Omnidirectional intelligent metasurface-assisted inductive joint beamforming and resource allocation method

Abstract

This invention belongs to the field of integrated communication and sensing technology, and relates to an omnidirectional intelligent metasurface-assisted method for joint sensing beamforming and resource allocation. The specific process of this method is as follows: Constructing a joint optimization problem: Under the conditions of satisfying communication rate constraints, sensing accuracy constraints, power constraints, resource block allocation constraints, and STAR-RIS energy splitting constraints, jointly optimize the resource block allocation variables, resource block power variables, and STAR-RIS transmission and reflection vectors to maximize the total communication rate of the system; Solving the optimization problem: Using an alternating optimization framework, the joint optimization problem is decomposed into a time-delay-Doppler domain resource block allocation and power allocation sub-problem and a STAR-RIS transmission / reflection beamforming sub-problem. After iterative solving, the resource block allocation results, power allocation results, and STAR-RIS transmission and reflection vectors are output.

Description

Technical Field

[0001] This invention belongs to the field of communication and sensing integration technology, and relates to an omnidirectional intelligent metasurface-assisted method for combined sensing beamforming and resource allocation. Background Technology

[0002] Integrated Sensing and Communication (ISAC) systems can simultaneously achieve wireless communication and target sensing on a unified spectrum and hardware platform, offering significant advantages in spectrum utilization, hardware reuse, and system integration. This has become a crucial research direction for next-generation wireless networks such as 6G. Especially in high-speed mobile scenarios, sensing systems not only need reliable information transmission but also high target parameter sensing capabilities, thus placing higher demands on system design. Omnidirectional intelligent metasurfaces (STAR-RIS, Simultaneous Transmission and Reflection Reconfigurable Intelligent Surface) can simultaneously transmit and reflect incident signals, modulating transmitted and reflected waves to achieve full-space coverage and improve the propagation environment under non-line-of-sight conditions. On the other hand, traditional sensing systems are susceptible to Doppler spread and time-varying channels in high-mobility scenarios, leading to degraded communication and sensing performance. Orthogonal Time-Frequency Space (OTFS) modulation technology, by characterizing the channel in the time-delay-Doppler domain, can better adapt to high-speed mobile environments.

[0003] Currently, in the beamforming and resource allocation design of ISAC systems, there are two main existing implementation schemes that are closest to each other:

[0004] Option 1: Joint Signal Design and Processing Scheme Based on OTFS Sensing System. This type of scheme typically targets integrated sensing systems in highly mobile scenarios. It utilizes OTFS modulation to characterize time-varying channels in the time-delay-Doppler domain and focuses on waveform design, receiver signal processing, target parameter estimation, and communication and sensing resource allocation. The typical implementation process involves first constructing a unified communication and sensing signal model based on OTFS, and then designing and optimizing resources such as transmit waveforms, receiver processing methods, bandwidth, or power according to the system's communication or sensing requirements to improve transmission reliability and target sensing performance in high-speed mobile environments. This type of scheme can adapt well to dynamic scenarios with significant Doppler spread, but it typically focuses primarily on signal design and processing in the time-delay-Doppler domain, with relatively insufficient consideration for spatial beamforming capabilities in complex propagation environments.

[0005] Option 2: STAR-RIS-based beamforming scheme. This type of scheme typically deploys STAR-RIS between the base station and the communication user / sensing target. By adjusting the phase and amplitude of the transmitted and reflected waves, it controls the signal propagation direction and energy distribution in different spatial areas. The typical implementation process involves jointly designing the base station's transmitted beam and the transmission / reflection coefficients of STAR-RIS based on the spatial distribution relationship between the communication and sensing objects, thereby enhancing communication link quality and improving target sensing performance. This type of scheme can improve coverage in non-line-of-sight propagation environments and supports full-space beamforming, but it usually focuses primarily on spatial domain beamforming design, with insufficient consideration for the coordinated optimization of latency-Doppler domain resource allocation and spatial domain beam control in high-mobility scenarios. Summary of the Invention

[0006] In view of this, an omnidirectional intelligent metasurface-assisted joint beamforming and resource allocation method for communication and sensing is proposed. This method integrates time-delay-Doppler domain resource allocation, power control, and STAR-RIS transmission / reflection beamforming into the same optimization framework, and explicitly incorporates sensing accuracy constraints into the joint optimization model, thereby achieving coordinated optimization of communication and sensing performance in high-speed mobile non-line-of-sight scenarios.

