Method for optimizing secure transmission of active STAR-RIS assisted RSMA communication system

Abstract

The invention discloses an optimization method for secure transmission of an active STAR-RIS assisted RSMA communication system. The optimization method mainly comprises the following steps: S1, constructing an active STAR-RIS assisted multi-input single-output RSMA communication system; s2, the base station transmits the information to legal users in the system through the active STAR-RIS; s3, the base station sends the artificial interference noise while sending the useful signal; constructing an optimization equation for maximizing system private information and safety rate based on beam forming, artificial noise, an active STAR-RIS power amplification matrix, an amplitude matrix and a phase shift matrix as optimization parameters; s4, performing proper conversion on a non-convex objective function and constraints by using a WMMSE criterion, and performing decoupling solution on an optimization problem by using an AO algorithm; s5, updating an auxiliary variable introduced by the WMMSE algorithm according to the closed expression; s6, using an SCA algorithm to optimize and update beamforming and artificial noise of the base station, and a power amplification matrix and an amplitude matrix of the active STAR-RIS; and S7, adopting an SCA algorithm and a PDD algorithm to optimize and update the phase shift matrix of the active STAR-RIS. According to the invention, the security of the communication system is obviously improved.

Description

Technical Field

[0001] The invention belongs to the technical field of wireless communications, and in particular relates to an optimization method for secure transmission of an active STAR-RIS-assisted RSMA communication system. Background Art

[0002] As 5G communication networks enter the commercialization stage, more and more research efforts are devoted to the upcoming Beyond 5G (B5G) and 6th Generation (6G) wireless communication networks to obtain faster and more reliable data transmission. Reconfigurable smart surfaces (RIS, also known as intelligent reflective surfaces IRS) can utilize a large number of passive reflective elements integrated on a plane to intelligently configure the wireless propagation environment through software programming to improve the performance of wireless communication systems. Compared with traditional relay systems, they have the characteristics of low cost, low power consumption, and easy deployment. Therefore, they have attracted great attention from the industry and are regarded as a potential solution for the next generation of communication systems.

[0003] At the same time, with the deepening of the study of RIS, some of the latest studies have revealed some inherent defects of RIS, mainly two points: 1. Traditional RIS can only serve mobile users in half space. 2. The signal reflected by RIS is severely attenuated due to the multiplicative fading of the cascade channel. In order to solve these two defects of the traditional RIS-assisted communication system, the present invention proposes an active STAR-RIS-assisted communication system. In addition, the issue of communication information security has also been highly valued by people. To this end, in combination with specific communication scenarios and security requirements, the present invention proposes an active STAR-RIS-assisted RSMA secure transmission communication system, which achieves maximum system and security rate by jointly designing the beamforming, artificial noise and related parameters of active STAR-RIS at the base station end. Since the problem is non-convex and there is serious variable coupling, the present invention decomposes the original non-convex optimization problem into several non-convex optimization sub-problems by using the alternating optimization (AO) method, and then uses the continuous convex approximation (SCA) and penalized dual decomposition (PDD) method to develop an effective optimization method to solve the problem. Summary of the invention

[0004] In order to solve the shortcomings and defects of the prior art, the present invention proposes an optimization method for secure transmission of an active STAR-RIS-assisted RSMA communication system. The present invention introduces active STAR-RIS into a multi-user wireless system, and by jointly optimizing the beamforming, artificial noise, active STAR-RIS power amplification matrix, amplitude matrix and phase shift matrix of the base station, the received signal of the eavesdropper is weakened and the signal of the legitimate user is enhanced, thereby maximizing the system and security rate (SSR, SystemSecrecy Rate).

