Joint communication and perception method based on FIM-CRB optimization

Abstract

The invention discloses a joint communication and perception method based on FIM-CRB optimization. The joint communication and perception method comprises the following steps: establishing an RIS-assisted joint communication and perception system model; constructing an echo signal model of a sensing link in the system, and deducing vectorization expression of an echo signal; constructing a Fisher information matrix FIM, and further deducing a lower limit CRB of perception parameter estimation precision; constructing a joint optimization problem, taking the minimum CRB as a target function, introducing the minimum signal to interference plus noise ratio constraint and the total transmitting power constraint of a communication user, and jointly optimizing a transmitting beam matrix of the BS and a reflection matrix of the RIS; and equivalently reconstructing the joint optimization problem into a solvable convex optimization problem by using a positive semidefinite programming and fractional programming method, and iteratively updating the transmitted beam, the RIS reflection matrix, the sensing beam matrix and the auxiliary variables by using an alternate optimization algorithm to finally obtain an optimization result. The method can effectively improve the sensing precision and robustness of the system, and is suitable for a 6G sensing communication network and an intelligent environment construction scene.

Description

Technical Field

[0001] The present invention belongs to the technical field of parameter estimation and beam design, and relates to signal processing and resource optimization of wireless communication and radar perception fusion systems, and specifically to a joint communication and perception method based on FIM-CRB optimization. Background Art

[0002] Integrated Sensing and Communication (ISAC) is a key development direction for next-generation wireless communication systems, showing great potential particularly in millimeter-wave and terahertz bands and in intelligent environments. By simultaneously performing communication and perception tasks within the same spectrum resources and hardware platform, ISAC systems can significantly improve spectrum utilization, reduce device power consumption, and achieve a deep integration of environmental perception and data transmission.

[0003] Reconfigurable Intelligent Surface (RIS), a novel wireless channel reshaping technology, has garnered widespread attention in recent years within ISAC systems. RIS, composed of a large number of programmable passive reflective elements, can reconfigure and control wireless channel propagation characteristics by manipulating the phase and amplitude of incident waves, independent of traditional RF links. Its low power consumption, high flexibility, and simple construction make it ideal for future ultra-dense deployments, green communications, and high-precision sensing scenarios.

[0004] In RIS-assisted ISAC systems, the key performance bottleneck lies in the joint design of communication and perception beams. Traditional perception systems typically design radar beams independently, ignoring the coupling effect between the communication beam and RIS reflection behavior. Most communication optimization algorithms only consider channel gain or SINR metrics, making it difficult to effectively model and control perception performance.

[0005] The Cramer-Rao bound (CRB) is an important tool for measuring the lower bound of target parameter estimation accuracy (such as azimuth and elevation) and is widely used in radar and perception theory. However, in the RIS-assisted architecture, due to the complex multi-hop channel structure and the BS-RIS-target-sensor link, traditional CRB modeling methods struggle to accurately capture the multidimensional parameter relationships in joint communication and perception scenarios. Furthermore, because CRB involves matrix inversion and structural optimization of the Fisher Information Matrix (FIM), the original problem is strongly non-convex, making it difficult to solve directly.

[0006] Currently, existing research focuses primarily on beam optimization for single-link radar or pure communication. Research on RIS-assisted SIAC communication optimization, which combines multi-objective constraints to minimize the perceived CRB and ensure minimum communication quality for users, remains limited. In particular, existing two-dimensional modeling and decoupling design strategies are insufficient when modeling UPA-RIS and ULA-sensor structures, potentially leading to unstable system performance or insufficient perception accuracy. Summary of the Invention

[0007] Purpose of the invention: In order to overcome the shortcomings of the existing technology, a joint communication and perception method based on FIM-CRB optimization is provided to minimize the CRB of the perception system and improve the perception accuracy while meeting the communication service quality constraints and total power constraints. It can effectively improve the system perception accuracy and robustness, and is suitable for 6G perception-based communication networks and smart environment construction scenarios.

