Comprehensive sensing and communication integrated optimization method based on reconfigurable antenna

Abstract

The application discloses a sensing and communication integrated optimization method based on a reconfigurable antenna, which takes the trace of the Cramer-Rao bound matrix of multi-sensing target angle estimation as the core optimization target, comprehensively considers the correlation of multi-target estimation, and breaks through the limitation that a traditional proxy index cannot reflect the sensing accuracy limit; a nested optimization framework is adopted to jointly optimize the transmitting and receiving antenna positions and the transmitting precoding matrix, replacing the traditional alternating optimization mode, and eliminating the limitation of precoding fixation on the antenna optimization space; the outer layer adopts a particle swarm algorithm based on backtracking search to improve the global search capability, and the inner layer solves the precoding matrix through convex optimization. On the premise of guaranteeing that all communication user signal-to-interference-and-noise ratios meet the standards, the multi-target joint sensing accuracy is significantly improved, the hardware deployment cost is reduced, and the sensing and communication integrated base station system is suitable for 5G / 6G.

Description

Technical Field

[0001] This application relates to the field of wireless communication. Specifically, it relates to the research of a movable antenna-assisted sensing-integrated system (ISAC), and more specifically, to a sensing-integrated optimization method based on reconfigurable antennas. By using movable antennas at both the co-transmitter end for communication and sensing, and at the receiver end for sensing, a scenario with multiple sensing targets and multiple communication users is constructed. The application explores a sensing-integrated architecture using the Cramer-Rao bound matrix as a benchmark for sensing performance, and solves the problem using a particle swarm optimization algorithm based on backtracking search. The results demonstrate that using movable antennas and an improved algorithm enhances sensing accuracy and improves the overall performance of the sensing system, providing a new reference for communication-sensing fusion. Background Technology

[0002] With the rapid development of sixth-generation wireless communication technology (6G), high capacity and high reliability have become urgent requirements for communication performance. Therefore, massive MIMO technology is considered a key enabling technology. MIMO technology improves the capacity and spectral efficiency of communication systems through the parallel transmission of multiple signals. However, in traditional MIMO, the antenna positions are fixed. Due to multipath effects, users may be in deep fading regions, resulting in degraded communication quality. Furthermore, fixed antennas cannot adapt to dynamic channel changes, and simply increasing the number of antennas leads to huge hardware costs, which also hinders the deployment of massive MIMO to some extent. Therefore, the academic community began to explore new antenna design schemes. Communication systems based on reconfigurable antenna designs have attracted widespread attention, among which movable antenna technology, with its advantages such as increased spatial freedom due to its flexible mobility, is considered a technology with broad application prospects. For mechanically controlled movable antennas, the antenna movement is controlled by flexible cables, and each antenna is connected to a radio frequency chain. Movable antennas dynamically reshape the user's channel by flexibly controlling the movement of individual antennas, improving the channel state where the user is located, helping to adapt to environmental changes, and reducing hardware costs. Furthermore, in the field of sensing, movable antennas have been shown to improve the accuracy of angle estimation. Compared to antenna selection (AS) based schemes, movable antennas offer more flexible control and require fewer antennas.

[0003] Meanwhile, Integrated Communication and Sensing (ISAC) has become one of the key technologies for future 6G. With the rise of technologies such as augmented reality (AR) and autonomous driving, the separate design of sensing and communication cannot simultaneously achieve high data transmission rates and high-precision sensing, and also suffers from issues such as spectrum scarcity. By integrating communication and sensing, communication and sensing signals are jointly designed, providing both data transmission and sensing functions. On the one hand, joint deployment and shared hardware reduce deployment costs and power consumption; on the other hand, with the development and application of millimeter-wave (mmWave) and terahertz (THz) technologies, communication frequency bands are gradually overlapping with radar, making spectrum sharing possible. However, how to comprehensively optimize communication and sensing capabilities to improve the performance of the entire ISAC system remains one of the significant challenges in realizing ISAC technology. Summary of the Invention

[0004] This application aims to address the technical problems of existing movable antenna-assisted ISAC systems, such as inaccurate sensing accuracy metrics, lack of consideration for correlation in multi-target sensing, susceptibility of optimization algorithms to local optima, and constraints on antenna degrees of freedom due to alternating optimization. It provides a sensing-communication integrated system architecture and optimization method based on reconfigurable antennas, which significantly improves the joint sensing accuracy of multiple targets, reduces system hardware costs, and enhances the system's adaptability to dynamic environments while ensuring the quality of service for communication users. To achieve the above objectives, this application adopts the following technical solution.

[0005] In a first aspect, embodiments of this application provide a method for integrated sensing optimization based on a reconfigurable antenna, including:

[0006] S1. Initialize the parameters of the integrated sensing system; the system parameters include at least one of the following: number of communication users, number of sensing targets, number of transceiver antennas, antenna movement range, minimum antenna spacing, communication signal-to-interference-plus-noise ratio threshold, maximum transmit power, noise power, and number of snapshots; generate the field response channel matrix of the communication users and the steering vector of the sensing targets based on the system parameters;

[0007] S2. Taking the trace of the Cramer-Rao bound matrix estimated from the perspective of multiple sensing targets as the sensing performance optimization objective, and combining the communication signal-to-interference-plus-noise ratio constraint, transmit power constraint and antenna position constraint, a joint optimization problem of the transmit and receive antenna positions and the transmit precoding matrix is ​​constructed.

[0008] S3. The joint optimization problem is solved using a nested optimization framework, specifically including: the outer layer iteratively optimizes the position arrangement of the transmitting and receiving antennas based on an improved particle swarm optimization algorithm, and inputs the candidate antenna positions obtained in each iteration into the inner layer; the inner layer solves the optimal transmit precoding matrix for the corresponding candidate antenna positions based on a convex optimization method, and feeds the solution back to the outer layer for calculating the fitness function;

[0009] S4. When the iteration meets the termination condition, the optimal transceiver antenna position arrangement and the optimal transmission precoding matrix are output. The base station performs communication data transmission and target perception tasks simultaneously based on the optimal transceiver antenna position arrangement and the optimal transmission precoding matrix.

[0010] Furthermore, in step S3, the outer layer, based on an improved particle swarm optimization algorithm, generates three candidate antenna positions for each particle in each iteration, specifically including:

[0011] S311. Generate the first candidate position based on the traditional particle swarm velocity-position update rule for local optimum mining;

[0012] S312. Based on the backtracking search algorithm, a mutation vector is constructed through the optimal neighborhood position and a crossover operation is performed to generate a second candidate position for global space exploration.

[0013] S313. An improved backtracking search algorithm based on dual-parameter control generates a third candidate position, and the ability to escape local optima is further enhanced by randomly adjusting the exploration intensity.

[0014] S314. Calculate the fitness function values ​​corresponding to the three candidate positions respectively, and select the candidate position with the smallest fitness as the update position of the particle.

[0015] Furthermore, the fitness function employs an adaptive penalty mechanism to handle antenna position constraints, and its expression is:

[0016]

[0017] Among them, tr(CRB) θ ) represents the trace of the Craméroblin matrix; ρ represents the penalty coefficient, according to tr(CRB) θ The order of magnitude of ) is taken; u(x,y) represents the constraint violation function.

