MISO system secrecy rate maximization method and system based on active RIS and artificial noise assistance
By introducing active RIS and artificial noise into the MISO system and optimizing relevant parameters, the problem of restricted confidentiality rate improvement caused by signal fading in passive RIS auxiliary systems is solved, and higher confidentiality rate and system security are achieved.
Patent Information
- Application Number
- CN202510247033.4
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- Filing Date
- 2025-03-04
- Publication Date
- 2025-06-20
AI Technical Summary
The existing passive RIS-assisted MISO system in the base station-reconfigurable intelligent surface-user link has limited confidentiality improvement due to signal fading problems, and the research on active RIS-assisted is not yet mature.
Using a MISO system based on active RIS and artificial noise, the confidentiality rate of the system is maximized by optimizing beamforming vectors, artificial noise vectors, active RIS reflection coefficient matrix and auxiliary variables.
Overcoming the "double fading" effect in wireless transmission, significantly improving the confidentiality rate of the MISO system, thereby enhancing the security of the system.
Smart Images

Figure CN120186619A_ABST
Abstract
Description
Technical Field
[0001] The present invention belongs to the technical field of physical layer security in wireless communication, and particularly relates to a method and system for maximizing the secrecy rate of a Multiple Input Single Output (MISO) system based on active RIS and artificial noise assistance. Background Art
[0002] A Reconfigurable Intelligent Surface (RIS) is a device manufactured using metamaterial technology and consists of a large number of passive reflecting elements that can be adjusted by a pre-programmed controller. The RIS technology can change the wireless transmission environment, thereby achieving the expansion of wireless coverage, the elimination of interference, and the improvement of spectral efficiency and energy efficiency. Application scenarios include deterministic wireless transmission, blind spot filling of wireless coverage, wireless coverage expansion, enhancement of wireless system capacity, indoor carriages, etc., and have broad application prospects. Currently, passive RIS is mainly used to assist the physical layer security of wireless communication. However, in the base station - reconfigurable intelligent surface - user link, since the signal is affected by two fading effects when transmitting between the RIS and the user, the signal quality is reduced, thereby limiting the improvement of the secrecy rate by passive RIS. Considering that the research on active RIS-assisted wireless communication systems is still in its infancy and there is no implementation case. Based on this, the present invention provides a method and system for maximizing the secrecy rate of a MISO system based on active RIS and artificial noise assistance. Summary of the Invention
[0003] Aiming at the defects in the existing physical layer security technology of passive RIS-assisted MISO systems, the present invention provides a method and system for maximizing the secrecy rate of a MISO system based on active RIS and artificial noise assistance.
[0004] To solve the above problems, the technical solution of the present invention is as follows:
[0005] A method for maximizing the secrecy rate of a MISO system based on active RIS and artificial noise assistance, comprising the following steps:
[0006] S1: Calculate the secrecy rate of the system according to the information rates of legitimate users and eavesdroppers where R B and R E are the information rates at the legitimate user and the eavesdropper respectively; represents the value of the nth secrecy rate iteration, n is a constant; the initial value of the maximum secrecy rate ε is the convergence threshold, usually taking the value of 10 -3 ;
[0007] S2: Determine whether the result obtained in step S1 or S6 satisfies If not, execute step S3; if so, jump to step S7 to output the result;
[0008] S3: Take the result obtained in step S2 as the input parameter, execute the beamforming vector and artificial noise vector optimization method, and output an M×1-dimensional beamforming vector whose elements are all complex numbers and artificial noise vector where M is a natural number representing the number of rows of the vector; the superscript n represents the value of the corresponding nth iteration, and n is a constant;
[0009] S4: Take the result obtained in step S3 as the input parameter, execute the active RIS reflection coefficient matrix optimization method, and output an N×N active RIS reflection coefficient matrix Θ whose elements are complex numbers; where N is a natural number representing the number of rows and columns of the matrix;
[0010] S5: Take Θ obtained in step S4 as the input parameter, execute the auxiliary variable optimization and update method, and update the real numbers and The superscript n represents the value of the corresponding nth iteration, and n is a constant;
[0011] S6: Take Θ, and as the input parameters, calculate the secrecy rate of the current system Then let n = n + 1, and jump to step S2;
[0012] S7: The value that satisfies in step S2 is the maximum secrecy rate of the system under the current conditions of the system, which is used as the output result.
