Safety communication perception integrated system optimization method based on rate segmentation multiple access and reconfigurable holographic surface

Abstract

The invention relates to the technical field of wireless communication, in particular to a secure communication perception integrated system optimization method based on rate segmentation multiple access and a reconfigurable holographic surface, which comprises the following steps: constructing a secure communication perception integrated system model; constructing a multi-objective joint optimization problem; a multi-objective joint optimization problem is decomposed into three sub-problems, and solving algorithms are designed respectively; and sequentially updating the solutions of the three sub-problems in an alternate iteration mode until a convergence condition is met, outputting an optimal parameter, and carrying out multi-dimensional verification on the system performance. According to the invention, through a rate segmentation multiple access public stream-private stream layered coding mechanism, the system can intelligently convert part of signal energy into man-made interference to potential eavesdroppers, and at the same time, a high-dimensional beam forming degree of freedom provided by a reconfigurable holographic surface is utilized; accurate orientation of legal user signals and accurate control of null depth of an eavesdropping direction are realized.

Description

Technical Field

[0001] This invention relates to the field of wireless communication technology, and in particular to an optimization method for a secure communication sensing integrated system based on rate segmentation multiple access and reconfigurable holographic surfaces. Background Technology

[0002] The development of sixth-generation mobile communication networks is driven by emerging applications such as smart homes and the Industrial Internet of Things (IIoT). These applications not only require high-speed data transmission but also high-precision environmental sensing capabilities. This exacerbates long-standing challenges such as spectrum scarcity and energy efficiency. In response, integrated sensing and communication technologies have become a key enabling technology for 6G networks, capable of providing communication and sensing services simultaneously through shared hardware and spectrum resources. Compared to traditional systems, integrated sensing and communication systems offer significant gains in spectrum utilization, energy efficiency, and cost-effectiveness.

[0003] However, the broadcast nature and openness of wireless channels make sensor-integrated systems particularly vulnerable to eavesdropping, a challenge faced by conventional wireless systems. Using a uniform waveform for sensing and communication further exacerbates the risk of information leakage, especially when the sensed target itself may act as a potential eavesdropper. These inherent vulnerabilities underscore the importance of embedding physical layer security mechanisms in the design of sensor-integrated systems.

[0004] In terms of hardware implementation, advancements in radio frequency microelectromechanical systems (MEMS) have accelerated the adoption of programmable metasurfaces in integrated sensing and communication systems. Among these, smart reflectors with reconfigurable elements have attracted significant attention due to their ability to enhance the performance of wireless systems. Although smart reflector structures are more energy-efficient than traditional phased arrays, their design is still limited by the half-wavelength spacing requirements between elements.

[0005] To overcome these hardware limitations, reconfigurable holographic surfaces have recently been proposed as a novel metamaterial-based antenna technology that alleviates the drawbacks of traditional phased arrays and smart reflectors, such as high power consumption, cost, and large physical size. By controlling the bias voltage of embedded diodes, reconfigurable holographic surfaces can achieve directional beamforming without high-power phase shifters. More importantly, the reconfigurable holographic surface architecture eliminates the requirement for half-wavelength spacing, enabling highly compact and scalable designs. These characteristics make reconfigurable holographic surfaces a promising hardware platform for future sonic systems requiring high-resolution sensing and agile beam management.

[0006] At the signal processing level, Rate Division Multiple Access (RDMA) is widely recognized as a powerful and versatile multiple access technology for multi-antenna multi-user systems, demonstrating significant performance improvements over traditional schemes such as Space Division Multiple Access (SDMA) and Non-orthogonal Multiple Access (NOAMA). The basic principle of RDMA is to split user messages into public and private parts. The public parts are combined into a shared stream that can be decoded by multiple users, while the private parts are encoded separately. This allows for a flexible interference management approach, where some interference is decoded and the remainder is treated as noise, achieving a balance between SDMA and NOAMA. In security-sensitive scenarios where the perceived target may be an eavesdropper, RDMA can improve confidentiality by transmitting non-critical information through the public stream and protecting sensitive data in the private stream. Simultaneously, a dedicated radar beam can detect the target, and both the public and radar beams can act as artificial interference to confuse eavesdroppers.

[0007] Although rate division multiple access (RDMA) and reconfigurable holographic surfaces have made significant progress in their respective fields, innovative design schemes that integrate them synergistically for use in integrated communication and sensing systems remain a gap in current technology. Most publicly available solutions study and apply these two technologies in isolation, failing to fully explore their synergistic benefits in improving communication security and sensing performance.

[0008] This invention addresses this technological gap by creatively integrating the compact wavebase beamforming capability of reconfigurable holographic surfaces with the hierarchical interference management mechanism of rate-division multiple access, thus constructing a novel secure dual-function system architecture. Specifically targeting key challenges in practical applications—including multi-antenna eavesdropping threats and some known target channel conditions—this invention provides an efficient and reliable systematic solution, effectively overcoming the limitations of existing technologies in terms of security, sensing accuracy, and resource utilization. Summary of the Invention

[0009] The purpose of this invention is to provide an optimization method for a secure communication and sensing integrated system based on rate segmentation multiple access and reconfigurable holographic surfaces, so as to improve the problems existing in the prior art.

[0010] This invention is implemented as follows: an optimization method for a secure communication and sensing integrated system based on rate segmentation multiple access and reconfigurable holographic surfaces, comprising the following steps:

[0011] S1: Construct a secure communication and sensing integrated system model of rate segmentation multiple access and reconfigurable holographic surface, and clarify the system hardware composition, signal transmission mechanism and channel uncertainty characterization method;

[0012] S2: Construct a multi-objective joint optimization problem with the goal of jointly maximizing the weighted total security rate and radar sensing mutual information, determine the set of optimization variables and constraints, and use a constrained optimization framework to characterize the Pareto boundary of communication and sensing performance;

[0013] S3: Decompose the multi-objective joint optimization problem into three sub-problems: receiving beamforming optimization, digital transmitting beamforming optimization, and reconfigurable holographic surface simulation beamforming optimization, and design solution algorithms for each sub-problem;

[0014] S4: Update the solutions to the three subproblems sequentially through alternating iterations until the convergence condition is met, output the optimal parameters, and verify the system performance in multiple dimensions.