[0008] Technical Solution: To achieve the above objectives, the present invention provides a joint communication and perception method based on FIM-CRB optimization, comprising the following steps:

[0009] S1: Establish a RIS-assisted joint communication and perception system model;

[0010] S2: Based on the joint communication and perception system model, the echo signal model of the perception link in the system is constructed. Based on the target direction parameters and the geometric arrangement of the RIS and sensor array, the vector representation of the echo signal is derived.

[0011] S3: Based on the echo signal model, the Fisher information matrix FIM is constructed, and the lower bound of the perception parameter estimation accuracy CRB is further derived;

[0012] S4: Construct a joint optimization problem with the objective function of minimizing CRB. Simultaneously, the minimum signal-to-interference-and-noise ratio (SINR) constraint and the total transmit power constraint of the communication user are introduced to jointly optimize the BS's transmit beam matrix and the RIS's reflection matrix.

[0013] S5: The joint optimization problem is reconstructed into a solvable convex optimization problem using semidefinite programming (SDP) and fractional programming (FP) methods. The alternating optimization (AO) algorithm is then used to iteratively update the transmit beam, RIS reflection matrix, sensing beam matrix, and auxiliary variables. Finally, the optimization result of minimizing the CRB under the communication constraints is obtained.

[0014] Furthermore, the joint communication and perception system in step S1 includes a base station, a reconfigurable intelligent surface (RIS), a sensor array, and multiple communication users, the communication path is base station-RIS-user, and the perception path is base station-RIS-target-sensor;

[0015] RIS is a uniform planar array (UPA) structure, while the base station (BS) and radar sensor array (sensor) are both uniform linear array (ULA) structures. The sensor receives echo signals including information such as pitch angle and azimuth angle.

[0016] The downlink communication link between the base station BS and the user UE is reflected by the RIS composed of the uniform planar array UPA. The sensing link is that after the BS illuminates the target through the RIS, the sensor array composed of the uniform linear array ULA receives the echo signal.

[0017] As an intermediary structure, RIS has dynamic programmability and can accurately control the propagation path of the incident electromagnetic wave, thereby simultaneously improving the beam gain of the communication signal and the directional resolution ability of the perception signal.

[0018] Furthermore, in step S2, it is assumed that the RIS antenna array in the UPA structure is deployed two-dimensionally along the yz plane, capable of control in both azimuth and elevation. The base station and sensor are both arranged linearly in one dimension, forming directional beams in specific directions. These factors were considered during modeling, and combined with statistical signal processing theory, a suitable RIS-assisted ISAC communication model was designed to simulate a system with simultaneous communication and perception.

[0019] The joint communication and perception system includes a communication channel G between the base station and RIS, and a communication channel h between RIS and user k. k , the reflection coefficient matrix of RIS is Θ = diag(Φ), where Then the downlink communication channel from BS to user k is:

[0020]

[0021] Among them, L c is the channel gain between BS and user k, is additive Gaussian white noise (AGWN), which has a mean of 0 and a variance of Gaussian distribution; at the same time, in the sensing channel, the echo channel received by the target in the time block T is:

[0022]

[0023] Where α is the complex gain affected by the round-trip path loss and the target's complex reflection factor, and are the direction vectors of the sensor and RIS respectively; let the communication signal be c(t), the radar signal at time t be s(t), and set W and V to be the beamforming matrices of the communication signal and radar respectively, then the transmission signal at time t is expressed as:

[0024] X(t)=Wc(t)+Vs(t)

[0025] Based on the established communication channel model, the SINR of the communication channel is obtained based on the communication and perception channels:

[0026]

[0027] And the beam gain:

[0028]

[0029] Where α is the complex gain affected by the round-trip path loss and the target's complex reflection factor, and are the direction vectors of the sensor and RIS, respectively.