[0007] The technical solution for implementing the present invention is as follows: In a first aspect, the present invention provides an omnidirectional intelligent metasurface-assisted method for combined sensing beamforming and resource allocation. This method is applicable to system models including a base station, an omnidirectional intelligent metasurface STAR-RIS, communication users, sensing targets, and sensors; the specific process is as follows: The joint optimization problem is constructed as follows: under the conditions of communication rate constraints, sensing accuracy constraints, power constraints, resource block allocation constraints, and STAR-RIS energy splitting constraints, the resource block allocation variables, resource block power variables, and STAR-RIS transmission and reflection vectors are jointly optimized to maximize the total communication rate of the system. Solution of the optimization problem: Using an alternating optimization framework, the joint optimization problem is decomposed into a time-delay-Doppler domain resource block allocation and power allocation subproblem and a STAR-RIS transmission / reflection beamforming subproblem. After iterative solution, the resource block allocation result, power allocation result, and STAR-RIS transmission vector and reflection vector are output.

[0008] Optionally, the joint optimization problem described in this invention is: , In the formula, The total communication rate of the system. For the first Communication rate of each communication user This is the minimum communication rate threshold. For the first The transmit power corresponding to each resource block This represents the upper limit of the total transmission power of the base station. The maximum transmit power for a single resource block. and These are the STAR-RIS numbers The transmission amplitude and reflection amplitude of each unit. The threshold for constraining perception accuracy is denoted as follows. and These are the weighting coefficients of the CRB corresponding to the Doppler parameters and the time delay parameters, respectively.

[0009] Optionally, under the condition of fixing the STAR-RIS transmission vector and reflection vector, the present invention calculates the first... Equivalent gain coefficient on each resource block Simultaneously calculate the time delay estimate involved in the perception constraints. Doppler estimation and their coupling terms The joint optimization problem is decomposed into a time-delay-Doppler domain resource block allocation and power allocation subproblem as follows:

[0011] Optionally, the solution process for the time-delay-Doppler domain resource block allocation and power allocation subproblem described in this invention is as follows: First, under the condition of fixing the STAR-RIS transmission vector and reflection vector, a power-gated variable is introduced to uniformly handle the product coupling relationship between resource allocation variables and power variables; Secondly, the binary resource allocation variable is relaxed into an interval variable, and a corresponding penalty term is introduced to make the relaxed resource allocation variable gradually approach the discrete solution during the iteration process; at the same time, the non-convex coupling term in the communication rate expression is reconstructed by perspective transformation; for the fractional and bilinear terms in the perception CRB constraint, auxiliary variables are introduced, and the original non-convex subproblem is transformed into an iteratively solvable convex optimization problem by combining convex approximation and constraint reconstruction methods. Finally, by solving the above convex optimization problem, the updated resource block allocation variables and gated power variables are obtained, and then the actual transmit power of each resource block is obtained. Thus, the DD domain resource block allocation results and power allocation results under the current iteration can be obtained, and input can be provided for the subsequent STAR-RIS transmission and reflection beamforming problem.

[0012] Introducing power-gated variables

[0013] in, Indicates the first The resource block is allocated to the first The gated power variable corresponding to each communication user Indicates the first The gating power variable corresponding to the allocation of a resource block to the sensing target.

[0014] Optionally, the penalty item described in this invention is:

[0015] in, and This is the point of the previous iteration.

[0016] Optionally, the present invention decomposes the joint optimization problem into a STAR-RIS transmission / reflection beamforming sub-problem, the specific process of which is as follows: First, introduce auxiliary quantities:

[0017] in, This represents the variance of the additive Gaussian noise at the sensor array receiver. and These represent the scale balance coefficients of the CRB terms corresponding to the Doppler parameter and the time delay parameter, respectively, used to balance the differences between the two in terms of physical dimensions and numerical scale; Secondly, the objective function, the minimum rate constraint function for single-user communication, and the perception constraint function are defined as follows: 、 、 The minimum communication rate constraint violation and the perception constraint violation are incorporated as penalty terms into the objective function, constructing the penalty objective function as follows:

[0018] in, and As a penalty factor, When the communication rate constraint and the perception constraint are satisfied, the corresponding penalty term is zero; when the constraint is violated, the penalty term is positive.

[0020] Optionally, this invention employs Riemannian manifold optimization to solve the STAR-RIS transmission / reflection beamformer problem.

