Secure communication perception integrated design method

Abstract

The invention discloses a secure communication and perception integrated design method, which introduces RIS and RSMA into ISAC, aims at maximizing beam gain in a perception direction, establishes system communication and perception performance indexes, further designs optimal beam forming at a base station and an active RIS, and achieves performance balance of secure communication and perception. The method disclosed by the invention not only ensures the balance of safety communication and sensing performance, but also is superior to a conventional sensing and communication integrated scheme based on space division multiple access assisted by a passive intelligent reflecting surface.

Description

Technical Field

[0001] The present invention belongs to the field of communication technology, and in particular relates to a secure communication perception integrated design method. Background Art

[0002] As more multifunctional radio devices appear in 5G networks, spectrum congestion is hindering the further development of wireless networks. In order to solve the problem of spectrum congestion, communication and perception integration technology has been proposed, which allows communication and radar to use the same spectrum resources and hardware facilities to achieve the reuse of the same electromagnetic wave signals. Compared with independent radar and communication coexistence systems, this technology can reduce hardware costs, reduce mutual interference between communication and radar signals, and improve the utilization of spectrum, energy and other resources. Therefore, the communication and perception integration (ISAC) technology has attracted the interest of many researchers.

[0003] On the other hand, the intelligent reflection surface (RIS) technology has also been proposed to improve the reception quality of signals and expand the coverage of the system. Specifically, RIS is composed of a large number of reflection elements, and the wireless propagation environment can be reconfigured by controlling the reflected signal. In communication, RIS improves the system capacity and is particularly suitable for environments with channel degradation; in perception, RIS can establish a virtual direct link for non-direct targets and expand the perception range. RIS includes passive RIS and active RIS, among which passive RIS can reflect signals to the maximum extent and adjust the phase of the reflected signal; active RIS can control the amplitude and phase of the reflected signal, realize active beamforming at RIS, and optimize the transmission direction of signal energy. Therefore, the application of RIS in ISAC systems has also received widespread attention. In addition to RIS, advanced multiple access technologies such as non-orthogonal multiple access (NOMA) and rate splitting (RSMA) are also used in ISAC systems. This technology can further manage the interference between communication and perception to obtain better performance balance.

[0004] However, since the detection target and the radar share signals, the target can be used as a potential eavesdropper to steal user information. In addition, the detection target requires a higher signal energy to point to the target, which also increases the eavesdropping signal-to-noise ratio (SNR) of the target, which seriously threatens the user's information security. Therefore, it is necessary to study the performance trade-off between secure communication and perception. Summary of the invention

[0005] In order to overcome the shortcomings of the prior art, the present invention provides a secure communication and perception integrated design method, which introduces RIS and RSMA into ISAC, establishes system communication and perception performance indicators with the goal of maximizing the beam gain in the perception direction, and then designs the optimal beamforming at the base station and active RIS to achieve performance balance between secure communication and perception. The method of the present invention not only ensures the balance between secure communication and perception performance, but also is superior to the traditional perception communication integrated solution based on space division multiple access assisted by passive intelligent reflectors.

[0006] The technical solution adopted by the present invention to solve the technical problem is as follows:

[0007] Step 1: Construct RIS-assisted RSMA ISAC system model:

[0008] Step 2: Determine the objective function and optimization variables, and list the optimization problem;

[0009] Step 3: Design an algorithm to solve the optimization problem.

[0010] Furthermore, the step 1 is specifically as follows:

[0011] Step 1-1: The RSMA ISAC system model includes a base station BS and a RIS, serving K single-antenna users and M single-antenna targets, where the targets are all potential eavesdroppers. represents the user set, represents the target set;

[0012] Step 1-2: The RSMA ISAC system model adopts RSMA to divide user information into public information c and private information k , Precoding is performed separately; at the same time, the radar sequence is added to the transmitted signal, and the precoded signal of the transmitted signal is expressed as:

[0013]

[0014] Among them, w c Indicates public information c The corresponding precoding vector, w k Represents the private information s of the kth user k The corresponding precoding vector, s r Indicates a perceived signal;