[0013] The present invention overcomes the "double fading" effect in wireless transmission and realizes the maximization of the secrecy rate of the MISO system.
[0014] Preferably, in step S1, the system maximum secrecy rate model is constructed as follows:
[0015] The system model includes a base station equipped with M antennas, an active RIS including N reflection units, a single-antenna legitimate user (Bob), and a single-antenna eavesdropper (Eve). The base station sends the reflected signal to the legitimate user through the active RIS to prevent the eavesdropper from eavesdropping. The channel from the base station to the active RIS is The channels from the base station to the legitimate user and the eavesdropper are respectively and The channels from the active RIS to the legitimate user and the eavesdropper are and wherein, represents an N×M dimensional complex matrix.
[0016] The transmission signal of the base station is: x = w1s1 + w2s2, wherein, and respectively represent the information signal and the noise signal transmitted by the base station, and are the corresponding beamforming and noise vectors, and satisfy E{|s1| 2} = 1, E{|s2| 2} = 1. Wherein, E(·) represents the expectation of a random variable.
[0017] The received signal y B at the legitimate user and the received signal y E at the eavesdropper can be respectively expressed as:
[0018]
[0019] wherein, n k ~CN(0,σ 2 ), k∈(B,E), respectively represent the additive white Gaussian noise with a mean of 0 and a variance of σ 2 at the legitimate user and the eavesdropper, n I ~CN(0,σ I I N ) represents the thermal noise introduced by the active RIS, (·) H represents the conjugate transpose.
[0020] The information rate R B at the legitimate user and the information rate R E at the eavesdropper can be respectively expressed as:
[0021] R k = ln(1 + γ k ), k∈(B,E) (2)
[0022] wherein, |·| and ||·|| respectively represent the modulus of a vector and the norm of a matrix,
[0023]
[0024] Under the conditions of the maximum transmit power constraint of the base station, the amplification power constraint of the active RIS, and the maximum transmit multiple constraint of the active RIS reflection unit, jointly optimize the transmit beamforming vector, the artificial noise beam vector, the active RIS reflection coefficient matrix, and the auxiliary variables k1, k2 to maximize the secrecy rate. The optimization expression formula is:
[0025]
[0026] where P T is the maximum power of the base station, P I is the maximum power of the active RIS, and η i is the maximum amplification factor of the i-th reflecting element. |Θ[i,i]| and arg(Θ[i,i]) represent the amplification factor and the phase shift coefficient at the i-th reflecting element, respectively, and arg(·) represents the argument of a complex number.
[0027] Preferably, in step S3, the beamforming vector and artificial noise vector optimization method is implemented by the following steps:
[0028] S3.1: By using the properties of matrix functions and optimization lemma 1 in the prior art, auxiliary variables k1 and k2 are introduced. For any given active RIS reflection coefficient matrix and auxiliary variables k1 and k2, the transmit beamforming vector and artificial noise beam vector are optimized, and this optimization problem is equivalently transformed into:
[0029]
[0030] where the intermediate variable matrix is defined as:
[0031] S3.2: Applying the semidefinite relaxation method, the problem is simplified to:
[0032]
[0033] where the intermediate variable matrix and values are defined as: W1 = w1*w1 H , W2 = w2*w2 H , W1 is the transmit beamforming matrix, W2 is the artificial noise covariance matrix, and are both positive semidefinite matrices. tr(·) represents the trace of a matrix, and ln(·) represents the natural logarithm function.
[0034] S3.3: The objective function and constraints in equation (6) are both convex functions, which is a convex optimization problem. The existing convex optimization solver CVX can be used for effective solution to obtain If has a rank of 1, then and are recovered by eigenvalue decomposition, otherwise and
[0035] Preferably, in step S4, the active RIS reflection coefficient matrix Θ optimization method is implemented by the following steps:
[0036] S4.1: For any given transmit beamforming vector, artificial noise vector, and auxiliary variables k1 and k2, optimize the active RIS reflection coefficient matrix. This optimization problem is equivalently transformed into:
[0037]
[0038] where the intermediate variable matrices and values are defined: H B = diag(h IB )H AI , H E = diag(h IE )H AI ,
[0039] Θ = diag(q). (·) * denotes the conjugate of a vector, and diag(·) denotes a diagonal matrix.
