An isac system security transmission method for illegal ris scene

Abstract

The application discloses an ISAC system security transmission method for an illegal RIS scene. The illegal intelligent reconfigurable surface (IRIS) is derived from the RIS, and the deployment of the IRIS brings potential security risks. To this end, a security transmission method for an ISAC system in an illegal RIS scene is built, the security communication is assisted by using sensing technology, and the security threat brought by the IRIS is alleviated. In the method, an ISAC base station (BS) first emits an omnidirectional beam to the surrounding environment, then analyzes the received echo signals by using Capon and approximate maximum likelihood (AML) technologies, and estimates the position parameters of potential targets. Subsequently, a weighted optimization problem is constructed, the normalized Cramer-Rao bound (CRB) of the target angle of departure (AoD) is minimized and the normalized secrecy rate is maximized under the condition of meeting the power and beam pattern constraint conditions. It is verified that the method can effectively resist the eavesdropping threat of the IRIS.

Description

Technical Field

[0001] This invention relates to a secure transmission method for an ISAC system in illegal RIS scenarios, belonging to the field of wireless communication technology. Background Technology

[0002] In recent years, driven by the demand for high-precision sensing in fields such as autonomous driving and smart homes, the industry has begun to explore Integrated Sensing and Communication (ISAC) systems. These systems allow communication and radar sensing to share spectrum resources on a unified platform. By coordinating communication and radar sensing functions, ISAC systems can significantly improve spectrum and hardware efficiency, while also providing additional coordination gains.

[0003] Despite the performance improvements offered by ISAC systems compared to traditional communication systems, they still face the same data transmission security challenges. The security of wireless transmission systems depends on the channel gain difference between the base station (BS) and legitimate users / eavesdroppers. To enhance physical layer security (PLS), techniques such as artificial noise (AN) jamming and relaying have been developed to increase the gain difference. However, these methods lead to unsustainable energy consumption and high hardware costs, while their impact on system security performance is limited, especially in harsh wireless propagation environments.

[0004] Recent studies have found that deploying Reconfigurable Intelligent Surfaces (RIS) in ISAC systems can significantly improve security performance. However, the benefits of RIS can also be exploited by malicious actors. Specifically, attackers can use unauthorized RIS (IRIS) to assist in jamming attacks, which can severely degrade system security. For example, in a fully passive jammer based on DIRS, the jammer can severely disrupt the normal operation of a communication system; in schemes based on AN jamming to suppress useful signal leakage, the location information of potential targets cannot be captured, limiting the overall performance of the system.

[0005] Therefore, our goal is to invent a secure transmission method based on the ISAC system. Summary of the Invention

[0006] The purpose of this invention is to address the shortcomings and deficiencies of existing technologies by proposing a secure transmission method for ISAC systems in illegal RIS scenarios. This method utilizes the sensing capabilities of the ISAC system to estimate the position parameters of potential targets. Based on the estimated position information, a weighted optimization problem is formulated to adjust the transmit beamforming matrix of the BS (Band of Battle) to minimize the normalized cramér-Rao bound (CRB) of the target's deviation angle (AoD) and maximize the normalized security rate, while satisfying power and beam pattern constraints. The ISAC system achieves synergy between communication and radar sensing by sharing spectrum resources, improving spectrum and hardware efficiency. However, the ISAC system still faces challenges in data transmission security, especially when illegally reconfigurable smart surfaces (IRIS) are maliciously exploited. IRIS may assist in jamming attacks or eavesdropping activities, seriously threatening system security. Traditional physical layer security technologies (such as artificial noise) suffer from high energy consumption, high hardware costs, and limited effectiveness. Therefore, we propose using the sensing capabilities of the ISAC system to estimate the position parameters of potential threats (such as IRIS and eavesdroppers), and based on this, design a secure transmission method to improve communication security and radar sensing performance.