[0005] An optimization method for secure transmission of an active STAR-RIS-assisted RSMA communication system, the method comprising the following steps:

[0006] S1: Construct an active STAR-RIS assisted RSMA communication system, which consists of a base station with M antennas, two single-antenna users (one located in the reflection area and the other in the transmission area), a single-antenna eavesdropper (located in the reflection area but can eavesdrop on the information of two legitimate users) and an active STAR-RIS, where the active STAR-RIS contains N units;

[0007] S2: The communication scenario is to send public and private information from the base station to both the reflection user and the transmission user. Due to obstacles or long distances, the base station transmits the information to the legitimate users in the system with the assistance of active STAR-RIS. In addition, the public information is transmitted in the form of broadcast, that is, both the eavesdropper and the legitimate user can receive the public information;

[0008] S3: In order to improve the security of the entire communication system, the base station sends artificial noise while sending useful signals; in the scenario where the total transmission energy of the base station and active STAR-RIS is limited, an optimization equation for maximizing the system private information and security rate is constructed based on beamforming, artificial noise, active STAR-RIS power amplification matrix, amplitude matrix and phase shift matrix as optimization parameters;

[0009] S4: First, some slack variables are introduced into the objective function to simplify the optimization objective function. Then, the three important facts contained in the WMMSE criterion are used to appropriately transform the original non-convex problem. Then, based on the alternating optimization algorithm, the auxiliary variables introduced by the WMMSE criterion, the beamforming of the base station, the artificial noise and the active STAR-RIS related parameters are decoupled and optimized.

[0010] S5: Given the beamforming, artificial noise, active STAR-RIS power amplification matrix, amplitude matrix and phase shift matrix parameters, update the auxiliary variables introduced by the WMMSE algorithm according to the closed-form expression;

[0011] S6: Given the auxiliary variable values ​​introduced by the WMMSE algorithm, the power amplification matrix, amplitude matrix and phase shift matrix parameters of the active STAR-RIS, optimize and update the beamforming and artificial noise of the base station: use the SCA algorithm for solution.

[0012] S7: Given the auxiliary variable values ​​introduced by the WMMSE algorithm, the beamforming of the base station, artificial noise, the amplitude matrix of the active STAR-RIS, and the phase shift matrix parameters, optimize and update the power amplification matrix of the active STAR-RIS: use the SCA algorithm to solve.

[0013] S8: Given the auxiliary variable values ​​introduced by the WMMSE algorithm, the beamforming of the base station, artificial noise, the power amplification matrix of the active STAR-RIS, and the phase shift matrix parameters, optimize and update the amplitude matrix of the active STAR-RIS: use the SCA algorithm to solve.

[0014] S9: Given the auxiliary variable values ​​introduced by the WMMSE algorithm, the beamforming of the base station, artificial noise, the power amplification matrix of the active STAR-RIS, and the amplitude matrix parameters, optimize and update the phase shift matrix of the active STAR-RIS: use the SCA algorithm and the PDD algorithm for optimization and solution.

[0015] S10: Repeat steps S5-S9 until the algorithm converges and the optimal solution to the optimization problem is obtained.

[0016] The received signals of the legitimate user and the eavesdropper in step S2 are respectively:

[0017]

[0018] Among them, h d,j ∈￡ M×1 、h s,j ∈￡ N×1 , G∈￡ N×M 、h d,e ∈￡ M×1 、h s,e ∈￡ N×1 They represent the channel from the base station to the legitimate user j, the channel from the active STAR-RIS to the legitimate user j, the channel from the base station to the active STAR-RIS, the channel from the base station to the eavesdropper, and the channel from the active STAR-RIS to the eavesdropper. A@Diag(a), is the amplification factor matrix of active STAR-RIS, E r @Diag(ζ),ζ@[z1,z2,...,z N ]and is the reflection and transmission amplitude coefficient matrix of active STAR-RIS, Φ r @Diag(f r ), and Φ t @Diag(f t ), is the reflection-transmission phase shift coefficient matrix of Active STAR-RIS. v is the thermal noise of Active STAR-RIS, and v is: The additive Gaussian white noise at the legitimate user j is n j : The additive white Gaussian noise at Eve is n e The signal sent by the base station is represented by x=w c s c +∑ k∈κ w k s k +z, where K@{r,t}, s c :CN(0,1) and s k :CN(0,1) represents the public stream of all LUs and the private stream of only LU k, respectively, and is represented by the precoding vector w c ∈C M×1 and w k ∈C M×1 In addition, z:CN(0,Z) is an artificial noise vector, where Z∈H M , Z f 0 is the covariance matrix of the artificial noise vector.