[0015] More preferably, in step S1, the hardware components of the system model include: equipped with A reconfigurable holographic surface composed of metamaterial radiative units and A base station with a planar receiving antenna array serves K base stations, each equipped with... Secure communication with a root antenna user, simultaneously sensing one equipped with The potential multi-antenna eavesdropping target of the root antenna; the signal transmission mechanism adopts rate division multiple access, which splits the user message into a public part and a private part and encodes them separately. Together with the radar signal, they form a composite transmission vector, which is transmitted after linear precoding and reconfigurable holographic surface beamforming; the channel uncertainty is characterized by a bounded error model, which includes the estimation error of the eavesdropping channel and the target response matrix and the uncertainty radius constraint.

[0016] More preferably, in step S1, the emission signal of the reconfigurable holographic surface satisfies: ,in For holographic beamforming matrix, The radiation amplitude includes all metamaterial units. This is the phase offset matrix. Represents the transpose of a vector. Let x represent the dimension of the digital transmit beam, and x be the baseband transmit signal of the base station; the base station baseband transmit signal satisfies: ,in Represents the precoded vector of a public message. The precoding vector corresponding to user k, Corresponding to the precoding vector of the radar signal, , , These are private messages, public messages, and radar signals, respectively.

[0017] More preferably, in step S2, the set of optimization variables is: ,in For holographic beamforming matrix, , , For the transmitter precoding vector, The beamforming vector at the receiver is defined as follows: the constraints include common message decodeability constraints, reconfigurable holographic surface unit amplitude constraints, base station transmit power constraints, and minimum threshold constraints for perceptual mutual information; the Pareto boundary representation is achieved through two constraint optimization modes: one is to maximize the weighted total security rate under the minimum perceptual mutual information constraint, and the other is to maximize the perceptual mutual information under the minimum weighted security rate constraint.

[0018] More preferably, in step S3, the solution to the receiving beamforming optimization subproblem is as follows: fix the transmitting end precoding vector and the holographic beamforming matrix, and with the goal of maximizing the user signal-to-interference-plus-noise ratio, transform the problem into a generalized Rayleigh quotient optimization, and obtain the closed-form solution of the receiving beamforming vector by solving the generalized eigenvector.

[0019] More preferably, in step S3, the solution process for the digital transmit beamforming optimization sub-problem includes:

[0020] By introducing auxiliary variables and using the generalized S-lemma, the infinite number of inequalities caused by the channel uncertainty set are transformed into a finite number of linear matrix inequalities.

[0021] definition , , The precoded vector is transformed into a positive semidefinite matrix, and a semidefinite relaxation technique is used to ignore the rank-one constraint.

[0022] For non-convex rate functions, a continuous convex approximation method is adopted, which reformulates them as the difference of convex functions, and the concave lower bound is obtained at a given feasible point using a first-order Taylor expansion.

[0023] If the solution obtained after solving the transformed convex semidefinite programming problem does not satisfy the rank-one constraint, then the Gaussian randomization method is applied to find a feasible rank-one solution.

[0024] More preferably, in step S3, the key to solving the beamforming optimization sub-problem of reconfigurable holographic surface simulation includes:

[0025] Set the optimization variable as ,definition Perform singular value decomposition operation: , Let U represent the terms in the objective function and constraints; use a first-order Taylor expansion to obtain the rate function. about The concave lower boundary; using the MM method for The item establishes a lower bound. Describing the 2-norm of a matrix and introducing a penalty factor. Then Substitute the values ​​into the objective function and perform iterative optimization, where... .

[0026] More preferably, in step S4, the convergence condition is that the difference between two adjacent iterations of the objective function is less than a preset precision. Or the number of iterations reaches the maximum number of iterations. The performance verification metrics include weighted total safety rate, radar sensing mutual information, Pareto boundary, robustness, benchmark comparison with traditional schemes, and system parameter impact analysis.

[0027] Compared with the prior art, the present invention has the following advantages:

[0028] 1. This invention achieves deep synergy between rate segmented multiple access (RSMA) and reconfigurable holographic surfaces in a secure sensing integrated system. Unlike existing technologies that study these two technologies in isolation, this invention is the first to deeply integrate the advanced signal processing capabilities of RMA with the beamforming capabilities of reconfigurable holographic surfaces. Through the public-private flow hierarchical coding mechanism of RMA, the system can intelligently convert a portion of the signal energy into artificial interference against potential eavesdroppers. Simultaneously, by utilizing the high-dimensional beamforming freedom provided by the reconfigurable holographic surface, it achieves precise orientation of legitimate user signals and precise control of the null depth of the eavesdropping direction. This innovative hardware and software synergy significantly enhances the system's anti-eavesdropping capability from the signal source, effectively improving security performance.

[0029] 2. Unlike traditional simple weighted sum optimization methods, this invention employs Pareto optimization theory based on constraint transformation, which can fully characterize and realize the optimal performance boundary between the weighted total security rate and radar sensing mutual information. Through two flexibly switchable optimization modes—communication-centralized and radar-centralized—the system can dynamically adjust resource allocation strategies according to real-time task requirements (such as prioritizing confidential communication or high-precision sensing). This design provides the system with unprecedented operational flexibility, enabling it to maintain optimal performance in complex and ever-changing combat environments, solving the core problem of traditional integrated sensing systems struggling to achieve an effective balance between security and sensing.