[0040] S4.2: Apply the semidefinite relaxation method, successive convex approximation, and min-max algorithm to simplify the problem to:
[0041]
[0042] s.t
[0043] tr(H A V) ≤ P I
[0044]
[0045] V[n+1,n+1] = 1, V ≥ 0
[0046] where the intermediate variable matrix is defined:
[0047]
[0048] δ j = diag(H AI w j )diag(H AI w j ) H , j ∈ (1,2) (13)
[0049]
[0050] S4.3: The objective function and constraints in Equation (8) are both convex functions, which is a convex optimization problem. The existing convex optimization solver CVX can be used for effective solution. If the rank of V is 1, then v is recovered by eigenvalue decomposition; otherwise, v is recovered by the existing Gaussian randomization method. Then, from Θ = diag(q) to obtain the optimal active RIS reflection coefficient matrix Θ.
[0051] Preferably, in step S5, for the auxiliary variable optimization and update method, the following steps are adopted:
[0052] S5.1 For any given active RIS reflection coefficient matrix, transmit beamforming vector, and artificial noise beam vector, update the auxiliary variables k1 and k2 by solving:
[0053]
[0054] to update the values of the auxiliary variables k1 and k2.
[0055] Preferably, in step S6, substitute the optimized Θ, k1, and k2 into
[0056] and calculate the secrecy rate value. Meeting the condition in step S2 is the maximum secrecy rate of the system under the current condition.
[0057] The present invention also discloses a secrecy rate maximization system for a MISO system based on active RIS and artificial noise assistance, which is used to execute the above method, including the following modules:
[0058] Secrecy rate calculation module: Calculate the secrecy rate of the system according to the information rates of legitimate users and eavesdroppers where R B and R E are the information rate at the legitimate user and the information rate at the eavesdropper respectively; represents the value of the nth secrecy rate iteration, where n is a constant; the initial value of the maximum secrecy rate ε is the convergence threshold;
[0059] Judgment module: Judge whether satisfies If not, it is executed by the beamforming and artificial noise vector optimization module; if satisfied, it is executed by the system maximum secrecy rate output module;
[0060] Beamforming and artificial noise vector optimization module: Take the obtained by the judgment module as input parameters, execute the beamforming and artificial noise vector optimization method, and output an M×1-dimensional beamforming vector and an artificial noise vector where M is a natural number representing the number of rows of the vector; the superscript n represents the value of the nth iteration;
[0061] Active RIS reflection coefficient matrix optimization module: Take the Taking the active RIS reflection coefficient matrix optimization method as an input parameter, an N×N active RIS reflection coefficient matrix Θ with complex elements is output; where N is a natural number representing the number of rows and columns of the matrix;
[0062] Auxiliary variable optimization and update module: Taking Θ obtained by the active RIS reflection coefficient matrix optimization module as an input parameter, performing the auxiliary variable optimization and update method to update the real numbers and
[0063] Current system secrecy rate calculation module: Taking Θ, and as input parameters to calculate the secrecy rate of the current system Then let n = n + 1, which is executed by the judgment module;
[0064] System maximum secrecy rate output module: Taking the value that satisfies in the judgment module as the system maximum secrecy rate under the current condition, as the output result.
[0065] Some existing technologies related to the present invention are introduced as follows:
[0066] 1. Matrix function properties and optimization lemma 1
[0067] Define an m - order matrix function where N represents any positive semi - definite matrix, then the following 3 facts always hold:
[0068] (1) For any positive semi - definite matrix there is always:
[0069]
[0070] (2) For any positive semi - definite matrix W, there is always:
[0071]
[0072] (3)
[0073] 2. Positive semi - definite relaxation method
[0074] Semidefinite Relaxation (SDR) is a technique widely used in optimization problems, especially non-convex optimization problems. In many practical optimization scenarios, such as combinatorial optimization, signal processing and other fields, the original problem is often non-convex, which makes it very difficult to find the global optimal solution. The semidefinite relaxation method solves the problem by relaxing the original non-convex problem into a convex semidefinite programming (SDP) problem. Semidefinite programming is a special convex optimization problem, where the variables are a semidefinite matrix, and there are linear objective functions and linear constraint conditions.
[0075] 3. Successive Convex Approximation Method
[0076] Successive Convex Approximation (SCA) is an iterative optimization method for solving non-convex optimization problems. Its core idea is that in each iteration, the non-convex objective function and / or constraint conditions are approximated by convex functions, thus transforming the original non-convex problem into a series of relatively easy-to-solve convex problems.
