Method for maximizing secrecy rate of RSMA-assisted ISAC system under imperfect CSI

Abstract

The application claims a method for maximizing the secrecy rate of a rate splitting multiple access (RSMA) assisted integrated sensing and communication (ISAC) system under imperfect channel state information (CSI), which considers the influence of multiple eavesdroppers with imperfect CSI on the RSMA assisted ISAC system, and realizes the maximization of the system secrecy rate under the constraints of the public secrecy rate, the maximum transmission power of the base station, the radar echo signal-to-noise ratio (SNR) and the like. In order to solve the non-convex optimization problem, an alternating optimization algorithm is designed to decouple the original problem into two sub-problems, and the system secrecy rate is maximized by introducing auxiliary variables and relaxation variables, successive convex approximation (SCA), semidefinite relaxation (SDR), S-Procedure and the like. The application can improve the security and robustness of the ISAC system in the presence of multiple eavesdroppers and under the condition that the CSI of the eavesdroppers is imperfect, and has better practicability and feasibility.

Description

Technical Field

[0001] This invention belongs to the field of communication sensing integration technology, specifically relating to a method for maximizing the security rate of an ISAC system based on RSMA assistance under imperfect CSI. Background Technology

[0002] With the rapid development of 6th Generation Mobile Networks (6G), the massive demand for concurrent data from multiple users poses a severe challenge to the spectrum resource allocation of next-generation wireless communication systems. Simultaneously, the increasing prevalence of intelligent technologies and the vision of the Internet of Things (IoT) have driven the development of technologies that enable base stations to achieve higher environmental perception accuracy. These technologies, through shared spectrum and hardware resources, integrate communication and sensing on the same platform, achieving mutual benefit and collaborative cooperation. This significantly alleviates the shortage of spectrum resources and reduces the signaling overhead and hardware costs of independent radar deployments. However, due to the broadcast nature of wireless communication channels, data symbols carried on the same spectrum channel are easily intercepted by malicious eavesdroppers, reducing the security performance of the ISAC (Multiple-Input Single-Output Integrated Sensing and Communication) system. Therefore, effectively improving the physical layer security performance of wireless communication systems, especially in Multiple-Input Single-Output Integrated Sensing and Communication (MISO-ISAC) networks, has become a critical issue that urgently needs to be addressed. Since eavesdroppers generally remain silent, base stations cannot obtain perfect CSI of the eavesdropping channel. Therefore, in scenarios where multiple eavesdroppers coexist, improving the security rate of the ISAC system under imperfect eavesdropping CSI is a key challenge.

[0003] RSMA technology is a novel multiple access technology. Studies have shown that under the same system conditions, RSMA technology outperforms Non-Orthogonal Multiple Access (NOMA) and Space Division Multiple Access (SDMA) in terms of both spectral efficiency and energy efficiency. At the same time, RSMA technology also demonstrates excellent performance in terms of physical layer security. While carrying shared data stream information among users, the common stream can also act as artificial noise to reduce the decoding ability of eavesdroppers, thereby ensuring the security performance of the ISAC system.

[0004] Currently, many studies have been conducted on resource allocation for ISAC systems based on RSMA. ZHAO B et al. published an article entitled "RSMA-enhanced physical layer security for ISAC systems" in IEEE Wireless Communications Letters, 2025, 14(4):1064-1068, which enhanced the physical layer security of ISAC systems using RSMA technology. By jointly optimizing the coding matrix and common rate allocation, the communication-awareness weighted security total rate was maximized. QIAN D et al. published an article entitled "Beamforming for secureRSMA-aided ISAC systems" in IEEE Transactions on Cognitive Communications and Networking, 2025, 11(5):2970-2983, which studied an ISAC security system based on RSMA, considering both known and unknown scenarios of eavesdroppers, and maximizing system energy efficiency under CRB constraints. As can be seen from the above research status, most existing research on RSMA-based ISAC security systems considers a single eavesdropper and assumes that the eavesdropping channel state information is known or perfectly obtainable. There is insufficient research on the complex security design for imperfect eavesdropping CSI. Therefore, for RSMA-ISAC systems, considering the actual eavesdropping scenario with multiple eavesdroppers, researching the method of maximizing the confidentiality rate of ISAC systems based on RSMA assistance under imperfect CSI has important practical application value and significance. Summary of the Invention