[0015] Step 1-3: Assume that all links from BS to the target are blocked and the direction of the target is known. The channels from BS to the kth user, BS to RIS, RIS to the kth user, and RIS to the tth target are denoted as h d,k ∈CN ,G∈C L×N ,h r,k ∈C L and h t,m ∈C L , where both the user and RIS channel models adopt the Rice fading channel model:

[0016]

[0017] Among them, d BS,k is the distance from the base station BS to the kth user, κ BS,k is the channel h d,k The Rice index, and are the direct and indirect channel vectors respectively. BS,RIS is the distance from the base station BS to RIS, κ BS,RIS is the channel Rice index, G L,S and G NL are the direct and indirect channel matrices respectively; C represents a complex number set, N represents the number of transmitting antennas, L represents the number of RIS elements, ε0 represents the path loss at the reference distance, σ BS,k represents the Rice index of the user channel, σ BS,RIS represents the Rice index of the RIS channel;

[0018] Using the channel model with error, it is expressed as:

[0019]

[0020] in, Denoted as a direct link channel, Δh t,m represents channel uncertainty; θ m represents the angle of the mth target;

[0021] Step 1-4: Assume that each user uses serial interference cancellation (SIC) to remove interference from the perceived signal and uses it to decode user information; specifically, it means that the perceived signal is removed before the user decodes its own signal, and when decoding the user information, s is decoded first. c , then remove it and decode s k , The decoding rate of user information is expressed as:

[0022]

[0023] where γ c,k and γ k To decode c and k , Signal to Interference and Noise Ratio (SINR) at hk represents the channel from BS to the kth user, σ k and σ e denote the Gaussian white noise energy at the kth user and the Gaussian white noise energy at the RIS, Φ denotes the RIS amplitude and phase shift matrix, and w j The beamforming vector representing the j-th user information;

[0024] If the received signal at the target is still decoded using SIC, the eavesdropping rate of target m on user k can be expressed as:

[0025] R m,c =log2(1+γ m,c ) (7)

[0026] R m,k =log2(1+γ m,k ) (8)

[0027] Among them, γ m,c and γ m,k The target m and the eavesdropping s are c and k , SINR at the time

[0028] Step 1-5: Given the statistical status information of the target, the security interruption probability SOP constraint is used as the criterion for measuring communication security, which is expressed as:

[0029]

[0030] in, It is represented as the eavesdropping rate threshold, Indicates the corresponding SOP value; Pr m,k represents the security interruption probability of target m eavesdropping on user k;

[0031] Under this SOP constraint, the safety capacity at target m is expressed as:

[0032]

[0033] Among them, R k represents the communication rate of user k;

[0034] Step 1-6: Install the sensor at the RIS to process the echo signal, and use the beam pattern gain of the RIS as the perception performance criterion, which is expressed as:

[0035]

[0036] Where x represents the transmitted signal, G represents the channel matrix from the base station to the RIS, and w cThe beamforming vector representing the common information, w k represents the beamforming vector corresponding to the private information of user k, R v Represents the perception signal s r The corresponding covariance matrix is, represents the steering vector, R x They represent the covariance matrix of the transmitted signal x respectively; L represents the number of RIS elements;

[0037] Furthermore, the step 2 is specifically as follows:

[0038] Step 2-1: By jointly optimizing the base station precoding matrix and the RIS reflection coefficient, the following optimization problem is formed:

[0039]

[0040] In this optimization problem, C1 represents the user’s c The decoding rate constraint, C2 and C4 represent the user security performance constraints, C3 represents the target perception performance constraint, C5 represents the signal amplification capability constraint of RIS, and C6 and C7 represent the power constraints of active RIS and BS respectively;

[0041] Step 2-2: Simplify the probability constraint C4 and convert it into a deterministic form:

[0042] Ignore the noise caused by RIS and enlarge the target eavesdropping SINR, which is expressed as:

[0043]

[0044] in Substituting formula (4) into formula (14), we get the following form:

[0045]

