Beamforming method, apparatus and medium for irs assisted cognitive radio system

Abstract

The application discloses a beamforming method, device and medium of an IRS-assisted cognitive radio system, to solve the beamforming design problem of the IRS-assisted CR system under the conditions of hardware impairment and imperfect channel state information. The method constructs a transmission signal model and a channel uncertainty model of the CR system based on the hardware impairment of the assisting base station, constructs a problem of jointly optimizing the transmission precoding matrix of the secondary user and the reflection coefficient of the intelligent reflecting surface with the target of maximizing the weighted sum transmission rate of the secondary user. The problem is converted into an equivalent weighted minimum mean square error problem, and then decoupled into a first sub-problem of transmit beamforming and a second sub-problem of intelligent reflecting surface beamforming which are solved alternately according to the precoding matrix and the reflection coefficient of the intelligent reflecting surface, and are solved iteratively by successively optimizing the reflection coefficient. The method can reduce the interference of the CR system to the primary user and improve the signal transmission rate under the conditions of imperfect channel state information and hardware impairment.

Description

Technical Field

[0001] This application relates to the field of communications, and more particularly to a beamforming method, apparatus and medium for an IRS-assisted cognitive radio system. Background Technology

[0002] Cognitive radio (CR) is considered an effective method to improve radio spectrum efficiency and energy efficiency, and it has great potential to reduce the cost and complexity of future 5G technologies, such as Multiple-Input Multiple-Output (MIMO). In CR systems, primary users (PUs) represent users with high-priority spectrum licenses and access to the spectrum, while secondary users (SUs) typically represent unlicensed users who can share the spectrum without causing harmful interference to the PUs. However, a problem exists in CR systems where performance improvements for PUs and SUs are conflicting. Specifically, increasing the transmit power at an SU base station can enhance its signal strength, but this will cause greater interference to the PUs, and the channel between SUs and PUs is difficult to estimate based on their non-cooperative relationship.

[0003] Intelligent Reflecting Surfaces (IRS) can improve the spectral and energy efficiency of wireless communication systems through pre-programmed controllers. These controllers precisely control the direction of electromagnetic wave reflection by manipulating the reflective elements on the IRS. This allows the reflected signal to be reconfigured to propagate in the desired direction. The signal reflected by the IRS can be added along with other signal paths to increase the signal strength of the desired receiver or mitigate co-channel interference for unintended users.

[0004] Therefore, combining the IRS and CR systems can help enhance the expected signal strength of the SU and reduce interference to the PU.

[0005] However, most current combinations of IRS and CR systems are based on the assumption of perfect transceiver hardware and perfect Channel State Information (CSI). In actual communication systems, however, transceiver hardware conditions and channel state information are usually not perfect. Unavoidable hardware damage can lead to a decrease in system performance and impair the signal quality of the receiver. Summary of the Invention

[0006] This application provides a beamforming method, apparatus, and medium for an IRS-assisted cognitive radio system to solve the beamforming design problem of an IRS-assisted cognitive radio system under conditions of hardware impairment and imperfect channel state information.

[0007] This application provides a beamforming method for an IRS-assisted cognitive radio system, the system including primary users, secondary users, a primary base station, and an auxiliary base station, the method comprising:

[0008] Based on the hardware impairments of the auxiliary base station, a transmission signal model and a channel uncertainty model of the system are constructed.

[0009] Based on the aforementioned transmission signal model and channel uncertainty model, with the goal of maximizing the weighted sum transmission rate of secondary users, a first problem is constructed to jointly optimize the secondary user transmission precoding matrix and the intelligent reflector reflection coefficient.

[0010] The first problem is transformed into a second problem of minimizing the mean square error using an equivalent weighted average.

[0011] Based on the precoding matrix and the reflection coefficient of the smart reflector, the second problem is decoupled into a first sub-problem of transmitted beamforming and a second sub-problem of smart reflector beamforming, which are solved alternately.

[0012] The first subproblem is transformed into a quadratic constrained quadratic programming problem, and the precoding matrix is ​​optimized and solved. The reflection coefficients corresponding to the second subproblem are then iteratively optimized and solved.

[0013] Furthermore, based on the hardware impairment of the auxiliary base station, a transmission signal model of the system is constructed, specifically including: determining the first transmission signal of the auxiliary base station and the second transmission signal of the main base station based on the secondary user precoding matrix, data symbol vector, and noise corresponding to the hardware impairment of the auxiliary base station; determining the signal received by the main user based on the interference of the auxiliary base station to the main user, and the first and second transmission signals; and determining the signal received by the secondary user based on the interference and thermal noise of the main base station to the secondary user, and the first and second transmission signals.