[0077] 4. Convex Optimization Solver CVX
[0078] Among the numerous tools and methods for solving convex optimization problems, the convex optimization solver CVX is a widely used and recognized tool. Users only need to input the objective function and constraint conditions in an easy-to-understand form into CVX according to specific syntax rules, and CVX can automatically transform these problems into the standard convex optimization form and call efficient underlying solving algorithms to solve them. CVX supports various types of convex optimization problems, including linear programming, quadratic programming, semidefinite programming, etc., and can handle complex problems with a large number of variables and constraint conditions.
[0079] 5. Minimize-Maximize (MM) Algorithm
[0080] The Minimize-Maximize (MM) algorithm is an iterative method for dealing with non-convex optimization problems. When faced with complex non-convex objective functions or constraint conditions, the MM algorithm gradually finds the optimal solution of the problem through ingenious approximation and iterative processes. In each iteration, the MM algorithm constructs a surrogate function to approximate the original objective function or constraint conditions. For the objective function, approximation is usually based on mathematical tools such as Taylor expansion.
[0081] 6. Gaussian Randomization Method
[0082] In convex optimization problems, when it comes to using the results of eigenvalue decomposition to solve certain problems (such as matrix rank constraint problems), direct solution may be difficult because the rank-1 constraint problem is usually non-convex. The Gaussian randomization method provides an effective means of approximate solution. The core idea of the Gaussian randomization method is to construct an approximation of the optimal solution by generating Gaussian random vectors.
[0083] In summary, the present invention can greatly improve the secrecy rate of the system by using active RIS and artificial noise assistance, thereby enhancing the security of the system. BRIEF DESCRIPTION OF THE DRAWINGS
[0084] Figure 1 It is a model diagram related to a method for maximizing the secrecy rate of a MISO system based on active RIS and artificial noise assistance in a preferred embodiment of the present invention;
[0085] Figure 2 It is a flowchart of a method for maximizing the secrecy rate of a MISO system based on active RIS and artificial noise assistance in a preferred embodiment of the present invention;
[0086] Figure 3 It is a convergence performance diagram of the algorithm under different numbers of base station antennas and numbers of active RIS reflection units in a preferred embodiment of the present invention;
[0087] Figure 4 It is a comparison diagram of secrecy performance when the number of active RIS reflection units increases under different schemes;
[0088] Figure 5 It is a system block diagram of a method for maximizing the secrecy rate of a MISO system based on active RIS and artificial noise assistance in a preferred embodiment of the present invention. DETAILED DESCRIPTION OF THE INVENTION
[0089] The present invention will be further described below in conjunction with specific embodiments. The specific implementation manner of the present invention can be detailedly illustrated by the following embodiment diagrams.
[0090] Figure 1 It is a model diagram related to a method for maximizing the secrecy rate of a MISO system based on active RIS and artificial noise assistance in a preferred embodiment of the present invention. The system includes a base station equipped with M antennas, an active RIS including N reflection units, a single-antenna legitimate user (Bob), and a single-antenna eavesdropper (Eve).
[0091] Figure 2 It is a flowchart of a method for maximizing the secrecy rate of a MISO system based on active RIS and artificial noise assistance in a preferred embodiment of the present invention, which is mainly completed through the following steps:
[0092] Step 1: Given an initial feasible solution Θ, set Initialize the number of iterations as \(n = 1\). Calculate the system secrecy rate according to the information rates of legitimate users and eavesdroppers.
[0093] Step 2: Judge Whether it holds, where \(\epsilon\) is the convergence threshold, and in this embodiment, its value is 10 -3 , where, Refers to the secrecy rate obtained in the first calculation. If it is satisfied, jump to step S7 to output the result; otherwise, execute step S3; here, Can characterize the convergence performance of the method;
[0094] Step 3: Take the Obtained in step S2 as the input parameter. Introduce auxiliary variables \(k_1\) and \(k_2\) through the properties of matrix functions and optimization lemma 1, and set the auxiliary variables \(k_1 = 1\), \(k_2 = 1\). Execute the beamforming and artificial noise vector optimization method to obtain
[0095] Step 4: Take the Obtained in step S3 as the input parameter, and execute the active RIS reflection coefficient matrix optimization method to obtain \(\Theta\);
[0096] Step 5: Take \(\Theta\) obtained in step S4 as the input parameter, execute the auxiliary variable optimization and update method to update and
[0097] Step 6: Calculate \(\Theta\), and Calculate And let \(n=n + 1\), then jump to step 2;
[0098] Step 7: Output the maximum secrecy rate of the system
[0099] In step S1 of this embodiment, the base station sends the reflected signal to the legitimate user through the active RIS to prevent eavesdropping by eavesdroppers. The channel from the base station to the active RIS is The channels from the base station to the legitimate user and the eavesdropper are respectively and The channels from the active RIS to the legitimate user and the eavesdropper are and where, Represents an \(N\times M\) - dimensional complex matrix.