[0019] The SINR of the public information decoded at j is:

[0020]

[0021] Among them, j∈KU{e},

[0022] The SINR of the private information decoded by the legitimate user k is:

[0023]

[0024] Among them, j∈KU{e},

[0025] Decoding at Eve k The SINR is:

[0026]

[0027] According to Shannon's formula, we have the public information rate at the legitimate user j as:

[0028] R c,j =log2(1+g c,j ) (6)

[0029] The private information rate at the legitimate user k is:

[0030] R p,k =log2(1+g p,k ) (7)

[0031] The rate at which an eavesdropper eavesdrops on private information k is:

[0032] R p,e→k =log2(1+g p,e→k ) (8)

[0033] The achievable confidentiality rate for user k is:

[0034]

[0035] where k∈K, [x] + =max{0,x}.

[0036] In order to improve the security of the entire communication system in step S3, the following optimization method is adopted:

[0037] (P1):

[0038] stC1:R c,k ≥r c,k

[0039] C2:

[0040] C3:

[0041] C4:

[0042] C5:

[0043] in, K∈{r,t}, is the maximum transmit power of the base station, is the maximum transmit power of the active STAR-RIS.

[0044] In step S4, the objective function is transformed and then the optimization problem is decoupled based on the alternating optimization algorithm:

[0045] First, we introduce auxiliary variables and use the WEESE criterion to transform the original non-convex problem into:

[0046] (P2):

[0047] stC1:

[0048] C2:R c,k ≥r c,k

[0049] C3:

[0050] C4:

[0051] C5:

[0052] C6:

[0053] in,

[0054]

[0055] At this time, the optimization variables of the optimization problem (P2) are coupled with each other, which is extremely difficult to solve. Therefore, the alternating optimization algorithm is used to decouple the problem (P2) into multiple sub-problems for alternating optimization:

[0056] (1) Sub-problem 1: Optimizing the auxiliary variables introduced by the WMMSE criterion:

[0057]

[0058] (2) Sub-problem 2: Optimizing beamforming and artificial noise of base stations:

[0059]

[0060] stC1:

[0061] C2:

[0062] C3:

[0063] C4:

[0064] (3) Sub-problem 3: Optimize the amplification factor matrix of active STAR-RIS:

[0065]

[0066] stC1:

[0067] C2:

[0068] C3:

[0069] (4) Sub-problem 4: Optimize the amplitude coefficient matrix of active STAR-RIS:

[0070]

[0071] stC1:

[0072] C2:

[0073] C3:

[0074] (5) Sub-problem 5: Optimize the phase shift matrix of active STAR-RIS:

[0075]

[0076] stC1:

[0077] C2:

[0078] C3:

[0079] In step S5, given the base station beamforming, artificial noise, active STAR-RIS power amplification matrix, amplitude matrix and phase shift matrix parameters, the auxiliary variables introduced by the WMMSE algorithm are updated according to the closed-form expression:

[0080]

[0081] In step S6, given the auxiliary variable values ​​introduced by the WMMSE algorithm, the power amplification matrix, amplitude matrix and phase shift matrix parameters of the active STAR-RIS, the beamforming and artificial noise of the base station are optimized and updated, and SCA is used to convert it into a convex optimization problem:

[0082]

[0083] stC1:

[0084] C2:

[0085] C3:

[0086] C4:

[0087] in,

[0088]

[0089]

[0090] In S7, given the auxiliary variable values ​​introduced by the WMMSE algorithm, the beamforming of the base station, artificial noise, the amplitude matrix and phase shift matrix parameters of the active STAR-RIS, the power amplification matrix of the active STAR-RIS is optimized and updated, and SCA is used to convert it into a convex optimization problem:

[0091]

[0092] in,

[0093]

[0094] In S8, given the auxiliary variable values ​​introduced by the WMMSE algorithm, the beamforming of the base station, artificial noise, the power amplification matrix of the active STAR-RIS, and the phase shift matrix parameters, the amplitude matrix of the active STAR-RIS is optimized and updated, and SCA is used to convert it into a convex optimization problem:

[0095]

[0096]

[0097] in,

[0098] is the solution of the last iteration, b p,t,n , c p,t,n , b c,t,n , c c,t,n Is a positive number

[0099]

[0100] In the S9, given the auxiliary variable values ​​introduced by the WMMSE algorithm, the beamforming of the base station, artificial noise, the power amplification matrix of the active STAR-RIS, and the amplitude matrix parameters, the phase shift matrix of the active STAR-RIS is optimized and updated, and the SCA algorithm and the PDD algorithm are used for optimization and solution:

[0101] PDD algorithm inner layer:

[0102] Given ψ r ,ψ t , the original problem becomes:

[0103]

[0104] Among them, ψ r ,ψ t is the auxiliary variable introduced, is φ k The solution of the last iteration

[0105]

[0106] This problem is a convex optimization problem and can be solved using cvx.