[0014] Furthermore, the noise corresponding to the hardware impairment of the auxiliary base station is obtained by determining the corresponding noise based on the hardware impairment factor of the auxiliary base station; the power loss corresponding to the noise is proportional to the power of the transmitted signal.

[0015] Furthermore, based on the hardware impairments of the auxiliary base station, a transmission signal model and a channel uncertainty model of the system are constructed, specifically including: determining the channel through which the auxiliary base station transmits information based on the estimation error, and obtaining an equivalent channel matrix with uncertainty; representing the equivalent channel matrix using the Gauss-Schronek model; and using the least squares estimation method to assume that the estimation error is uncorrelated with the channel coefficients, thereby obtaining the channel uncertainty model corresponding to the final channel matrix.

[0016] Furthermore, the first problem of jointly optimizing the secondary user transmission precoding matrix and the reflection coefficient of the smart reflector specifically includes: determining the transmission power constraints of the auxiliary base station, the interference constraints of the primary user, and the reflection coefficient constraints of the smart reflector; and based on the constraints, constructing the first problem of jointly optimizing the secondary user transmission precoding matrix and the reflection coefficient of the smart reflector.

[0017] Furthermore, the first problem is transformed into a second problem of minimizing the mean square error using an equivalent weighted method. Specifically, this includes: using a minimum mean square error algorithm, introducing an auxiliary matrix and a decoding matrix to obtain the mean square error of the auxiliary base station, and expressing the first problem as a second problem of minimizing the mean square error; differentiating the auxiliary matrix and the decoding matrix to obtain the optimal auxiliary matrix and the optimal decoding matrix; transforming the main user interference constraint of the second problem into an interruption probability constraint, and obtaining the re-represented third problem through DBLDI-type inequalities.

[0018] Furthermore, the first subproblem is transformed into a quadratic constrained quadratic programming problem, and the precoding matrix is ​​optimized and solved. Specifically, this includes: optimizing the precoding matrix based on the decoding matrix, the auxiliary matrix, and the given reflection coefficients; substituting the mean square error into the first subproblem to transform it into a quadratic constrained quadratic programming problem; and solving the quadratic constrained quadratic programming problem using the standard interior point method.

[0019] Furthermore, the reflection coefficients corresponding to the second sub-problem are successively optimized and iteratively solved, specifically including: optimizing the reflection coefficients based on the decoding matrix, auxiliary matrix, and given precoding matrix; substituting the mean square error into the third problem, and iteratively updating the optimal solution for the phase shift based on the derivative of the objective function.

[0020] This application provides a beamforming apparatus for an IRS-assisted cognitive radio system, the system including primary users, secondary users, a primary base station, and an auxiliary base station, the apparatus comprising:

[0021] The model building module constructs the transmission signal model and channel uncertainty model of the system based on the hardware impairment of the auxiliary base station;

[0022] The problem construction module, based on the transmission signal model and the channel uncertainty model, aims to maximize the weighted sum transmission rate of secondary users and constructs a first problem that jointly optimizes the transmission precoding matrix of secondary users and the reflection coefficient of the smart reflector.

[0023] The transformation module transforms the first problem into a second problem of minimizing the mean square error using an equivalent weighted average.

[0024] The decoupling module, based on the precoding matrix and the reflection coefficient of the smart reflector, decouples the second problem into a first sub-problem of transmitted beamforming that is solved alternately and a second sub-problem of smart reflector beamforming.

[0025] The solution module transforms the first subproblem into a quadratic constrained quadratic programming problem, performs precoding matrix optimization, and iteratively optimizes the reflection coefficients corresponding to the second subproblem.

[0026] This application provides a storage medium for storing computer-executable instructions, which, when executed, implement the method described in any of the above-described embodiments.

[0027] The beneficial effects of this invention are as follows:

[0028] This application provides a beamforming method, apparatus, and medium for an IRS-assisted cognitive radio system. An IRS-assisted cognitive radio system based on a statistical channel error model is designed. This system model considers multiple primary and secondary users. It assumes that the channel through which the primary base station transmits information is known, while the channel through which the auxiliary base station transmits information is imperfect. To describe the statistical characteristics of the channel imperfection error in the IRS-assisted MIMO system, the classic Gauss-Kronecker model is adopted. Furthermore, to handle the complex objective function, the problem is reformulated using a weighted minimum mean square error method. Specifically, for the auxiliary matrix, decoding matrix, transmit beamforming matrix, and IRS reflection coefficient matrix, while keeping other matrices unchanged, one of the matrices is iteratively solved until convergence.