[0021] Optionally, the specific process of solving the STAR-RIS transmission / reflection beamforming problem using Riemannian manifold optimization as described in this invention is as follows: First, the reflection coefficient and transmission coefficient of each STAR-RIS unit are represented as a point on a complex two-dimensional unit sphere. All STAR-RIS units together constitute a product manifold of the complex sphere, thus modeling the original STAR-RIS beamforming problem as a constrained optimization problem on the product manifold. Secondly, the Euclidean gradient of the objective function with respect to the reflection and transmission vectors is calculated on the product manifold, and the Euclidean gradient is projected onto the tangent space of the manifold to obtain the corresponding Riemann gradient; Finally, the variables of each STAR-RIS unit are updated in the opposite direction of the Riemann gradient, and the updated variables are remapped back to the feasible manifold through the shrinkage operator, thereby ensuring that the energy conservation constraint of the STAR-RIS unit is satisfied after each update.

[0022] Beneficial effects: First, compared with existing schemes that only optimize OTFS sensing from the time-delay-Doppler domain or only design STAR-RIS beamforming from the spatial domain, this application considers DD domain resource allocation, power control and STAR-RIS transmission / reflection beamforming in a unified manner, which can simultaneously take into account the time-varying channel characteristics in high-speed mobile scenarios and the spatial control requirements in complex propagation environments, thereby improving the overall performance of the system in high-mobility, non-line-of-sight scenarios.

[0023] Second, compared with existing schemes that, while incorporating sensing performance into optimization objectives or constraints, often employ relatively indirect sensing metrics or fail to establish explicit CRB constraints for target delay and Doppler parameters, this application constructs a Fisher information matrix based on the target echo signal and derives the closed-form CRB expressions corresponding to the target delay and Doppler parameters, directly introducing them as sensing accuracy constraints into the joint optimization problem. Thus, sensing accuracy requirements can be modeled and co-optimized in a more explicit and computable form, unified with communication rate targets, power allocation, resource block allocation, and STAR-RIS beamforming, thereby enhancing the specificity of the model expression and the feasibility of joint optimization.

[0024] Third, compared with existing schemes that handle resource selection, power control, and STAR-RIS beamforming in separate stages, or that adopt fixed designs for some variables, this application proposes an alternating iterative solution framework for joint resource allocation and STAR-RIS beamforming. In the DD domain resource allocation and power allocation subproblems, by introducing power-gated variables, the product coupling relationship between discrete resource allocation variables and continuous power variables is uniformly processed. Combined with variable relaxation, penalty term construction, and convex approximation methods, the original non-convex subproblem is transformed into an iteratively solvable convex optimization problem. In the STAR-RIS beamforming subproblem, transmission and reflection variables are modeled on a product manifold satisfying energy splitting constraints, and the Riemannian manifold method is used to iteratively update the transmission and reflection vectors. Therefore, this application can more effectively handle the non-convex coupling relationship between resource allocation, power control, and STAR-RIS beamforming, and has good engineering feasibility. Attached Figure Description

[0025] To more clearly illustrate the technical solutions of the embodiments of the present invention, the drawings used in the embodiments will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present invention. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0026] Figure 1 The curve represents the convergence of the objective function.

[0027] Figure 2 This represents the resource allocation results for the time-delay-Doppler domain.

[0028] Figure 3 The curves showing the influence of transmission rate on base station transmit power are shown.

[0029] Figure 4 This is a flowchart of the present invention. Detailed Implementation

[0030] The embodiments of the present invention will now be described in detail with reference to the accompanying drawings.

[0031] It should be noted that, in the absence of conflict, the following embodiments and features can be combined with each other; and, based on the embodiments of this disclosure, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of this disclosure.