[0007] The technical solution adopted by this invention to solve its technical problem is: a secure transmission method for an ISAC system in an illegal RIS scenario, the method comprising the following steps:

[0008] Step 1: Establish a system model including an ISAC BS, UAV, IRIS, Target / Eve, and users. The ISAC BS is equipped with M transmit and receive antennas, and each antenna is a uniform linear array with half-wavelength spacing. The BS serves K users and simultaneously estimates the position parameters of two potential targets: a hovering UAV and an Eve. The UAV is equipped with N IRIS to assist in eavesdropping activities.

[0009] Step 2: The base station transmits an omnidirectional radar beam covering potential target areas around the base station, including legitimate users, illicit reconfigurable smart surfaces (IRIS), and eavesdroppers (Eve). The base station receives echo signals from potential targets and preprocesses the echo signals, including noise filtering and signal normalization.

[0010] Step 3: Estimate the position parameters of the potential target using Capon and Approximate Maximum Likelihood Ratio (AML) techniques. Based on the estimated target position parameters, design a weighted optimization problem, adjust the beamforming matrix, and simultaneously minimize the normalized cramér-Rao bound (CRB) of the target angle and maximize the normalized security ratio.

[0011] Step 4: Introduce an auxiliary variable J to represent the lower bound of the CRB, introduce v1 and v2 to handle the non-convexity of the confidentiality rate term, and use Schur complement to transform the positive semidefinite constraint into a simpler form.

[0012] Step 5: Optimize the auxiliary variables v1 and v2 and the beamforming matrix alternately. Then use SDR relaxation techniques The rank constraint is used to transform the problem into a convex problem, which is then solved using CVX. Finally, simulation and analysis are performed to determine how to design the ISAC system to achieve better security performance.

[0013] Furthermore, step 1 of this invention establishes an ISAC system consisting of an ISAC BS with M transceiver antennas, a UAV equipped with Ne-element IRIS, a sensing target (eavesdropper), and K single-antenna users. This ISAC BS simultaneously performs the tasks of communicating with users and sensing the location parameters of two potential targets. The dual-function signal transmitted by this model BS in the Lth time slot is:

[0014] X = W c S c +w r s r =WS (1)

[0015] in Indicates the transmitted waveform. The dual-function beamforming matrix W is represented by the j-th column w. j . Indicates user The data stream S is k = 1, ..., K. The K+1 rows of S correspond to sensing symbols. Assuming the data stream, sensing symbols, and their intersections are approximately orthogonal, we can obtain SS. H ≈LI K+1 Therefore, the signal matrix received at the k-th user can be represented as:

[0016] Y c =H c X+Z c (2)

[0017] in Let represent the communication channel matrix, assuming that the matrix is ​​known at the BS point and each component is independently distributed. It is an additive white Gaussian noise (AWGN) matrix with variance of Based on a typical millimeter-wave channel model, we assume that the channel for the k-th user follows a slow, time-varying block fading Rician fading channel, expressed as:

[0018]

[0019] Where, k k h is the Rician factor for the k-th user. LoS,k =a(θ) k,0 ) represents the most deterministic line-of-sight (LoS) channel vector, a(θ) k,0 ) represents the array steering vector. The AoD represents the LoS from BS to user k. It is the multipath scattering channel vector, where L p Indicates the number of propagation paths, This is the AoD associated with the (k,l)th scattering path. Therefore, the SINR of the kth user is given as:

[0020]

[0021] in,

[0022] Furthermore, step 2 of this invention analyzes the system from a radar positioning perspective. The target direction is determined by θ. BT =[θ BE ,θ BU ] T ,θ BE This represents the AoD between BS and Eve / UAV. This paper defines θ. BT =θ = [θ1, θ2] T The echo received by the BS array from the target is written as:

[0023]

[0024] Where Λ(θ)=diag(β(θ)),β(θ)=[β(θ1),β(θ2)] T This represents the complex amplitude of the reflected signal from the target, which is proportional to the radar cross section (RCS). It consists of noise terms, clutter reflected beyond θ, and additional interference components. This paper considers Z... r Each column is independent and identically distributed, with zero mean and covariance matrix. A circularly symmetric complex Gaussian random vector.