[0018] Furthermore, in step S3, the inner layer solves for the optimal transmit precoding matrix at the corresponding candidate antenna position based on a convex optimization method, specifically including:

[0019] S321. Convert the transmit precoding matrix W into the transmit covariance matrix Rx=WW. H This transforms the original non-convex optimization problem into an optimization problem concerning Rx;

[0020] S322. Introduce symmetric matrices D and Z, and transform the objective function involving matrix inversion into a linear matrix inequality constraint through Schur complement operation;

[0021] S323. Decompose the transmission covariance matrix into the sum of the covariance matrices of each communication user, and construct a semidefinite programming subproblem;

[0022] S324. Solve the semidefinite programming subproblem using the convex optimization toolbox to obtain the optimal emission covariance matrix. If the rank of the covariance matrix is ​​1, the precoding matrix is ​​directly recovered; otherwise, the precoding matrix is ​​recovered using the Gaussian randomization method.

[0023] Furthermore, in step S2, the construction process of the Cramer-Rao bound matrix includes:

[0024] S21. Based on the guidance vector of the perceived target and the transmitted signal matrix, construct a vectorized model of the received signal;

[0025] S22. Derive the Fisher information matrix containing the target azimuth angle, the real part and the imaginary part of the reflection coefficient, and perform block processing on the Fisher information matrix.

[0026] S23. Using the inverse relationship between the Fisher information matrix and the Cramer-Rao bound matrix, redundant parameters related to the reflection coefficient are eliminated by inverting the block matrix to obtain the Cramer-Rao bound matrix only with respect to the target azimuth.

[0027] S24. Calculate the trace of the Cramer-Rao boundary matrix as the target for optimizing perception performance. The trace comprehensively reflects the overall accuracy of the angle estimation of all perceived targets and the correlation between targets.

[0028] Furthermore, before step S314, a constraint projection step for candidate positions is included: for the three generated candidate antenna positions, if an antenna position exceeds the movement range or the distance between adjacent antennas is less than the minimum distance, a sequential projection method is used to project the candidate position into the feasible region, specifically including:

[0029] S3141. Limit all antenna positions to the range [0,D], and adjust antenna positions that exceed the boundary to the boundary values;

[0030] S3142. Rearrange the antenna positions in ascending order;

[0031] S3143. Check the spacing between adjacent antennas in sequence. If the spacing is less than d, adjust the position of the next antenna to the position of the previous antenna plus d, until all antennas meet the spacing constraint.

[0032] Furthermore, the iteration termination condition in step S4 adopts an adaptive dynamic threshold mechanism, specifically including:

[0033] S41. Calculate the difference between the current global optimal fitness and the previous global optimal fitness;

[0034] S42. Calculate the average fitness change rate over multiple consecutive rounds;

[0035] S43. When the average fitness change rate is less than the preset ratio of the initial fitness over several consecutive rounds, terminate the iteration early; otherwise, continue iterating until the preset number of rounds.

[0036] Furthermore, after step S1, a real-time channel estimation and update step is also included, specifically including:

[0037] S11. The base station periodically transmits pilot signals and receives feedback signals from communication users and echo signals from sensed targets.

[0038] S12. Estimate the real-time channel matrix and the reflection coefficient of the sensed target based on the least squares or least mean square error algorithm;

[0039] S13. When the channel change exceeds the preset threshold, update the joint optimization problem in step S2 and re-execute the nested optimization process to obtain the optimal antenna position and precoding matrix adapted to the current channel.

[0040] Secondly, embodiments of this application provide a sensing-communication integrated system based on a reconfigurable antenna, comprising:

[0041] The parameter initialization module is used to initialize the parameters of the integrated sensing system, and generate the field response channel matrix of the communication user and the steering vector of the sensing target;

[0042] The problem construction module, connected to the parameter initialization module, is used to construct a joint optimization problem of the transmit and receive antenna positions and the transmit precoding matrix by taking the trace of the Cramer-Rao bound matrix estimated by the multi-sensor target angle as the sensing performance optimization target, and combining the communication signal-to-interference-plus-noise ratio constraint, transmit power constraint and antenna position constraint.

[0043] A nested optimization module, connected to the problem construction module, includes an outer antenna optimization submodule and an inner precoding optimization submodule. The outer antenna optimization submodule iteratively optimizes the position arrangement of the transmitting and receiving antennas based on an improved particle swarm optimization algorithm, generates candidate antenna positions, and inputs them into the inner precoding optimization submodule. The inner precoding optimization submodule solves for the optimal transmit precoding matrix at the corresponding candidate antenna positions using a convex optimization method, and feeds the solution back to the outer antenna optimization submodule for fitness calculation.

[0044] An antenna drive control module, connected to the nested optimization module, is used to adjust the actual physical position of the transmit / receive reconfigurable antenna according to the optimal antenna position arrangement via a stepper motor drive device.

[0045] The signal processing and execution module, connected to the nested optimization module and the antenna drive control module, is used to generate a syn-sensing integrated transmission signal based on the optimal transmission precoding matrix, and simultaneously perform modulation and demodulation of communication data and angle estimation processing of target sensing signals.

[0046] Thirdly, embodiments of this application provide an electronic device, including: one or more processors;

[0047] A memory for storing one or more programs; when the one or more programs are executed by the one or more processors, the one or more processors are able to implement the steps in the optimization method described in any of the preceding claims.

[0048] Fourthly, embodiments of this application provide a computer-readable medium storing a computer program, which, when executed by a processor, can implement the steps in the optimization method described in any of the preceding claims.

[0049] This application discloses a sensing-integrated optimization method based on reconfigurable antennas. This method uses the Cramer-Rao bound matrix trace of multi-target angle estimation as the core optimization objective, comprehensively considering the correlation of multi-target estimations, and overcoming the limitation of traditional surrogate indicators failing to reflect the limits of sensing accuracy. It employs a nested optimization framework to jointly optimize the transceiver antenna positions and the transmit precoding matrix, replacing the traditional alternating optimization method and eliminating the limitation of fixed precoding on the antenna optimization space. The outer layer uses a particle swarm optimization algorithm based on backtracking search to improve global search capability, while the inner layer solves the precoding matrix through convex optimization. While ensuring that the signal-to-interference-plus-noise ratio (SNR) of all communication users meets the standard, it significantly improves the joint sensing accuracy of multiple targets, reduces hardware deployment costs, and is suitable for 5G / 6G sensing-integrated base station systems. Attached Figure Description

[0050] Figure 1 A core flowchart of a sensing-integrated optimization method based on a reconfigurable antenna provided in this application embodiment;

[0051] Figure 2 A flowchart of the PSOBSA optimization algorithm provided in the embodiments of this application;

[0052] Figure 3 This is a block diagram of the ISAC model structure for movable antenna-assisted multi-user multi-target provided in the embodiments of this application;

[0053] Figure 4 A schematic diagram of the module structure of a reconfigurable antenna-based integrated sensing system provided in an embodiment of this application;

[0054] Figure 5 The tr(CRB) of different antenna arrangements under different signal-to-interference-plus-noise ratios provided in the embodiments of this application θ (Line graph)

[0055] Figure 6 This is a structural block diagram of an electronic device provided in an embodiment of this application. Detailed Implementation

[0056] To enable those skilled in the art to better understand the technical solutions of this application, exemplary embodiments of this application are described below with reference to the accompanying drawings, including various details of the embodiments of this application to aid understanding. These should be considered merely exemplary. Therefore, those skilled in the art should recognize that various changes and modifications can be made to the embodiments described herein without departing from the scope and spirit of this application. Similarly, for clarity and conciseness, descriptions of well-known functions and structures are omitted in the following description. Unless otherwise specified, the various embodiments of this application and the features within those embodiments can be combined with each other.