[0005] This invention aims to solve the problems of the prior art mentioned above, and proposes a method for maximizing the security rate of an ISAC system based on RSMA assistance under imperfect CSI. The technical solution of this invention is as follows:

[0006] A method for maximizing the security rate of an ISAC system based on RSMA assistance under imperfect CSI, characterized by the following steps:

[0007] Step 1) Dual-function single base station usage The uniform linear array of the transmitting antennas is ground-based. Each single-antenna user provides communication services, while using A uniform linear array of root receiving antennas receives echo signals from a single sensing point target. Multiple single-antenna eavesdroppers attempt to eavesdrop on messages from legitimate users. Considering the impact of the eavesdroppers' imperfect CSI on the ISAC system, an RSMA-assisted model for maximizing the security rate of the ISAC system under imperfect CSI is established.

[0008] Step 2) Using the SCA method, the mixed integer non-convex objective function in the original optimization problem is linearized to obtain an iteratively solvable convex approximation. The original optimization problem is then transformed into two subproblems: solving the beamforming vector, the public security rate allocation, and the radar receiving filter. These subproblems are then solved iteratively using an alternating optimization approach.

[0009] Step 3) Fix the radar receiving filter, introduce slack variables, and combine the SDR and S-Procedure methods to jointly solve the beamforming vector and public security rate allocation.

[0010] Step 4) Fix the beamforming vector and construct an optimization subproblem to maximize the radar echo signal-to-noise ratio. Obtain the closed-form solution of the radar receiving filter based on the generalized Rayleigh quotient criterion.

[0011] Step 5) Convergence judgment of the security rate update: If the absolute value of the difference between the updated security rate and the previous security rate is not greater than the algorithm tolerance factor, the security rate is judged to be converged, the maximum security rate is given, and the method ends; if the absolute value of the difference between the two security rates is greater than the algorithm tolerance factor, the current security rate value is saved, and the process jumps to step 2) until the security rate meets the condition and the maximum security rate is given.

[0012] Furthermore, in step 1), the security rate maximization model P1 of the ISAC system based on RSMA assistance under imperfect CSI is established as follows:

[0013]

[0014] in Represents the public-private rate allocation vector. This represents the transmit beamforming matrix of the base station. The beamforming vector for the public message. For the first Beamforming vectors for individual user private messages The beamforming vector representing the radar message. For the set of legitimate users, specifically ; A group of eavesdroppers, specifically , Represents the system's security rate. Indicates assignment to the first Public confidentiality rate for each user This represents the lower bound of the common rate that all users can reach. This indicates the achievable rate at which an eavesdropper can decode public messages. The error in eavesdropping channel vector estimation is expressed as follows: , This indicates the upper bound of the error norm. Describes the uncertain set of eavesdropping channels. This indicates the radar echo SNR received by the base station. This indicates the preset threshold for radar echo SNR. The maximum transmit power of the base station is represented by the beamforming vector of the base station. The optimization variable for problem P1 is the base station beamforming vector. , , Public-private rate allocation vector and radar receiving filter Constraint C1 ensures that the sum of the public confidentiality rates obtained by each user is less than the system's public confidentiality rate; constraint C2 ensures that an eavesdropper cannot steal all confidential information in the public data stream; constraint C3 is a non-negative constraint on the allocation of public confidentiality rates; constraint C4 guarantees that the radar's detection performance is not lower than a minimum threshold. Constraint C5 is the base station transmit power budget, ensuring that the maximum transmit power of the base station does not exceed [the specified limit]. .

[0015] Furthermore, in step 2), the objective function is first expanded into a convex form, and based on successive convex approximation, the non-convex part is linearized using a lower bound surrogate function. The original optimization problem is transformed into a two-part alternating optimization problem of solving the beamforming vector, the common security rate allocation, and the radar receiving filter. The first part is to optimize the beamforming vector and the common security rate allocation vector; the second part is to update the radar receiving filter according to the generalized Rayleigh quotient criterion, given the beamforming vector.