[0107] Given φ r ,φ t , solve the introduced auxiliary variable ψ r ,ψ t Values:

[0108]

[0109] in, is r ,ψ t The optimal solution of .

[0110] PDD outer layer algorithm:

[0111] 1. If the equation φ r =ψ r ,φ t =ψ t Approximately, the dual variable λ r ,λ t It will be updated in a gradient ascent manner, that is:

[0112] λ k =λ k +ρ -1 (φ k -ψ k ),k∈K

[0113] 2. If the equation φ r =ψ r ,φ t =ψ t Far from being "true", that is, very different, then:

[0114] The outer layer will choose to increase the penalty parameter ρ -1 , which will force the equation φ r =ψ r ,φ t =ψ t In the subsequent iteration process, it is better satisfied, namely:

[0115] Wherein, c is a predetermined constant.

[0116] In S10, alternating optimization iterations are used until the algorithm converges to obtain an optimal solution to the optimization problem.

[0117] Compared with the prior art, the present invention has the following beneficial technical effects:

[0118] 1. The present invention provides an active STAR-RIS auxiliary communication system model, which solves the problem that the legal users of the traditional RIS communication system can only move in half space and the multiplicative fading problem of the cascade channel.

[0119] 2. The present invention provides an active STAR-RIS assisted RSMA communication system security transmission solution, which solves the information security problem of the communication system.

[0120] 3. The present invention provides an optimization method for secure transmission of an active STAR-RIS-assisted RSMA communication system. Through algorithm optimization, a higher system and security rate is obtained. BRIEF DESCRIPTION OF THE DRAWINGS

[0121] Figure 1 It is a system model diagram of the present invention;

[0122] Figure 2 It is the algorithm flow chart of the present invention;

[0123] Figure 3 It is a convergence comparison diagram of the algorithm of the present invention under different numbers of units;

[0124] Figure 4 It is a convergence comparison diagram of the algorithm of the present invention under different STAR-RIS; DETAILED DESCRIPTION

[0125] Now the specific implementation of the present invention is further described in detail with reference to the accompanying drawings:

[0126] like Figure 1 As shown, the active STAR-RIS assisted RSMA communication system includes: a base station with M antennas, 2 single-antenna users (one in the reflection area and the other in the transmission area), a single-antenna eavesdropper (located in the reflection area but can eavesdrop on the information of two legitimate users) and an active STAR-RIS, where the active STAR-RIS contains N units. Downlink transmission is adopted. Let the channel from the base station to legitimate user j, the channel from active STAR-RIS to legitimate user j, the channel from the base station to active STAR-RIS, the channel from the base station to the eavesdropper and the channel from active STAR-RIS to the eavesdropper be: h d,j ∈￡ M×1 、h s,j ∈￡ N×1 , G∈￡ N×M 、h d,e ∈￡ M×1 、h s,e ∈￡ N×1 , the channel from the base station to the active STAR-RIS obeys the Rice fading model, and the other channels obey the Rayleigh fading model. Assume that the bandwidth of each channel in the system is unit bandwidth.

[0127] The amplification factor matrix of active STAR-RIS is: A@Diag(a), The reflection-transmission amplitude coefficient matrix of active STAR-RIS is: r @Diag(ζ),ζ@[z1,z2,...,z N ]and The reflection-transmission phase shift coefficient matrix of active STAR-RIS is: Φ r @Diag(f r ), and Φ t @Diag(f t ), The signal sent by the base station is represented by x=w c s c +∑ k∈κ w k s k +z, where K@{r,t}, s c :CN(0,1) and s k :CN(0,1) represents the public stream of all LUs and the private stream of only LU k, respectively, and is represented by the precoding vector w c ∈C M×1 and w k ∈C M×1 In addition, z:CN(0,Z) is an artificial noise vector, where Z∈H M , Z f 0 is the covariance matrix of the artificial noise vector. The received signals of the legitimate user and the eavesdropper are:

[0128]

[0129] Among them, n j : is the additive Gaussian white noise at the legitimate user j, n e : is the additive white Gaussian noise at Eve.