[0030] 3. This invention addresses the unavoidable channel estimation errors and feedback delays inherent in real-world systems. It employs a bounded error model to characterize channel uncertainty and, based on the worst-case criterion and the generalized S-lemma, transforms the original non-convex problem, composed of infinitely many constraints, into a linear matrix inequality problem that can be efficiently solved using convex optimization tools. This series of mathematical transformations ensures the algorithm's convergence and stability under harsh conditions, enabling the system to exhibit excellent robustness in the face of channel uncertainty. This effectively overcomes the common bottleneck of many existing studies that rely on ideal channel assumptions, thus limiting their practical application value.

[0031] 4. The reconfigurable holographic surface hardware platform upon which this invention is based provides a scalable implementation path for future large-scale MIMO systems due to its characteristics of requiring no half-wavelength spacing, low power consumption, and easy conformal design. Simultaneously, the optimization algorithm framework proposed in this invention has high universality and can be adapted to various key application scenarios, such as ensuring vehicle communication security while performing high-precision perception of the road environment in intelligent transportation systems, or providing anti-interference secure links for ground users during reconnaissance missions in UAV networks. This universal design of hardware and algorithms greatly expands the application boundaries and industrialization prospects of this invention. Attached Figure Description

[0032] Figure 1 This is a flowchart illustrating the overall process of the optimization method for a secure communication and sensing integrated system based on rate segmentation multiple access and reconfigurable holographic surfaces according to the present invention.

[0033] Figure 2 This is a schematic diagram illustrating the rate segmentation multiple access and reconfigurable holographic surface signal generation and demodulation of the present invention;

[0034] Figure 3 A convergence verification diagram for the algorithm proposed in this invention;

[0035] Figure 4 A convergence verification diagram for the algorithm proposed in this invention;

[0036] Figure 5 The simulation verification diagram shows the Pareto boundary obtained by the optimization of this invention.

[0037] Figure 6 The simulation verification diagram shows the Pareto boundary obtained by the optimization of this invention.

[0038] Figure 7 These are simulation verification diagrams of the present invention under different basic parameters;

[0039] Figure 8 The figures show simulation verification of the present invention under different basic parameters. Detailed Implementation

[0040] To make the objectives, technical solutions, and advantages of the embodiments of the present invention clearer, the technical solutions in the embodiments of the present invention will be clearly and completely described below. Obviously, the described embodiments are only some embodiments of the present invention, not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention. Unless otherwise defined, the technical or scientific terms used herein should have the ordinary meaning understood by those skilled in the art. The terms "comprising" and similar expressions used herein mean that the element or object preceding the word covers the element or object listed after the word and its equivalents, but does not exclude other elements or objects.

[0041] Example 1: This example provides an optimization method for a secure communication and sensing integrated system based on rate segmentation multiple access and reconfigurable holographic surfaces, such as... Figure 1 As shown, it includes the following steps:

[0042] S1: Construct a secure communication and sensing integrated system model of rate segmentation multiple access and reconfigurable holographic surface, and clarify the system hardware composition, signal transmission mechanism and channel uncertainty characterization method;

[0043] In some embodiments, in step S1, the hardware components of the system model include: equipped with A reconfigurable holographic surface composed of metamaterial radiative units and A base station with a planar receiving antenna array serves K base stations, each equipped with... Secure communication with a root antenna user, simultaneously sensing one equipped with A potential multi-antenna eavesdropping target with a single antenna, attempting to intercept legitimate communication between a base station and a secure user; the signal transmission mechanism employs rate division multiple access, assigning the k-th antenna to a specific antenna. Individual user messages are split into common parts. With private part , The set of K users, the common part of all users Combined encoding into a single public message , while private parts Encoded independently into corresponding private messages To form a private stream set , , With radar signals Together they form a composite emission vector , This represents the transpose of a vector, processed by a linear preencoder. Transmission following reconfigurable holographic surface beamforming, wherein, Represents the precoded vector of a public message. The precoding vector corresponding to user k, Corresponding to the precoding vector of the radar signal, This indicates the dimension of the digital transmission beam.

[0044] Specifically, base station baseband transmission signal , , , These are private messages, public messages, and radar signals, respectively. The hybrid beamforming in RHS integrates digital and analog methods. Analog-side holographic beamforming is achieved by individually adjusting the bias voltage of each radiating element, thereby controlling the amplitude of the leaked wave and its transmitted signal. (That is, the reconstructed holographic surface emission signal) is represented as ,in M is the holographic beamforming matrix, and M is the number of its array elements (i.e., metamaterial radiating units). Including the radiation amplitude of all metamaterial units on the RHS, The phase shift matrix of the reconfigurable holographic surface represents the phase shift from the feed source to the metamaterial unit; the corresponding signals received by the k-th user and the eavesdropper can be expressed as follows: ; ;in, and These are the channel matrices for the user and the eavesdropper, respectively. and The additive white Gaussian noise from the user and the eavesdropper are respectively. The radar signal received at the base station can be represented as: ,in, It is the target response matrix. It is Gaussian white noise received by the radar.

[0045] In some embodiments, considering the imperfections of CSI (Channel State Information) in real-world systems, the channel uncertainty is characterized using a bounded error model, which includes the estimation error of the eavesdropping channel and the target response matrix, as well as an uncertainty radius constraint. Specifically:

[0046] ;

[0047] ;

[0048] in, and The base station is in a known estimated channel. and This is the corresponding CSI estimation error. Denotes the F-norm of a matrix. and Let radii represent the uncertainty radii of the eavesdropping channel and the echo response matrix, respectively. Therefore, the decoding common stream of the k-th user... and private stream The signal-to-interference-plus-noise ratio can be expressed as follows:

[0049] ;

[0050] ;

[0051] in and These represent the received beamforming vectors when the k-th user decodes the public stream and their own private stream, respectively. This represents Gaussian white noise at the user's location. This represents the interference energy between different users. This indicates the interference energy of the radar. This represents the self-interference cancellation coefficient of the radar. This represents the conjugate transpose of a vector; therefore, the k-th user decodes... and The achievable rates are expressed as follows:

[0052] ;

[0053] ;

[0054] To ensure all users successfully decode the public stream , The rate must not exceed The total common rate Shared among all K users, i.e. ,in This indicates that user k is assigned to transmit public messages. Therefore, the total achievable rate for user k includes this common rate component. and transmission Related private rates .