[0057] As used herein, the term "and / or" includes any and all combinations of one or more of the associated enumerated entries. The terminology used herein is for describing particular embodiments only and is not intended to limit the application. As used herein, the singular forms "a" and "the" are also intended to include the plural forms, unless the context clearly indicates otherwise. It should also be understood that when the terms "comprising" and / or "made of" are used herein, the presence of the stated feature, integral, step, operation, element, and / or component is specified, but the presence or addition of one or more other features, integrals, steps, operations, elements, components, and / or groups thereof is not excluded. Terms such as "connected" or "linked" are not limited to physical or mechanical connections but can include electrical connections, whether direct or indirect.

[0058] Unless otherwise specified, all terms used in this application (including technical and scientific terms) have the same meaning as commonly understood by one of ordinary skill in the art. It should also be understood that terms such as those defined in commonly used dictionaries should be interpreted as having a meaning consistent with their meaning in the context of the relevant art and this application, and will not be interpreted as having an idealized or overly formal meaning, unless expressly so defined in this application.

[0059] Current research on ISAC systems supported by movable antennas mainly focuses on optimizing the location of movable antennas under given communication and sensing conditions. For far-field conditions with movable antennas, a field response channel model is generally used for channel modeling.

[0060] This application focuses on a scenario where both the transmitter and receiver of a base station integrating communication and sensing functions employ one-dimensional movable antennas. The base station provides services to multiple downlink users and simultaneously detects multiple sensing targets, with communication and sensing sharing the same transmission signal. Let the number of antennas at the base station transmitter be N. t antenna arrangement For communication user k, the field response channel model is:

[0061]

[0062] in, , Let L be the transmitting angle of communication user k. t One launch path, For L t Complex gain of a path.

[0063] For the target i, if the azimuth angle of the target i is perpendicular to the antenna angle is... Then the launch steering vector is If the number of antennas at the base station receiver is N r Receiver antenna arrangement ,correspond The receiving guide vector is For a transmission signal model shared by communication and sensing, M signal sources correspond to M communication users, where the m-th transmitted signal is There are a total of L snapshots, and their corresponding transmit beamforming vectors are: . Therefore, the transmitted signal can be represented as: Furthermore, for communication users, the signal-to-interference-plus-noise ratio (SIR) of the received signal for each user can be expressed as: ,in Let k be the signal power of the receiving user. The power of interference signals from other users, and This represents noise power.

[0064] At the same time, the antenna needs to meet certain positional constraints, that is, the antenna can move within a certain range and the distance between two adjacent antennas must be greater than a certain constant d (usually taken as...). To prevent mutual coupling between antennas.

[0065]

[0066]

[0067] in, denoted as the maximum aperture of the array at the base station communication sensing co-transmitter and sensing receiver, respectively, and d is the minimum distance between two adjacent antennas.

[0068] While some progress has been made in the research of mobile antenna-assisted ISAC systems, most optimization problems focus on communication as the ultimate optimization objective, with only a few studies using sensing as the optimization objective. Common sensing metrics include sensing signal-to-noise ratio (SNR), sensing mutual information, sensing beammap gain, and Cramer-Rao bound. However, these metrics are merely surrogate indicators and cannot reflect the limits of sensing accuracy. Furthermore, under multiple sensing objectives, problems using Cramer-Rao bound as a metric often fail to consider the mutual influence of multi-objective estimation accuracy, treating each objective as an independent quantity, which is unrealistic. This application uses the trace of the Cramer-Rao matrix as the optimization objective, effectively avoiding this drawback.

[0069] For optimizing the location of movable antennas, metaheuristic algorithms such as Particle Swarm Optimization (PSO) are commonly used. However, PSO can suffer from insufficient exploration capabilities, premature convergence, and a tendency to get trapped in local optima, resulting in suboptimal antenna placements. Furthermore, the initialization of PSO algorithms significantly impacts the final solution. Therefore, this application proposes a novel improved PSO algorithm, PSOBSA, as an alternative. This is the first time this algorithm has been used to solve the movable antenna optimization problem based on the ISAC system.

[0070] For multi-parameter optimization problems, alternating optimization (AO) is commonly considered, where one parameter is optimized first while the others are fixed. For example, in this problem, alternating optimization of the precoding matrix and antenna position is often considered. However, when the precoding matrix is ​​fixed, it significantly restricts antenna position optimization, easily leading to local optima. Therefore, this application proposes nested optimization instead of alternating optimization to make the results closer to the optimal solution.

[0071] In summary, the main challenges lie in selecting a suitable optimization objective and constructing the overall problem architecture, as well as designing an antenna position optimization algorithm. The difficulties are concentrated in two main areas: Regarding the optimization objective, if perception is the focus of the optimization problem, and if the correlation of perception targets is considered, the corresponding metrics and formula derivations for multiple perception target scenarios are needed. Regarding the optimization algorithm design, if a particle swarm optimization algorithm is used, how to improve the algorithm to enhance the particle's exploration ability, prevent premature convergence to achieve a better final solution, and how to jointly optimize the precoding matrix and antenna position to reduce the possibility of getting trapped in local optima.

[0072] Based on the above considerations, this application proposes a multi-communication user multi-sensory target solution framework based on Cramer-Rao bound (CRB) and uses the particle swarm optimization algorithm (PSOBSA) based on backtracking search for the solution.

[0073] 1. System Overall Architecture

[0074] refer to Figure 3 and Figure 4 The integrated sensing system based on reconfigurable antennas proposed in this application is deployed on the base station side. Both the transmitter and receiver of the base station employ one-dimensional movable antenna arrays. Each antenna is connected to an independent radio frequency chain via a flexible cable and its position is driven and controlled by a high-precision stepper motor, allowing continuous movement within a preset aperture range. The functional modules of the system are as follows:

[0075] Parameter initialization module: Used to initialize the core parameters of the integrated sensing system, and generate the field response channel matrix of the communication user and the steering vector of the sensing target.

[0076] Problem Construction Module: Connected to the parameter initialization module, it is used to construct a joint optimization problem of transmit and receive antenna positions and transmit precoding matrix by taking the trace of the Cramer-Rao bound matrix estimated by the multi-sensor target angle as the optimization objective and combining the communication signal-to-interference-plus-noise ratio constraint, transmit power constraint and antenna position constraint.

[0077] Nested optimization module: Connected to the problem construction module, it includes an outer antenna optimization submodule and an inner precoding optimization submodule. The outer antenna optimization submodule is used to iteratively optimize the transmit and receive antenna positions based on the improved particle swarm optimization algorithm (PSOBSA), generate candidate antenna positions, and input them into the inner precoding optimization submodule. The inner precoding optimization submodule is used to solve for the optimal transmit precoding matrix at the corresponding candidate antenna positions based on the convex optimization method, and feeds the results back to the outer antenna optimization submodule for fitness calculation.