[0029] Then, the non-convex problem was decoupled into two subproblems that can be solved alternately using an alternating optimization algorithm. The transmit beam path problem was solved using CVX, and the reflection coefficient was solved sequentially using an iterative algorithm, optimizing only one coefficient at a time. The optimality condition of the IRS coefficient was given.

[0030] This reduces interference to primary users from cognitive radio systems, increases signal transmission rates, and improves system performance, even with imperfect channel state information (CSI) and transceiver hardware impairments. Attached Figure Description

[0031] The accompanying drawings, which are included to provide a further understanding of this application and form part of this application, illustrate exemplary embodiments and are used to explain this application, but do not constitute an undue limitation of this application. In the drawings:

[0032] Figure 1 A schematic diagram of an IRS-assisted cognitive radio system model provided for one or more embodiments of this specification;

[0033] Figure 2 A flowchart of a beamforming method for an IRS-assisted cognitive radio system provided in one or more embodiments of this specification;

[0034] Figure 3 A schematic diagram illustrating the beamforming design principle of an IRS-assisted cognitive radio system provided for one or more embodiments of this specification;

[0035] Figure 4 This is a schematic diagram of the beamforming apparatus structure for one or more embodiments of the IRS-assisted cognitive radio system provided in this specification. Detailed Implementation

[0036] To make the objectives, technical solutions, and advantages of this application clearer, the technical solutions of this application will be clearly and completely described below in conjunction with specific embodiments and corresponding drawings. Obviously, the described embodiments are only a part of the embodiments of this application, and not all of them. Based on the embodiments in this application, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of this application.

[0037] Figure 1 This is a schematic diagram of an IRS-assisted cognitive radio system model provided for one or more embodiments of this specification.

[0038] like Figure 1 As shown, the system model of the intelligent reflector (IRS) assisted downlink multiple-input multiple-output (MIMO) cognitive radio (CR) system consists of several nodes: K primary users (PU), L secondary users (SU), intelligent reflector (IRS), primary base station (PT), and auxiliary base station (ST).

[0039] The main transmitter (i.e., the primary base station) serves multiple secondary users without affecting communication between the primary users. PT, ST, PU, ​​and SU each have M PT M ST M PU M SU A single antenna. Assuming the IRS consists of N passive reflective elements, the IRS can accept and passively reflect reflected signals, requiring no additional RF circuitry. Each element is composed of... Indicates. βn ∈[0,1] represents the amplitude. Assuming total internal reflection of the signal, then β n =1. θ n ∈[0,2π] represents the phase shift of the nth element. definition For the main user group, This is a collection of secondary users.

[0040] The channels of PT-IRS, PT-PU, PT-SU, ST-IRS, ST-PU, and ST-SU are respectively controlled by...

[0041] This indicates that the reflection channels IRS-PU and IRS-SU are... and The term is represented as follows. Here, PT-IRS represents the channel between PT and IRS; other similar forms follow the same logic.

[0042] Figure 2 A flowchart of a beamforming method for an IRS-assisted cognitive radio system provided in one or more embodiments of this specification, specifically including the following steps:

[0043] S201: Based on the hardware impairment of the auxiliary base station, construct the system's transmission signal model and channel uncertainty model.

[0044] This application takes into account the hardware defects of the base station and the imperfections of the channel, and requires the construction of corresponding transmission signal models and channel uncertainty models.

[0045] When constructing the transmission signal model, the first transmission signal of the auxiliary base station and the second transmission signal of the main base station are determined based on the secondary user precoding matrix, data symbol vector, and noise corresponding to hardware impairments of the auxiliary base station. The signal received by the main user is determined based on the interference of the auxiliary base station to the main user, as well as the first and second transmission signals. The signal received by the secondary user is determined based on the interference and thermal noise of the main base station to the secondary user, as well as the first and second transmission signals. The noise corresponding to hardware impairments of the auxiliary base station is determined according to the hardware impairment factor of the auxiliary base station, and the power loss corresponding to the noise is proportional to the transmitted signal power.

[0046] Specifically, in the IRS-assisted CR system model, ST-SU considers transceiver hardware impairments, while PT-PU does not. Assume the number of data streams per SU is d, satisfying 1 ≤ d ≤ min{M}. ST M SU Therefore, the transmitted signal of ST is: The PT's transmission signal is The parentheses “(1)” and the numbers inside them are used to mark the aforementioned formula and facilitate its reference in the following text. The same applies to similar forms in the following text.