[0025] A r (θ)=[a r (θ1),a r [(θ2)],A t (θ)=[a t (θ1),a t [(θ2)],a r (θ i ) and a t (θ i ) indicates that the receiving and transmitting arrays are at θ iThe steering vector in the direction is defined in this paper as:

[0026]

[0027] In the formula for the echo received from the target by the BS array, the time of arrival (ToAs) is known a priori. This data model can be extended to the case where ToAs is unknown.

[0028] Furthermore, step 3 of this invention estimates the position parameters of the potential target using Capon and Approximate Maximum Likelihood Ratio (AML) techniques. Before estimating the target's position parameters, the ISAC BS transmits an omnidirectional radar beam with a covariance matrix. To detect the target. Then, a Capon beamformer is used to process the received echo to obtain the output spatial spectrum.

[0029]

[0030] Subsequently, a spectral peak search is performed, where the peak index corresponds to the estimated target AoDs, using... This indicates that, to avoid false peaks, the search interval is set to... Δθ represents the possible prior estimation error. The Approximate Maximum Likelihood (AML) method is used to estimate... The target amplitude β is estimated to be:

[0031]

[0032] Vecd(·) denotes a column vector formed by the diagonal elements of a matrix, and:

[0033]

[0034] in It is the covariance of the observed data sample.

[0035] Since the distance between the drone and the IRIS is much smaller than the distance between the base station and the drone, we approximate the AoD between the base station and the drone to be the same as the AoD between the base station and the IRIS. Therefore, θ2≈θ BI ,θ BI This represents the AoD between BS and IRIS. The channel matrix of the BS-IRIS link can be represented as:

[0036]

[0037] Where α BIThis represents the large-scale path loss. Once θ1 and θ2 are determined, the AoD between IRIS and Eve can be calculated using geometric relationships. Due to estimation errors, some design beams may leak onto IRIS and then reflect back to Eve. Therefore, the actual SINR received by Eve is SINR. e ≤(SINR e ) UB (SINR) e ) UB Written as:

[0038]

[0039] in, h BE =α BE a t (θ2) represents the channel from BS to Eve, h IE This represents the channel from IRIS to Eve. This represents the IRIS reflection matrix.

[0040] The target estimate is then evaluated using CRB, which provides a lower bound on the variance of the unbiased estimator, while communication security is assessed using the confidentiality rate. The unknown parameters to be estimated are denoted as η = [θ, β]. T , β=[Re{β},Im{β}]. This paper sets A r (θ)=A r ,Λ r (θ)=Λ r A t (θ)=A t , and The Fisher information matrix can be written as:

[0041]

[0042] Each θ i The CRB, as a radar detection performance indicator, can be obtained through calculation:

[0043]

[0044] in,

[0045] F θθ =2LRe(F 11 (18)

[0046] F θβ =2L[Re(F 12 ),-Im(F 12 (19)

[0047]

[0048] CRBs at each angle can be obtained through the covariance matrix. Substituting this, we assume that the probability density function (PDF) of the AoD estimation error for the i-th target follows... The radar beam has a Gaussian distribution. Therefore, the initial main lobe width of the radar beam is: The probability of the target being located within the main lobe is 99.73%.

[0049] The achievable level of security is defined as the difference between the speed of a legitimate user and the speed of an eavesdropper. Therefore, the worst-case security level is:

[0050]

[0051] Ultimately, the weighted problem of communication and sensing is formulated as follows:

[0052] and (R) sr ) UB This is the normalization factor for CRB and security rate. 0 ≤ ρ ≤ 1 is a weighting factor used to balance communication security performance and radar detection performance. γ r This is the sidelobe power threshold, used to limit sidelobe power. S(Ψ) i ) and S(Ω i ) represents the sidelobe beam region Ψ i and the main region Ω i The cardinality. Constraint (23) indicates that it is a positive semidefinite constraint, (24) limits the power to no more than the set total power, (25) limits the sidelobe power, and (26) and (27) enhance the robustness of the system.