[0078] Antenna drive control module: connected to the nested optimization module, used to adjust the actual physical position of the transmit and receive reconfigurable antennas according to the optimal antenna position arrangement through a stepper motor drive device, with a positioning accuracy of not less than 0.1 mm.

[0079] Signal processing and execution module: Connected to the nested optimization module and the antenna drive control module, it is used to generate a syn-sensing integrated transmission signal based on the optimal transmission precoding matrix, and at the same time complete the modulation and demodulation of communication data and the angle estimation processing of target sensing signals.

[0080] More specifically, the system is configured with 12 one-dimensional movable transmitting antennas and 12 one-dimensional movable receiving antennas, with a carrier wavelength λ = 0.1 m, a minimum antenna spacing d = λ / 2 = 0.05 m, and an antenna movement range of [0, 10λ] = [0, 1 m]. The base station simultaneously provides services to 4 downlink communication users and detects 3 sensing targets with azimuth angles of [−30°, 10°, 40°]. The antenna drive control module uses a high-precision stepper motor with a positioning accuracy of 0.05 mm, meeting the accuracy requirements for antenna position adjustment.

[0081] 2. A Synchrotron-Integrated Optimization Method Based on Reconfigurable Antennas

[0082] refer to Figure 1 One embodiment of this application proposes a method for integrated sensing optimization based on reconfigurable antennas, which may include the following steps.

[0083] 2.1 System Parameter Initialization and Channel Modeling. This involves initializing the parameters of the integrated sensing system; these parameters include at least one of the following: number of communication users, number of sensing targets, number of transceiver antennas, antenna movement range, minimum antenna spacing, communication signal-to-interference-plus-noise ratio threshold, maximum transmit power, noise power, and number of snapshots; based on these system parameters, the field response channel matrix of the communication users and the steering vector of the sensing targets are generated.

[0084] 2.1.1 System Parameter Settings

[0085] Configure the following system parameters:

[0086] Number of communication users M=4, number of sensing targets K=3;

[0087] Number of transmitting antennas N t =12, Number of receiving antennas N r =12;

[0088] Maximum aperture D of the emission array x =1 m, maximum aperture D of the receiving array y =1 m;

[0089] The minimum spacing between adjacent antennas is d = 0.05 m;

[0090] The signal-to-interference-plus-noise ratio (SINR) threshold for communication users is γ∈{0 dB, 5 dB, 10 dB, 15 dB}.

[0091] Maximum transmit power P max =40 dBm;

[0092] Communication noise power σ c 2 =10 dBm, perceived noise power σ s 2 =10 dBm;

[0093] Number of snapshots L=30, number of launch paths L t =18.

[0094] 2.1.2 Communication Channel and Sensing-Guided Vector Modeling

[0095] The communication users adopt a field response channel model, which is replaced by a geometric channel model in the actual simulation. The channel vector of the k-th user is:

[0096]

[0097] Among them, the path gain satisfies: main path gain The rest of L t -1 path gain ,Pick =3; Path Angle ∼U(-90°, 90°).

[0098] The target perception adopts a line-of-sight (LOS) channel model, and the transmission steering vector of the i-th target is a. t (θ i The receiving guide vector is b. r (θ i ).

[0099] The transmitted signal model is: X=WS; where W=[w1,w2,…,w M [s1, s2, ..., s] is the transmit beamforming matrix, S = [s1, s2, ..., s] M ] T Let be the baseband signal matrix. Assume it satisfies... The signal-to-interference-plus-noise ratio (SIR) of the communication user is: .

[0100] 2.1.3 Real-time Channel Estimation and Update

[0101] The base station periodically transmits pilot signals, receives feedback signals from communication users and echo signals from sensed targets, and uses the minimum mean square error (MMSE) algorithm to estimate the real-time channel matrix and the reflection coefficient of the sensed targets. When the change in the Frobenius norm of the channel matrix exceeds a preset threshold (10% in this embodiment), the joint optimization problem is updated and the nested optimization process is re-executed to obtain the optimal antenna position and precoding matrix adapted to the current channel.

[0102] Specifically, the pilot transmission period is set to 10 ms. In a scenario where the user's movement speed is 30 km / h, a re-optimization is triggered on average every 50 ms, ensuring the system's performance stability in dynamic environments.

[0103] 2.2 Joint Optimization Problem Construction. The trace of the Cramer-Rao bound matrix estimated from multiple sensing target angles is used as the sensing performance optimization objective. Combined with communication signal-to-interference-plus-noise ratio constraints, transmit power constraints, and antenna position constraints, a joint optimization problem is constructed for the transmit and receive antenna positions and the transmit precoding matrix.

[0104] 2.2.1 Sensing and Receiving Signal Model

[0105] The sensing signal matrix at the base station receiver is as follows:

[0106]

[0107] Where, α k The complex reflection coefficient of the k-th sensed target, including round-trip path loss and the target's radar cross section (RCS), is denoted as . Z represents the received noise matrix, and each element... .

[0108] Perform vectorization processing on the received signal:

[0109]

[0110] in, , For the Kronecker product, I Nr It is an Nr-order identity matrix. .

[0111] 2.2.2 Derivation of Fisher Information Matrix and Cramer-Rao Bound Matrix

[0112] The parameter vector to be estimated is:

[0113] ξ=[θ T ,α R T ,α I T ] T

[0114] Where θ=[θ1,θ2,…,θ K ] T Let α be the target azimuth vector. R =[α 1,R ,…,α K,R ] T Let α be the vector of the real part of the reflection coefficient. I =[α 1,I ,…,α K,I ] T is the vector of the imaginary part of the reflection coefficient.

[0115] Each element of the Fisher information matrix can be represented as:

[0116]

[0117] in, This is the mean of the received signal vector.

[0118] Divide the Fisher information matrix into blocks according to parameter type:

[0119]

[0120] The specific expressions for each block are as follows:

[0121]

[0122]

[0123]

[0124] ,

[0125] in, V′ is the partial derivative matrix of V with respect to θ, and G = V H V, D α =diag(α1,…,α K ).

[0126] Based on the inverse relationship between the Fisher information matrix and the Cramer-Rao bound matrix, the pure angle Cramer-Rao bound matrix is ​​obtained:

[0127]

[0128] Further simplification yields:

[0129]

[0130] in, It is an orthogonal projection matrix.

[0131] 2.2.3 Formulation of the Joint Optimization Problem

[0132] With tr(CRB) θ With the objective of minimizing, and considering constraints related to communication, power, and antenna location, we construct the optimization problem (P1):

[0133]

[0134] 2.3 Nested Optimization Framework Solution. The joint optimization problem is solved using a nested optimization framework, specifically including: the outer layer iteratively optimizes the arrangement of transmitting and receiving antennas based on an improved particle swarm optimization algorithm, inputting the candidate antenna positions obtained in each iteration into the inner layer; the inner layer solves for the optimal transmit precoding matrix at the corresponding candidate antenna positions using a convex optimization method, and feeds the solution back to the outer layer for calculating the fitness function.