[0047] Let ST be the linear precoding matrix from the l-th SU, and define... It is a set of linear precoding matrices.

[0048] ν l The data symbol vector of the l-th SU and follows and This noise is caused by ST hardware damage, and its power loss is proportional to its transmitted signal power, that is: β t ∈[0,1] represents the hardware impairment factor of the base station.

[0049] Considering the interference of ST on PU, the signal received by the k-th PU can be expressed as:

[0050]

[0051] Based on the interference and thermal noise of PT on SU The signal received by the l-th SU is:

[0052]

[0053] in

[0054]

[0055]

[0056] Where h n，l g n，k and f sn They are H r，l G r，k and The nth column, and The mean is 0 at the k-th PU, and the variance is . Additive white Gaussian noise, Indicates the interference and thermal noise at PT. s，l The combined effect. l n represents the received distortion noise at the first SU. l It follows a zero-mean Gaussian distribution, and its variance is proportional to the received signal power, i.e.:

[0057] βl ∈[0,1] represents the hardware impairment factor of the first SU.

[0058] When constructing the channel uncertainty model, the channel through which the auxiliary base station transmits information is determined based on the estimation error, and the equivalent channel matrix of uncertainty is obtained. The Gauss-Schronek model is used to represent the equivalent channel matrix. The least squares estimation method is used to assume that the estimation error is uncorrelated with the channel coefficients, and the channel uncertainty model corresponding to the final channel matrix is ​​obtained.

[0059] Specifically, assuming the channel through which PT transmits information is known, that is: the channel H between PT-SU, PT-PU, PT-IRS-PU, and PT-IRS-SU. p，l G p，k G r，k F pr H r，l F pr It is known that the channel through which ST transmits information is uncertain.

[0060] Robust design based on estimated cascaded channels, where ST-IRS-SU and ST-IRS-PU are modeled as follows: and Direct channel H s，l G s，k The corresponding estimation errors are: ΔH s，l ΔG s，k The actual channel can be represented as:

[0061] in Let be the known estimated channel matrices of the channels between ST-SU, ST-PU, ST-IRS-SU, and ST-IRS-PU, respectively. Then, the equivalent channel matrix with uncertainty can be obtained from (4) and (5):

[0062]

[0063]

[0064]

[0065] Therefore, we can conclude that:

[0066]

[0067] Since discussing robust design under worst-case channel conditions may lead to low resource utilization, and robust design with statistical CSI error generally performs better in enhancing the global stability of the system, this application adopts a robust design with statistical CSI error.

[0068] To more concisely describe the statistical channel error in IRS-assisted MIMO scenarios, the classic Gauss-Schronek model is adopted:

[0069]

[0070] Where A s,l A s,k A n,l A n,k The covariance matrix seen from the receiving end, B s,l B s,k B n,l B n,k This is the covariance matrix as seen from the transmitting end. Define the set of spatial correlation matrices. Estimating the CSI set The estimated CSI error terms all follow a zero-mean circular symmetric complex Gaussian (CSCG) distribution with independent and identically distributed (i.e., i ... Then set The spatial correlation matrix in the equation can be represented by the following formula:

[0071]

[0072] The least squares (LS) estimation method assumes that the statistical error is uncorrelated with the estimated channel coefficients. Therefore:

[0073]

[0074] S202: Based on the transmission signal model and the channel uncertainty model, this paper addresses the first problem of constructing a joint optimization of the secondary user transmission precoding matrix and the intelligent reflector reflection coefficient, with the goal of maximizing the weighted sum transmission rate of secondary users.

[0075] In this application embodiment, it is necessary to determine the transmission power constraints of the auxiliary base station, the interference constraints of the main user, and the reflection coefficient constraints of the smart reflector. Based on the above constraints, a first problem is to jointly optimize the secondary user transmission precoding matrix and the reflection coefficient of the smart reflector.

[0076] Specifically, the transmission rate of the l-th SU is:

[0077]

[0078]

[0079] in:

[0080] Depend on have: Therefore, the above formula can be transformed as follows:

[0081]

[0082] Where A1=β l WW H +1+β l )β t diag(WW H ).