[0053] Furthermore, the optimization process for problem P1 in step 4 of this invention is as follows: we introduce an auxiliary variable. definition The transformed problem P1 becomes:

[0054]

[0055] (23)-(27) By applying Schur's algorithm, constraint (29) can be written as:

[0056]

[0057] The nonconvexity of the transformed P1 problem stems from the secrecy rate R. srBased on other literature, we obtained the following: When t = 1 / x, we have -lnx ≥ -tx + lnt + 1. By introducing two auxiliary variables v1 > 0 and v2 > 0, we obtain the following formula:

[0058]

[0059] Based on the above formula, R sr The lower limit was determined to be Problem P2 is further equivalent to:

[0060]

[0061] st(23)-(27)(29)

[0062] Furthermore, in step 5 of this invention, based on the proposed algorithm, since in problem (23) The coupling problem between (v1, v2) and P3 remains nonconvex. To address this issue, we propose an adaptive algorithm based on SDR. We first fix... Optimize (v1,v2). The optimal solution for (v1,v2) can be expressed as:

[0063]

[0064] Then, we fix (v1, v2) and optimize. By relaxing the rank constraint through SDR, problem P3 becomes a convex problem, which can be solved efficiently using CVX.

[0065] Beneficial effects:

[0066] 1. This invention verifies the effectiveness of the proposed sensing-assisted security beamforming design. This design effectively mitigates the eavesdropping threat posed by IRIS while achieving a good balance between communication and sensing tasks. Furthermore, the introduction of sensing capabilities significantly enhances the system's security performance, especially when the number of IRIS reflector units is large or when random phases are used.

[0067] 2. The sensing-assisted security beamforming design proposed in this invention has significant beneficial effects on improving system security, enhancing radar sensing performance, optimizing resource allocation, and reducing computational complexity. These achievements provide important theoretical support and technical guidance for the deployment and optimization of Integrated Sensing and Communication (ISAC) systems in practical applications. Attached Figure Description

[0068] Figure 1 A flowchart of an entire system provided by the present invention.

[0069] Figure 2This invention provides a model diagram of a secure ISAC system in an illegal RIS scenario.

[0070] Figure 3 The present invention provides an overall beam diagram of a sensing-assisted safety ISAC system.

[0071] Figure 4 This invention provides a convergence graph of the proposed optimization algorithm's performance under various conditional parameters.

[0072] Figure 5 A diagram showing the relationship between the confidentiality rate, the number of IRIS elements, and CRB provided for this invention.

[0073] Figure 6 This invention provides a trade-off between sensing and communication. Detailed Implementation

[0074] To make the objectives, technical solutions, and advantages of this invention clearer, the invention will be described in detail below with reference to the accompanying drawings and specific embodiments.

[0075] like Figure 2 The diagram illustrates a system model comprising an ISAC BS, a UAV, IRIS, a target / Eve, and users. The ISAC BS is equipped with M transmit and receive antennas, each arranged in a uniform linear array with half-wavelength spacing. The BS serves K users while simultaneously estimating the position parameters of two potential targets: a hovering UAV and an Eve. The UAV is equipped with N IRIS to assist in eavesdropping activities.

[0076] In step 1, the dual-function signal transmitted by the model BS in the l-th time slot is: X = W c S c +w r s r =WS. Where Indicates the transmitted waveform. The dual-function beamforming matrix W is represented by the j-th column w. j . Indicates user The data stream S has rows K+1 corresponding to sensing symbols. Assuming the data stream, sensing symbols, and their intersections are approximately orthogonal, we can obtain SS. H ≈LI K+1 .