[0046] in

[0047] Applying Bayesian inequality, equation (15) is transformed into:

[0048]

[0049] Substituting (16), (17) and (18) into P1, we form the optimization problem P2, which can be expressed as:

[0050]

[0051] Furthermore, the step 3 is specifically as follows:

[0052] Using the idea of ​​alternating iteration, the optimization problem shown in formula (19) is decomposed into two sub-problems. For the BS beamforming optimization problem, the semi-positive definite relaxation technique is used to solve it; for the RIS reflection coefficient optimization problem, a series of transformations are performed on its constraints, and then the continuous convex approximation method is used to solve it; the details are as follows:

[0053] Step 3-1: Fix the RIS reflection coefficient and optimize the BS precoding;

[0054] With the RIS reflection coefficient PHI fixed, the optimization problem is expressed as:

[0055]

[0056] in Ignoring the rank 1 constraint C8, the optimization problem can be solved by using the semi-positive definite relaxation SDR. i )=1, the eigenvalue decomposition is used to obtain the precoding vector; when rank(w i )>1, a Gaussian randomization method is used to obtain a precoding vector that satisfies the constraint conditions;

[0057] Step 3-2: Fix BS precoding and optimize RIS reflection coefficient;

[0058] Given a precoding vector w c 、w i and v i , the optimization problem is expressed as:

[0059]

[0060] Separating Φ from the coupled variables, it is expressed as:

[0061]

[0062] in:

[0063]

[0064] Simplify constraints (C4-1), (C4-2) and (C4-3) by approximation; according to Tr(A(B⊙C))=Tr((A⊙B)C) and The first and last terms of the left-hand side of the constraint (C4-1) can be expressed as:

[0065]

[0066]

[0067] Substitute equations (26) and (27) into the constraint (C4-1) and transform (C4-1) into:

[0068]

[0069] then, The equivalent expression is:

[0070]

[0071] According to equations (28) and (29), the (C4-2) constraint is equivalently expressed as:

[0072]

[0073] in

[0074]

[0075] The upper bound of equation (30) can be written using the second-order Taylor expansion at point φ0 as:

[0076]

[0077] in is a semi-positive definite matrix, d m,k Yes D m,k The maximum eigenvalue of

[0078] Since the signal amplification capability of RIS is limited, that is, |φ l |≤a max , The upper bound of is expressed as:

[0079]

[0080] Substituting equations (34) and (36) into the constraint (C4-2), the upper bound expression of (C4-2) is converted into:

[0081]

[0082] The semi-positive definite constraint (C4-3) is equivalently expressed as:

[0083]

[0084] in, It can be equivalently expressed as The minimum eigenvalue of . According to the Weyl inequality in the eigenvalue form in matrix theory, The lower bound of is expressed as:

[0085]

[0086] Substituting it into equation (38), the (C4-3) constraint can be further expressed as:

[0087] C4"-3:-x m,k,2 -Tr(P m,k φφ H )≤0,-x m,k,2 ≤0 (40)

[0088] Combining equations (28), (37) and (40), the optimization problem P4' can be reformulated as:

[0089]

[0090] By using continuous convex approximation to transform the non-convex terms in all constraints in P4″ into the form of linear functions, the problem P4″ can be rewritten as:

[0091]

[0092] P5 is a convex optimization problem that can be solved using CVX;

[0093] Step 3-3: Solve the optimization problem P2 through an alternating iterative algorithm. In each iteration, fix PHI and use SDR to solve the precoding vector optimization problem. Secondly, fix w c 、w i and v i , simplify the C1-C7 constraints, use the continuous convex approximation algorithm to solve, and finally update the value of the parameter to perform the next iteration until the algorithm converges; the specific process is as follows:

[0094] 1) Set the convergence tolerance>0, the number of iterations t=1, and initialize the RIS reflection coefficient Φ0;

[0095] 2) Given Φ t-1 , solve the optimization problem P3 and obtain the BS beamforming result of the tth iteration, and

[0096] 3) Given and The convex optimization problem P5 is solved iteratively for many times to obtain the RIS beamforming result, Φ t ;

[0097] 4) Update the number of iterations t=t+1;

[0098] 5) Check whether it converges. If t n -t n-1 ≥∈, return to step 2) and perform the next iterative solution; otherwise, return the solution result of the optimization problem P1. and Φ * .