[0083] The interference power applied by ST to the k-th PU is:

[0084]

[0085] Given the constraints of total ST transmission power, PU interference, IRS reflection coefficient, and considering imperfect CSI, we obtain the first problem P1: jointly optimizing the ST transmission precoding matrix W and IRS reflection coefficient to maximize the weighted sum-rate (WSR) of SUs.

[0086] P1:

[0087]

[0088]

[0089]

[0090]

[0091]

[0092]

[0093] Among them, P in (15b) t The set rated power represents the total transmission power of ST. (15c) represents the interference constraint at the k-th PU, 0 ≤ μ k ≤1 represents the interruption probability when the interference constraint applied to the k-th PU exceeds the threshold. (15d) and (15e) represent the reflection coefficient constraint at the IRS.

[0094] S203: Transform the first problem into the second problem of minimizing the mean square error using equivalent weighting.

[0095] In this embodiment, for ease of solution, a minimum mean square error algorithm is employed, introducing an auxiliary matrix and a decoding matrix to obtain the mean square error of the auxiliary base station. The first problem of maximizing the joint WSR is expressed as a second problem of weighted minimum mean square error (WMMSE). Differentiation of the auxiliary matrix and the decoding matrix yields the optimal auxiliary matrix and the optimal decoding matrix. Since interference power constraints will affect the solution results under the statistical error channel model, the interference constraint is transformed into an outage probability constraint. Furthermore, through a Decomposition-Based Large Deviation Inequality (DBLDI) type inequality, a third problem, which is easier to handle after re-representation, is obtained.

[0096] Specifically,

[0097] The first problem is reconstructed using the WMMSE method, introducing an auxiliary matrix set. Decoding Matrix Set S I Let E be the decoding matrix of the l-th SU. Therefore, the mean square error E of the l-th SU is... l for:

[0098]

[0099] in:

[0100]

[0101] The first problem P1 can be reformulated as the second problem P2:

[0102]

[0103] st(15b),(15c),(15d),(15e)(18b)

[0104]

[0105]

[0106] Taking the derivatives of the decoding matrix S and the auxiliary matrix Z respectively and setting them to 0, we obtain the optimal decoding matrix S. ° and the optimal auxiliary matrix Z ° :

[0107]

[0108]

[0109] Decoding matrix Substituting into (16), we get:

[0110] Furthermore, since the interruption probability constraint in (15c) is non-convex and does not have a simple closed-form expression, it can be transformed into another easily handled expression by applying DBLDI-type inequalities.

[0111] The DBLDI lemma is as follows: Assume It is a standard complex Gaussian random vector. Then for any μ k ∈(0,1], the DBLDI type inequality is:

[0112]

[0113] Where x and y are slack variables. This means A equals B.

[0114] Using lemmas of DBLDI type inequalities and matrix transformations

[0115] The interruption probability constraint can be equivalent to:

[0116]

[0117] in Introduce auxiliary variables x = [x1, ..., x k ],y=[y1,…,y k Therefore, it can be approximated as:

[0118]

[0119] Therefore, the problem can be reformulated as the third problem, P3:

[0120]

[0121]

[0122]

[0123]

[0124]

[0125] in:

[0126]

[0127] Then, the optimization problem is solved by using the given decoding matrix S and auxiliary matrix Z to optimize the precoding matrix W and the reflection coefficient matrix Φ.

[0128] S204: Based on the precoding matrix and the reflection coefficients of the smart reflector, the second problem is decoupled into the first subproblem of transmit beamforming and the second subproblem of smart reflector beamforming, which are solved alternately.

[0129] The algorithm employs an alternating optimization approach, transforming the joint optimization problem into an alternating optimization problem for solution.

[0130] S205: Transform the first subproblem into a quadratic constrained quadratic programming problem, perform precoding matrix optimization, and iteratively optimize the reflection coefficients corresponding to the second subproblem.

[0131] In optimizing the precoding matrix, this application optimizes the precoding matrix based on the decoding matrix, auxiliary matrix, and given reflection coefficients. The mean square error is substituted into the first subproblem, which is then transformed into a quadratically constrained quadratic programming (QCQP) problem. The standard interior-point method is then used to solve the quadratically constrained quadratic programming problem.