[0055] Furthermore, in the presence of potential eavesdroppers, a worst-case approach is used to assess eavesdropping capabilities. This assumes the eavesdropper possesses unlimited computing resources and can perfectly eliminate all multi-user interference before decoding the intended signal for the legitimate user. Under this assumption, the communication between the base station and the eavesdropper for decoding the signal intended for the k-th user is... The channel capacity is:

[0056] ;

[0057] in, The interference matrix represents the eavesdropper's eavesdropping rate. It refers to the noise power of the eavesdropper; in radar sensing, a key performance indicator is the reception of radar signals. With the target response matrix Mutual information between them is essential for fully understanding their transmitted waveforms. The integrated sensing base station, given hour and The mutual information between radars is represented as follows:

[0058] ;

[0059] in, It is the interference matrix of radar mutual information. It is the noise power at the base station. This represents the inverse of a matrix.

[0060] S2: Construct a multi-objective joint optimization problem with the goal of jointly maximizing the weighted total security rate and radar sensing mutual information, determine the set of optimization variables and constraints, and use a constrained optimization framework to characterize the Pareto boundary of communication and sensing performance;

[0061] Specifically, in step S2, the set of optimization variables is: ,in For holographic beamforming matrix, , , For the transmitter precoding vector, The beamforming vector at the receiver is defined as follows: the constraints include common message decodeability constraints, reconfigurable holographic surface unit amplitude constraints, base station transmit power constraints, and minimum threshold constraints for perceptual mutual information; the Pareto boundary representation is achieved through two constraint optimization modes: one is to maximize the weighted total security rate under the minimum perceptual mutual information constraint, and the other is to maximize the perceptual mutual information under the minimum weighted security rate constraint.

[0062] Furthermore, the optimization objective is to jointly maximize the weighted total safe rate and radar sensing mutual information under the worst-case scenario of channel uncertainty, which requires the co-design of the receiver beamformer. Digital beamformer And holographic metasurface-based simulated beamformers Considering the norm constraint of channel uncertainty, the robust multi-objective optimization problem can be formulated as follows:

[0063]

[0064] Where the formula and To ensure all users can successfully decode public information and achieve feasible rate allocation, the formula is as follows. The amplitude constraint of the m-th holographic beamforming unit is considered, and the formula... This is the total emission power constraint of the holographic metasurface. Represents the trace of a matrix. It is the set of optimization variables for an optimization problem. In the security system of this invention, the inherent dual-function transmit beamforming technology introduces a fundamental trade-off between communication performance and sensing performance. Accurately characterizing and realizing the optimal operating point in this trade-off is crucial for designing a practical integrated communication and sensing security system—especially when there is incomplete channel state information in the presence of eavesdroppers. However, due to the close coupling between secure communication and sensing performance, the analysis of this trade-off is extremely complex. Its root causes can be traced back to factors such as the selection of performance indicators, system configuration, resource constraints, and the uncertainty characteristics of channel state information. To address this challenge and map the achievable performance region, we adopt the Pareto optimization framework.

[0065] In some embodiments, in the field of multi-objective optimization, the concept of optimality is redefined by Pareto optimality. A solution is Pareto optimal when it cannot be further optimized without compromising at least one other objective. However, since the structure and boundaries of the achievable performance region cannot be predetermined, identifying the complete set of Pareto optimal solutions or determining the optimal trade-offs among the objectives remains a challenge. Therefore, to find the most ideal solution, a systematic and iterative exploration of the solution space is required.

[0066] In some embodiments, to characterize the Pareto boundary of the performance region, we employ a constrained optimization framework. This method mainly includes two variants: one is a modeling method centered on secure communication, aiming to maximize the weighted sum security rate under the constraint of minimum perceived mutual information. The mathematical expression of this modeling method is:

[0067]

[0068] in The threshold requirement for perceptual mutual information;

[0069] Another problem is the optimization problem of the radar center, which aims to maximize the perceived mutual information under the constraint of secure communication performance. Its mathematical expression is as follows:

[0070]

[0071] in For weighted security and rate thresholds.

[0072] In some embodiments, the constraint method employed in this invention can circumvent the scaling problem of weight selection. Furthermore, this method can fully obtain the Pareto boundary.

[0073] S3: Decompose the multi-objective joint optimization problem into three sub-problems: receiving beamforming optimization, digital transmitting beamforming optimization, and reconfigurable holographic surface simulation beamforming optimization, and design solution algorithms for each sub-problem;

[0074] Specifically, in step S3, the solution to the receiving beamforming optimization subproblem is as follows: fix the transmitting end precoding vector and the holographic beamforming matrix, and with the goal of maximizing the user signal-to-interference-plus-noise ratio, transform the problem into a generalized Rayleigh quotient optimization, and obtain the closed-form solution of the receiving beamforming vector by solving the generalized eigenvector.

[0075] Furthermore, fixed digital transmit beamforming Holographic metasurface simulation of beamforming Optimize the received beamforming vector This subproblem is only relevant to the user, and the only optimization variable is maximizing the weighted sum rate. In, and satisfying the common rate constraint, because Since it is a monotonically increasing function, the signal-to-interference-plus-noise ratio (SINR) maximization criterion is used for optimization:

[0076] ;

[0077] By formulating it as a generalized Rayleigh quotient problem, the optimal solution can be directly obtained as:

[0078] ;

[0079] in:

[0080] ;

[0081] ;

[0082] ;

[0083] .