[0030] Furthermore, the construction of the Fisher information matrix FIM in step S3 includes:

[0031] The echo channel received by the sensor array is vectorized to obtain the channel vector form:

[0032]

[0033] Assume that the unknown parameters are in and According to the likelihood function expression of the complex Gaussian distribution, the FIM of the system parameter ξ is constructed, where each element [J ξ ] i,j It is obtained by the following formula:

[0034]

[0035] Among them, u is the signal mean, C y is the covariance matrix;

[0036] On this basis, the FIM for estimating the unknown parameter vector ξ is constructed as:

[0037]

[0038] Furthermore, the derivation of the perception parameter estimation accuracy lower bound CRB in step S3 includes:

[0039] After the FIM is constructed, the CRB can be given by its inverse matrix, that is, the lower bound of the error of the estimated parameter ξ is:

[0040]

[0041] Then, using Schur complement, CRB becomes:

[0042]

[0043] Furthermore, in step S4, to minimize the CRB in the perception system while ensuring communication quality, the present invention combines the aforementioned CRB constraint with the communication performance constraint to form a multi-objective optimization problem. The optimization goal is to minimize the perceived CRB derived from the FIM while satisfying the minimum SINR constraint and transmit power limit of the communication user.

[0044] The formulation of the joint optimization problem includes:

[0045] Introducing SINR constraints, minimum sensing requirement constraints and maximum energy constraints:

[0046]

[0047] G s ≥χ

[0048] tr(WW H +VV H )≤P

[0049] According to the monotonicity of the matrix, the function tr(X -1 ) decreases in the semi-positive definite matrix space; therefore, the auxiliary variable U is introduced, and the optimization of CRB is expressed as:

[0050]

[0051] G s ≥χ,

[0052] tr(WW H +VV H )≤P,

[0053] U±0.

[0054] Furthermore, in step S5, in order to effectively deal with the non-convex structure in the original joint communication and perception optimization problem, auxiliary variables and Schur complement conditions are introduced, and the non-convex objective function related to CRB is converted into a linear matrix inequality (LMI) form. Combined with the semi-definite programming (SDP) and fractional programming (FP) methods, the original problem is equivalently reconstructed into a solvable convex optimization problem, and the alternating optimization (AO) algorithm is further used to iteratively update each optimization variable, and finally the beam and RIS configuration scheme that achieves the minimum CRB under communication constraints is obtained.

[0055] Furthermore, the specific process of equivalently reconstructing the joint optimization problem into a solvable convex optimization problem using the semidefinite programming (SDP) and fractional programming (FP) methods in step S5 includes:

[0056] Introduce auxiliary variable t and impose LMI constraints At this time, the optimized objective function becomes:

[0057]

[0058] G s ≥χ,

[0059] tr(WW H +VV H )≤P,

[0060] U±0.

[0061] Furthermore, in step S5, an alternating optimization (AO) algorithm is used to iteratively update the transmit beam, RIS reflection matrix, sensing beam matrix, and auxiliary variables, and finally obtain an optimization result that minimizes the CRB under the communication constraints, specifically including:

[0062] Optimize the communication beam under the condition of fixed RIS reflection matrix and perception beam covariance matrix; optimize the RIS reflection matrix under the condition of fixed communication beam and perception beam; optimize the perception beam under the condition of fixed communication beam and RIS; finally, update the auxiliary variables and iterate until convergence;

[0063] When optimizing the RIS reflection matrix, in order to ensure the minimum communication SINR requirement constraint, FP is adopted and an auxiliary variable τ is introduced to convert the non-convex constraint into a convex optimization:

[0064]

[0065] The optimal solution The value of For the non-convex constraint G on minimum radar perception s , introduce auxiliary variable B=VV H , the non-convex constraint becomes:

[0066] G s =tr(V H q H qV)=tr(BQ)

[0067] Finally, EVD is used to restore the radar signal;

[0068] When optimizing the beamforming matrix, since the non-convex condition has been converted into a convex function in the first step, the optimal solution can be directly iterated;

[0069] When optimizing auxiliary variables, for LMI conditions, use Schur complement to convert LMI constraints into SDP form:

[0070] Ut -1 I±0

[0071]

[0072] For the transformed LMI condition, the interior point method is used for iteration:

[0073]

[0074] Until convergence, the optimal solution can be obtained.

[0075] The present invention constructs a dual-link system model of the BS-RIS-UE and BS-RIS-target-sensor, and derives the Cramérau bound (CRB) expression for target parameter estimation based on the Fisher Information Matrix (FIM). Combined with the user's minimum SINR constraint and power limit, a joint optimization problem is constructed. By introducing auxiliary variables, the Schur complement method, and fractional programming theory, the original non-convex problem is transformed into a solvable semi-definite programming (SDP) form. An alternating optimization algorithm is used to update the transmit beam matrix, RIS reflection matrix, and radar perception covariance matrix in stages to minimize the perception CRB and ensure communication quality.