[0077] Therefore, the signal matrix received at the k-th user can be represented as: Y c =H c X+Z c ,in Representing the communication channel matrix, based on a typical millimeter-wave channel model, we assume that the channel of the k-th user follows a slow time-varying block fading Rician fading channel, expressed as: h LoS,k =a(θ) k,0 ) represents the most deterministic line-of-sight (LoS) channel vector, a(θ) k,0 ) represents the array steering vector, and represents the AoD of the LoS from BS to user k. This is the multipath scattering channel vector. Therefore, the SINR of the k-th user is given as:

[0078]

[0079] In step 2, the system is analyzed from the perspective of radar positioning. The target direction is determined by θ. BT =[θ BE ,θ BU ] T ,θ BE This represents the AoD between BS and Eve / UAV. This paper defines θ. BT =θ = [θ1, θ2] T The echo received by the BS array from the target is written as:

[0080]

[0081] Where Λ(θ)=diag(β(θ)),β(θ)=[β(θ1),β(θ2)] T This represents the complex amplitude of the reflected signal from the target, which is proportional to the radar cross section (RCS). It consists of noise terms, clutter reflected beyond θ, and additional interference components. This paper considers Z... r Each column is independent and identically distributed, with zero mean and covariance matrix. A circularly symmetric complex Gaussian random vector.

[0082] A r (θ)=[a r (θ1),a r [(θ2)],A t (θ)=[a t (θ1),a t [(θ2)],a r (θ i ) and a t (θ i ) indicates that the receiving and transmitting arrays are at θ i The steering vector in the direction is defined in this paper as:

[0083]

[0084] In the formula for the echo received from the target by the BS array, the time of arrival (ToAs) is known a priori. This data model can be extended to the case where ToAs is unknown.

[0085] In step 3, the position parameters of the potential target are estimated using Capon and Approximate Maximum Likelihood Ratio (AML) techniques. Before estimating the target's position parameters, the ISAC BS transmits an omnidirectional radar beam with a covariance matrix. To detect the target. Then, a Capon beamformer is used to process the received echo to obtain the output spatial spectrum.

[0086] Subsequently, a spectral peak search was performed, and the approximate maximum likelihood (AML) method was used for estimation. The target amplitude β is estimated to be: Vecd(·) denotes a column vector formed by the diagonal elements of a matrix, and: in It is the covariance of the observed data sample.

[0087] Due to estimation errors, some of the design beam may leak onto IRIS and then reflect back to Eve. Therefore, the actual SINR received by Eve is SINR. e ≤(SINR e ) UB (SINR) e ) UB Written as:

[0088]

[0089] The target estimate is then evaluated using CRB, which provides a lower bound on the variance of the unbiased estimator, while communication security is assessed using the secrecy rate. The achievable secrecy rate is defined as the difference between the rates of legitimate users and eavesdroppers. Therefore, the worst-case secrecy rate is:

[0090]

[0091] Ultimately, the weighted problem of communication and sensing is formulated as follows:

[0092]

[0093] In step 4, problem P1 is optimized. An auxiliary variable is introduced. Problem P1 is transformed into P2. By applying Schur's algorithm and rewriting constraint C6, the nonconvexity of problem P2 stems from the secrecy rate R. sr Based on other literature, we introduce [a new approach / method]. When t = 1 / x, we have -lnx ≥ -tx + lnt + 1. By introducing two auxiliary variables v1 > 0 and v2 > 0, we can determine R. sr The lower bound is then determined. Problem P2 is further equivalent to P3, and an adaptive algorithm based on SDR is used to find the optimal solution. Problem P3 then becomes a convex problem, which can be efficiently solved using CVX.

[0094] Step 5, based on the proposed algorithm, demonstrates that combining the transmit / receive beamforming matrix of the BS with the RIS phase shift design can achieve better security performance. It also reveals the crucial role of the sensing beamforming matrix design in mitigating the impact of IRIS on radar beamforming design.