[0132] Specifically, given the reflection coefficients, decoding matrix, and auxiliary matrix at IRS, the precoding matrix at ST is optimized, and E... l Substitute these terms into the objective function of problem P3 and discard irrelevant terms:

[0133] Based on the properties of the trace of a matrix, we can obtain: tr(WW) H )=tr{diag(WW H )}

[0134] therefore:

[0135] tr(WW H +A1)=tr(WW H +β l WW H +(1+β l )β t diag(WW H ))

[0136] =(1+β) l ){tr(WW H )+β t tr[diag(WW H )]}

[0137] =(1+β) l (1+β) t)tr(WW H (28)

[0138] The precoding matrix optimization problem can be transformed into the first subproblem P4:

[0139]

[0140]

[0141]

[0142]

[0143]

[0144] in,

[0145] WW H +A1=(1+β l )[WW H +β t diag(WW H (30a)

[0146]

[0147]

[0148]

[0149] This problem is a QCQP problem, which can be solved using the standard interior-point method with the CVX tool.

[0150] When optimizing the IRS reflection coefficient, the reflection coefficient is optimized based on the decoding matrix, auxiliary matrix and given precoding matrix. The mean square error is substituted into the third problem, and the optimal solution of phase shift is obtained by iterative update based on the derivative of the objective function.

[0151] Specifically, given the precoding matrix W, decoding matrix S, and auxiliary matrix Z at ST, we discuss the optimization of the reflection coefficient matrix Φ at IRS. Let E l Substitute these terms into the objective function of problem P3 and discard terms irrelevant to the phase shift:

[0152]

[0153] in:

[0154]

[0155]

[0156] Then set The second subproblem P5 is represented as follows:

[0157]

[0158]

[0159]

[0160] Φ=diag(φ1,φ2,…,φ N (33d)

[0161] First, combining (32), we have:

[0162] objective function

[0163] rank(A n ) = M SU ,rank(B n )≤rank(C′ n ) = 1, then Diagonalizable if and only if

[0164] Therefore, we can discuss different situations, when... hour, At this time, any condition that satisfies |φ n Both |=1 and (33b) are optimal solutions.

[0165] when There are two situations at that time.

[0166] ① Can be Perform singular value decomposition, therefore T is a unitary matrix. n =

[0167] diagλ n ,0,…,0},λ n It is T n The only non-zero singular value.

[0168]

[0169]

[0170] in

[0171] Then we can obtain: make

[0172]

[0173] Differentiate the objective function Solving for:

[0174]

[0175] ② It can be represented as two non-zero vectors The product of, i.e. and

[0176]

[0177]

[0178] The optimal phase shift is obtained as follows:

[0179] In summary, the optimal solution for phase shift in the objective function is:

[0180]

[0181] Substitute (36) into constraint (33b) to solve for x and y, and iteratively update the phase shift.

[0182] In this embodiment, the joint WSR maximization problem is first transformed into an equivalent weighted sum and mean square error minimization problem. Secondly, since interference power constraints will affect the solution results under the statistical error channel model, this constraint is transformed into an outage probability constraint. Then, by applying a DBLDI-type inequality, it is transformed into another easily tractable expression. Under the new optimization problem, the problem is decoupled into two alternately solvable subproblems: solving the transmit beamforming problem and solving the IRS beamforming problem. In solving the transmit beamforming problem, the subproblem is transformed into a quadratic constraint quadratic programming problem, solved using the CVX toolkit. In solving the IRS reflection beamforming problem, a successive optimization method for reflection coefficients is used to solve the reflection coefficients in the cascaded signal. Only one reflection coefficient is solved at a time, and then the optimal solution is obtained through an iterative algorithm. This reduces interference to the primary user from the cognitive radio system, increases signal transmission rate, and improves system performance under conditions of imperfect channel state information and transceiver hardware impairments.

[0183] Figure 3 Provided for embodiments of this application Figure 2A schematic diagram illustrating the beamforming design principle of the IRS-assisted cognitive radio system.

[0184] like Figure 3 As shown, the transmission channel model and robust transmission technology are applied to the IRS-assisted cognitive radio system. Given preset channel state information, user location information, communication model information, and IRS location information, the goal is to maximize the weighted sum rate of secondary users. In the problem-solving process, the problem is first simplified by transforming the joint weighted sum problem into an equivalent weighted sum minimization of mean square error problem. After further decomposition, a successive iterative optimization algorithm for reflection coefficients is used to solve the problem, ultimately obtaining a local optimum.

[0185] The above describes the beamforming method for an IRS-assisted cognitive radio system provided in this application. Based on the same inventive concept, this application also provides a corresponding beamforming apparatus for an IRS-assisted cognitive radio system, such as... Figure 4 As shown.