[0099] The beneficial effects of the present invention are as follows:

[0100] The present invention introduces rate splitting multiple access technology into the communication perception integrated system assisted by the intelligent reflector, and provides a solution for maximizing the beam gain in the direction of the target of interest under the constraint of the probability of safe interruption of user communication by jointly optimizing the base station precoding vector and the transmission coefficient matrix of the intelligent reflector. This method not only ensures the balance between secure communication and perception performance, but also outperforms the traditional perception communication integrated solution based on space division multiple access assisted by passive intelligent reflectors. The present invention provides a reference design method for realizing secure communication perception integration assisted by intelligent reflectors. BRIEF DESCRIPTION OF THE DRAWINGS

[0101] Figure 1 A schematic diagram of a system used in the integrated method for secure communication perception assisted by an intelligent reflective surface of the present invention;

[0102] Figure 2 Schematic diagram of iteration number VS minimum beam pattern gain according to an embodiment of the present invention;

[0103] Figure 3 This is a schematic diagram of maximum power VS minimum beam pattern gain of a base station according to an embodiment of the present invention;

[0104] Figure 4 Schematic diagram of beam pattern gain at different angles according to an embodiment of the present invention;

[0105] Figure 5 Schematic diagram of safety rate vs. minimum beam pattern gain according to an embodiment of the present invention. DETAILED DESCRIPTION

[0106] The present invention is further described below in conjunction with the accompanying drawings and embodiments.

[0107] The purpose of this invention is to maximize the perception performance of the RIS-assisted ISAC system under the constraints of secure communication. In the network model, the base station serves K users and M targets, where the targets act as potential eavesdroppers to steal user information. The specific scheme diagram is shown in Figure 1 Based on this model, the present invention provides an active beamforming design scheme that jointly optimizes power distribution and RIS reflection coefficient to maximize the perception performance under the constraints of secure communication.

[0108] In order to achieve the above purpose, the technical solution adopted by the present invention is as follows:

[0109] The first step is to build the RIS-assisted RSMA ISAC system model:

[0110] (1) The system consists of a base station BS and a RIS, serving K single-antenna users and M single-antenna targets, where the targets are all potential eavesdroppers. represents the user set, Represents the target set.

[0111] (2) The system uses RSMA to divide user information into public information s c and private information k , At the same time, in order to ensure detection and safety performance, the radar sequence is added to the transmitted signal, which can be used to detect the target and can also be used as an artificial noise to interfere with the target. The precoded signal of the transmitted signal is expressed as:

[0112]

[0113] (3) Assuming that all links from BS to the target are blocked and the direction of the target is known, the channels from BS to the kth user, BS to RIS, RIS to the kth user, and RIS to the tth target are denoted as h d,k ∈C N ,G∈C L×N ,h r,k ∈C L and h t,m ∈C L , where both the user and RIS channel models adopt the Rice fading channel model, such as

[0114]

[0115] Among them, d BS,k is the distance from the base station BS to the kth user, κ BS,k is the channel h d,k The Rice index, and are the direct and indirect channel vectors respectively. BS,RIS is the distance from the base station BS to RIS, κ BS,RIS is the channel Rice index, G L,S and G NL are the direct and indirect channel matrices respectively.

[0116] In addition, since the user and BS interact frequently, the channel between the user and RIS can be accurately estimated, while the target interacts less with the BS and its channel cannot be accurately estimated. A channel model with errors can be used, which is expressed as:

[0117]

[0118] in, Denoted as a direct link channel, Δh t,m represents the channel uncertainty, and its statistical distribution information is available.