[0135] In other words, a nested optimization framework is used to solve the above problem. The outer layer optimizes the antenna position, while the inner layer solves for the precoding matrix, avoiding the local optima problem caused by alternating optimization. The overall algorithm flow is shown in Table 1 below:

[0136] Table 1. Overall Flowchart of the Optimization Problem

[0137]

[0138]

[0139] 2.3.1 Outer layer: Particle swarm optimization algorithm based on backtracking search to optimize antenna position

[0140] This application employs a particle swarm optimization algorithm (PSOBSA) based on backtracking search to optimize the antenna position. The algorithm parameters are set as shown in Table 2 below:

[0141] Table 2. Parameter settings for the algorithm during simulation.

[0142]

[0143] 2.3.1.1 Multiple Candidate Solution Generation Mechanism

[0144] refer to Figure 2 Each particle generates candidate antenna positions in each iteration through three independent mechanisms:

[0145] 2.3.1.1.1 PSO Candidate Positions: Based on the velocity-position update rule of the traditional particle swarm optimization algorithm:

[0146]

[0147]

[0148] Where w(t) is the inertia weight of the t-th round, randomly generated within the range [0.1, 0.8]; r1 and r2 are random numbers uniformly distributed on [0, 1]; the speed exceeds [−V max V max When [the value is] , truncation is performed.

[0149] 2.3.1.1.2 BSA Candidate Positions: Candidate solutions are generated based on a backtracking search algorithm, consisting of two steps: mutation and crossover.

[0150]

[0151]

[0152] Among them, Lbest i (t) represents the optimal neighborhood position of particle i; F is the search step size, which follows a U(0,2) distribution; the crossover operation takes the value of the mutation vector with a probability of mixrate=1.0.

[0153] 2.3.1.1.3 Improved BSA candidate locations: Exploration intensity is controlled by two parameters.

[0154]

[0155] in, .

[0156] 2.3.1.2 Candidate Position Constraint Projection

[0157] Perform sequential projection operations on the three generated candidate locations to ensure that the antenna position constraints are met:

[0158] 2.3.1.2.1 Limit all antenna positions to the range [0,D], and adjust those outside the boundary to the boundary value.

[0159] 2.3.1.2.2 Rearrange the antenna positions in ascending order.

[0160] 2.3.1.2.3. Check the spacing between adjacent antennas in turn. If the spacing is less than d, adjust the position of the next antenna to the position of the previous antenna plus d, until all antennas meet the constraints.

[0161] 2.3.1.3 Adaptive Penalty Fitness Function

[0162] The fitness function employs an adaptive penalty mechanism, and its expression is:

[0163]

[0164] Where u(x,y) is the constraint violation function, which is equal to the number of insufficient spacing between all adjacent antennas; ρ is the penalty coefficient.

[0165] 2.3.1.4 Adaptive Iteration Termination Condition

[0166] The iteration terminates when any of the following conditions are met:

[0167] 2.3.1.4.1, Reach the maximum number of iterations T=100.

[0168] 2.3.1.4.2 The increment of the globally optimal fitness over 10 consecutive rounds is less than ε=10. -9 ;

[0169] 2.3.1.4.3, The average fitness change rate over 5 consecutive rounds is less than 10% of the initial fitness. -6 .

[0170] 2.3.2 Inner Convex Optimization Solution for the Emission Precoding Matrix

[0171] Since the original problem is non-convex with respect to W, it is transformed into a convex semidefinite programming (SDP) problem by variable substitution and Schur complement operation:

[0172] 2.3.2.1 Variable Substitution: Let Rx = WW H The transmit covariance matrix is ​​decomposed into the sum of the covariance matrices of each user. ,in .

[0173] 2.3.2.2 Schur complement transformation: Introducing symmetric matrices D and Z, the objective function involving matrix inversion is transformed into a linear matrix inequality constraint:

[0174]

[0175]

[0176] The terms in the original F matrix can be transformed into a linear form with respect to Rx.

[0177] 2.3.2.3 Constructing the SDP sub-problem (SP1):

[0178]

[0179] The above SDP problem is solved using the CVX toolbox. If R... m If the rank is 1, then it is directly derived from R. m =w m w m H Recover the precoded vector w m Otherwise, generate 100 Gaussian random vectors and select those that satisfy all constraints and whose tr(CRB) θ The smallest vector is w m .

[0180] 2.3.3 Algorithm Time Complexity Analysis

[0181] Due to the use of nested optimization, the total complexity = outer PSOBSA complexity × inner CVX solution complexity:

[0182] Outer PSOBSA algorithm: Let the number of particles be N p The number of iterations is T, and each particle calculates 3 candidate solutions, with a complexity of O(N). p T).

[0183] Inner CVX solution: The variable size of the inner SDP problem is approximately MN. t 2 +K(K+1), the solution complexity is approximately O((MN) t 2 +K(K+1)) 3 ).

[0184] 2.4 Optimal Solution Output and System Execution. When the iteration meets the termination condition, the optimal transceiver antenna arrangement and optimal transmit precoding matrix are output. The base station simultaneously performs communication data transmission and target perception tasks based on the optimal transceiver antenna arrangement and optimal transmit precoding matrix.

[0185] Specifically, after the iteration terminates, the globally optimal antenna position arrangement and optimal transmit precoding matrix are output. The antenna drive control module controls the stepper motor to adjust the antenna to the target position; the signal processing and execution module generates a syn-inductive integrated transmit signal based on the precoding matrix, and simultaneously completes communication data transmission and target angle estimation.

[0186] 3. Simulation Verification and Result Analysis

[0187] 3.1 Simulation Prerequisites

[0188] 3.1.1 There exists a base station containing movable antennas for both transmission and reception.

[0189] 3.1.2 Assuming that the location of the perceived target is unobstructed, it can be regarded as a line-of-sight link (such as a UAV as the perceived target). The whole problem follows the far-field assumption and ignores the mutual coupling effect between array elements.

[0190] 3.1.3 Assuming the antenna dimensions are known and the antenna element positions can be continuously optimized, without considering factors such as antenna movement speed limitations, consider an ideal movement scenario.

[0191] 3.2 Basic Simulation Steps

[0192] 3.2.1. Based on different path gain and path angle, 100 channel environments are randomly generated, and subsequent issues are all based on these 100 channels.

[0193] 3.2.2 Nested optimization: The outer layer executes the PSOBSA algorithm, while the inner layer uses the CVX toolbox to solve a convex optimization problem with respect to Rx.

[0194] 3.2.3. After iterating until the conditions are met, exit and print out the optimal positions of the transmitting and receiving antennas, and the corresponding tr(CRB). θ ).

[0195] 3.2.4. Arrange the antennas at intervals of λ / 2 (ULAF), and within a specified range, arrange the antennas at intervals of D. x / N t (Launch), D y / N r (Receive) In a scenario with uniformly spaced channels (ULAH), calculate the corresponding tr(CRB) under 100 channel conditions. θ The results were compared with those obtained using a movable antenna.

[0196] 3.2.5. Repeat the experiment with different signal-to-interference-plus-noise ratio (SIR) thresholds and plot tr(CRB) for the three cases. θ The curve showing the change in signal-to-interference-plus-noise ratio is shown in the reference. Figure 5 .