[0032] It should be noted that various aspects of embodiments within the scope of the appended claims are described below. It will be apparent that the aspects described herein can be embodied in a wide variety of forms, and any particular structure and / or function described herein is merely illustrative. Based on this disclosure, those skilled in the art will understand that one aspect described herein can be implemented independently of any other aspect, and two or more of these aspects can be combined in various ways. For example, any number of aspects set forth herein can be used to implement the device and / or practice the method. Additionally, this device and / or method can be implemented using structures and / or functionalities other than one or more of the aspects set forth herein. (See accompanying drawings.) This invention, in constructing an omnidirectional intelligent metasurface-assisted sensing system, unifies the modeling of time-delay-Doppler domain resource block allocation, power control, and STAR-RIS transmission / reflection beamforming. Under communication rate constraints, sensing CRB constraints, and STAR-RIS energy splitting constraints, an alternating optimization method is used for iterative solution, ultimately outputting resource block allocation results, power allocation results, and STAR-RIS transmission and reflection vectors. Figure 4 As shown, the specific implementation steps are as follows: Step 1: Construct the joint optimization problem as follows: Under the conditions of satisfying communication rate constraints, sensing accuracy constraints, power constraints, resource block allocation constraints, and STAR-RIS energy splitting constraints, jointly optimize the resource block allocation variables, resource block power variables, and STAR-RIS transmission and reflection vectors to maximize the total system communication rate; the specific implementation process of this step is as follows: Consider a system consisting of a base station, STAR-RIS, A downlink system consisting of a communication user, a sensing target, and a sensor array. The base station transmits a dual-function OTFS waveform. STAR-RIS establishes a transmission link and a reflection link for the communication user and the sensing target, respectively. The sensor array receives the target echo signal and estimates the time delay and Doppler parameters. STAR-RIS employs an energy splitting protocol, with its reflection and transmission vectors denoted as follows: and The corresponding reflection and transmission matrices are respectively and and satisfy

[0033] in, and These are the STAR-RIS numbers The transmission amplitude and reflection amplitude of each unit.

[0034] In the time-delay-Doppler domain, the two-dimensional resource blocks are uniformly numbered as the [number]. Each resource block is defined, and a binary variable is defined. and They represent the first The resource block is allocated to the first The relationship between a communication user and a sensing target, satisfying , in, and They represent the first The resource block is allocated to the first Resources for individual communication users and sensing targets; Furthermore, a corresponding transmit power is allocated to each resource block. Furthermore, by combining the base station–STAR-RIS–user transmission link and the base station–STAR-RIS–target–sensor reflection link, the equivalent input-output relationship between the communication signal and the sensing echo signal in the time delay-Doppler domain is established, providing a foundation for the subsequent construction of the communication rate expression, sensing accuracy constraints, and joint optimization problem.

[0035] First, on the sensing side, consider constructing a parameter vector to be estimated using the target time delay parameter and the Doppler parameter. Based on the target echo signal received by the sensor array, calculate the partial derivative of the echo signal with respect to the parameter vector to be estimated, construct the corresponding Fisher information matrix, and further obtain the Cramer-Rhodes bounds for the target time delay parameter and the Doppler parameter. For ease of unified expression, the statistics related to time delay estimation, Doppler estimation, and their coupling terms are denoted as... , and The coefficients related to the STAR-RIS reflected beam gain and sensor receiving noise are denoted as... Based on this, the CRB expressions corresponding to the target Doppler parameters and time delay parameters can be obtained, expressed as follows:

[0036]

[0037] in, This represents the CRB corresponding to the target Doppler parameters. This represents the CRB corresponding to the time delay Doppler parameters.

[0038] Secondly, on the communication side, based on the equivalent input-output relationship of the base station–STAR-RIS–communication user transmission link in the time delay-Doppler domain, the communication rate expression of each communication user under given resource block allocation, power allocation, and STAR-RIS transmission beamforming conditions can be obtained.

[0039] Let the first The communication rate of each communication user is for

[0040] in, For the first Each resource block is allocated a corresponding transmit power. Indicates the number of resource blocks. Indicates the first The resource block is allocated to the first Resources for individual communication users Indicates STAR-RIS to the 1st Array response vectors in the direction of each communication user. This represents the transmission coefficient matrix of STAR-RIS. This represents the STAR-RIS array response vector when a base station signal is incident on STAR-RIS. Indicates the first On the resource block The equivalent delay-Doppler domain communication channel vector corresponding to each communication user Indicates the base station at the The resource block is the first Transmit beamforming vectors designed by individual communication users Indicates the first The variance of additive Gaussian noise at the receiver of each communication user. The total communication rate of the system can be expressed as the sum of the rates of each communication user. Therefore, the total communication rate of the system can be used as the objective function of the joint optimization problem.

[0041] Furthermore, considering the decision variables to be optimized in this invention, these include: delay-Doppler domain resource block allocation variables, transmit power variables of each resource block, and the transmission and reflection vectors of STAR-RIS. The joint optimization problem can be expressed as problem (P1): Under the conditions of satisfying communication rate constraints, sensing accuracy constraints, power constraints, resource block allocation constraints, and STAR-RIS energy splitting constraints, jointly optimize the resource block allocation variables. Resource block power variables and STAR-RIS transmission vector and reflection vector In order to maximize the total communication rate of the system.