[0119] (4) Assume that each user uses serial interference cancellation (SIC) to remove interference from the perceived signal and use it to decode the user information. Specifically, it means that the perceived signal is removed before the user decodes its own signal. When decoding the user information, the perceived signal is first decoded. c , then remove it and decode s k , Then the decoding rate of user information can be expressed as:

[0120]

[0121] Where s is the decoded c and k , The signal-to-noise ratio (SNR) at .

[0122] If the received signal at the target is still decoded using SIC, the eavesdropping rate of target m on user k can be expressed as:

[0123] R m,c =log2(1+γ m,c ) (7)

[0124] R m,k =log2(1+γ m,k ) (8)

[0125] Among them, they are target m and eavesdrop s c and k , The SNR at that time.

[0126] (5) Given the statistical status information of the target, the security outage probability (SOP) constraint can be used as a criterion for measuring communication security, which is expressed as:

[0127]

[0128] in, It is represented as the eavesdropping rate threshold, represents the corresponding SOP value. Therefore, under this SOP constraint, the safety capacity at target m can be expressed as:

[0129]

[0130] (6) Installing sensors at the RIS can realize the processing of echo signals. The beam pattern gain of the RIS is used as the perception performance criterion, which is expressed as:

[0131]

[0132] The second step is to determine the objective function and optimization variables and list the optimization problem

[0133] (1) By jointly optimizing the base station precoding matrix and the RIS reflection coefficient, the following optimization problem is formed:

[0134]

[0135] In this optimization problem, (C1) represents the user's c (C2) and (C4) represent the user security performance constraints, (C3) represents the target perception performance constraints, (C5) represents the signal amplification capability constraints of RIS, and (C6) and (C7) represent the power constraints of active RIS and BS, respectively.

[0136] (2) Simplify the probabilistic constraint (C4) and convert it into a deterministic form

[0137] Ignoring the noise caused by RIS and enlarging the target eavesdropping SINR, it can be expressed as:

[0138]

[0139] in Substituting formula (4) into formula (14), we get the following form:

[0140]

[0141] in Applying Bayesian inequality, equation (15) is transformed into

[0142]

[0143] Substituting (16), (17) and (18) into P1, we can form the optimization problem P2, which can be expressed as

[0144]

[0145] The third step is to design an algorithm to solve the optimization problem

[0146] Using the idea of ​​alternating iteration, the optimization problem shown in equation (19) is decomposed into two sub-problems. For the BS beamforming optimization problem, the semi-positive definite relaxation technique is used to solve it; for the RIS reflection coefficient optimization problem, a series of transformations are performed on its constraints, and then the continuous convex approximation method is used to solve it. The specific steps are as follows:

[0147] (1) Fixing the RIS reflection coefficient and optimizing BS precoding

[0148] With the RIS reflection coefficient PHI fixed, the optimization problem is expressed as:

[0149]

[0150] in Ignoring the rank 1 constraint (C8), the optimization problem can be solved using semidefinite relaxation (SDR). i )=1, the eigenvalue decomposition is used to obtain the precoding vector. i )>1, a Gaussian randomization method is used to obtain a precoding vector that meets the constraints.

[0151] (2) Fixed BS precoding and optimized RIS reflection coefficient

[0152] Given a precoding vector w c 、w i and v i , the optimization problem can be expressed as

[0153]

[0154] This problem is difficult to solve directly for the RIS reflection coefficient PHI, because w c 、w i 、v i and Φ are coupled in constraints (C1)-(C7). To design an algorithm to directly optimize Φ, Φ is first separated from the coupled variables, which can be expressed as:

[0155]

[0156] in

[0157]

[0158] Furthermore, constraints (C4-1), (C4-2) and (C4-3) are simplified by approximation. According to Tr(A(B⊙C))=Tr((A⊙B)C) and The first and last terms of the left-hand side of the constraint (C4-1) can be expressed as:

[0159]

[0160] Substituting equations (26) and (27) into the constraint (C4-1), (C4-1) can be transformed into:

[0161]

[0162] then, It can be equivalently expressed as:

[0163]

[0164] According to equations (28) and (29), the (C4-2) constraint can be equivalently expressed as:

[0165]