[0076] Beneficial effects: Compared with the existing technology, the present invention introduces the Fisher information matrix and CRB theory to establish a perception performance measurement criterion with target direction estimation accuracy as the core, effectively overcoming the limitation of ignoring perception performance in traditional communication-dominated optimization. The method of the present invention systematically unifies the array structure and multipath propagation characteristics of base stations, RIS and radar sensors into a unified model. It is particularly suitable for near-field scenarios where the wavefront curvature is significant and the angle and distance are highly coupled. It can accurately characterize the statistical relationship between target parameters and received signals. By using Schur complement and auxiliary variable introduction technology, the non-convex expression of perception CRB is converted into LMI. Combined with SDP and FP technology, the feasibility and efficiency of problem solving are significantly improved. At the same time, the designed alternating optimization algorithm iteratively updates the communication beam, RIS reflection matrix and perception beam in stages, realizing dynamic coordination of communication quality assurance and perception accuracy improvement. The method of the present invention not only improves the spatial resource utilization and target resolution capability of the system, but also has good convergence and engineering feasibility. It can effectively improve the robustness and accuracy of the joint communication and perception system in complex environments, and has important application value for building a new generation of efficient, low-power, and highly reliable 6G intelligent communication perception network. BRIEF DESCRIPTION OF THE DRAWINGS

[0077] Figure 1 is a flow chart of the method of the present invention;

[0078] Figure 2 A model diagram of the RIS-assisted ISAC communication system in the present invention;

[0079] Figure 3 This is a flow chart of the convex optimization algorithm of the present invention;

[0080] Figure 4 Flowchart of the AO algorithm in the present invention. DETAILED DESCRIPTION

[0081] The present invention is further illustrated below with reference to the accompanying drawings and specific embodiments. It should be understood that these embodiments are only used to illustrate the present invention and are not used to limit the scope of the present invention. After reading the present invention, modifications of various equivalent forms of the present invention made by those skilled in the art all fall within the scope defined by the claims attached to this application.

[0082] like Figure 1 As shown, the present invention provides a joint communication and perception method based on FIM-CRB optimization, comprising the following steps:

[0083] S1: Establish a RIS-assisted joint communication and perception system model;

[0084] S2: Based on the joint communication and perception system model, the echo signal model of the perception link in the system is constructed. Based on the target direction parameters and the geometric arrangement of the RIS and sensor array, the vector representation of the echo signal is derived.

[0085] S3: Based on the echo signal model, the Fisher information matrix FIM is constructed, and the lower bound of the perception parameter estimation accuracy CRB is further derived;

[0086] S4: Construct a joint optimization problem with the objective function of minimizing CRB. Simultaneously, the minimum signal-to-interference-and-noise ratio (SINR) constraint and the total transmit power constraint of the communication user are introduced to jointly optimize the BS's transmit beam matrix and the RIS's reflection matrix.

[0087] S5: The joint optimization problem is reconstructed into a solvable convex optimization problem using semidefinite programming (SDP) and fractional programming (FP) methods. The alternating optimization (AO) algorithm is then used to iteratively update the transmit beam, RIS reflection matrix, sensing beam matrix, and auxiliary variables. Finally, the optimization result of minimizing the CRB under the communication constraints is obtained.

[0088] The following combination Figures 2 to 4 , the above process is explained in detail:

[0089] Reference Figure 2 The joint communication and perception system includes a base station, a reconfigurable intelligent surface RIS, a sensor array and multiple communication users. The communication path is base station-RIS-user, and the perception path is base station-RIS-target-sensor;

[0090] RIS is a uniform planar array (UPA) structure, while the base station (BS) and radar sensor array (sensor) are both uniform linear array (ULA) structures. The sensor receives echo signals including information such as pitch angle and azimuth angle.

[0091] The downlink communication link between the base station BS and the user UE is reflected by the RIS composed of the uniform planar array UPA. The sensing link is that after the BS illuminates the target through the RIS, the sensor array composed of the uniform linear array ULA receives the echo signal.