[0100] The transmission signal of the base station is: \(x = w_1s_1+w_2s_2\), where, and Represent the information signal and the noise signal sent by the base station respectively, and are the corresponding beamforming and noise vectors, and satisfy E{|s1| 2} = 1, E{|s2| 2} = 1. Among them, E(·) represents the expectation of a random variable.
[0101] The received signal y of the legitimate user B and the received signal y at the eavesdropper E can be respectively expressed as:
[0102]
[0103] where, n k ~CN(0,σ 2 ), k∈(B,E), respectively represent the additive white Gaussian noise with a mean of 0 and a variance of σ 2 at the legitimate user and the eavesdropper, n I ~CN(0,σ I I N ) represents the thermal noise introduced by the active RIS, (·) H represents the conjugate transpose.
[0104] The information rate R of the legitimate user B and the information rate R at the eavesdropper E can be respectively expressed as:
[0105] R k = ln(1 + γ k ), k∈(B,E) (2)
[0106] where, |·| and ||·|| respectively represent the norm of a vector and the norm of a matrix,
[0107]
[0108] Under the conditions of the maximum transmit power constraint of the base station, the amplification power constraint of the active RIS, and the maximum transmit multiple constraint of the active RIS reflection unit, jointly optimize the transmit beamforming vector, the artificial noise beam vector, the active RIS reflection coefficient matrix, and the auxiliary variables k1, k2 to maximize the secrecy rate. The optimization expression formula is:
[0109]
[0110] where, P T is the maximum power of the base station, P I is the maximum power of the active RIS, η iis the maximum amplification factor of the i-th reflection unit. |Θ[i,i]| and arg(Θ[i,i]) represent the amplification factor and the phase shift coefficient at the i-th reflection element respectively, and arg(·) represents the argument of a complex number.
[0111] In step S3, the beamforming vector and artificial noise vector optimization method is implemented as follows:
[0112] S3.1: By using the properties of the existing technology matrix function and the optimization lemma 1, introduce auxiliary variables k1 and k2. For any given active RIS reflection coefficient matrix and auxiliary variables k1 and k2, optimize the transmit beamforming vector and the artificial noise beam vector. This optimization problem is equivalently transformed into:
[0113]
[0114] Among them, define the intermediate variable matrix:
[0115] S3.2: Apply the semidefinite relaxation method to simplify the problem to:
[0116]
[0117] Among them, define the intermediate variable matrix and value: W1 = w1*w1 H , W2 = w2*w2 H , W1 is the transmit beamforming matrix, W2 is the artificial noise covariance matrix, and are both positive semidefinite matrices. tr(·) represents the trace of a matrix, and ln(·) represents the natural logarithm function.
[0118] S3.3: The objective function and the constraint conditions in equation (6) are both convex functions, which is a convex optimization problem. The existing technology convex optimization solver CVX can be used for effective solution to obtain If has a rank of 1, then and are recovered by eigenvalue decomposition and
[0119] In step S4, the active RIS reflection coefficient matrix Θ optimization method is implemented as follows:
[0120] S4.1: For any given transmit beamforming vector, artificial noise vector, and auxiliary variables k1 and k2, optimize the active RIS reflection coefficient matrix. This optimization problem is equivalently transformed into:
[0121]
[0122] Among them, define the intermediate variable matrix and value: H B = diag(h IB )H AI , H E = diag(h IE )H AI ,
[0123] Θ = diag(q). (·) * denotes the conjugate of a vector, and diag(·) denotes a diagonal matrix.
[0124] S4.2: Apply the semidefinite relaxation method, continuous convex approximation, and min-max algorithm to simplify the problem to:
[0125]
[0126] s.t
[0127] tr(H A V) ≤ P I
[0128]
[0129] V[n + 1, n + 1] = 1, V ≥ 0
[0130] Among them, define the intermediate variable matrix:
[0131]
[0132] δ j = diag(H AI w j )diag(H AI w j ) H , j ∈ (1, 2) (13)
[0133]
[0134] S4.3: The objective function and constraint conditions of Equation (8) are both convex functions, which is a convex optimization problem. The existing convex optimization solver CVX can be used for effective solution. If the rank of V is 1, then v is recovered by eigenvalue decomposition, otherwise v is recovered by the existing Gaussian randomization method. Then, from Θ = diag(q), the optimal active RIS reflection coefficient matrix Θ is obtained.