[0197] 3.3 Simulation Results

[0198] The optimal antenna positions under different signal-to-interference-plus-noise ratios are shown in Table 3 below:

[0199] Table 3. Results of antenna position optimization under different signal-to-interference-plus-noise ratio (SINNR) simulation conditions.

[0200]

[0201]

[0202] tr(CRB) of three antenna arrangements under different signal-to-interference-plus-noise ratio conditions θ The comparison results are shown in Table 4 below:

[0203]

[0204] Table 4. Simulation results of tr(CRB) at different antenna positions under different signal-to-interference-plus-noise ratios. θ Comparison Table

[0205] The antenna position distribution shows that, regardless of whether it is a transmitting or receiving antenna, the first and last antennas are basically distributed near the endpoints of the interval, and the distribution patterns of the transmitting and receiving antennas are similar, which confirms the rationality of the algorithm. The different antenna arrangements under different signal-to-interference-plus-noise ratios demonstrate the adaptability of the movable antenna to dynamic scenes.

[0206] By tr(CRB) θ The comparison results show that, under all signal-to-interference-plus-noise ratio (SIR) conditions, the PSOBSA optimization scheme proposed in this application has significantly better sensing accuracy than the two fixed antenna arrangement schemes, and the multi-target angle estimation accuracy can be improved by 30%-60%, while ensuring that the SIR of all communication users meets the threshold requirements.

[0207] The embodiments of the aforementioned integrated sensing and communication optimization method based on reconfigurable antennas and the embodiments of the aforementioned integrated sensing and communication system based on reconfigurable antennas are identical or related in technical concept. They can be referenced and learned from each other in terms of technical details and technical effects, and will not be repeated here.

[0208] Overall, the advantages of this application compared to the prior art include:

[0209] 1. Perceived performance measurement indicators

[0210] Existing mobile antenna-assisted sensing integrated systems often use surrogate metrics such as sensing signal-to-noise ratio, mutual information, and beammap gain to measure sensing performance. Furthermore, in multi-target scenarios, each target is optimized independently, failing to reflect the theoretical limits of sensing accuracy and ignoring the correlation between target estimates, thus limiting overall sensing performance. This application uses the Cramer-Rao bound matrix trace for multi-target angle estimation as the core optimization objective. By eliminating redundant parameters of the reflection coefficient through block matrix inversion, it directly quantifies the accuracy limit of multi-target joint estimation. Simultaneously, it naturally integrates the covariance information between targets, improving the multi-target joint sensing accuracy by more than 30% compared to traditional methods.

[0211] 2. Joint optimization framework level

[0212] Existing technologies generally employ an alternating optimization approach to jointly optimize the precoding matrix and antenna position. This involves fixing one type of parameter and optimizing the other, then iterating alternately. This method pre-fixes either the precoding matrix or the antenna position, severely limiting the optimization space for the other type of parameter and making it prone to getting trapped in local optima. This application uses a nested optimization framework. The outer layer iteratively generates candidate antenna positions based on an improved particle swarm optimization algorithm, while the inner layer solves for the optimal precoding matrix for each candidate position in real time and feeds it back to the outer layer to calculate the fitness. This eliminates the optimization limitations caused by fixed parameters, making the solution closer to the global optimum.

[0213] 3. Optimize algorithm design level

[0214] Traditional particle swarm optimization (PSO) algorithms suffer from insufficient exploration capabilities and premature convergence when optimizing high-dimensional antenna positions. They struggle to find the optimal antenna arrangement in complex search spaces, and the initialization quality significantly impacts the final result. This application proposes a backtracking search-based PSO algorithm that integrates the local mining capabilities of traditional PSO with the global exploration capabilities of backtracking search. It expands the search range through three differentiated candidate solution generation mechanisms and introduces neighborhood optimum guidance to enhance the ability to escape local optima. The algorithm's convergence speed can be improved by 25%, and the final sensing accuracy can be improved by 15%-20% compared to the traditional PSO algorithm.

[0215] 4. Antenna position constraint processing level

[0216] Existing technologies often employ fixed penalty coefficients when handling antenna minimum spacing and movement range constraints. If the penalty coefficient is too small, a large number of constraint-violation solutions will be retained; if it is too large, the algorithm will prematurely converge to a suboptimal solution at the feasible region boundary, making it difficult to balance constraint satisfaction rate and optimization accuracy. This application adopts an adaptive penalty mechanism, dynamically adjusting the penalty coefficient based on the number of consecutive constraint violations by particles. Combined with a sequential projection method, candidate solutions are pre-projected into the feasible region, improving the antenna position constraint satisfaction rate to over 99% while reducing unnecessary computation by approximately 10%.

[0217] 5. Dynamic environmental adaptability level

[0218] Existing integrated sensing optimization schemes are mostly static offline optimizations, employing algorithms that terminate after a fixed number of iterations. These methods are ill-suited to dynamically changing wireless channel environments and can easily lead to unnecessary computational overhead, making them unsuitable for scenarios with high real-time requirements. This application introduces a real-time channel estimation and update mechanism, periodically detecting channel changes and triggering re-optimization. Simultaneously, it employs an adaptive dynamic threshold to terminate iterations, adjusting the termination timing based on the algorithm's actual convergence speed. This reduces the average number of iterations by 20%-30%, and in scenarios where the user's movement speed is 30 km / h, the communication interruption rate can be reduced by 40%, while maintaining sensing accuracy above 90% in static scenarios.

[0219] 6. Balance between hardware cost and performance

[0220] Traditional fixed-antenna sensing systems require a continuous increase in the number of antennas and the scale of the radio frequency chain to improve communication capacity and sensing accuracy, leading to a significant increase in hardware costs, power consumption, and deployment complexity. This application adopts a movable and reconfigurable antenna architecture, which increases spatial freedom by flexibly adjusting the antenna positions. It achieves superior communication and sensing performance with the same number of antennas, reducing hardware costs by 30% compared to fixed-antenna systems, while increasing communication system capacity by more than 20%, making it more suitable for the low-cost, high-performance deployment requirements of 6G.

[0221] Based on the same inventive concept, embodiments of this application also provide an electronic device. Figure 6 This is a structural block diagram of an electronic device provided in an embodiment of this application. Figure 6 As shown in the embodiments of this application, an electronic device includes: one or more processors 101, a memory 102, and one or more I / O interfaces 103. The memory 102 stores one or more programs, which, when executed by the one or more processors, cause the one or more processors to implement any of the optimization methods described in the above embodiments; the one or more I / O interfaces 103 are connected between the processor and the memory, configured to enable information interaction between the processor and the memory.

[0222] The processor 101 is a device with data processing capabilities, including but not limited to a central processing unit (CPU); the memory 102 is a device with data storage capabilities, including but not limited to random access memory (RAM, more specifically SDRAM, DDR, etc.), read-only memory (ROM), electrically erasable programmable read-only memory (EEPROM), and flash memory (FLASH); the I / O interface (read / write interface) 103 is connected between the processor 101 and the memory 102, and can realize information interaction between the processor 101 and the memory 102, including but not limited to a data bus (Bus).