[0092] As an intermediary structure, RIS has dynamic programmability and can accurately control the propagation path of the incident electromagnetic wave, thereby simultaneously improving the beam gain of the communication signal and the directional resolution ability of the perception signal.

[0093] Assume that the RIS antenna array in a UPA structure is deployed two-dimensionally along the yz plane, capable of azimuth and elevation control. The base station and sensor are arranged linearly in one dimension, forming a directional beam in a specific direction. Considering these factors and incorporating statistical signal processing theory, a suitable RIS-assisted ISAC communication model is designed to simulate a system with simultaneous communication and perception.

[0094] The joint communication and perception system in step S2 includes a communication channel G between the base station and the RIS, and a communication channel h between the RIS and the user k. k , the reflection coefficient matrix of RIS is Θ = diag(Φ), where Then the downlink communication channel from BS to user k is:

[0095]

[0096] Among them, L c is the channel gain between BS and user k, is additive Gaussian white noise (AGWN), which has a mean of 0 and a variance of Gaussian distribution; at the same time, in the sensing channel, the echo channel received by the target in the time block T is:

[0097]

[0098] Where α is the complex gain affected by the round-trip path loss and the target's complex reflection factor, and are the direction vectors of the sensor and RIS respectively; the above echo channels constitute the echo signal model, which represents the signal process from the base station to the RIS, to the sensing target, and then to the sensing target echo to the sensing module in the RIS;

[0099] Assume that the communication signal is c(t), the radar signal at time t is s(t), and W and V are the beamforming matrices of the communication signal and radar, respectively. The transmission signal at time t is expressed as:

[0100] X(t)=Wc(t)+Vs(t)

[0101] Based on the established communication channel model, the SINR of the communication channel is obtained based on the communication and perception channels:

[0102]

[0103] And the beam gain:

[0104]

[0105] Where α is the complex gain affected by the round-trip path loss and the target's complex reflection factor, and are the direction vectors of the sensor and RIS, respectively.

[0106] like Figure 3 As shown, in step S3, the Fisher Information Matrix (FIM) of the target parameters is constructed for evaluation, and the perception error lower bound CRB is derived from this, providing an optimization criterion for system design. Specifically, the echo channel received by the sensor array is first vectorized to obtain the channel vector form:

[0107]

[0108] Assume that the unknown parameters are in and From the vectorized channel, we can get set up Then about y s The complex probability density function of is:

[0109]

[0110] To this end, the likelihood function expression of the complex Gaussian distribution can be derived:

[0111]

[0112] According to the likelihood function expression of the complex Gaussian distribution, the definition of the FIM matrix is:

[0113]

[0114] Construct the FIM of system parameters ξ, where each element [J ξ ] i,j It can be obtained by the following formula:

[0115]

[0116] Where u is the signal mean, C y is the covariance matrix. On this basis, the FIM for estimating the unknown parameter vector ξ can be constructed as:

[0117]

[0118] After the FIM is constructed, the CRB can be given by its inverse matrix, that is, the lower bound of the error of the estimated parameter ξ is:

[0119]

[0120] Then, using the Schur complement method, CRB is transformed into:

[0121]

[0122] In step S4, to minimize the CRB in the perception system while ensuring communication quality, the present invention combines the aforementioned CRB constraint with the communication performance constraint to form a multi-objective optimization problem. The optimization goal is to minimize the perceived CRB derived from the FIM while satisfying the minimum SINR constraint and transmit power limit of the communicating user.

[0123] The formulation of the joint optimization problem includes:

[0124] Introducing SINR constraints, minimum sensing requirement constraints and maximum energy constraints:

[0125]

[0126] G s ≥χ

[0127] tr(WW H +VV H )≤P

[0128] According to the monotonicity of the matrix, the function tr(X -1 ) decreases in the semi-positive definite matrix space; therefore, the auxiliary variable U is introduced, and the optimization of CRB is expressed as:

[0129]

[0130] G s ≥χ,

[0131] tr(WW H +VV H )≤P,

[0132] U±0.