[0166] in

[0167]

[0168]

[0169] It can be seen that since equation (30) is a quartic function of φ, the optimization problem is difficult to solve. In order to solve this problem, the upper bound expression of equation (30) is written using the second-order Taylor expansion at point φ0:

[0170]

[0171] in is a semi-positive definite matrix, d m,k Yes D m,k The maximum eigenvalue of . Since the signal amplification capability of RIS is limited, that is, φ l |≤a max , The upper bound of can be expressed as

[0172]

[0173] Substituting equations (34) and (36) into the constraint (C4-2), the upper bound expression of (C4-2) is converted into:

[0174]

[0175] The semi-positive definite constraint (C4-3) can be equivalently expressed as:

[0176]

[0177] in, It can be equivalently expressed as The minimum eigenvalue of . According to the Weyl inequality in the eigenvalue form in matrix theory, The lower bound of is expressed as:

[0178]

[0179] Substituting it into equation (38), the (C4-3) constraint can be further expressed as:

[0180]

[0181] Combining equations (28), (37) and (40), the optimization problem P4' can be reformulated as:

[0182]

[0183] Although the variable Φ is separated from the coupled variables and the constraints (C4-1), (C4-2) and (C4-3) are simplified, the constraints C1″-C6″ are still non-convex and it is difficult to find a global solution. To solve this problem, a continuous convex approximation is used to convert the non-convex terms in all constraints in P4″ into the form of linear functions. Problem P4″ can be rewritten as:

[0184]

[0185] P5 is a convex optimization problem that can be solved using CVX.

[0186] (3) Solve the optimization problem P2 through the alternating iterative algorithm. In each iteration, fix PHI and use SDR to solve the precoding vector optimization problem. c 、w i and v i , simplify the constraints (C1)-(C7), solve them using the continuous convex approximation algorithm, and finally update the values ​​of the parameters for the next iteration until the algorithm converges.

[0187] The specific process is as follows:

[0188] 6) Set the convergence tolerance ∈>0, the number of iterations t=1, and initialize the RIS reflection coefficient Φ0;

[0189] 7) Given Φ t-1 , solve the optimization problem P3 and obtain the BS beamforming result of the tth iteration, and

[0190] 8) Given and The convex optimization problem P5 is solved iteratively for many times to obtain the RIS beamforming result, Φ t ;

[0191] 9) Update the number of iterations t=t+1;

[0192] 10) Check whether it converges. If t n -t n-1 ≥∈, return to step 2) and perform the next iterative solution; otherwise, return the solution result of the optimization problem P1. and Φ * .

[0193] Figure 1The architecture diagram of the RSMA ISAC system assisted by active RIS. The system serves K users and M targets. The base station BS and active RIS will implement beamforming according to the location of the target and user. The target, as a potential eavesdropper, poses a security threat to user information. All links from the base station BS to the target are blocked, and only RIS can be used for target perception.

[0194] Figure 2 The iterative performance of the algorithm is demonstrated, and it can be seen that the proposed algorithm can converge within 25 iterations under different parameter configurations.

[0195] Figure 3 The effect of base station power on beam gain is demonstrated, and the beam gains generated by active and passive RIS under different parameter configurations are compared.

[0196] Figure 4 The beam gains obtained by the proposed scheme at different angles are demonstrated, and the performances of active and passive RIS are compared under different signal amplification capabilities and different numbers of RIS elements.

[0197] Figure 5 The relationship between the safety rate and beam pattern gain is demonstrated, and the performance of RSMA and SDMA schemes under different RIS amplification signal capabilities is compared, indicating the superiority of the proposed scheme.

Claims

1. A secure communication perception integrated design method, characterized in that: The steps include: Step 1: Construct RIS-assisted RSMA ISAC system model: Step 2: Determine the objective function and optimization variables, and list the optimization problem; Step 3: Design an algorithm to solve the optimization problem.