[0042]

[0043] In the formula, The total communication rate of the system. For the first Communication rate of each communication user This is the minimum communication rate threshold. For the first The transmit power corresponding to each resource block This represents the upper limit of the total transmission power of the base station. The maximum transmit power for a single resource block. and These are the STAR-RIS numbers The transmission amplitude and reflection amplitude of each unit. The threshold for constraining perception accuracy is denoted as follows. and These are the weighting coefficients of the CRB corresponding to the Doppler parameters and the time delay parameters, respectively.

[0044] Thus, a unified joint optimization model that simultaneously characterizes communication performance, sensing performance, and STAR-RIS physical realizability is obtained, laying the foundation for subsequent alternating optimization decomposition and iterative solution.

[0045] Step 2: Decompose the joint optimization problem In step one, a joint optimization problem was established, which comprehensively considers multiple factors such as communication rate, sensing accuracy constraints, and STAR-RIS energy allocation. To solve this problem, an alternating optimization method was adopted to decompose the joint optimization problem into two sub-problems: the resource block allocation and power allocation sub-problem (sub-problem 1) and the STAR-RIS transmitted beam and reflected beam optimization sub-problem (sub-problem 2).

[0046] Step 3: Solve the DD domain resource block allocation and power allocation subproblems Under the condition of fixed STAR-RIS transmission and reflection vectors, calculate the first... Equivalent gain coefficient on each resource block And simultaneously calculate the perceptual constraints involved , and Intermediate quantities. Therefore, the atomic problem can be written as... and power variables Sub-problem 1: , To uniformly handle the product coupling relationship between resource allocation variables and power variables, a power-gated variable is introduced.

[0047] in, Indicates the first The resource block is allocated to the first The gated power variable corresponding to each communication user Indicates the first The gating power variable corresponding to the allocation of a resource block to the sensing target.

[0048] To ensure consistency between the gated power variable and the resource allocation variable, the following power-related constraints apply:

[0049] Therefore, the total transmit power constraint of the system can be further rewritten as follows: , because and Since it is a binary variable, it is first relaxed into a continuous interval variable.

[0050] This is then included as a penalty term in the objective function to encourage the relaxed resource allocation variable to gradually approach the 0-1 discrete solution during the iteration process. The penalty term is as follows:

[0051] in, and This is the point of the previous iteration.

[0052] For the non-convex coupling terms in the communication rate constraint, a perspective transformation is used for reconstruction. Specifically, the original rate term is... Rewritten as:

[0053] Among them, when At that time, the above equation is defined as zero according to continuous extension. After the above rewriting, the communication rate term can be expressed as about The perspective function form is convenient for unified processing within the convex optimization framework.

[0054] For the perceived CRB constraint

[0055] Because it contains fractional terms, bilinear terms, and square terms, two auxiliary variables are introduced. and ,in, Represents the product term The lower bound, Represents the squared term Upper bound:

[0056] Therefore, the original CRB constraint can be conservatively transformed into

[0057] Among them, constraints It can be further transformed into a second-order cone constraint.

[0058] For bilinear terms Using identity transformation

[0059] make Then in the first In the next inner iteration, based on the previous iteration point... , for item By performing a first-order linear approximation, we obtain The lower bound is approximately equal to . Therefore, the constraint It can be further transformed into

[0060] After the above processing, the squared terms and bilinear terms in the original perceptual CRB constraints are transformed into second-order cone constraints and convex approximation constraints at the current iteration point, respectively, which can then be combined with the other constraints to form an iteratively solvable convex optimization subproblem.

[0061] By solving the above convex optimization problem, the updated resource block allocation variables and gated power variables can be obtained. Then, the actual transmit power of each resource block can be recovered according to the following formula:

[0062] Thus, the resource block allocation results and power allocation results of the DD domain under the current iteration can be obtained, and input can be provided for the subsequent STAR-RIS transmission and reflection beamforming problem.

[0063] Step 4: After fixing the resource block allocation results and power allocation results obtained in Step 4, further optimize the transmission vector and reflection vector of STAR-RIS.