[0016] Furthermore, in step 3), with the radar receiving filter fixed, the beamforming vector and the common security rate allocation vector are solved by processing the objective function and constraints of the original problem using methods such as SCA, SDR, and S-Procedure. Specifically, this is the optimization problem P2:

[0017]

[0018] Specifically, the function expressions involved in the rewritten objective function and constraints are as follows:

[0019]

[0020] in, It is the upper bound that the eavesdropper can reach regarding the rate of private messages; It is the upper bound that the eavesdropper can reach regarding the rate of public messages; It is a set of slack variables. It is the set of auxiliary variables introduced by the S-Procedure method when dealing with quadratic inequality constraints. This represents the product of the beamforming vector of the public message and its conjugate transpose. This represents the product of the beamforming vector of the private message and its conjugate transpose. This represents the product of the beamforming vector of the radar message and its conjugate transpose. , and The function is for the user Private messages, eavesdroppers about private messages and users The expansion of the original non-convex rate function of the public message is based on the lower bound function obtained after SCA, where , and It is the affine function after linearization. , and All are linearized auxiliary variables, in the first... In each alternating iteration, to ensure close similarity, the optimal solution of the beamforming matrix from the previous iteration is substituted in. Update auxiliary variables , and , Indicates base station to user Channel gain, Represent a An identity matrix of order 1. and They represent A column vector with 1 row and 1 column and 1 row The row vector of the column, This indicates the noise power at the user's receiver. This indicates the noise power at the receiver of the eavesdropper. This indicates the noise power at the radar receiver. Represents the target radar cross-section coefficient. This represents the direction vector of the receiving array. This represents the direction vector of the transmission array.

[0021] Furthermore, in step 4), with the beamforming vector fixed, the radar receiving filter is updated, specifically to optimize problem P3:

[0022]

[0023] in, express An identity matrix of order 1, based on the optimal solution of the beamforming vector obtained in the previous iteration. ,like Beamforming vectors can be extracted through eigenvalue decomposition. Otherwise, it can be obtained through singular value decomposition. The approximate solution, based on the assumption that only noise exists in the sensing environment, and according to the generalized Rayleigh quotient criterion, yields the update expression for the radar receiving filter, specifically: .

[0024] Furthermore, in step 5), the comparison... With algorithm tolerance factor The size, where, For the iteration number The system's security rate at this level For the () iteration The system security rate is [number] times, if Not greater than If the sum and the secrecy rate converge, the maximum sum and the secrecy rate are given, and the method ends; if Greater than If the current sum and confidentiality rate are saved, then jump to step 3) until the sum and confidentiality rate meet the condition, and give the maximum sum and confidentiality rate.

[0025] The advantages and beneficial effects of this invention are as follows:

[0026] This invention combines RSMA with ISAC systems, proposing a scheme to maximize the system's security rate under conditions of multiple eavesdroppers and imperfect eavesdropping CSI. This is achieved by jointly optimizing the base station transmit beamforming vector, common security rate allocation, and radar receive filter. The innovation of this invention lies in introducing RSMA technology, with its flexible interference management capabilities, into ISAC systems with imperfect eavesdropping CSI. It systematically studies the physical layer security issues and proposes a robust and security rate maximization method, achieving an effective trade-off between the dual requirements of sensing and secure communication. The ingenuity of this invention lies in fully leveraging the dual attributes of the RSMA common flow: it carries shared information from legitimate users while simultaneously suppressing eavesdroppers, thus significantly improving the system's security robustness under channel uncertainty conditions without additional transmit power consumption. Compared to traditional NOMA-ISAC and SDMA-ISAC schemes, this invention offers a significant improvement in security rate, demonstrating better practicality and feasibility. Attached Figure Description

[0027] Figure 1 This is a schematic diagram of an ISAC system model based on RSMA assistance under the condition of imperfect CSI, as provided in the preferred embodiment of the present invention;

[0028] Figure 2 This is the iterative convergence graph of the confidentiality rate of this invention;

[0029] Figure 3 This is a diagram showing the security rate of the present invention under different radar SNR thresholds.