[0135] In step S5, for the auxiliary variable optimization update method, the following steps are adopted:
[0136] S5.1 Update the auxiliary variables k1 and k2 for any given active RIS reflection coefficient matrix, transmit beamforming vector, and artificial noise beam vector by solving:
[0137]
[0138] to update the values of the auxiliary variables k1 and k2.
[0139] In step S6, substitute the optimized Θ, k1, and k2 and calculate the secrecy rate value. Meeting the conditions in step S2 is the maximum secrecy rate of the system under the current conditions.
[0140] Figure 3 This is the convergence performance graph of the algorithm of this scheme under different numbers of base station antennas and active RIS reflection units. Among them, the number of base station antennas and active RIS reflection units are divided into three groups, namely M = 6, N = 12; M = 8, N = 16; M = 10, N = 20. The base station transmit power is 30 dBm, and the maximum power of the active RIS is 30 dBm. It can be seen from the figure that the method proposed in this invention has a fast convergence speed and basically reaches convergence after five iterations, verifying the effectiveness and convergence of this invention. And the maximum secrecy rate is proportional to the number of transmit antennas of the base station and the number of active RIS reflection units. (The curves in the figure are all the convergence graphs of this invention under different conditions)
[0141] Figure 4 This is the comparison graph of the secrecy performance under different schemes with the increase of the number of active RIS reflection units. The secrecy rate of the system increases with the increase of the number of active RIS reflection units. Compared with the method of this invention (active RIS + AN), the secrecy rate of deploying passive RIS is much lower, which indicates that active RIS and artificial noise assistance can greatly improve the secrecy rate of the system, thus enhancing the security of the system.
[0142] As Figure 5 shown, this embodiment discloses a secrecy rate maximization system for a MISO system based on active RIS and artificial noise assistance for performing the above method, including the following modules:
[0143] Secrecy rate calculation module: Calculate the secrecy rate of the system according to the information rates of legitimate users and eavesdroppers where, R B and R E are the information rate at the legitimate user and the information rate at the eavesdropper respectively; represents the value of the nth secrecy rate iteration, n is a constant; the initial value of the maximum secrecy rate ε is the convergence threshold;
[0144] Judgment module: Judge Whether it is satisfied If not, it is executed by the beamforming and artificial noise vector optimization module; if satisfied, it is executed by the system maximum secrecy rate output module;
[0145] Beamforming and artificial noise vector optimization module: Take the obtained by the judgment module as input parameters, execute the beamforming and artificial noise vector optimization method, and output an M×1 dimensional beamforming vector with all elements being complex numbers and an artificial noise vector where M is a natural number representing the number of rows of the vector; the superscript n represents the value of the corresponding nth iteration;
[0146] Active RIS reflection coefficient matrix optimization module: Take the obtained by the beamforming and artificial noise vector optimization module as input parameters, execute the active RIS reflection coefficient matrix optimization method, and output an N×N dimensional active RIS reflection coefficient matrix Θ with elements being complex numbers; where N is a natural number representing the number of rows and columns of the matrix;
[0147] Auxiliary variable optimization and update module: Take Θ obtained by the active RIS reflection coefficient matrix optimization module as input parameters, execute the auxiliary variable optimization and update method, and update the real numbers and
[0148] Current system secrecy rate calculation module: Take Θ, and as input parameters, and calculate the secrecy rate of the current system Then let n=n + 1, and it is executed by the judgment module;
[0149] System maximum secrecy rate output module: Take the value that satisfies in the judgment module as the system maximum secrecy rate under the current conditions, and use it as the output result.
[0150] For other contents of this embodiment, reference can be made to the above method embodiment.
[0151] In summary, the present invention provides a method for overcoming the "double fading" (referring to the fact that wireless communication signals are affected by large-scale fading and small-scale fading simultaneously) effect in the link, and improves the secrecy rate of the system.