[0130] The SINR of the public information decoded at j is:

[0131]

[0132] Among them, j∈KU{e},

[0133] The SINR of the private information decoded by the legitimate user k is:

[0134]

[0135] Among them, j∈KU{e},

[0136] Decoding at Eve k The SINR is:

[0137]

[0138] According to Shannon's formula, we have the public information rate at the legitimate user j as:

[0139] R c,j =log2(1+g c,j )

[0140] The private information rate at the legitimate user k is:

[0141] R p,k =log2(1+g p,k )

[0142] The rate at which an eavesdropper eavesdrops on private information k is:

[0143] R p,e→k =log2(1+g p,e→k )

[0144] The achievable confidentiality rate for user k is:

[0145]

[0146] where k∈K, [x] + =max{0,x}.

[0147] In order to improve the secure transmission rate of private information in the system, an optimization problem is proposed:

[0148] (P1)

[0149] stC1:R c,k ≥r c,k

[0150] C2:

[0151] C3:

[0152] C4:

[0153] C5:

[0154] in, K∈{r,t}, r c,k is the threshold for decoding public information, is the maximum transmit power of the base station, is the maximum transmit power of active STAR-RIS, C1 represents the user public information service quality (QOS) constraint; C2 represents the maximum transmit power constraint of the base station; C3 represents the maximum transmit power constraint of active STAR-RIS; C4 represents the amplitude limit of STAR-RIS, and C5 represents the phase shift constant modulus limit of STAR-RIS.

[0155] Since P1 is a non-convex optimization problem, the objective function and constraints are complex and the variables are coupled with each other. It is difficult to directly use the existing algorithms to obtain the global optimal solution to the optimization problem. Therefore, alternating iterative optimization is adopted for this problem. The specific ideas are as follows: 1) Introduce some slack variables to the objective function, simplify the optimized objective function, and then use the three important facts contained in the WMMSE criterion to appropriately transform the original non-convex problem; 2) Given the beamforming, artificial noise, active STAR-RIS power amplification matrix, amplitude matrix and phase shift matrix parameters, update the auxiliary variables introduced by the WMMSE algorithm according to the closed expression; 3) Given the auxiliary variable values ​​introduced by the WMMSE algorithm, the power amplification matrix, amplitude matrix and phase shift matrix parameters of the active STAR-RIS, optimize and update the beamforming and artificial noise of the base station: use the SCA algorithm for solution; 4) Given the auxiliary variable values ​​introduced by the WMMSE algorithm, the beamforming, artificial noise, and the active STAR-RIS power amplification matrix, amplitude matrix and phase shift matrix parameters of the base station, optimize and update the beamforming and artificial noise of the base station: use the SCA algorithm for solution; The amplitude matrix and phase shift matrix parameters of the source STAR-RIS are given, and the power amplification matrix of the active STAR-RIS is optimized and updated: the SCA algorithm is used for solution; 5) Given the auxiliary variable values ​​introduced by the WMMSE algorithm, the beamforming of the base station, artificial noise, the power amplification matrix of the active STAR-RIS, and the phase shift matrix parameters, the amplitude matrix of the active STAR-RIS is optimized and updated: the SCA algorithm is used for solution; 6) Given the auxiliary variable values ​​introduced by the WMMSE algorithm, the beamforming of the base station, artificial noise, the power amplification matrix of the active STAR-RIS, and the amplitude matrix parameters, the phase shift matrix of the active STAR-RIS is optimized and updated: the SCA algorithm and the PDD algorithm are used for optimization and solution until the algorithm converges to obtain the solution to the optimization problem.