[0133] In step S5, in order to effectively deal with the non-convex structure of the original joint communication and perception optimization problem, auxiliary variables and Schur complement conditions are introduced, and the non-convex objective function related to CRB is transformed into a linear matrix inequality (LMI) form. Combined with the semi-definite programming (SDP) and fractional programming (FP) methods, the original problem is equivalently reconstructed into a solvable convex optimization problem. The alternating optimization (AO) algorithm is further used to iteratively update each optimization variable, and finally the beam and RIS configuration scheme that achieves the minimum CRB under communication constraints is obtained.

[0134] The specific process of using semidefinite programming (SDP) and fractional programming (FP) methods to equivalently reconstruct the joint optimization problem into a solvable convex optimization problem includes:

[0135] Introduce auxiliary variable t and impose LMI constraints At this time, the optimized objective function becomes:

[0136]

[0137] G s ≥χ,

[0138] tr(WW H +VV H )≤P,

[0139] U±0.

[0140] Reference Figure 4 , the alternating optimization (AO) algorithm is used to iteratively update the transmit beam, RIS reflection matrix, sensing beam matrix and auxiliary variables, and finally obtain the optimization result of minimizing CRB under the communication constraints, including:

[0141] Optimize the communication beam under the condition of fixed RIS reflection matrix and perception beam covariance matrix; optimize the RIS reflection matrix under the condition of fixed communication beam and perception beam; optimize the perception beam under the condition of fixed communication beam and RIS; finally, update the auxiliary variables and iterate until convergence;

[0142] When optimizing the RIS reflection matrix, in order to ensure the minimum communication SINR requirement constraint, FP is adopted and an auxiliary variable τ is introduced to convert the non-convex constraint into a convex optimization:

[0143]

[0144] The optimal solution The value of For the non-convex constraint G on minimum radar perception s , introduce auxiliary variable B=VV H , the non-convex constraint becomes:

[0145] G s =tr(V H q H qV)=tr(BQ)

[0146] Finally, EVD is used to restore the radar signal;

[0147] When optimizing the beamforming matrix, since the non-convex condition has been converted into a convex function in the first step, the optimal solution can be directly iterated;

[0148] When optimizing auxiliary variables, for LMI conditions, use Schur complement to convert LMI constraints into SDP form:

[0149] Ut -1 I±0

[0150]

[0151] For the transformed LMI condition, the interior point method is used for iteration:

[0152]

[0153] Until convergence, the optimal solution can be obtained.

Claims

1. A joint communication and perception method based on FIM-CRB optimization, characterized in that: The steps include: S1: Establish a RIS-assisted joint communication and perception system model; S2: Based on the joint communication and perception system model, the echo signal model of the perception link in the system is constructed. Based on the target direction parameters and the geometric arrangement of the RIS and sensor array, the vector representation of the echo signal is derived. S3: Based on the echo signal model, the Fisher information matrix FIM is constructed, and the lower bound of the perception parameter estimation accuracy CRB is further derived; S4: Construct a joint optimization problem with the objective function of minimizing CRB. Meanwhile, the minimum signal-to-interference-and-noise ratio constraint and the total transmit power constraint of the communication user are introduced to jointly optimize the BS's transmit beam matrix and the RIS's reflection matrix. S5: The joint optimization problem is reconstructed as a solvable convex optimization problem using semidefinite programming and fractional programming methods. The transmitting beam, RIS reflection matrix, sensing beam matrix, and auxiliary variables are iteratively updated using an alternating optimization algorithm. Finally, the optimization result of minimizing CRB under the communication constraints is obtained.

2. The joint communication and perception method based on FIM-CRB optimization according to claim 1, characterized in that: In step S1, the joint communication and perception system includes a base station, a reconfigurable intelligent surface (RIS), a sensor array, and multiple communication users. The communication path is base station-RIS-user, and the perception path is base station-RIS-target-sensor. RIS is a uniform planar array (UPA) structure, and both the base station (BS) and radar sensor array (sensor) are uniform linear array (ULA) structures. The downlink communication link between the base station BS and the user UE is reflected by the RIS composed of the uniform planar array UPA. The sensing link is that after the BS illuminates the target through the RIS, the sensor array composed of the uniform linear array ULA receives the echo signal.

