the physical structure of an electronic device, such as... Figure 4As shown, the electronic device may include a processor 410, a communications interface 420, a memory 430, and a communication bus 440. The processor 410, communications interface 420, and memory 430 communicate with each other via the communication bus 440. The processor 410 can call logical instructions in the memory 430 to execute a weighted sum rate optimization method for a multi-user co-existing radio network. This method includes: constructing a multi-user co-existing radio network model, which includes a base station, intelligent reflector nodes, primary users, and secondary users; determining an optimization problem based on the multi-user co-existing radio network model, with the goal of maximizing the system's weighted sum rate. The objective function of the optimization problem is determined based on the time slot access matrix, user association matrix, and active / passive beamforming, and satisfies the minimum transmission rate constraints for primary and secondary users and the base station transmit power constraints; iteratively solving the objective function until convergence, obtaining the optimal time slot allocation result, user association result, and optimal solution for active / passive beamforming.

[0113] Furthermore, the logical instructions in the aforementioned memory 430 can be implemented as software functional units and, when sold or used as independent products, can be stored in a computer-readable storage medium. Based on this understanding, the technical solution of the present invention, essentially, or the part that contributes to the prior art, or a part of the technical solution, can be embodied in the form of a software product. This computer software product is stored in a storage medium and includes several instructions to cause a computer device (which may be a personal computer, server, or network device, etc.) to execute all or part of the steps of the methods of the various embodiments of the present invention. The aforementioned storage medium includes various media capable of storing program code, such as USB flash drives, portable hard drives, read-only memory (ROM), random access memory (RAM), magnetic disks, or optical disks.

[0114] On the other hand, the present invention also provides a computer program product, which includes a computer program that can be stored on a non-transitory computer-readable storage medium. When the computer program is executed by a processor, the computer can execute the weighted sum rate optimization method for a multi-user coexisting radio network provided by the above methods. The method includes: constructing a multi-user coexisting radio network model, which includes a base station, a smart reflector node, a primary user, and a secondary user; determining an optimization problem based on the multi-user coexisting radio network model with the goal of maximizing the system weighted sum rate, wherein the objective function of the optimization problem is determined based on the time slot access matrix, the user association matrix, and active and passive beamforming, and satisfies the minimum transmission rate constraints of the primary user and the secondary user, and the base station transmit power constraints; iteratively solving the objective function until convergence, to obtain the optimal time slot allocation result, the user association result, and the optimal solution for active and passive beamforming.

[0115] In another aspect, the present invention also provides a non-transitory computer-readable storage medium storing a computer program thereon. When executed by a processor, the computer program implements a weighted sum rate optimization method for a multi-user co-existing radio network provided by the methods described above. The method includes: constructing a multi-user co-existing radio network model, the multi-user co-existing radio network model including a base station, a smart reflector node, a primary user, and a secondary user; determining an optimization problem based on the multi-user co-existing radio network model with the goal of maximizing the system weighted sum rate, the objective function of the optimization problem being determined based on the time slot access matrix, the user association matrix, and active / passive beamforming, and satisfying the minimum transmission rate constraints for the primary user and secondary user, and the base station transmit power constraints; iteratively solving the objective function until convergence, obtaining the optimal time slot allocation result, the user association result, and the optimal solution for active / passive beamforming.

[0116] The device embodiments described above are merely illustrative. The units described as separate components may or may not be physically separate, and the components shown as units may or may not be physical units; that is, they may be located in one place or distributed across multiple network units. Some or all of the modules can be selected to achieve the purpose of this embodiment according to actual needs. Those skilled in the art can understand and implement this without any creative effort.

[0117] Through the above description of the embodiments, those skilled in the art can clearly understand that each embodiment can be implemented by means of software plus necessary general-purpose hardware platforms, and of course, it can also be implemented by hardware. Based on this understanding, the above technical solutions, in essence or the part that contributes to the prior art, can be embodied in the form of a software product. This computer software product can be stored in a computer-readable storage medium, such as ROM / RAM, magnetic disk, optical disk, etc., including several instructions to cause a computer device (which may be a personal computer, server, or network device, etc.) to execute the methods of various embodiments or some parts of embodiments.

[0118] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention, and not to limit them; although the present invention has been described in detail with reference to the foregoing embodiments, those skilled in the art should understand that modifications can still be made to the technical solutions described in the foregoing embodiments, or equivalent substitutions can be made to some of the technical features; and these modifications or substitutions do not cause the essence of the corresponding technical solutions to deviate from the spirit and scope of the technical solutions of the embodiments of the present invention.