2. A secure communication perception integrated design method according to claim 1, characterized in that: The step 1 is specifically as follows: Step 1-1: The RSMA ISAC system model includes a base station BS and a RIS, serving K single-antenna users and M single-antenna targets, where the targets are all potential eavesdroppers. represents the user set, represents the target set; Step 1-2: The RSMA ISAC system model adopts RSMA to divide user information into public information c and private information Precoding is performed separately; at the same time, the radar sequence is added to the transmitted signal, and the precoded signal of the transmitted signal is expressed as: Among them, w c Indicates public information c The corresponding precoding vector, w k Represents the private information s of the kth user k The corresponding precoding vector, s r Indicates a perceived signal; Step 1-3: Assume that all links from BS to the target are blocked and the direction of the target is known. The channels from BS to the kth user, BS to RIS, RIS to the kth user, and RIS to the tth target are denoted as h d,k ∈C N ,G∈C L×N ,h r,k ∈C L and h t,m ∈C L , where both the user and RIS channel models adopt the Rice fading channel model: Among them, d BS,k is the distance from the base station BS to the kth user, κ BS,k is the channel h d,k The Rice index, and are the direct and indirect channel vectors respectively; d BS,RIS is the distance from the base station BS to RIS, κ BS,RIS is the channel Rice index, G L,S and G NL are the direct and indirect channel matrices respectively; C represents a complex set, N represents the number of transmitting antennas, L represents the number of RIS elements, ε0 represents the path loss at the reference distance, σ BS,k represents the Rice index of the user channel, σ BS,RIS represents the Rice index of the RIS channel; Using the channel model with error, it is expressed as: in, Denoted as a direct link channel, Δh t,m represents the channel uncertainty; θ m represents the angle of the mth target; Step 1-4: Assume that each user uses serial interference cancellation (SIC) to remove interference from the perceived signal and uses it to decode user information; specifically, it means that the perceived signal is removed before the user decodes its own signal, and when decoding the user information, s is decoded first. c , then remove it and decode it The decoding rate of user information is expressed as: where γ c,k and γ k To decode c and Signal to Interference and Noise Ratio (SINR) at time h k represents the channel from BS to the kth user, σ k and σ e denote the Gaussian white noise energy at the kth user and the Gaussian white noise energy at the RIS, Φ denotes the RIS amplitude and phase shift matrix, and w j The beamforming vector representing the j-th user information; If the received signal at the target is still decoded using SIC, the eavesdropping rate of target m on user k can be expressed as: R m,c =log2(1+γ m,c ) (7) R m,k =log2(1+γ m,k ) (8) Among them, γ m,c and γ m,k The target m and the eavesdropping s are c and SINR at the time Step 1-5: Given the statistical status information of the target, the security interruption probability SOP constraint is used as the criterion for measuring communication security, which is expressed as: in, It is represented as the eavesdropping rate threshold, Indicates the corresponding SOP value; Pr m,k represents the security interruption probability of target m eavesdropping on user k; Under this SOP constraint, the safety capacity at target m is expressed as: Among them, R k represents the communication rate of user k; Step 1-6: Install the sensor at the RIS to process the echo signal, and use the beam pattern gain of the RIS as the perception performance criterion, which is expressed as: Where x represents the transmitted signal, G represents the channel matrix from the base station to the RIS, and w c The beamforming vector representing the common information, w k represents the beamforming vector corresponding to the private information of user k, R v Represents the perception signal s r The corresponding covariance matrix is, represents the steering vector, R x They represent the covariance matrix of the transmitted signal x respectively; L represents the number of RIS elements.