[0064] To simplify the expression, an auxiliary quantity is defined:

[0065] Furthermore, the objective function, the minimum rate constraint function for single-user communication, and the perception constraint function are defined as follows: 、 、 The minimum communication rate constraint violation and the perception constraint violation are incorporated as penalty terms into the objective function, constructing the penalty objective function as follows:

[0066] in, and As a penalty factor, When the communication rate constraint and the perception constraint are satisfied, the corresponding penalty term is zero; when the constraint is violated, the penalty term is positive, thus driving the iterative result to gradually approach the feasible region. Therefore, subproblem 2 can be written as:

[0067] Among them, the minimum communication rate constraint and the sensing constraint have been incorporated into the objective function through a penalty term, while the energy conservation constraint of the STAR-RIS unit is still retained in the form of an explicit constraint.

[0068] Based on this, the subproblem is solved using Riemannian manifold optimization. First, the reflection and transmission coefficients of each STAR-RIS element are represented as a point on a complex two-dimensional unit sphere. For the ... Each STAR-RIS unit is defined.

[0069] Since each STAR-RIS unit satisfies the energy conservation constraint Therefore, local variables satisfy Therefore, it lies on the complex two-dimensional unit sphere. (The rest of the text appears to be incomplete and requires further context.) The overall feasible region composed of all units can be represented as a product manifold. ,Right now

[0070] Subsequently, the Euclidean gradient of the objective function with respect to the reflection and transmission vectors is calculated on the product manifold, and the Euclidean gradient is projected onto the tangent space of the manifold to obtain the corresponding Riemann gradient. Then, the variables of each STAR-RIS unit are updated in the opposite direction of the Riemann gradient, and the updated variables are remapped back to the feasible manifold through the shrinkage operator, thereby ensuring that the energy conservation constraint of the STAR-RIS unit is satisfied after each update.

[0071] Specifically, for the first There are 1 STAR-RIS unit, and its corresponding Euclidean gradient component is denoted as .

[0072] By the The reflection and transmission coefficients of each element are composed of Euclidean gradient components. The Riemann gradient obtained by projection can be written as...

[0073] After obtaining the Riemann gradient, proceed along the descent direction of the Riemann gradient to the first... Each STAR-RIS cell is updated, and the update result is remapped back to the manifold using a shrinkage operator to ensure that the cell-level energy conservation constraint is still satisfied after the update. The update formula can be expressed as follows:

[0074] in, Indicates step size, Indicates the first The Riemann gradient corresponding to each unit This represents the shrinkage operator. For complex spherical manifolds, the shrinkage operator employs a normalized mapping, i.e.

[0075] Using the above method, the transmission and reflection beamforming results of STAR-RIS can be iteratively updated while ensuring that the constraints of the STAR-RIS unit are always met.

[0076] Step 5: Alternately execute DD domain resource block allocation and power allocation updates, STAR-RIS transmission and reflection beamforming updates, and penalty factor updates until the convergence condition is met, and finally output the resource block allocation variable. Power allocation variables STAR-RIS transmission vector and reflection vector This enables an omnidirectional intelligent metasurface-assisted inductive beamforming and resource allocation method for high-mobility scenarios.

[0077] The embodiments are further illustrated by MATLAB simulations to demonstrate the present invention's method for omnidirectional intelligent metasurface-assisted inductive joint beamforming and resource allocation.

[0078] In this embodiment, an omnidirectional intelligent metasurface-assisted sensing downlink system is considered in a high-speed mobile scenario. In the system, the base station is configured... A uniform linear array antenna, STAR-RIS configuration Individual units, sensor array configuration One receiving unit, the number of communication users is set to [number]. The number of sensed targets is set to 1. The OTFS frame structure is divided into delay dimension and Doppler dimension respectively. and There are [number] resource units, therefore the entire delay-Doppler domain has [number] resource units. One resource block. Subcarrier spacing is set to... kHz, OTFS symbol duration set to .

[0079] Regarding channel parameters, a time-frequency dual-selection channel model is used between STAR-RIS and the communication user, as well as between STAR-RIS and the sensing target. A quasi-static line-of-sight link is configured between the base station and STAR-RIS, while single-path propagation links are configured between STAR-RIS and the communication user and sensing target. The complex channel gain of each link is generated according to a range-dependent fading model, and the path loss exponent is set to 2.0. Consider a STAR-RIS-assisted OTFS integrated sensing downlink system in a three-dimensional coordinate system. The base station location is set as follows: STAR-RIS location set The target has a non-zero radial velocity to generate a significant Doppler shift. Communication users are randomly distributed within the STAR-RIS transmission side region, while sensing targets are distributed within the STAR-RIS reflection side region, thus demonstrating the system characteristic of STAR-RIS that supports both transmission and reflection coverage.