Claims

1. A weighted sum rate optimization method for a multi-user coexisting radio network, characterized in that, include: A multi-user coexisting radio network model is constructed, which includes a base station, a smart reflector node, a primary user, and secondary users; Based on the multi-user co-existing radio network model, an optimization problem is constructed with the goal of maximizing the system weighted sum rate. The objective function of the optimization problem is determined based on the time slot access matrix, the user correlation matrix, and active and passive beamforming, and satisfies the minimum transmission rate constraints for primary and secondary users and the base station transmit power constraints. The objective function is iteratively solved until convergence, yielding the optimal time slot allocation result, user association result, and optimal solution for active and passive beamforming. The iterative solution of the objective function until convergence is obtained to obtain the optimal time slot allocation result, user association result and optimal solution of active and passive beamforming. Specifically, this includes: optimizing the time slot access matrix based on Stackelberg game to determine the optimal time slot allocation strategy. The user association matrix is ​​optimized based on a many-to-one matching algorithm to determine the user association results. Based on the optimal time slot allocation strategy and the user association results, the active and passive beamforming is jointly optimized using a semidefinite relaxation and continuous convex approximation method to obtain the optimal solution for active and passive beamforming.

2. The weighted sum rate optimization method for multi-user co-existing radio networks according to claim 1, characterized in that, The optimization of the time slot access matrix based on Stackelberg game specifically includes: The time slot access problem is modeled as a Stackelberg game problem, in which the base station acts as the leader and dynamically adjusts the time slot access price through a pricing mechanism, while secondary users act as followers and choose the optimal time slot. The Particle Swarm Optimization (PSO) algorithm is used to solve the Stackelberg game problem to find the Stackelberg equilibrium and determine the optimal time slot allocation strategy.

3. The weighted sum rate optimization method for multi-user co-existing radio networks according to claim 1, characterized in that, The optimization of the user association matrix based on the many-to-one matching algorithm specifically includes: Model the user association problem as a many-to-one matching problem; Based on the many-to-one matching problem, preference lists are constructed for secondary users and intelligent reflective surface nodes respectively, and a preliminary matching relationship between secondary users and intelligent reflective surface nodes is established. Based on the preliminary matching relationship, an exchange matching is performed, and the user association result is determined according to the matching result.

4. The weighted sum rate optimization method for multi-user co-existing radio networks according to claim 1, characterized in that, The method of jointly optimizing active and passive beamforming using semi-definite relaxation and continuous convex approximation to obtain the optimal solution for active and passive beamforming specifically includes: The non-convex beamforming optimization problem is transformed into a convex optimization problem by using a semi-definite relaxation method; The convex optimization problem is solved by continuous convex approximation and Gaussian randomization method to obtain the local optimal solution of active beamforming; Based on the local optimal solution of active beamforming, the phase shift of each smart reflector is optimized to obtain the optimal solution of passive beamforming.

5. The weighted sum rate optimization method for multi-user co-existing radio networks according to claim 4, characterized in that, The non-convex beamforming optimization problem is transformed into a convex optimization problem using a semi-definite relaxation method; the convex optimization problem is solved using a continuous convex approximation and Gaussian randomization method to obtain a local optimum solution for active beamforming, specifically including: Transform the active beamforming vector into a positive semi-definite matrix; The non-convex beamforming optimization problem is transformed based on the positive semi-definite matrix to obtain an intermediate objective function; Obtain the defined slack variables, relax the intermediate objective function based on the slack variables, and determine the constraints satisfied by the slack variables; The constraints are processed using a continuous convex approximation method to obtain a convex optimization problem; The optimal solution for the positive semi-definite matrix is ​​obtained by solving the convex optimization problem. By using Gaussian randomization, the local optimum of active beamforming is obtained based on the optimum solution of the positive semi-definite matrix.

6. The weighted sum rate optimization method for multi-user co-existing radio networks according to claim 4, characterized in that, The optimization of the phase shift for each smart reflector to obtain the optimal solution for passive beamforming specifically includes: Obtain the phase shift matrix and auxiliary variables of the defined smart reflector; The transmission rates of the primary user and the secondary user are determined as functions of the auxiliary variables to obtain the rate expression; Relaxation variables are introduced into the fractional terms in the rate expression, and the non-convex constraint is determined as a linear inequality based on the relaxation variables. Solve the linear inequality to obtain the optimal solution for the auxiliary variable; The optimal solution for passive beamforming is obtained by using Gaussian randomization based on the optimal solution of the auxiliary variables.