3. A secure communication perception integrated design method according to claim 2, characterized in that: The step 2 is specifically as follows: Step 2-1: By jointly optimizing the base station precoding matrix and the RIS reflection coefficient, the following optimization problem is formed: In this optimization problem, C1 represents the user’s c The decoding rate constraint, C2 and C4 represent the user security performance constraints, C3 represents the target perception performance constraint, C5 represents the signal amplification capability constraint of RIS, and C6 and C7 represent the power constraints of active RIS and BS respectively; Step 2-2: Simplify the probability constraint C4 and convert it into a deterministic form: Ignore the noise caused by RIS and enlarge the target eavesdropping SINR, which is expressed as: in Substituting formula (4) into formula (14), we get the following form: in Applying Bayesian inequality, equation (15) is transformed into: C4-1: C4-2: C4-3: Substituting (16), (17) and (18) into P1, we form the optimization problem P2, which can be expressed as:

4. A secure communication perception integrated design method according to claim 3, characterized in that: The step 3 is specifically as follows: Using the idea of ​​alternating iteration, the optimization problem shown in formula (19) is decomposed into two sub-problems. For the BS beamforming optimization problem, the semi-positive definite relaxation technique is used to solve it; for the RIS reflection coefficient optimization problem, a series of transformations are performed on its constraints, and then the continuous convex approximation method is used to solve it; the details are as follows: Step 3-1: Fix the RIS reflection coefficient and optimize the BS precoding; With the RIS reflection coefficient PHI fixed, the optimization problem is expressed as: in Ignoring the rank 1 constraint C8, the optimization problem can be solved by using the semi-positive definite relaxation SDR. i )=1, the eigenvalue decomposition is used to obtain the precoding vector; when rank(w i )>1, a Gaussian randomization method is used to obtain a precoding vector that satisfies the constraint conditions; Step 3-2: Fix BS precoding and optimize RIS reflection coefficient; Given a precoding vector w c 、w i and v i , the optimization problem is expressed as: Separating Φ from the coupled variables, it is expressed as: in: P′ m =φ H diag(a(θ m ))R x diag(a H (i m ))φ (25) Simplify constraints (C4-1), (C4-2) and (C4-3) by approximation; according to Tr(A(B⊙C))=Tr((A⊙B)C) and The first and last terms of the left-hand side of the constraint (C4-1) can be expressed as: Substitute equations (26) and (27) into the constraint (C4-1) and transform (C4-1) into: C4”-1: then, The equivalent expression is: According to equations (28) and (29), the (C4-2) constraint is equivalently expressed as: in The upper bound of equation (30) can be written as follows using the second-order Taylor expansion at point φ0: in is a semi-positive definite matrix, d m,k Yes D m,k The maximum eigenvalue of Since the signal amplification capability of RIS is limited, i.e., φ l |≤a max , The upper bound of is expressed as: Substituting equations (34) and (36) into the constraint (C4-2), the upper bound expression of (C4-2) is converted into: C4”-2: The semi-positive definite constraint (C4-3) is equivalently expressed as: in, It can be equivalently expressed as The minimum eigenvalue of ; According to the Weyl inequality in the eigenvalue form in matrix theory, The lower bound of is expressed as: Substituting it into equation (38), the (C4-3) constraint can be further expressed as: C4”-3:-x m,k,2 -Tr(P m,k φφ H )≤0,-x m,k,2 ≤0 (40) Combining equations (28), (37) and (40), the optimization problem P4' can be reformulated as: By using continuous convex approximation to transform the non-convex terms in all constraints in P4″ into the form of linear functions, the problem P4″ can be rewritten as: P5 is a convex optimization problem that can be solved using CVX; Step 3-3: Solve the optimization problem P2 through the alternating iterative algorithm. In each iteration, fix PHI and use SDR to solve the precoding vector optimization problem. Secondly, fix w c 、w i and v i , simplify the C1-C7 constraints, use the continuous convex approximation algorithm to solve, and finally update the value of the parameter to perform the next iteration until the algorithm converges; the specific process is as follows: 1) Set the convergence tolerance ∈>0, the number of iterations t=1, and initialize the RIS reflection coefficient Φ0; 2) Given Φ t-1 , solve the optimization problem P3 and obtain the BS beamforming result of the tth iteration, and 3) Given and The convex optimization problem P5 is solved iteratively for many times to obtain the RIS beamforming result, Φ t ; 4) Update the number of iterations t=t+1; 5) Check whether it converges; if t n -t n-1 ≥∈, return to step 2) and perform the next iterative solution; otherwise, return the solution result of the optimization problem P1. and Φ * .

Images