[0084] Specifically, in step S3, the solution process for the digital transmit beamforming optimization sub-problem includes:

[0085] By introducing auxiliary variables and using the generalized S-lemma, the infinite number of inequalities caused by the channel uncertainty set are transformed into a finite number of linear matrix inequalities.

[0086] definition , , The precoded vector is transformed into a positive semidefinite matrix, and a semidefinite relaxation technique is used to ignore the rank-one constraint.

[0087] For non-convex rate functions, a continuous convex approximation method is adopted, which reformulates them as the difference of convex functions, and the concave lower bound is obtained at a given feasible point using a first-order Taylor expansion.

[0088] If the solution obtained after solving the transformed convex semidefinite programming problem does not satisfy the rank-one constraint, then the Gaussian randomization method is applied to find a feasible rank-one solution.

[0089] Furthermore, fixed holographic metasurface simulation beamforming and receiving beamforming Optimize private flow beamforming vector Common Stream Beamforming and radar signal beamforming To facilitate the solution, auxiliary variables are introduced. This transforms the problem into an equivalent form. The core challenge lies in handling the set of channel uncertainties. and The resulting semi-infinite constraint. First, using the generalized S-lemma and matrix inequality transformations, the infinite number of inequalities in the perceptual mutual information constraint and the eavesdropping rate constraint are transformed into a finite number of linear matrix inequalities. For example, the perceptual mutual information constraint can be equivalently transformed into a linear matrix inequality as follows:

[0090] ;

[0091] in, , , , These are the introduced non-negative auxiliary variables. The transformed radar mutual information constraints now include a linear matrix inequality constraint, making it more suitable for algorithm design than the original constraints, which involved an infinite number of constraints and were non-convex. Similarly, the eavesdropping rate constraint can also be addressed by introducing non-negative relaxation variables. Applying the generalized S-lemma, we can transform this into a set of linear matrix inequalities: ;

[0092] in, , It is a positive semidefinite slack variable matrix. However, due to the optimization variables... , and All variables appear in quadratic form, causing the optimization problem to remain non-convex under the combined effect of these variables. Therefore, we reconstruct it as a semidefinite programming problem with rank constraints: Define , , Furthermore, the defined matrix must satisfy the rank-one constraint.

[0093] Next, to address the non-convexity of the objective function and the common rate constraint, a continuous convex approximation method is employed. The rate function... Rephrased as the difference of convex functions: ;

[0094] in, Then, using a first-order Taylor expansion, its concave lower bound at a given feasible point is obtained:

[0095]

[0096] in, It is about optimization variables affine function, , , The first Taylor expansion represents the... The approximate point is then obtained. A semidefinite relaxation technique is then applied to ignore the rank-one constraint, resulting in a convex semidefinite programming problem, which can be efficiently solved using convex optimization tools such as CVX. If the solution obtained using the semidefinite relaxation technique does not satisfy the rank-one constraint, a Gaussian randomization method is applied to find a feasible rank-one solution.

[0097] Specifically, in step S3, the key to solving the beamforming optimization sub-problem of reconfigurable holographic surface simulation includes:

[0098] Set the optimization variable as ,definition Perform singular value decomposition operation: , Let U represent the terms in the objective function and constraints; use a first-order Taylor expansion to obtain the rate function. about The concave lower bound; the MM (Majorization–Minimization) method is used to... The item establishes a lower bound. Describing the 2-norm of a matrix and introducing a penalty factor. Then Substitute the values ​​into the objective function and perform iterative optimization, where... .

[0099] Furthermore, regarding the analog beamforming matrix at the receiving end... Optimization is performed, given the beamforming vector at the transmitter. and the beamforming vector at the receiver By setting the optimization variable to Instead of using directly To facilitate semidefinite programming, we define By performing singular value decomposition: , Radar mutual information constraints and eavesdropping constraints can be converted into:

[0100] ;

[0101] ;

[0102] in, Similarly, by performing singular value decomposition on the correlation matrix, the terms in the objective function and constraints are separated. This is represented by the expression. Then, the rate function is obtained again using a first-order Taylor expansion. about The concave lower bound. For the rank-one constraint, it is equivalently expressed as: ;

[0103] because The existence of the term, the objective function is The variables are not concave. Therefore, we adopt a MM-based approach, as follows: Establish a lower bound for the item:

[0104] ;

[0105] in, To represent the largest eigenvalue of the matrix, we finally introduce a penalty factor. Then Substitute the values ​​into the objective function and perform iterative optimization.

[0106] S4: Update the solutions to the three subproblems sequentially through alternating iterations until the convergence condition is met, output the optimal parameters, and verify the system performance in multiple dimensions.

[0107] Specifically, in step S4, the convergence condition is that the difference between two adjacent iterations of the objective function is less than a preset precision. Or the number of iterations reaches the maximum number of iterations. The performance verification metrics include weighted total safety rate, radar sensing mutual information, Pareto boundary, robustness, benchmark comparison with traditional schemes, and system parameter impact analysis.

[0108] Furthermore, the solutions to the three subproblems are updated sequentially through alternating iterations. The specific process is as follows:

[0109] Initialize the transmit beamforming vector , Simulated beamforming matrix of holographic metasurface Set target precision Maximum number of iterations Iterative index .

[0110] repeat:

[0111] a. Given , , Calculate the optimal receiving beamforming vector based on S31. and .

[0112] b. Given , The convex semidefinite programming problem, after being processed by semidefinite relaxation and continuous convex approximation, is solved in step S32, and the problem is updated. , .

[0113] c. Given , By solving the convex problem in S33 after processing with the penalty function method and the MM method, the update is performed. .

[0114] d. .

[0115] Until the convergence condition meets the target accuracy or .

[0116] Output the optimal solution , .

[0117] This alternating iterative process ensures the monotonic convergence of the objective function and eventually reaches a stable point in the original problem.