[0030] Figure 4 This is a schematic diagram of the process of the present invention. Detailed Implementation

[0031] The technical solutions of the embodiments of the present invention will be clearly and thoroughly described below with reference to the accompanying drawings. The described embodiments are merely some embodiments of the present invention.

[0032] The technical solution of the present invention to solve the above-mentioned technical problems is:

[0033] This invention discloses a method for maximizing the security rate of an ISAC system based on RSMA assistance under imperfect CSI. It includes the following steps:

[0034] Step 1: Using a dual-function single base station The uniform linear array of the transmitting antennas is ground-based. Each single-antenna user provides communication services, while using A uniform linear array of root receiving antennas receives echo signals from a single sensing point target. Multiple single-antenna eavesdroppers attempt to eavesdrop on messages from legitimate users. Considering the impact of the eavesdroppers' imperfect CSI on the ISAC system, an RSMA-assisted model for maximizing the security rate of the ISAC system under imperfect CSI is established.

[0035] The second step involves using the SCA method to linearize the mixed-integer non-convex objective function in the original optimization problem to obtain an iteratively solvable convex approximation. The original optimization problem is then transformed into two subproblems: solving the beamforming vector, the public security rate allocation vector, and the radar receiving filter. These subproblems are then solved iteratively using an alternating optimization approach.

[0036] Step 3: Fix the radar receiving filter, introduce relaxation variables, and combine SDR and S-Procedure methods to jointly solve the beamforming vector and the public security rate allocation vector.

[0037] Step 4: Fix the beamforming vector and construct the problem of maximizing the radar echo signal-to-noise ratio. Based on the generalized Rayleigh quotient, obtain the closed-form solution of the radar receiving filter.

[0038] Step 5: Convergence judgment of the security rate update. If the absolute value of the difference between the updated security rate and the previous security rate is not greater than the algorithm tolerance factor, the security rate is judged to have converged, the maximum security rate is given, and the method ends. If the absolute value of the difference between the two security rates is greater than the algorithm tolerance factor, the current security rate value is saved, and the process jumps to step 3 until the security rate meets the condition and the maximum security rate is given.

[0039] Furthermore, the security rate maximization model P1 for the ISAC system based on RSMA assistance under imperfect CSI described in the first step is as follows:

[0040]

[0041] in Represents the public-private rate allocation vector. This represents the transmit beamforming matrix of the base station. The beamforming vector for the public message. For the first Beamforming vectors for individual user private messages The beamforming vector representing the radar message. For the set of legitimate users, specifically ; A group of eavesdroppers, specifically , Represents the system's security rate. Indicates assignment to the first Public confidentiality rate for each user This represents the lower bound of the common rate that all users can reach. This indicates the achievable rate at which an eavesdropper can decode public messages. The error in eavesdropping channel vector estimation is expressed as follows: , This indicates the upper bound of the error norm. Describes the uncertain set of eavesdropping channels. This indicates the radar echo SNR received by the base station. This indicates the preset threshold for radar echo SNR. The maximum transmit power of the base station is represented by the beamforming vector of the base station. The optimization variable for problem P1 is the base station beamforming vector. , , Public-private rate allocation vector and radar receiving filter Constraint C1 ensures that the sum of the public confidentiality rates obtained by each user is less than the system's public confidentiality rate; constraint C2 ensures that an eavesdropper cannot steal all confidential information in the public data stream; constraint C3 is a non-negative constraint on the allocation of public confidentiality rates; constraint C4 guarantees that the radar's detection performance is not lower than a minimum threshold. Constraint C5 is the base station transmit power budget, ensuring that the maximum transmit power of the base station does not exceed [the specified limit]. .