[0080] Regarding noise and power settings, the noise power setting at the communication user receiver is... dBm, sensor receiver noise power set to dBm. The maximum total transmit power of the base station is set to dBm. The minimum rate threshold for communication users is set to... bit / s / Hz. In the sensing performance constraints, the weighting coefficients for the time delay parameter and the Doppler parameter are set to... and .

[0081] Results Analysis: This embodiment verifies the effectiveness of the proposed method through MATLAB simulation. The simulation content includes algorithm convergence verification, display of time delay-Doppler domain resource allocation results, and system performance analysis under different base station transmit power conditions. Figure 1 The convergence curve of the objective function of the optimization algorithm proposed in this invention is shown. Figure 2 The results of the time-delay-Doppler domain resource allocation are shown; Figure 3 The curves showing the effect of system and rate variations on base station transmit power are presented.

[0082] Figure 1Convergence curves of the objective function of the proposed optimization algorithm as a function of the number of iterations are presented under different sensing accuracy constraints. As can be seen from the figures, under different CRB threshold conditions, each curve gradually increases with the number of iterations and tends to stabilize after a finite number of iterations, indicating that the proposed alternating optimization algorithm has good convergence. It can also be seen that the final converged system and speed differ under different sensing accuracy constraints. When the sensing constraints become stricter, to meet the higher accuracy requirements for objective parameter estimation, the system needs to allocate more optimization degrees of freedom to the sensing side in resource allocation, power control, and STAR-RIS beam design, thus limiting communication performance; while when the sensing constraints are relatively relaxed, the system can achieve higher communication and speed.

[0083] Figure 2 The resource allocation results of the proposed joint optimization method in the delay-Doppler domain are presented, where the horizontal axis represents the resource block index and the vertical axis represents the allocation status of the corresponding resource block. Different subgraphs correspond to four communication users and one sensing target, thus intuitively reflecting the allocation results of DD domain resource blocks under given channel conditions and sensing constraints. As can be seen from the figure, this invention can coordinate and schedule delay-Doppler domain resources according to the coupling relationship between communication requirements and sensing requirements, thereby providing a foundation for subsequent power allocation and STAR-RIS beamforming.

[0084] Figure 3 The trends of system and rate as base station transmit power are shown. It can be seen that as base station transmit power increases, system and rate generally increase, indicating that a higher transmit power budget helps improve communication link quality and enhance joint optimization gains. Meanwhile, the curves show certain differences under different numbers of STAR-RIS units: the more STAR-RIS units, the higher the achievable system and rate. This is because more STAR-RIS units provide higher degrees of freedom in transmission and reflection beam control, thereby enhancing cascaded link gain, improving communication link quality, and improving joint optimization performance. These results demonstrate that the proposed method can achieve good joint optimization results under different base station transmit power and different STAR-RIS configurations.

[0085] In summary, the above are merely preferred embodiments of the present invention and are not intended to limit the scope of protection of the present invention. Any modifications, equivalent substitutions, improvements, etc., made within the spirit and principles of the present invention should be included within the scope of protection of the present invention.

Claims

1. A method for omnidirectional intelligent metasurface-assisted sensing-based joint beamforming and resource allocation, applicable to system models including base stations, the omnidirectional intelligent metasurface STAR-RIS, communication users, sensing targets, and sensors; characterized in that, The specific process is as follows: The joint optimization problem is constructed as follows: under the conditions of communication rate constraints, sensing accuracy constraints, power constraints, resource block allocation constraints, and STAR-RIS energy splitting constraints, the resource block allocation variables, resource block power variables, and STAR-RIS transmission and reflection vectors are jointly optimized to maximize the total communication rate of the system. Solution of the optimization problem: Using an alternating optimization framework, the joint optimization problem is decomposed into a time-delay-Doppler domain resource block allocation and power allocation subproblem and a STAR-RIS transmission / reflection beamforming subproblem. After iterative solution, the resource block allocation result, power allocation result, and STAR-RIS transmission vector and reflection vector are output.