[0118] In some embodiments, the final optimized parameters undergo comprehensive performance verification, including but not limited to the following key metrics:

[0119] 1. Weighted Total Security Rate: Evaluates the secure communication performance of the system under different channel uncertainty levels.

[0120] 2. Radar sensing mutual information: assesses the system's ability to detect targets and estimate parameters.

[0121] 3. Pareto Boundary: By solving the communication-centralized or radar-centralized problem and systematically varying the performance threshold... or The Pareto boundary between the weighted total security rate and radar mutual information is plotted to visually demonstrate the performance trade-off between communication and sensing.

[0122] 4. Robustness analysis: Compare the system performance under perfect channel state information and imperfect channel state information conditions to verify the ability of the proposed robust optimization algorithm to combat channel uncertainty.

[0123] 5. Benchmark Comparison: The proposed optimization scheme for a secure communication and sensing integrated system based on rate segmentation multiple access and reconfigurable holographic surface is compared with traditional space division multiple access schemes and traditional planar array schemes, highlighting the synergistic gain of rate segmentation multiple access and reconfigurable holographic surface.

[0124] 6. Parameter Influence Analysis: Analyze the impact of system parameters such as the number of reconfigurable holographic surface units, base station transmit power, and number of receiving antennas on performance.

[0125] Furthermore, simulation results demonstrate that the proposed scheme achieves superior performance in terms of both security rate and sensing accuracy compared to traditional spatial division multiple access (SDMA) and traditional planar array schemes, while maintaining strong robustness under channel uncertainty. This highlights the synergistic gains of integrating rate division multiple access and reconfigurable holographic surfaces into a secure sensing system.

[0126] Example 2: Based on Example 1, and with reference to the accompanying drawings, the technical solution of the present invention will be described in detail. The implementation process of the present invention covers key aspects such as system modeling, algorithm design, and simulation verification.

[0127] I. System Hardware and Signal Processing Flow Implementation

[0128] Hardware platform setup:

[0129] like Figure 2 As shown, the base station employs a millimeter-wave communication platform supporting hybrid beamforming. The digital baseband section is equipped with a multi-channel digital signal processor for generating the rate-division multiple access encoded data stream (common stream). K private streams ) and dedicated radar sensing sequences The radio frequency section passes through Simulated beamforming is achieved using reconfigurable holographic surface units. Each holographic metasurface unit integrates a biasable diode, and its radiation amplitude can be controlled by applying a programmable bias voltage. Each legitimate user is equipped with a multi-antenna receiving module and an integrated signal processing unit, capable of performing continuous interference cancellation technology and first decoding the common stream. Then decode their respective private streams. Control and Feedback: The base station is equipped with a high-performance computing unit to run the alternating optimization algorithm proposed in this invention. The system needs to establish a limited feedback link with the user to obtain an estimate of the channel state information.

[0130] II. Signal Generation and Transmission Process

[0131] Step 1: Based on service requirements, the base station determines the current optimization objective (communication centralization or radar centralization) and the corresponding performance threshold. or ).

[0132] Step 2: Run the alternating optimization algorithm described in this invention to solve for the optimal digital precoder. Holographic beamforming matrix of reconfigurable holographic metasurface and receiving beamformer .

[0133] Step 3: Load the optimized precoded vector and reconfigurable holographic metasurface configuration parameters into the hardware. Digital precoded vector Used in baseband for data streams Weighting is applied. The reconfigurable holographic metasurface control unit calculates the amplitude vector. This is converted into a specific bias voltage and applied to each reconfigurable holographic metasurface unit.

[0134] Step 4: The base station generates and transmits a composite signal. ).

[0135] Step 5: Legitimate user k receives the received signal. And utilize the optimized receiving beamforming vector and Then, the public stream and the private stream are decoded in sequence.

[0136] Step 6: The base station simultaneously receives radar echo signals. It is used for target parameter estimation.

[0137] III. Implementation Details of Core Algorithm

[0138] The alternating optimization algorithm proposed in this invention is the core of the system implementation, and its specific implementation steps are as follows:

[0139] Initialization: Set algorithm parameters: maximum number of iterations = 50, convergence accuracy = 10^(-3), initial value of the penalty factor = 1.0.

[0140] Initialize variables: Randomly generate or initialize the transmit beamforming vector using a classic beamforming algorithm (such as maximum ratio transmission). , The reconfigurable holographic metasurface will be used to simulate the beamforming matrix. Initialize as an all-1 matrix or a random phase matrix.

[0141] Alternating optimization iterative loop (for the t-th iteration):

[0142] A. Update the receiver beamformer : Fix the current iteration , , .

[0143] For each user k, the following can be calculated directly from the closed-form solution: This step is highly efficient and can be completed quickly.

[0144] B. Update the transmit beamformer : Fix other optimization variables. This subproblem is the most complex and needs to be transformed into a convex problem to be solved through the following steps:

[0145] B1. Variable Substitution and Relaxation: Definition , , This elevates the problem to the matrix space.

[0146] B2. Handling Channel Uncertainty: Utilizing a generalized S-Procedure to address channel uncertainty during eavesdropping. and sensing channels The infinite number of constraints can be transformed into a finite number of linear matrix inequalities.

[0147] B3. Handling non-convex objective functions and constraints: User rate and The logarithmic function is used, employing a continuous convex approximation method. A first-order Taylor expansion is performed at the current feasible point to construct its global concave lower bound, thus transforming the non-convex problem into one concerning... , , The convex approximation problem.

[0148] B4. Semidefinite Programming Solution and Gaussian Randomization: Ignoring the Rank Constraint (Rank( )=1, Rank( )=1,Rank( If )=1, the problem is relaxed to a convex semidefinite programming problem.