[0042] Furthermore, in the second step, the objective function is first expanded into a convex form, and based on SCA, the non-convex part is linearized using a lower bound surrogate function. The original optimization problem is transformed into a two-part alternating optimization problem of solving the beamforming vector, the common security rate allocation vector, and the radar receiving filter. The first part is to optimize the beamforming vector and the common security rate allocation vector; the second part is to update the radar receiving filter according to the generalized Rayleigh quotient criterion, given the beamforming vector.

[0043] Furthermore, in the third step, with a fixed radar receiving filter, methods such as SCA, SDR, and S-Procedure are used to process the objective function and constraints of the original problem, and then the beamforming vector and common security rate allocation vector are solved. Specifically, this is the optimization problem P2:

[0044]

[0045] Specifically, the function expressions involved in the rewritten objective function and constraints are as follows:

[0046]

[0047] in, It is the upper bound that the eavesdropper can reach regarding the rate of private messages; It is the upper bound that the eavesdropper can reach regarding the rate of public messages; It is a set of slack variables. It is the set of auxiliary variables introduced by the S-Procedure method when dealing with quadratic inequality constraints. This represents the product of the beamforming vector of the public message and its conjugate transpose. This represents the product of the beamforming vector of the private message and its conjugate transpose. This represents the product of the beamforming vector of the radar message and its conjugate transpose. , and The function is for the user Private messages, eavesdroppers about private messages and users The expansion of the original non-convex rate function of the public message is based on the lower bound function obtained after SCA, where , and It is the affine function after linearization. , and All are linearized auxiliary variables, in the first... In each alternating iteration, to ensure close similarity, the optimal solution of the beamforming matrix from the previous iteration is substituted in. Update auxiliary variables , and , Indicates base station to user Channel gain, Represent a An identity matrix of order 1. and They represent A column vector with 1 row and 1 column and 1 row The row vector of the column, This indicates the noise power at the user's receiver. This indicates the noise power at the receiver of the eavesdropper. This indicates the noise power at the radar receiver. Represents the target radar cross-section coefficient. This represents the direction vector of the receiving array. This represents the direction vector of the transmission array.

[0048] Furthermore, the fourth step, under the condition of a fixed beamforming vector, updates the radar receiving filter, specifically addressing optimization problem P3:

[0049]

[0050] in, express An identity matrix of order 1, based on the optimal solution of the beamforming vector obtained in the previous iteration. ,like Beamforming vectors can be extracted through eigenvalue decomposition. Otherwise, it can be obtained through singular value decomposition. The approximate solution, based on the assumption that only noise exists in the sensing environment, and according to the generalized Rayleigh quotient criterion, yields the update expression for the radar receiving filter, specifically: .

[0051] Furthermore, the fifth step of comparison With algorithm tolerance factor The size, where, For the iteration number The system's security rate at this level For the () iteration The system security rate is [number] times, if Not greater than If the sum and the secrecy rate converge, the maximum sum and the secrecy rate are given, and the method ends; if Greater than If the current sum and confidentiality rate are saved, then jump to step 3) until the sum and confidentiality rate meet the condition, and give the maximum sum and confidentiality rate.

[0052] This embodiment describes a method for maximizing the security rate of an ISAC system based on RSMA assistance under imperfect CSI. In the RSMA-ISAC system, the number of legitimate users is... The number of eavesdroppers is The base station is located at (20, 0) meters. Legitimate users are randomly distributed within a circle with a center at (20, 50) meters and a radius of 10 meters. Eavesdroppers are randomly distributed within a circle with a center at (20, 60) meters and a radius of 5 meters. The system noise is... dBm, target radar cross-section factor The path loss exponent and Rice factor for the base station-to-user channel and the eavesdropper channel are respectively... and , dB, channel gain at a reference distance of 1 meter dB, maximum transmit power dBm, radar threshold dB, the maximum normalized channel estimation error of the eavesdropper is defined as .