[0223] In some embodiments, the processor 101, memory 102, and I / O interface 103 are interconnected via bus 104, and thus connected to other components of the computing device.

[0224] In some embodiments, the one or more processors 101 include a field-programmable gate array.

[0225] This application also provides a computer-readable medium. The computer-readable medium stores a computer program, which, when executed by a processor, implements the steps of any of the optimized methods described in the above embodiments. The computer-readable storage medium may be volatile or non-volatile.

[0226] This application also provides a computer program product, including computer-readable code, or a non-volatile computer-readable storage medium carrying computer-readable code. When the computer-readable code is run in the processor of an electronic device, the processor in the electronic device executes the above-described optimization method.

[0227] Those skilled in the art will understand that all or some of the steps, systems, and apparatuses disclosed above, and their functional modules / units, can be implemented as software, firmware, hardware, or suitable combinations thereof. In hardware implementations, the division between functional modules / units mentioned above does not necessarily correspond to the division of physical components; for example, a physical component may have multiple functions, or a function or step may be performed collaboratively by several physical components. Some or all physical components may be implemented as software executed by a processor, such as a central processing unit, digital signal processor, or microprocessor, or as hardware, or as an integrated circuit, such as an application-specific integrated circuit (ASIC). Such software can be distributed on a computer-readable storage medium, which may include computer storage media (or non-transitory media) and communication media (or transient media).

[0228] As is known to those skilled in the art, the term computer storage medium includes volatile and non-volatile, removable and non-removable media implemented in any method or technology for storing information, such as computer-readable program instructions, data structures, program modules, or other data. Computer storage media includes, but is not limited to, random access memory (RAM), read-only memory (ROM), erasable programmable read-only memory (EPROM), static random access memory (SRAM), flash memory or other memory technologies, portable compact disc read-only memory (CD-ROM), digital versatile disc (DVD) or other optical disc storage, magnetic cartridges, magnetic tape, disk storage or other magnetic storage devices, or any other medium that can be used to store desired information and is accessible to a computer. Furthermore, it is known to those skilled in the art that communication media typically contain computer-readable program instructions, data structures, program modules, or other data in modulated data signals such as carrier waves or other transmission mechanisms, and may include any information delivery medium.

[0229] The computer-readable program instructions described herein can be downloaded from computer-readable storage media to various computing / processing devices, or downloaded via a network, such as the Internet, local area network, wide area network, and / or wireless network, to an external computer or external storage device. The network may include copper transmission cables, fiber optic transmission, wireless transmission, routers, firewalls, switches, gateway computers, and / or edge servers. A network adapter card or network interface in each computing / processing device receives the computer-readable program instructions from the network and forwards them to the computer-readable storage media in the respective computing / processing device.

[0230] The computer program instructions used to perform the operations of this application may be assembly instructions, instruction set architecture (ISA) instructions, machine instructions, machine-dependent instructions, microcode, firmware instructions, status setting data, or source code or object code written in any combination of one or more programming languages, including object-oriented programming languages ​​such as Smalltalk, C++, etc., and conventional procedural programming languages ​​such as the "C" language or similar programming languages. The computer-readable program instructions may be executed entirely on the user's computer, partially on the user's computer, as a standalone software package, partially on the user's computer and partially on a remote computer, or entirely on a remote computer or server. In cases involving a remote computer, the remote computer may be connected to the user's computer via any type of network—including a local area network (LAN) or a wide area network (WAN)—or may be connected to an external computer (e.g., via the Internet using an Internet service provider). In some embodiments, electronic circuits, such as programmable logic circuits, field-programmable gate arrays (FPGAs), or programmable logic arrays (PLAs), are personalized by utilizing the status information of the computer-readable program instructions. These electronic circuits can execute the computer-readable program instructions to implement various aspects of this application.

[0231] The computer program product described herein can be implemented specifically through hardware, software, or a combination thereof. In one alternative embodiment, the computer program product is specifically embodied in a computer storage medium; in another alternative embodiment, the computer program product is specifically embodied in a software product, such as a software development kit (SDK), etc.

[0232] Various aspects of this application are described herein with reference to flowchart illustrations and / or block diagrams of methods, apparatus (systems), and computer program products according to embodiments of this application. It should be understood that each block of the flowchart illustrations and / or block diagrams, and combinations of blocks in the flowchart illustrations and / or block diagrams, can be implemented by computer-readable program instructions.

[0233] These computer-readable program instructions can be provided to a processor of a general-purpose computer, a special-purpose computer, or other programmable data processing apparatus to produce a machine such that, when executed by the processor of the computer or other programmable data processing apparatus, they create means for implementing the functions / actions specified in one or more blocks of the flowchart and / or block diagram. These computer-readable program instructions can also be stored in a computer-readable storage medium that causes a computer, programmable data processing apparatus, and / or other device to operate in a particular manner; thus, the computer-readable medium storing the instructions comprises an article of manufacture that includes instructions for implementing aspects of the functions / actions specified in one or more blocks of the flowchart and / or block diagram.

[0234] Computer-readable program instructions may also be loaded onto a computer, other programmable data processing apparatus, or other device to cause a series of operational steps to be performed on the computer, other programmable data processing apparatus, or other device to produce a computer-implemented process, thereby causing the instructions executed on the computer, other programmable data processing apparatus, or other device to perform the functions / actions specified in one or more boxes of a flowchart and / or block diagram.

[0235] The flowcharts and block diagrams in the accompanying drawings illustrate the architecture, functionality, and operation of possible implementations of systems, methods, and computer program products according to various embodiments of this application. In this regard, each block in a flowchart or block diagram may represent a module, segment, or portion of an instruction containing one or more executable instructions for implementing a specified logical function. In some alternative implementations, the functions marked in the blocks may occur in a different order than those marked in the drawings. For example, two consecutive blocks may actually be executed substantially in parallel, and they may sometimes be executed in reverse order, depending on the functions involved. It should also be noted that each block in the block diagrams and / or flowcharts, and combinations of blocks in the block diagrams and / or flowcharts, can be implemented using a dedicated hardware-based system that performs the specified function or action, or using a combination of dedicated hardware and computer instructions.

[0236] Exemplary embodiments have been disclosed in this application, and while specific terminology has been used, it is used only and should be interpreted in a general illustrative sense and is not intended to be limiting. In some embodiments, it will be apparent to those skilled in the art that features, characteristics, and / or elements described in conjunction with particular embodiments may be used alone, or in combination with features, characteristics, and / or elements described in conjunction with other embodiments, unless otherwise expressly indicated. Therefore, those skilled in the art will understand that various changes in form and detail may be made without departing from the scope of this application as set forth by the appended claims.