2. The omnidirectional intelligent metasurface-assisted inductive joint beamforming and resource allocation method according to claim 1, characterized in that, The joint optimization problem is: ， In the formula, The total communication rate of the system. For the first Communication rate of each communication user This is the minimum communication rate threshold. For the first The transmit power corresponding to each resource block This represents the upper limit of the total transmission power of the base station. The maximum transmit power for a single resource block. and These are the STAR-RIS numbers The transmission amplitude and reflection amplitude of each unit. The threshold for constraining perception accuracy is denoted as follows. and These are the weighting coefficients of the CRB corresponding to the Doppler parameters and the time delay parameters, respectively.

3. The omnidirectional intelligent metasurface-assisted inductive joint beamforming and resource allocation method according to claim 2, characterized in that, Under the condition of fixed STAR-RIS transmission and reflection vectors, calculate the first... Equivalent gain coefficient on each resource block Simultaneously calculate the time delay estimate involved in the perception constraints. Doppler estimation and their coupling terms The joint optimization problem is decomposed into a time-delay-Doppler domain resource block allocation and power allocation subproblem as follows:

4. The omnidirectional intelligent metasurface-assisted inductive joint beamforming and resource allocation method according to claim 3, characterized in that, The solution process for the time-delay-Doppler domain resource block allocation and power allocation subproblems is as follows: First, under the condition of fixing the STAR-RIS transmission vector and reflection vector, a power-gated variable is introduced to uniformly handle the product coupling relationship between resource allocation variables and power variables; Secondly, the binary resource allocation variable is relaxed into an interval variable, and a corresponding penalty term is introduced to make the relaxed resource allocation variable gradually approach the discrete solution during the iteration process; at the same time, the non-convex coupling term in the communication rate expression is reconstructed by perspective transformation; for the fractional and bilinear terms in the perception CRB constraint, auxiliary variables are introduced, and the original non-convex subproblem is transformed into an iteratively solvable convex optimization problem by combining convex approximation and constraint reconstruction methods. Finally, by solving the above convex optimization problem, the updated resource block allocation variables and gated power variables are obtained, and then the actual transmit power of each resource block is obtained. Thus, the DD domain resource block allocation results and power allocation results under the current iteration can be obtained, and input can be provided for the subsequent STAR-RIS transmission and reflection beamforming problem.

5. The omnidirectional intelligent metasurface-assisted inductive joint beamforming and resource allocation method according to claim 4, characterized in that, The penalty item is: in, and This is the point of the previous iteration.

6. The omnidirectional intelligent metasurface-assisted inductive joint beamforming and resource allocation method according to claim 2 or 3, characterized in that, The joint optimization problem is decomposed into a STAR-RIS transmission / reflection beamforming subproblem, specifically as follows: First, introduce auxiliary quantities: in, This represents the variance of the additive Gaussian noise at the sensor array receiver. and These represent the scale balance coefficients of the CRB terms corresponding to the Doppler parameter and the time delay parameter, respectively, used to balance the differences between the two in terms of physical dimensions and numerical scale; Secondly, the objective function, the minimum rate constraint function for single-user communication, and the perception constraint function are defined as follows: 、 、 The minimum communication rate constraint violation and the perception constraint violation are incorporated as penalty terms into the objective function, constructing the penalty objective function as follows: in, and As a penalty factor, When the communication rate constraint and the perception constraint are satisfied, the corresponding penalty term is zero; when the constraint is violated, the penalty term is positive.

7. The omnidirectional intelligent metasurface-assisted inductive joint beamforming and resource allocation method according to claim 6, characterized in that, The STAR-RIS transmission / reflection beamformer problem is solved using Riemannian manifold optimization.

8. The omnidirectional intelligent metasurface-assisted inductive joint beamforming and resource allocation method according to claim 7, characterized in that, The specific process of using Riemannian manifold optimization to solve the STAR-RIS transmission / reflection beamforming subproblem is as follows: First, the reflection coefficient and transmission coefficient of each STAR-RIS unit are represented as a point on a complex two-dimensional unit sphere. All STAR-RIS units together constitute a product manifold of the complex sphere, thus modeling the original STAR-RIS beamforming problem as a constrained optimization problem on the product manifold. Secondly, the Euclidean gradient of the objective function with respect to the reflection and transmission vectors is calculated on the product manifold, and the Euclidean gradient is projected onto the tangent space of the manifold to obtain the corresponding Riemann gradient; Finally, the variables of each STAR-RIS unit are updated in the opposite direction of the Riemann gradient, and the updated variables are remapped back to the feasible manifold through the shrinkage operator, thereby ensuring that the energy conservation constraint of the STAR-RIS unit is satisfied after each update.

Images