[0186] Figure 4 A schematic diagram of the beamforming apparatus structure for one or more embodiments of the IRS-assisted cognitive radio system provided in this specification is shown, specifically including:

[0187] Model building module 401 constructs the transmission signal model and channel uncertainty model of the system based on the hardware impairment of the auxiliary base station;

[0188] Problem construction module 402, based on the transmission signal model and channel uncertainty model, constructs a first problem to jointly optimize the secondary user transmission precoding matrix and the intelligent reflector reflection coefficient, with the goal of maximizing the weighted sum transmission rate of secondary users;

[0189] The transformation module 403 transforms the first problem into a second problem of minimizing the mean square error using an equivalent weighted average.

[0190] The decoupling module 404 decouples the second problem into a first sub-problem of transmitting beamforming and a second sub-problem of intelligent reflector beamforming, which are solved alternately, based on the precoding matrix and the reflection coefficient of the intelligent reflector.

[0191] The solution module 405 transforms the first subproblem into a quadratic constrained quadratic programming problem, performs precoding matrix optimization, and iteratively optimizes the reflection coefficients corresponding to the second subproblem.

[0192] The various embodiments in this application are described in a progressive manner. Similar or identical parts between embodiments can be referred to mutually. Each embodiment focuses on describing the differences from other embodiments. In particular, the device embodiments are basically similar to the method embodiments, so the description is relatively simple; relevant parts can be referred to the descriptions of the method embodiments.

[0193] It should also be noted that the terms "comprising," "including," or any other variations thereof are intended to cover non-exclusive inclusion, such that a process, method, article, or apparatus that comprises a list of elements includes not only those elements but also other elements not expressly listed, or elements inherent to such a process, method, article, or apparatus. Without further limitation, an element defined by the phrase "comprising one..." does not exclude the presence of other identical elements in the process, method, article, or apparatus that includes said element.

[0194] The above description is merely an embodiment of this application and is not intended to limit the scope of this application. Various modifications and variations can be made to this application by those skilled in the art. Any modifications, equivalent substitutions, improvements, etc., made within the spirit and principles of this application should be included within the scope of the claims of this application.

Claims

1. A beamforming method for an IRS-assisted cognitive radio system, the system comprising primary users, secondary users, a primary base station, and an auxiliary base station, characterized in that, The method includes: Based on the hardware impairments of the auxiliary base station, a transmission signal model and a channel uncertainty model of the system are constructed. Based on the aforementioned transmission signal model and channel uncertainty model, with the goal of maximizing the weighted sum transmission rate of secondary users, a first problem is constructed to jointly optimize the secondary user transmission precoding matrix and the intelligent reflector reflection coefficient. The first problem is transformed into a second problem of minimizing the mean square error using an equivalent weighted average. The minimum mean square error algorithm is adopted, and an auxiliary matrix and a decoding matrix are introduced to obtain the mean square error of the auxiliary base station. The first problem of maximizing the joint WSR is expressed as the second problem of weighted minimum mean square error. By taking the derivatives of the auxiliary matrix and the decoding matrix, the optimal auxiliary matrix and the optimal decoding matrix are obtained. Transform the disturbance constraint into an interruption probability constraint, and obtain a third problem that is easier to handle after being re-represented through inequalities; When optimizing the IRS reflection coefficient, the reflection coefficient is optimized based on the decoding matrix, auxiliary matrix and given precoding matrix. The mean square error is substituted into the third problem, and the optimal solution of phase shift is obtained by iterative update based on the derivative of the objective function. Based on the precoding matrix and the reflection coefficient of the smart reflector, the second problem is decoupled into a first sub-problem of transmitted beamforming and a second sub-problem of smart reflector beamforming, which are solved alternately. The first subproblem is transformed into a quadratic constrained quadratic programming problem, and the precoding matrix is ​​optimized and solved. The reflection coefficients corresponding to the second subproblem are then iteratively optimized and solved.

2. The method according to claim 1, characterized in that, Based on the hardware impairments of the auxiliary base station, a transmission signal model of the system is constructed, specifically including: Based on the secondary user precoding matrix, data symbol vector, and noise corresponding to hardware impairment of the auxiliary base station, the first transmission signal of the auxiliary base station and the second transmission signal of the main base station are determined. Based on the interference of the auxiliary base station to the main user, and the first transmission signal and the second transmission signal, determine the signal received by the main user; The signal received by the secondary user is determined based on the interference and thermal noise of the primary base station to the secondary user, as well as the first transmission signal and the second transmission signal.