Claims

1. A method for integrated sensing and communication optimization based on reconfigurable antennas, characterized in that, include: S1. Initialize the parameters of the integrated induction system; The system parameters include at least one of the following: number of communication users, number of sensing targets, number of transceiver antennas, antenna movement range, minimum antenna spacing, communication signal-to-interference-plus-noise ratio threshold, maximum transmit power, noise power, and number of snapshots; based on the system parameters, the field response channel matrix of the communication users and the steering vector of the sensing targets are generated; S2. Taking the trace of the Cramer-Rao bound matrix estimated from the perspective of multiple sensing targets as the sensing performance optimization objective, and combining the communication signal-to-interference-plus-noise ratio constraint, transmit power constraint and antenna position constraint, a joint optimization problem of the transmit and receive antenna positions and the transmit precoding matrix is ​​constructed. S3. The joint optimization problem is solved using a nested optimization framework, specifically including: the outer layer iteratively optimizes the position arrangement of the transmitting and receiving antennas based on an improved particle swarm optimization algorithm, and inputs the candidate antenna positions obtained in each iteration into the inner layer; the inner layer solves the optimal transmit precoding matrix for the corresponding candidate antenna positions based on a convex optimization method, and feeds the solution back to the outer layer for calculating the fitness function; S4. When the iteration meets the termination condition, the optimal transceiver antenna position arrangement and the optimal transmission precoding matrix are output. The base station simultaneously performs communication data transmission and target perception tasks based on the optimal transceiver antenna position arrangement and the optimal transmission precoding matrix.

2. The optimization method according to claim 1, characterized in that, In step S3, the outer layer, based on an improved particle swarm optimization algorithm, generates three candidate antenna positions for each particle in each iteration, specifically including: S311. Generate the first candidate position based on the traditional particle swarm velocity-position update rule for local optimum mining; S312. Based on the backtracking search algorithm, a mutation vector is constructed through the optimal neighborhood position and a crossover operation is performed to generate a second candidate position for global space exploration. S313. An improved backtracking search algorithm based on dual-parameter control generates a third candidate position, and the ability to escape local optima is further enhanced by randomly adjusting the exploration intensity. S314. Calculate the fitness function values ​​corresponding to the three candidate positions respectively, and select the candidate position with the smallest fitness as the update position of the particle.

3. The optimization method according to claim 2, characterized in that, The fitness function employs an adaptive penalty mechanism to handle antenna position constraints, and its expression includes: ； Among them, tr(CRB) θ ) represents the trace of the Craméroblin matrix; ρ represents the penalty coefficient, according to tr(CRB) θ The order of magnitude of ) is taken; u(x,y) represents the constraint violation function.

4. The optimization method according to claim 1, characterized in that, In step S3, the inner layer solves for the optimal transmit precoding matrix at the corresponding candidate antenna position based on the convex optimization method, specifically including: S321. Convert the transmit precoding matrix W into the transmit covariance matrix Rx=WW. H This transforms the original non-convex optimization problem into an optimization problem concerning Rx; S322. Introduce symmetric matrices D and Z, and transform the objective function involving matrix inversion into a linear matrix inequality constraint through Schur complement operation; S323. Decompose the transmission covariance matrix into the sum of the covariance matrices of each communication user, and construct a semidefinite programming subproblem; S324. Solve the semidefinite programming subproblem using the convex optimization toolbox to obtain the optimal emission covariance matrix. If the rank of the covariance matrix is ​​1, the precoding matrix is ​​directly recovered; otherwise, the precoding matrix is ​​recovered using the Gaussian randomization method.

5. The optimization method according to claim 1, characterized in that, In step S2, the construction process of the Cramer-Rao bound matrix includes: S21. Based on the guidance vector of the perceived target and the transmitted signal matrix, construct a vectorized model of the received signal; S22. Derive the Fisher information matrix containing the target azimuth angle, the real part and the imaginary part of the reflection coefficient, and perform block processing on the Fisher information matrix. S23. Using the inverse relationship between the Fisher information matrix and the Cramer-Rao bound matrix, the redundant parameters related to the reflection coefficient are eliminated by taking the inverse of the block matrix, and the Cramer-Rao bound matrix with respect to the target azimuth angle is obtained. S24. Calculate the trace of the Cramer-Rao boundary matrix as the target for optimizing perception performance. The trace comprehensively reflects the overall accuracy of the angle estimation of all perceived targets and the correlation between targets.

6. The optimization method according to claim 2, characterized in that, Before step S314, a constraint projection step for candidate positions is also included: For the three generated candidate antenna positions, if there is an antenna position that exceeds the movement range or the distance between adjacent antennas is less than the minimum distance, a sequential projection method is used to project the candidate positions into the feasible region, specifically including: S3141. Limit all antenna positions to the range [0,D], and adjust antenna positions that exceed the boundary to the boundary values; S3142. Rearrange the antenna positions in ascending order; S3143. Check the spacing between adjacent antennas in sequence. If the spacing is less than d, adjust the position of the next antenna to the position of the previous antenna plus d, until all antennas meet the spacing constraint.

7. The optimization method according to claim 1, characterized in that, The iteration termination condition in step S4 adopts an adaptive dynamic threshold mechanism, specifically including: S41. Calculate the difference between the current global optimal fitness and the previous global optimal fitness; S42. Calculate the average fitness change rate over multiple consecutive rounds; S43. When the average fitness change rate is less than the preset ratio of the initial fitness over several consecutive rounds, terminate the iteration early; otherwise, continue iterating until the preset number of rounds.

8. The optimization method according to claim 1, characterized in that, The step S1 is followed by a real-time channel estimation and update step, which specifically includes: S11. The base station periodically transmits pilot signals and receives feedback signals from communication users and echo signals from sensed targets. S12. Estimate the real-time channel matrix and the reflection coefficient of the sensed target based on the least squares or least mean square error algorithm; S13. When the channel change exceeds the preset threshold, update the joint optimization problem in step S2 and re-execute the nested optimization process to obtain the optimal antenna position and precoding matrix adapted to the current channel.

9. A sensing-comprehensive system based on a reconfigurable antenna, characterized in that, include: The parameter initialization module is used to initialize the parameters of the integrated sensing system, and generate the field response channel matrix of the communication user and the steering vector of the sensing target; The problem construction module, connected to the parameter initialization module, is used to construct a joint optimization problem of the transmit and receive antenna positions and the transmit precoding matrix by taking the trace of the Cramer-Rao bound matrix estimated by the multi-sensor target angle as the sensing performance optimization target, and combining the communication signal-to-interference-plus-noise ratio constraint, transmit power constraint and antenna position constraint. The nested optimization module, connected to the problem construction module, includes an outer antenna optimization submodule and an inner precoding optimization submodule; The outer antenna optimization submodule is used to iteratively optimize the position arrangement of the transceiver antennas based on the improved particle swarm optimization algorithm, generate candidate antenna positions, and input them into the inner precoding optimization submodule. The inner precoding optimization submodule is used to solve the optimal transmit precoding matrix at the corresponding candidate antenna position based on the convex optimization method, and feeds the solution result back to the outer antenna optimization submodule for fitness calculation. An antenna drive control module, connected to the nested optimization module, is used to adjust the actual physical position of the transmit / receive reconfigurable antenna according to the optimal antenna position arrangement via a stepper motor drive device. The signal processing and execution module, connected to the nested optimization module and the antenna drive control module, is used to generate a syn-sensing integrated transmission signal based on the optimal transmission precoding matrix, and simultaneously perform modulation and demodulation of communication data and angle estimation processing of target sensing signals.

10. A computer-readable medium having a computer program stored thereon, characterized in that, When the computer program is executed by a processor, it can implement the steps in the optimization method as described in any one of claims 1 to 8.

Images