[0149] B5. Solve the semidefinite programming problem using a convex optimization solver (such as CVX with MOSEK or SDPT3) to obtain the optimal solution. , , If the solution does not satisfy the rank-one constraint, then Gaussian randomization is used: for Eigenvalue decomposition is performed to generate a large number of random vectors following a complex Gaussian distribution. Candidate solutions that optimize the objective function and satisfy all constraints are then selected from these vectors. Finally, an approximately optimal pre-encoding vector that satisfies the rank-one requirement is recovered. .

[0150] C. Update the holographic beamforming matrix of the reconfigurable holographic metasurface. :fixed , This step optimizes the variables into vectors. and define .

[0151] C1. Problem Refactoring: Through singular value decomposition, rewrite the terms in the objective function and constraints as relating to... The linear form of .

[0152] C2. Continuous convex approximation: Similarly, for the rate function with respect to... Perform a first-order Taylor expansion to construct a concave lower bound.

[0153] C3. Handling rank-one constraints: [This likely refers to a specific constraint or rule, but without further context, a precise translation is not possible.] As a penalty term, it is added to the objective function, and the MM algorithm is used to apply it to non-convex terms. Perform a linear lower bound approximation.

[0154] C4. Solving convex problems: The final problem is transformed into solving convex problems. This is a convex semidefinite programming problem, solved using a solver. After solving, [the solution is...]. Perform rank-one factorization to obtain And then update .

[0155] Convergence criterion: Calculate the absolute difference between the weighted total safe rate of the current iteration and the previous iteration. If this difference is less than a preset threshold... or the number of iterations t exceeds If the condition is met, the algorithm terminates and outputs the current optimization variable as the optimal solution.

[0156] Otherwise, let t = t + 1 and return to step A to continue the iteration.

[0157] IV. Typical Application Scenarios and Implementation Examples

[0158] In intelligent transportation systems (ITS) applications within a vehicle-to-everything (V2X) environment, roadside units act as integrated sensing and communication base stations, simultaneously providing communication services to multiple vehicles and sensing the road environment. During implementation, vehicle users are categorized into high-security-requirement users (such as police vehicles) and ordinary users, with different rate weights assigned to each type. When the system detects a suspicious vehicle (potential eavesdropper) approaching a high-security-requirement user, it automatically enhances the beamforming gain pointing towards that user while simultaneously creating a deep null in the direction of the suspicious vehicle, ensuring uninterrupted communication.

[0159] An emergency communication network for unmanned aerial vehicles (UAVs) is constructed, with UAVs equipped with integrated sensing and communication base stations. While conducting reconnaissance and imaging of disaster areas, they also provide communication connections for rescue personnel. In this dynamic scenario, the proposed rapid adaptive mechanism updates the beamforming strategy every 100-200ms. When the UAV detects a new obstacle or moving target, it immediately re-optimizes the beam pattern, ensuring both accurate perception of the new target and reliable communication with ground rescue teams.

[0160] The park security monitoring system deploys multiple integrated sensing base stations within key areas, forming a sensing and communication network covering the entire region. Through collaborative optimization among the base stations, it achieves the tracking and location of intrusion targets while ensuring the confidentiality of internal communications. The system adopts a distributed computing architecture, with each base station independently performing local optimization, and a central server handling global coordination, effectively reducing computational complexity and communication overhead.

[0161] The effects of the present invention will be further illustrated by simulation below:

[0162] (1) Simulation conditions

[0163] In the simulation experiment, it is assumed that the integrated inductive base station is equipped with 8 RF links and a 64-antenna array, and the total transmit power budget of the base station is set to 30 dBm. The reconfigurable holographic surface is composed of 64 metamaterial radiating elements, and the radiation amplitude constraint of each element is... The system serves four legitimate users, each equipped with eight receiving antennas; potential eavesdropping targets are also equipped with eight antennas. The channel model uses Ricean fading, with a path loss exponent of 2.5 and a Ricean factor of 5 dB. The uncertainty radius of the channel estimation error is set to 0.02, and the noise power is -90 dBm. The perceived mutual information threshold and the weighted security rate threshold are dynamically set in the simulation based on the optimization mode.

[0164] (2) Simulation results and data analysis

[0165] Figure 3 and Figure 4 The uncertainty radius is shown under different eavesdropping channels and echo response matrices. and Below is the convergence curve of the system's weighted total security rate as a function of the number of alternating optimization iterations. It can be seen that as the number of iterations increases, the weighted total security rate rises rapidly and tends to stabilize, typically converging within 10–15 iterations, verifying the efficiency and stability of the proposed alternating optimization algorithm. Under perfect channel state information conditions, the system achieves the highest security rate; as channel uncertainty increases, the system performance decreases somewhat, but it still maintains a high level of secure communication capability, demonstrating the effectiveness of the proposed robust optimization algorithm.

[0166] Figure 5 and Figure 6 Pareto boundaries between the weighted total security rate and radar sensing mutual information were plotted under two optimization modes: communication-centralized and radar-centralized. Results show that the proposed rate-segmented multiple access (RSA) and reconfigurable holographic surface collaborative scheme achieves a better trade-off between security rate and sensing accuracy, with its Pareto boundary significantly superior to traditional spatial division multiple access (SDMA) and planar array schemes. Particularly in high-security scenarios, the synergistic effect of the common flow mechanism of RSA and the beamforming degrees of freedom of the reconfigurable holographic surface significantly enhances the system's anti-eavesdropping capability.

[0167] Figure 7 The impact of the number of reconfigurable holographic surface elements on the system's sensing mutual information performance was analyzed. As the number of reconfigurable holographic surface elements increased from 16 to 49, the system's sensing mutual information significantly improved, indicating that more radiating elements provide higher beamforming degrees of freedom, enhancing target detection and parameter estimation capabilities. Simultaneously, with the increase in transmit power, the sensing performance of the reconfigurable holographic surface element was significantly improved, highlighting its advantages in hardware architecture.