[0156] 1) Introduce some slack variables into the objective function to simplify the objective function to be optimized, and then use the three important facts contained in the WMMSE criterion to appropriately transform the original non-convex problem:

[0157] (P2):

[0158] stC1:

[0159] C2:R c,k ≥r c,k

[0160] C3:

[0161] C4:

[0162] C5:

[0163] C6:

[0164] in,

[0165]

[0166] 2) Given the beamforming, artificial noise, active STAR-RIS power amplification matrix, amplitude matrix, and phase shift matrix parameters, update the auxiliary variables introduced by the WMMSE algorithm according to the closed-form expression:

[0167]

[0168] 3) Given the auxiliary variable values ​​introduced by the WMMSE algorithm, the power amplification matrix, amplitude matrix, and phase shift matrix parameters of the active STAR-RIS, the beamforming and artificial noise of the base station are optimized and updated, and the SCA algorithm is used to solve:

[0169]

[0170] stC1: C2: C3: C4:

[0171] in,

[0172]

[0173] 4) Given the auxiliary variable values ​​introduced by the WMMSE algorithm, the beamforming of the base station, artificial noise, the amplitude matrix and phase shift matrix parameters of the active STAR-RIS, the power amplification matrix of the active STAR-RIS is optimized and updated, and the SCA algorithm is used to solve:

[0174]

[0175] in,

[0176]

[0177]

[0178] 5) Given the auxiliary variable values ​​introduced by the WMMSE algorithm, the beamforming of the base station, artificial noise, the power amplification matrix of the active STAR-RIS, and the phase shift matrix parameters, the amplitude matrix of the active STAR-RIS is optimized and updated, and the SCA algorithm is used to solve it:

[0179]

[0180] in, is the solution of the last iteration, b p,t,n , c p,t,n , b c,t,n , c c,t,n Is a positive number.

[0181]

[0182]

[0183] 6) Given the auxiliary variable values ​​introduced by the WMMSE algorithm, the beamforming of the base station, artificial noise, the power amplification matrix of the active STAR-RIS, and the amplitude matrix parameters, the phase shift matrix of the active STAR-RIS is optimized and updated, and the SCA algorithm and the PDD algorithm are used for optimization and solution:

[0184] PDD algorithm inner layer:

[0185] Given ψ r ,ψ t , the original problem becomes:

[0186]

[0187] Among them, ψ r ,ψ t is the auxiliary variable introduced, is φ k The solution from the last iteration.

[0188]

[0189] This problem is a convex optimization problem and can be solved using cvx.

[0190] Given φ r ,φ t , solve the introduced auxiliary variable ψ r ,ψ t Values:

[0191]

[0192] in, is r ,ψ t The optimal solution of .

[0193] PDD outer layer algorithm:

[0194] 1. If the equation φ r =ψ r ,φ t =ψ t The approximation holds, then the dual variable λ r ,λ t It will be updated in a gradient ascent manner, that is:

[0195] λ k =λ k +ρ -1 (φ k -ψ k ),k∈K

[0196] 2. If the equation φ r =ψ r ,φ t =ψ t Far from being "true", that is, very different, then:

[0197] The outer layer will choose to increase the penalty parameter ρ -1 , which will force the equation φ r =ψ r ,φ t =ψ t In the subsequent iteration process, it is better satisfied, namely:

[0198] Wherein, c is a predetermined constant.

[0199] Alternate iterations are performed until the algorithm converges and the solution to the optimization problem is obtained.

[0200] The specific algorithm process is as follows Figure 2 .

[0201] The practical effects of the present invention are described in detail below in conjunction with simulation.

[0202] 1) Simulation conditions

[0203] like Figure 1 As shown, the base station has 8 antennas and an active STAR-RIS. The active STAR-RIS has 16 intelligent reflection units. The path loss from the base station to the active STAR-RIS is 3.0, the path loss from the base station to the legitimate user and the eavesdropper is 3.6, and the path loss from the active STAR-RIS to the legitimate user and the eavesdropper is 2.8. Assume that the base station coordinates are (0,-40,0), the active STAR-RIS coordinates are (0,0,0), the reflection user and the transmission user are randomly distributed on both sides of the circle with a radius of 10m with the active STAR-RIS as the center, and the eavesdropper is on the same side as the reflection user, in a circle with a radius of 15m with the active STAR-RIS as the center.