[0152] The above has clearly described the embodiments of the present invention. However, for those skilled in the art, without departing from the principle and spirit of the method of the present invention, various changes, modifications, substitutions, and variations can be made to these embodiments. Then the scope of the present invention is defined by the appended claims and their equivalents, and still belongs to the scope of the method of the present invention and is still regarded as the protection scope of the present invention.
Claims
1. A method for maximizing the confidentiality rate of a MISO system based on active RIS and artificial noise assistance, characterized in that: The following steps are involved: S1. Calculate the confidentiality rate of the system based on the information rate of the legitimate user and the eavesdropper Among them, R B and R E are the information rate at the legitimate user and the information rate at the eavesdropper respectively; Indicates the value of the corresponding nth confidentiality rate iteration, where n is a constant; the initial value of the maximum confidentiality rate ε is the convergence threshold; S2. Judgment Is it satisfied? If not, execute step S3; if satisfied, jump to step S7; S3, the step S2 obtained As input parameters, the beamforming and artificial noise vector optimization methods are performed, and the output elements are all complex M×1-dimensional beamforming vectors and the artificial noise vector Where M is a natural number, indicating the number of rows of the vector; the superscript n indicates the value of the corresponding nth iteration; S4, the step S3 obtained As an input parameter, an active RIS reflection coefficient matrix optimization method is executed, and an active RIS reflection coefficient matrix Θ of N×N whose elements are complex numbers is output; wherein N is a natural number, indicating the number of rows and columns of the matrix; S5, using the Θ obtained in step S4 as an input parameter, executing the auxiliary variable optimization update method, and updating the real number and S6. Θ、 and As input parameter, calculate the current system's confidentiality rate Then let n=n+1 and jump to step S2; S7: Step S2 satisfies The value of is the maximum confidentiality rate of the system under the current conditions. As the output result.
2. The method for maximizing the confidentiality rate of a MISO system based on active RIS and artificial noise assistance as claimed in claim 1, characterized in that: In step S1, the system includes a base station equipped with M antennas, an active RIS with N reflection units, a single-antenna legitimate user, and a single-antenna eavesdropper. The base station sends the reflected signal to the legitimate user through the active RIS. The channel from the base station to the active RIS is The channels from the base station to the legitimate user and the eavesdropper are and The channel from active RIS to legitimate users and eavesdroppers is and in, represents a complex matrix of N×M dimensions; The base station transmits a signal: x = w1s1 + w2s2, where and Respectively represent the information signal and noise signal sent by the base station, and are the corresponding beamforming and noise vectors, and satisfy E{|s1| 2 }=1,E{|s2| 2 }=1; where E(·) represents the expectation of the random variable; Legitimate user receives signal y B and the eavesdropper receives the signal y E Respectively expressed as: Among them, n k ~CN(0,σ 2 ), k∈(B,E), which respectively represent the legitimate user and the eavesdropper with a mean of 0 and a variance of σ 2 Additive Gaussian white noise, n I ~CN(0,σ I I N ) represents the thermal noise introduced by active RIS, (·) H represents conjugate transpose; Legal user information rate R B and the eavesdropper's information rate R E Respectively expressed as: R k ln(1+γ k ),k∈(B,E) (2) Among them, |·| and ||·|| represent the vector modulus and matrix norm respectively. Under the constraints of the maximum transmission power of the base station, the active RIS amplification power and the maximum transmission multiple of the active RIS reflection unit, the transmit beamforming vector, the artificial noise beam vector, the active RIS reflection coefficient matrix and the auxiliary variables k1 and k2 are jointly optimized to maximize the confidentiality rate. The optimization formula is: Among them, P T is the maximum power of the base station, P I is the maximum power of active RIS, η i is the maximum amplification factor of the i-th reflection unit; |Θ[i,i]| and arg(Θ[i,i]) represent the amplification factor and phase shift coefficient at the i-th reflection element, respectively, and arg(·) represents the complex argument.
3. The method for maximizing the confidentiality rate of a MISO system based on active RIS and artificial noise assistance as claimed in claim 2, characterized in that: In step S3, the beamforming and artificial noise vector optimization method is implemented by the following steps: S3.