[0168] Figure 8 The weighted total security rate of the system was compared under different eavesdropping channel uncertainty radii. As the eavesdropper channel uncertainty increases, the performance of all schemes decreases, but the proposed rate segmentation multiple access (RSMA) and reconfigurable holographic surface schemes show the smallest decrease, demonstrating stronger robustness. Furthermore, introducing a dedicated radar sequence as artificial noise further enhances the system's security performance, especially under high channel uncertainty. Meanwhile, compared to traditional smart reflectors, the reconfigurable holographic surface exhibits superior integrated sensing and communication performance with the same number of elements, highlighting its advantages in hardware architecture.

[0169] The simulation results above fully demonstrate that the optimization method for a secure communication and sensing integrated system based on rate segmentation multiple access and reconfigurable holographic surfaces proposed in this invention is superior to existing technical solutions in terms of secure communication rate, radar sensing accuracy, and system robustness, and has good prospects for engineering applications.

[0170] While embodiments of the present invention have been described in detail above, it will be apparent to those skilled in the art that various modifications and variations can be made to these embodiments. However, it should be understood that such modifications and variations fall within the scope and spirit of the invention as set forth in the claims. Furthermore, the invention described herein may have other embodiments and can be implemented or carried out in various ways.

Claims

1. A security communication and perception integrated system optimization method based on rate division multiple access and reconfigurable holographic surface, characterized in that, The method comprises the following steps: S1: Construct a secure communication and perception integrated system model of rate division multiple access and reconfigurable holographic surface, and determine the system hardware composition, signal transmission mechanism and channel uncertainty representation method; S2: Construct a multi-objective joint optimization problem with the joint maximization of weighted security total rate and radar perception mutual information as the target, determine the optimization variable set and constraint condition, and use the constraint optimization framework to represent the Pareto boundary of communication and perception performance; S3: Decompose the multi-objective joint optimization problem into three sub-problems of receive beamforming optimization, digital transmit beamforming optimization and reconfigurable holographic surface analog beamforming optimization, and design the solving algorithm respectively; S4: Update the solutions of the three sub-problems in turn through alternating iteration until the convergence condition is met, output the optimal parameters, and verify the system performance in multiple dimensions.

2. The method of claim 1, wherein, In step S1, the hardware components of the system model include: equipped with A reconfigurable holographic surface composed of metamaterial radiative units and A base station with a planar receiving antenna array serves K base stations, each equipped with... Secure communication with a root antenna user, simultaneously sensing one equipped with The potential multi-antenna eavesdropping target of the root antenna; the signal transmission mechanism adopts rate division multiple access, which splits the user message into a public part and a private part and encodes them separately. Together with the radar signal, they form a composite transmission vector, which is transmitted after linear precoding and reconfigurable holographic surface beamforming; the channel uncertainty is characterized by a bounded error model, which includes the estimation error of the eavesdropping channel and the target response matrix and the uncertainty radius constraint.

3. The method of claim 2, wherein, In step S1, the emission signal of the reconfigurable holographic surface satisfies: wherein is a holographic beamforming matrix, contains the radiation amplitudes of all metamaterial units, is a phase shift matrix, denotes the transpose of a vector, denotes the dimension of the digital transmit beam, x is the base station baseband transmit signal; the base station baseband transmit signal satisfies: wherein denotes a precoding vector of the common message, denotes a precoding vector corresponding to the user k, denotes a precoding vector corresponding to the radar signal, , , are respectively a private message, a common message, a radar signal.

4. The method of claim 3, wherein, In step S2, the optimization variable set is wherein is a holographic beamforming matrix, , , is a transmit-end precoding vector, is a receive-end beamforming vector; the constraint conditions include a common message decodable constraint, a reconstructable holographic surface unit amplitude constraint, a base station transmit power constraint, and a minimum perceptual mutual information constraint; the Pareto boundary is characterized by two constraint optimization modes: one is to maximize the weighted secure total rate under the minimum perceptual mutual information constraint, and the other is to maximize the perceptual mutual information under the minimum weighted secure rate constraint.

5. The method of claim 4, wherein, In step S3, the solving method of the receive beamforming optimization sub-problem is: fixing the transmit end precoding vector and holographic beamforming matrix, taking the maximum user signal-to-interference-and-noise ratio as the target, converting the problem into a generalized Rayleigh quotient optimization, and obtaining the closed-form solution of the receive beamforming vector by solving the generalized eigenvector.

6. The method of claim 5, wherein, In step S3, the solving process of the digital transmit beamforming optimization sub-problem comprises: Introducing auxiliary variables, converting the infinite number of inequalities brought by the channel uncertainty set into a limited number of linear matrix inequalities by using the generalized S-lemma; Definitions , , transforming the precoding vectors into a semi-definite matrix, and ignoring the rank-one constraint using a semi-definite relaxation technique; Using the continuous convex approximation method for the non-convex rate function, reexpressing it as the difference form of convex functions, and obtaining its concave lower bound at a given feasible point by using the first-order Taylor expansion; Solving the transformed convex semi-definite programming problem, if the obtained solution does not satisfy the rank-one constraint, applying the Gaussian randomization method to find a feasible rank-one solution.

7. The method of claim 6, wherein, In step S3, the key of solving the reconfigurable holographic surface analog beamforming optimization sub-problem comprises: Set the optimization variable as , define , perform singular value decomposition operation: , , express the target function and the terms in the constraints with U; use the first-order Taylor expansion to obtain the rate function The concave lower bound of ; the MM method is used to establish a lower bound for , , represents the 2-norm of the matrix, introduces a penalty factor , then is substituted into the target function for iterative optimization, where .

8. The method of claim 7, wherein, In step S4, the convergence condition is that the difference between the objective functions of two adjacent iterations is less than a preset precision or the number of iterations reaches a maximum number of iterations The performance verification indicators include weighted safety sum rate, radar perception mutual information, Pareto boundary, robustness, benchmark comparison with traditional schemes, and system parameter influence analysis.

Images