[0204] 2) Simulation results

[0205] In this embodiment, Figure 3 This is a comparison chart of the convergence of the algorithm under different numbers of units. Figure 3It can be observed that when the number of active STAR-RIS units is M=16, 32, and 64, the algorithm proposed in the present invention can converge the sum safety rate in about the fifth time, and has good convergence. And we found that the larger the number of active STAR-RIS units, the larger the sum safety rate that is finally converged. Figure 4 This is a comparison chart of the convergence of the algorithm under different STAR-RIS. Figure 4 It can be observed that the algorithm proposed in the present invention also has good convergence in the passive STAR-RIS system.

[0206] The above description is only a preferred embodiment of the present invention and is not intended to limit the present invention. Any modifications, equivalent substitutions and improvements within the spirit and principle of the present invention should be included in the protection scope of the present invention.

Claims

1. An optimization method for secure transmission of an active STAR-RIS-assisted RSMA communication system, characterized in that: The following steps are involved: S1: Construct an active STAR-RIS assisted RSMA communication system, which consists of a base station with M antennas, two single-antenna users (one located in the reflection area and the other in the transmission area), a single-antenna eavesdropper (located in the reflection area but can eavesdrop on the information of two legitimate users) and an active STAR-RIS, where the active STAR-RIS contains N units; S2: The communication scenario is to send public information and private information from the base station to the reflective user and the transmissive user at the same time. Due to obstacles or long distances, the base station transmits the information to the legitimate users in the system with the assistance of active STAR-RIS. The public information is transmitted in the form of broadcast, and both the eavesdropper and the legitimate user can receive the public information. S3: In order to improve the security of the entire communication system, the base station sends artificial noise while sending useful signals; in the scenario where the total transmission energy of the base station and active STAR-RIS is limited, an optimization equation for maximizing the system private information and security rate is constructed based on beamforming, artificial noise, active STAR-RIS power amplification matrix, amplitude matrix and phase shift matrix as optimization parameters; S4: First, some slack variables are introduced into the objective function to simplify the optimization objective function. Then, the three important facts contained in the WMMSE criterion are used to appropriately transform the original non-convex problem. Then, based on the alternating optimization algorithm, the auxiliary variables introduced by the WMMSE criterion, the beamforming of the base station, the artificial noise and the active STAR-RIS related parameters are decoupled and optimized. S5: Given the beamforming, artificial noise, active STAR-RIS power amplification matrix, amplitude matrix and phase shift matrix parameters, update the auxiliary variables introduced by the WMMSE algorithm according to the closed-form expression; S6: Given the auxiliary variable values ​​introduced by the WMMSE algorithm, the power amplification matrix, amplitude matrix and phase shift matrix parameters of the active STAR-RIS, optimize and update the beamforming and artificial noise of the base station: use the SCA algorithm to solve; S7: Given the auxiliary variable values ​​introduced by the WMMSE algorithm, the beamforming of the base station, the artificial noise, the amplitude matrix of the active STAR-RIS, and the phase shift matrix parameters, the power amplification matrix of the active STAR-RIS is optimized and updated: the SCA algorithm is used for solution; S8: Given the auxiliary variable values ​​introduced by the WMMSE algorithm, the beamforming of the base station, the artificial noise, the power amplification matrix of the active STAR-RIS, and the phase shift matrix parameters, the amplitude matrix of the active STAR-RIS is optimized and updated: the SCA algorithm is used for solution; S9: Given the auxiliary variable values ​​introduced by the WMMSE algorithm, the beamforming of the base station, the artificial noise, the power amplification matrix of the active STAR-RIS and the amplitude matrix parameters, the phase shift matrix of the active STAR-RIS is optimized and updated: the SCA algorithm and the PDD algorithm are used for optimization and solution; S10: Repeat steps S5-S9 until the algorithm converges and an optimal solution to the optimization problem is obtained.