3. The joint communication and perception method based on FIM-CRB optimization according to claim 2, characterized in that: The joint communication and perception system in step S2 includes a communication channel G between the base station and the RIS, a communication channel h between the RIS and the user k k , the reflection coefficient matrix of RIS is Θ = diag(Φ), where Then the downlink communication channel from BS to user k is: Among them, L c is the channel gain between BS and user k, is additive Gaussian white noise with a mean of 0 and a variance of Gaussian distribution; at the same time, in the sensing channel, the echo channel received by the target in the time block T is: Where α is the complex gain affected by the round-trip path loss and the target's complex reflection factor, and are the direction vectors of the sensor and RIS respectively; let the communication signal be c(t), the radar signal at time t be s(t), and set W and V to be the beamforming matrices of the communication signal and radar respectively, then the transmission signal at time t is expressed as: X(t)=Wc(t)+Vs(t) Based on the established communication channel model, the SINR of the communication channel is obtained based on the communication and perception channels: And the beam gain: Where α is the complex gain affected by the round-trip path loss and the target's complex reflection factor, and are the direction vectors of the sensor and RIS, respectively.

4. The joint communication and perception method based on FIM-CRB optimization according to claim 3 is characterized in that: The construction of the Fisher information matrix FIM in step S3 includes: The echo channel received by the sensor array is vectorized to obtain the channel vector form: Assume that the unknown parameters are in and According to the likelihood function expression of the complex Gaussian distribution, the FIM of the system parameter ξ is constructed, where each element [J ξ ] i,j It is obtained by the following formula: Among them, u is the signal mean, C y is the covariance matrix; On this basis, the FIM for estimating the unknown parameter vector ξ is constructed as:

5. The joint communication and perception method based on FIM-CRB optimization according to claim 4 is characterized in that: The derivation of the perception parameter estimation accuracy lower limit CRB in step S3 includes: After the FIM is constructed, the CRB can be given by its inverse matrix, that is, the lower bound of the error of the estimated parameter ξ is: Then, using Schur complement, CRB becomes:

6. The joint communication and perception method based on FIM-CRB optimization according to claim 5, characterized in that: The construction of the joint optimization problem in step S4 includes: Introducing SINR constraints, minimum sensing requirement constraints and maximum energy constraints: G s ≥χ tr(WW H +VV H )≤P According to the monotonicity of the matrix, the function tr(X -1 ) decreases in the semi-positive definite matrix space; therefore, the auxiliary variable U is introduced, and the optimization of CRB is expressed as: G s ≥χ, tr(WW H +VV H )≤P, U±0.

7. The joint communication and perception method based on FIM-CRB optimization according to claim 6, characterized in that: The specific process of equivalently reconstructing the joint optimization problem into a solvable convex optimization problem by using the semi-positive programming and fractional programming methods in step S5 includes: Introduce auxiliary variable t and impose LMI constraints At this time, the optimized objective function becomes: G s ≥χ, tr(WW H +VV H )≤P, U±0.

8. The joint communication and perception method based on FIM-CRB optimization according to claim 7, characterized in that: In step S5, an alternating optimization algorithm is used to iteratively update the transmit beam, RIS reflection matrix, sensing beam matrix, and auxiliary variables, and finally an optimization result that minimizes CRB under communication constraints is obtained. The specific process includes: When optimizing the RIS reflection matrix, in order to ensure the minimum communication SINR requirement constraint, FP is used and an auxiliary variable μ is introduced to convert the non-convex constraint into a convex optimization: The optimal solution The value of For the non-convex constraint G on minimum radar perception s , introduce auxiliary variable B=VV H , the non-convex constraint becomes: G s =tr(V H q H qV)=tr(BQ) Finally, EVD is used to restore the radar signal; When optimizing the beamforming matrix, the optimal solution is obtained by direct iteration; When optimizing auxiliary variables, for LMI conditions, use Schur complement to convert LMI constraints into SDP form: U-t -1 I±0 For the transformed LMI condition, the interior point method is used for iteration: Until convergence, the optimal solution can be obtained.

Images