[0095] In this invention, we provide a numerical evaluation of the proposed secure beamforming design. Assuming θ1 = 10°, θ2 = 35°, β1 = 3, β2 = 5, L = 64. Assume all users are randomly distributed at a distance of 15m from the base station. The link distances between BS-IRIS and BS-Eve are set to 20m and 25m, respectively. According to the mathematical formula, the AoD between IRIS and Eve is: The distance between IRIS and Eve can then be calculated using a common formula. Assume the large-scale path loss α in this invention... BI and α BE It follows a complex Gaussian distribution with a mean of 0 and a variance of . The received SNR of the echo signal is expressed as: k k =0.1, α and γ r Set it to 0.05.

[0096] The ethical beam diagram of the sensing-assisted safety ISAC system proposed in this invention is as follows: Figure 3 As shown, the communication beam is primarily directed towards the communication user and the sensing target, achieving coordination between communication and sensing tasks. The radar beam is concentrated in the directions of Eve and IRIS, effectively interfering with potential eavesdroppers and reducing the risk of signal leakage. This result verifies that the present invention enhances system security by optimizing the beam direction.

[0097] Figure 4 Simulation results show that the security rate and the root value of CRB converge almost completely within approximately 10 iterations. This indicates that the proposed algorithm has good convergence performance, can reach a stable state within a finite number of iterations, and quickly provides an effective beamforming scheme. Figure 5The results show that as the number of IRIS elements increases, both the system security rate and the root value of the CRB increase. This indicates that increasing the IRIS scale can effectively improve the system's communication security and radar sensing performance. Furthermore, the proposed AO-SDR method performs comparably to the SCA-SDR method, but with lower computational complexity. Figure 6 This demonstrates that communication and sensing tasks are not simply in conflict. Through appropriate beamforming design, effective communication can be achieved while simultaneously improving radar sensing accuracy and enhancing system security. Conversely, the system cannot achieve optimal security when not performing sensing tasks, highlighting the crucial role of sensing in enhancing communication security.

Claims

1. A secure transmission method for an ISAC system in an illegal RIS scenario, characterized in that, Includes the following steps: Step 1: Establish a system model including ISAC BS, UAV, IRIS, Target / Eve, and users. The ISAC BS is equipped with M transmit and receive antennas, and each antenna is a uniform linear array with half-wavelength spacing. The BS provides services to K users. At the same time, estimate the position parameters of two potential targets: hovering UAV and Eve. The UAV is equipped with N IRIS to assist in eavesdropping activities. Step 2: The base station transmits an omnidirectional radar beam that covers the potential target area around the base station, including legitimate users, illegal reconfigurable smart surfaces (IRIS), and eavesdroppers (Eve). The base station receives echo signals from potential targets and preprocesses the echo signals, including noise filtering and signal normalization. Step 3: Estimate the position parameters of the potential target using Capon and Approximate Maximum Likelihood Ratio (AML) techniques. Based on the estimated target position parameters, design a weighted optimization problem, adjust the beamforming matrix, and simultaneously minimize the normalization of the target angle. CRB (Common Restricted Block) and maximizing normalized confidentiality rate; Step 4: Introduce an auxiliary variable J to represent the lower bound of CRB. and To handle the nonconvexity of the confidentiality rate term, Schur's complement is used to transform the positive semidefinite constraint into a simpler form: in definition ; Step 5: Optimize auxiliary variables alternately and and beamforming matrix Then use SDR relaxation to The rank constraint is used to transform the problem into a convex problem, and the optimization results of the system are solved. Finally, the simulation is conducted and the design of the ISAC system is analyzed to achieve better security performance. in and The optimal solution is: , , This indicates the channel from BS to Eve. This represents the channel from IRIS to Eve. Represents the IRIS reflection matrix. This is the channel matrix of the BS-IRIS link. It is the variance of additive white Gaussian noise. , , Indicates the communication channel. The dual-function beamforming matrix W is represented as follows: .

Images