1. Auxiliary variables k1 and k2 are introduced through matrix function properties and optimization Lemma 1. For any given active RIS reflection coefficient matrix and auxiliary variables k1 and k2, the transmit beamforming and artificial noise beam vector are optimized. The optimization problem is equivalently converted to: st |w1 2 ||+||w2 2 ||≤P T Among them, the intermediate variable matrix is defined as: S3.2, using the semidefinite relaxation method, the problem is simplified to: Among them, define the intermediate variable matrix and value: W1=w1w1 H , W2=w2w2 H , W1 is the transmit beamforming matrix, W2 is the artificial noise covariance matrix, W1≥0 and W2≥0 are both semi-positive definite matrices; tr(·) represents the trace of the matrix, and ln(·) represents the natural logarithm function; S3.3, Equation (6) The objective function and constraint conditions are both convex functions, and the solution is W1 * , If W1 * , The rank of is 1, then it is recovered by eigenvalue decomposition and Otherwise, recover by Gaussian randomization method and 4. The method for maximizing the confidentiality rate of a MISO system based on active RIS and artificial noise assistance as claimed in claim 3, characterized in that: In step S4, the active RIS reflection coefficient matrix θ optimization method is implemented by the following steps: S4.
1. For any given transmit beamforming vector, artificial noise vector and auxiliary variables k1 and k2, the active RIS reflection coefficient matrix is optimized. The optimization problem is equivalently converted to: Among them, define the intermediate variable matrix and value: H B =diag(h IB )H AI , H E =diag(h IE )H AI , Θ=diag(q); (·) * represents the conjugate of a vector, and diag(·) represents a diagonal matrix; S4.
2. Apply semi-positive relaxation, continuous convex approximation and minimization-maximization algorithm to simplify the problem to: Among them, the intermediate variable matrix is defined as: δ j =diag(H AI w j )diag(H AI w j ) H ,j∈(1,2) (13) S4.3, Equation (8) The objective function and the constraint conditions are both convex functions. If the rank of V is 1, then v is restored by eigenvalue decomposition, otherwise v is restored by Gaussian randomization method; then Θ=diag(q) obtains the optimal active RIS reflection coefficient matrix Θ.
5. The method for maximizing the confidentiality rate of a MISO system based on active RIS and artificial noise assistance as claimed in claim 4, characterized in that: In step S5, the auxiliary variable optimization update method is implemented by the following steps: For any given active RIS reflection coefficient matrix, transmit beamforming vector and artificial noise beam vector, the auxiliary variables k1 and k2 are updated by solving: To update the values of auxiliary variables k1 and k2.
6. The method for maximizing the confidentiality rate of a MISO system based on active RIS and artificial noise assistance as claimed in claim 5, characterized in that: In step S6, the optimized Θ, k1 and k2 are introduced And calculate the confidentiality rate value 7. A MISO system confidentiality rate maximization system based on active RIS and artificial noise assistance, used to execute the method according to any one of claims 1 to 6, characterized in that: Includes the following modules: Confidentiality rate calculation module: Calculates the confidentiality rate of the system based on the information rate of legitimate users and eavesdroppers Among them, R B and R E are the information rate at the legitimate user and the information rate at the eavesdropper respectively; Indicates the value of the corresponding nth confidentiality rate iteration, where n is a constant; the initial value of the maximum confidentiality rate ε is the convergence threshold; Judgment module: Judgment Is it satisfied? If not, the beamforming and artificial noise vector optimization module will execute; if satisfied, the system maximum confidentiality rate output module will execute; Beamforming and artificial noise vector optimization module: The As input parameters, the beamforming and artificial noise vector optimization methods are performed, and the output elements are all complex M×1-dimensional beamforming vectors and the artificial noise vector Where M is a natural number, indicating the number of rows of the vector; the superscript n indicates the value of the corresponding nth iteration; Active RIS reflection coefficient matrix optimization module: The beamforming and artificial noise vector optimization modules obtain As an input parameter, an active RIS reflection coefficient matrix optimization method is executed, and an active RIS reflection coefficient matrix Θ of N×N whose elements are complex numbers is output; wherein N is a natural number, indicating the number of rows and columns of the matrix; Auxiliary variable optimization update module: takes the Θ obtained by the active RIS reflection coefficient matrix optimization module as the input parameter, executes the auxiliary variable optimization update method, and updates the real number and Current system confidentiality rate calculation module: The beamforming and artificial noise vector optimization module obtains The active RIS reflection coefficient matrix optimization module obtains Θ, the auxiliary variable optimization update module obtains and As input parameter, calculate the current system's confidentiality rate Then let n=n+1, and the judgment module executes; System maximum confidentiality rate output module: the module that satisfies The value of is taken as the maximum confidentiality rate of the system under the current conditions. As the output result.