[0053] In this embodiment, Figure 1 This invention provides a schematic diagram of a secure rate model for an ISAC system based on RSMA assistance under a preferred implementation of imperfect CSI. The diagram shows a dual-function base station... A uniform linear array of transmitting antennas transmits integrated communication and sensing signals. Based on the RSMA transmission protocol, it provides communication services to legitimate users on the downlink while simultaneously sensing individual point targets. A uniform linear array of receiving antennas receives echo signals from the target, while unauthorized cooperating eavesdroppers in the system are randomly distributed in an attempt to steal confidential information from the transmitted signals. Figure 2 This is the iterative convergence graph of the confidentiality rate of this invention; Figure 3 This chart compares the encryption rate of the method in this embodiment with that of NOMA and SDMA under different radar echo SNR thresholds. Figure 2 as well as Figure 3As can be seen from the above, compared with the NOMA and SDMA protocols, the method in this embodiment achieves a higher security and confidentiality rate and stronger robustness in the ISAC network. Therefore, the method in this embodiment can provide a more secure guarantee for the entire system while ensuring the basic requirements of perception. Figure 4 This is a flowchart of the present invention.

[0054] The systems, devices, modules, or units described in the above embodiments may be implemented primarily by a computer chip or entity, or by a product having a certain function.

[0055] The above embodiments should be understood as illustrative only and not as limiting the scope of protection of the present invention. After reading the description of the present invention, those skilled in the art can make various alterations or modifications to the present invention, and these equivalent changes and modifications also fall within the scope defined by the claims of the present invention.

Claims

1. A method for maximizing the security rate of an ISAC system based on RSMA assistance under imperfect CSI, characterized in that, Includes the following steps: Step 1) Dual-function single base station usage The uniform linear array of the transmitting antennas is ground-based. Each single-antenna user provides communication services, while using A uniform linear array of root receiving antennas receives echo signals from a single sensing point target. Multiple single-antenna eavesdroppers attempt to eavesdrop on messages from legitimate users. Considering the impact of the eavesdroppers' imperfect CSI on the ISAC system, an RSMA-assisted model for maximizing the security rate of the ISAC system under imperfect CSI is established. Step 2) The mixed integer non-convex objective function in the original optimization problem is linearized using the SCA method to obtain an iteratively solvable convex approximation. The original optimization problem is then transformed into two subproblems: solving the beamforming vector, the public security rate allocation vector, and the radar receiving filter. The problem is then solved iteratively using an alternating optimization method. Step 3) Fix the radar receiving filter, introduce slack variables and auxiliary variables, and combine SDR and S-Procedure methods to jointly solve the transmit beamforming vector and the common security rate allocation vector; Step 4) Fix the beamforming vector and construct the problem of maximizing the radar echo signal-to-noise ratio. Based on the generalized Rayleigh quotient, obtain the closed-form solution of the radar receiving filter. Step 5) Convergence judgment of the security rate update: If the absolute value of the difference between the updated security rate and the previous security rate is not greater than the algorithm tolerance factor, the security rate is judged to be converged, the maximum security rate is given, and the method ends; if the absolute value of the difference between the two security rates is greater than the algorithm tolerance factor, the current security rate value is saved, and the process jumps to step 3) until the security rate meets the condition and the maximum security rate is given.

2. The method for maximizing security of an ISAC system based on RSMA assistance under imperfect CSI as described in claim 1, characterized in that, In step 1), the security rate maximization model P1 for the ISAC system based on RSMA assistance under imperfect CSI is established as follows: in Represents the public-private rate allocation vector. This represents the transmit beamforming matrix of the base station. The beamforming vector for the public message. For the first Beamforming vectors for individual user private messages The beamforming vector representing the radar message. For the set of legitimate users, specifically ; A group of eavesdroppers, specifically , Represents the system's security rate. Indicates assignment to the first Public confidentiality rate for each user This represents the lower bound of the common rate that all users can reach. This indicates the achievable rate at which an eavesdropper can decode public messages. The error in eavesdropping channel vector estimation is expressed as follows: , This represents the upper bound of the error norm. Represents an uncertain set of eavesdropping channels. This indicates the radar echo SNR received by the base station. This indicates the preset threshold for radar echo SNR. The maximum transmit power of the base station is represented by the beamforming vector of the base station. The optimization variable for problem P1 is the base station beamforming vector. , , Public-private rate allocation vector and radar receiving filter Constraint C1 ensures that the sum of the public confidentiality rates obtained by each user is less than the system's public confidentiality rate; constraint C2 ensures that an eavesdropper cannot steal all confidential information in the public data stream; constraint C3 is a non-negative constraint on the allocation of public confidentiality rates; constraint C4 guarantees that the radar's detection performance is not lower than a minimum threshold. ; Constraint C5 is the base station transmit power budget, ensuring that the maximum transmit power of the base station does not exceed [the specified limit]. .