2. The method for optimizing the secure transmission of an active STAR-RIS-assisted RSMA communication system according to claim 1, characterized in that: The received signals of the legitimate user and the eavesdropper in step S2 are: in, They represent the channel from the base station to the legitimate user j, the channel from the active STAR-RIS to the legitimate user j, the channel from the base station to the active STAR-RIS, the channel from the base station to the eavesdropper, and the channel from the active STAR-RIS to the eavesdropper, respectively. is the amplification factor matrix of active STAR-RIS, E r @Diag(ζ),ζ@[z1,z2,...,z N ]and is the reflection and transmission amplitude coefficient matrix of active STAR-RIS, and is the reflection-transmission phase shift coefficient matrix of Active STAR-RIS, v is the thermal noise of Active STAR-RIS, and v: The additive Gaussian white noise at the legitimate user j is n j : The additive white Gaussian noise at Eve is The signal sent by the base station is represented by x=w c s c +∑ k∈κ w k s k +z, where K@{r,t}, s c :CN(0,1) and s k :CN(0,1) represents the public stream of all LUs and the private stream of only LU k, respectively, and is represented by the precoding vector w c ∈C M×1 and w k ∈C M×1 Precoding is performed, z:CN(0,Z) is an artificial noise vector, where Z∈H M , is the covariance matrix of the artificial noise vector; The SINR of the public information decoded at j is: Among them, j∈KU{e}, The SINR of the private information decoded by the legitimate user k is: Among them, j∈KU{e}, Decoding at Eve k The SINR is: According to Shannon's formula, the public information rate at the legitimate user j is: R c,j =log2(1+g c,j ) (6) The private information rate at the legitimate user k is: R p,k =log2(1+g p,k ) (7) The rate at which an eavesdropper eavesdrops on private information k is: R p,e→k =log2(1+g p,e→k ) (8) The achievable confidentiality rate for user k is: where k∈K, [x] + =max{0,x}.

3. The method for optimizing the secure transmission of an active STAR-RIS-assisted RSMA communication system according to claim 1, characterized in that: In order to improve the security of the entire communication system in S3, the following optimization method is adopted: in, K∈{r,t}, is the maximum transmit power of the base station, is the maximum transmit power of the active STAR-RIS, r c,k The threshold for decoding public information.

4. The method for optimizing the secure transmission of an active STAR-RIS-assisted RSMA communication system according to claim 1, characterized in that: The specific operation of step S4 is: First, we introduce auxiliary variables and use the WEESE criterion to transform the original non-convex problem into: in, The alternating optimization algorithm is used to decouple the problem (P2) into multiple sub-problems for alternating optimization: (1) Sub-problem 1: Optimizing the auxiliary variables introduced by the WMMSE criterion: (2) Sub-problem 2: Optimizing beamforming and artificial noise of base stations: (3) Sub-problem 3: Optimize the amplification factor matrix of active STAR-RIS: (4) Sub-problem 4: Optimize the amplitude coefficient matrix of active STAR-RIS: (5) Sub-problem 5: Optimize the phase shift matrix of active STAR-RIS:

5. The method for optimizing the secure transmission of an active STAR-RIS-assisted RSMA communication system according to claim 1, characterized in that: The step S5 is specifically as follows:

6. The method for optimizing the secure transmission of an active STAR-RIS-assisted RSMA communication system according to claim 1, characterized in that: The step S6 is specifically as follows: in, 7. The method for optimizing the secure transmission of an active STAR-RIS-assisted RSMA communication system according to claim 1, characterized in that: The step S7 is specifically as follows: in, 8. The method for optimizing secure transmission of an active STAR-RIS-assisted RSMA communication system according to claim 1, characterized in that: The step S8 is specifically as follows: in, is the solution of the last iteration, b p,t,n , c p,t,n , b c,t,n , c c,t,n is a positive number; 9. The method for optimizing secure transmission of an active STAR-RIS-assisted RSMA communication system according to claim 1, characterized in that: The step S9 is specifically as follows: PDD algorithm inner layer: Given ψ r ,ψ t , the original problem becomes: Among them, ψ r ,ψ t is the auxiliary variable introduced, is φ k The solution of the last iteration; This problem is a convex optimization problem and can be solved using cvx; Given φ r ,φ t , solve the introduced auxiliary variable ψ r ,ψ t Values: in, is r ,ψ t The optimal solution of PDD outer layer algorithm: If the equation φ r =ψ r ,φ t =ψ t The approximation holds, then the dual variable λ r ,λ t It will be updated in a gradient ascent manner, that is: l k =λ k +r -1 (f k -ψ k ),k∈K: If the equation φ r =ψ r ,φ t =ψ t Far from "true", that is, very different, the outer layer will choose to increase the penalty parameter ρ -1 , which will force the equation φ r =ψ r ,φ t =ψ t In the subsequent iteration process, it is better satisfied, namely: Wherein, c is a predetermined constant.

Images