3. The method according to claim 2, characterized in that, The noise corresponding to the hardware damage of the auxiliary base station is obtained in the following way: Based on the hardware impairment factor of the auxiliary base station, the corresponding noise is determined; the power loss corresponding to the noise is proportional to the transmitted signal power.

4. The method according to claim 1, characterized in that, Based on the hardware impairments of the auxiliary base station, a transmission signal model and a channel uncertainty model for the system are constructed, specifically including: Based on the estimation error, the channel through which the auxiliary base station transmits information is determined, and the uncertain equivalent channel matrix is ​​obtained; The equivalent channel matrix is ​​represented using the Gauss-Schronek model; By employing the least squares estimation method and assuming that the estimation error is uncorrelated with the channel coefficients, the channel uncertainty model corresponding to the final channel matrix is ​​obtained.

5. The method according to claim 1, characterized in that, The first problem in constructing a joint optimized secondary user transport precoding matrix and smart reflector reflection coefficients specifically includes: Determine the transmission power constraints, main user interference constraints, and reflection coefficient constraints of the intelligent reflector for the auxiliary base station; Based on the aforementioned constraints, the first problem is to construct a joint optimization of the secondary user transport precoding matrix and the smart reflector reflection coefficients.

6. The method according to claim 1, characterized in that, The first problem is transformed into a second problem of equivalent weighted minimization of the mean square error, specifically including: The minimum mean square error algorithm is adopted, and an auxiliary matrix and a decoding matrix are introduced to obtain the mean square error of the auxiliary base station. The first problem is then expressed as the second problem of minimum mean square error. Taking the derivatives of the auxiliary matrix and the decoding matrix, we obtain the optimal auxiliary matrix and the optimal decoding matrix. The main user interference constraint of the second problem is transformed into an interruption probability constraint, and the third problem is obtained by using the DBLDI type inequality.

7. The method according to claim 1, characterized in that, The first subproblem is transformed into a quadratic constrained quadratic programming problem, which is then solved using precoding matrix optimization. Specifically, this includes: The precoding matrix is ​​optimized based on the decoding matrix, the auxiliary matrix, and the given reflection coefficients; Substituting the mean square error into the first subproblem, the first subproblem is transformed into a quadratic constrained quadratic programming problem; The standard interior point method is used to solve the quadratic constrained quadratic programming problem.

8. The method according to claim 6, characterized in that, The reflection coefficient corresponding to the second subproblem is solved iteratively and successively, specifically including: The reflection coefficients are optimized based on the decoding matrix, auxiliary matrix, and given precoding matrix. Substituting the mean square error into the third problem, and taking the derivative of the objective function, the optimal solution for the phase shift is obtained through iterative updates.

9. A beamforming apparatus for an IRS-assisted cognitive radio system, the system comprising a primary user, a secondary user, a primary base station, and an auxiliary base station, characterized in that, The device includes: The model building module constructs the transmission signal model and channel uncertainty model of the system based on the hardware impairment of the auxiliary base station; The problem construction module, based on the transmission signal model and the channel uncertainty model, aims to maximize the weighted sum transmission rate of secondary users and constructs a first problem that jointly optimizes the transmission precoding matrix of secondary users and the reflection coefficient of the smart reflector. The transformation module transforms the first problem into a second problem of minimizing the mean square error using an equivalent weighted average. The minimum mean square error algorithm is adopted, and an auxiliary matrix and a decoding matrix are introduced to obtain the mean square error of the auxiliary base station. The first problem of maximizing the joint WSR is expressed as the second problem of weighted minimum mean square error. By taking the derivatives of the auxiliary matrix and the decoding matrix, the optimal auxiliary matrix and the optimal decoding matrix are obtained. Transform the disturbance constraint into an interruption probability constraint, and obtain a third problem that is easier to handle after being re-represented through inequalities; When optimizing the IRS reflection coefficient, the reflection coefficient is optimized based on the decoding matrix, auxiliary matrix and given precoding matrix. The mean square error is substituted into the third problem, and the optimal solution of phase shift is obtained by iterative update based on the derivative of the objective function. The decoupling module, based on the precoding matrix and the reflection coefficient of the smart reflector, decouples the second problem into a first sub-problem of transmitted beamforming that is solved alternately and a second sub-problem of smart reflector beamforming. The solution module transforms the first subproblem into a quadratic constrained quadratic programming problem, performs precoding matrix optimization, and iteratively optimizes the reflection coefficients corresponding to the second subproblem.

10. A storage medium, characterized in that, Used to store computer-executable instructions, which, when executed, implement the method as described in any one of claims 1 to 8.

Images