7. A weighted sum rate optimization device for a multi-user coexisting radio network, characterized in that, include: The model building module is used to build a multi-user coexisting radio network model, which includes a base station, a smart reflector node, a primary user, and secondary users. The optimization problem construction module is used to determine the optimization problem based on the multi-user co-existing radio network model, with the goal of maximizing the system weighted sum rate. The objective function of the optimization problem is determined based on the time slot access matrix, user association matrix, and active and passive beamforming, and satisfies the minimum transmission rate constraints for primary and secondary users and the base station transmit power constraints. The solution module is used to iteratively solve the objective function until convergence, and obtain the optimal time slot allocation result, user association result and optimal solution for active and passive beamforming; The solution module includes: an optimal time slot allocation submodule, used to optimize the time slot access matrix based on Stackelberg game to determine the optimal time slot allocation strategy; an association submodule, used to optimize the user association matrix based on a many-to-one matching algorithm to determine the user association result; and an optimal solution solution submodule, used to jointly optimize active and passive beamforming using semidefinite relaxation and continuous convex approximation methods according to the optimal time slot allocation strategy and the user association result to obtain the optimal solution for active and passive beamforming.

8. The weighted sum rate optimization apparatus for a multi-user coexisting radio network according to claim 7, characterized in that, The optimal time slot allocation submodule includes: The optimal time slot selection unit is used to model the time slot access problem as a Stackelberg game problem, in which the base station, as the leader, dynamically adjusts the time slot access price through a pricing mechanism, and the secondary users, as followers, select the optimal time slot. The optimal time slot allocation unit is used to solve the Stackelberg equilibrium in the Stackelberg game problem using the particle swarm optimization algorithm to determine the optimal time slot allocation strategy.

9. The weighted sum rate optimization apparatus for a multi-user coexisting radio network according to claim 7, characterized in that, The associated sub-modules include: The problem modeling unit is used to model user-related problems as many-to-one matching problems; The preliminary matching unit is used to construct preference lists for secondary users and intelligent reflective surface nodes respectively based on the many-to-one matching problem, and establish a preliminary matching relationship between secondary users and intelligent reflective surface nodes; The result association unit is used to perform exchange matching based on the preliminary matching relationship and determine the user association result based on the matching result.

10. The weighted sum rate optimization apparatus for a multi-user coexisting radio network according to claim 7, characterized in that, The optimal solution solving submodule includes: The problem transformation and solution unit is used to transform the non-convex beamforming optimization problem into a convex optimization problem using a semidefinite relaxation method; and solves the convex optimization problem by using a continuous convex approximation and Gaussian randomization method to obtain the local optimal solution of active beamforming. The optimal solution solving unit is used to optimize the phase shift of each smart reflector based on the local optimal solution of active beamforming to obtain the optimal solution of passive beamforming.

11. The weighted sum rate optimization apparatus for a multi-user coexisting radio network according to claim 10, characterized in that, The problem transformation and solution unit includes: The vector transformation subunit is used to transform the active beamforming vector into a positive semi-definite matrix; The problem transformation subunit is used to transform the non-convex beamforming optimization problem according to the positive semi-definite matrix to obtain the intermediate objective function; The function relaxation subunit is used to obtain the defined relaxation variables, relax the intermediate objective function according to the relaxation variables, and determine the constraint conditions satisfied by the relaxation variables. The constraint processing subunit is used to process the constraints using a continuous convex approximation method to obtain a convex optimization problem; The problem-solving subunit is used to solve the convex optimization problem and obtain the optimal solution of the positive semi-definite matrix; The local optimal solution sub-unit is used to obtain the local optimal solution of active beamforming based on the optimal solution of the semi-positive definite matrix by means of Gaussian randomization.

12. The weighted sum rate optimization apparatus for a multi-user coexisting radio network according to claim 10, characterized in that, The optimal solution solving unit includes: The variable acquisition sub-unit is used to acquire the phase shift matrix and auxiliary variables of the defined smart reflector. The expression acquisition subunit is used to determine the transmission rate of the primary user and the transmission rate of the secondary user as a function of the auxiliary variable, thereby obtaining the rate expression; Inequality acquisition sub-units are used to introduce slack variables into the fractional terms in the rate expression, and the non-convex constraints are determined as linear inequalities based on the slack variables. The inequality solving subunit is used to solve the linear inequalities and obtain the optimal solution for the auxiliary variable; The optimal solution sub-unit is used to obtain the optimal solution for passive beamforming based on the optimal solution of the auxiliary variables through Gaussian randomization.

13. An electronic device comprising a memory, a processor, and a computer program stored in the memory and executable on the processor, characterized in that, When the processor executes the computer program, it implements the weighted sum rate optimization method for a multi-user coexisting radio network as described in any one of claims 1 to 6.

14. A non-transitory computer-readable storage medium having a computer program stored thereon, characterized in that, When the computer program is executed by a processor, it implements the weighted sum rate optimization method for a multi-user coexisting radio network as described in any one of claims 1 to 6.

Images