3. The method for maximizing the security rate of an ISAC system based on RSMA assistance under imperfect CSI as described in claim 2, characterized in that, In step 2), the objective function is first expanded into a convex form, and the non-convex part is linearized using a lower bound surrogate function based on SCA. The original optimization problem is transformed into a two-part alternating optimization problem of solving the beamforming vector, the common security rate allocation vector, and the radar receiving filter. The first part is to optimize the beamforming vector and the common security rate allocation vector; the second part is to update the radar receiving filter according to the generalized Rayleigh quotient criterion, given the beamforming vector.

4. The method for maximizing the security rate of an ISAC system based on RSMA assistance under imperfect CSI as described in claim 3, characterized in that, Step 3) specifically includes: with a fixed radar receiving filter, using methods such as SCA, S-Procedure, and SDR to process the objective function and constraints of the original problem, and then solving for the beamforming vector and the common security rate allocation vector, specifically optimization problem P2: Specifically, the function expressions involved in the rewritten objective function and constraints are as follows: in, It is the upper bound that the eavesdropper can reach regarding the rate of private messages; It is the upper bound that the eavesdropper can reach regarding the rate of public messages; It is a set of slack variables. It is the set of auxiliary variables introduced by the S-Procedure method when dealing with quadratic inequality constraints. This represents the product of the beamforming vector of the public message and its conjugate transpose. This represents the product of the beamforming vector of the private message and its conjugate transpose. This represents the product of the beamforming vector of the radar message and its conjugate transpose. , and The function is for the user Private messages, eavesdroppers about private messages and users The expansion of the original non-convex rate function of the public message is based on the lower bound function obtained after SCA, where , and It is the affine function after linearization. , and All are linearized auxiliary variables, in the first... In each alternating iteration, to ensure close similarity, the optimal solution of the beamforming matrix from the previous iteration is substituted in. Update auxiliary variables , and , Indicates base station to user Channel gain, Represent a An identity matrix of order 1. and They represent A column vector with 1 row and 1 column and 1 row The row vector of the column, This indicates the noise power at the user's receiver. This indicates the noise power at the receiver of the eavesdropper. This indicates the noise power at the radar receiver. Represents the target radar cross-section coefficient. This represents the direction vector of the receiving array. This represents the direction vector of the transmission array.

5. The method for maximizing the security rate of an ISAC system based on RSMA assistance under imperfect CSI as described in claim 4, characterized in that, Step 4) specifically includes: updating the radar receiving filter with a fixed beamforming vector, specifically optimizing problem P3: in, express An identity matrix of order 1, based on the optimal solution of the beamforming vector obtained in the previous iteration. ,like Beamforming vectors can be extracted through eigenvalue decomposition. Otherwise, it can be obtained through singular value decomposition. The approximate solution, based on the assumption that only noise exists in the sensing environment, and according to the generalized Rayleigh quotient criterion, yields the update expression for the radar receiving filter, specifically: .

6. The method for maximizing the security rate of an ISAC system based on RSMA assistance under imperfect CSI as described in claim 5, characterized in that, Step 5) specifically involves: comparing... With algorithm tolerance factor The size, where, For the iteration number The system's security rate at this level For the () iteration The system security rate is [number] times, if Not greater than If the sum and the secrecy rate converge, the maximum sum and the secrecy rate are given, and the method ends; if Greater than If the current sum and confidentiality rate are saved, then jump to step 3) until the sum and confidentiality rate meet the condition, and give the maximum sum and confidentiality rate.

Images