Omnidirectional Reconfigurable Metasurface-Assisted Array Radar Transmit Beamforming Method for Non-Line-of-Sight Target Sensing

Abstract

This invention discloses a beamforming method for omnidirectional reconfigurable metasurface-assisted array radar for non-line-of-sight target perception. For omnidirectional reconfigurable metasurface-assisted array radar, the method uses maximizing the main lobe gain in both the transmission and reflection regions as the objective function, and constrains the array transmission energy and the transmission and reflection coefficients of the omnidirectional reconfigurable metasurface as constraints. An optimization model is constructed and subjected to an equivalent transformation. A scaled augmented Lagrangian function of the transformed optimization model is then constructed. Based on the ADMM criterion, the transformed optimization model is decomposed into several sub-problems. Non-convex quadratic constrained quadratic programming and semidefinite relaxation techniques are used to solve each sub-problem. The optimized transmission weights and the transmission and reflection coefficients of the omnidirectional reconfigurable metasurface are obtained iteratively. The STARS-assisted array radar in this invention achieves full-space coverage and exhibits better performance in terms of main lobe gain compared to standard RIS-assisted array radar.

Description

TECHNICAL FIELD

[0001] The application belongs to the technical field of radar signal processing and array signal processing, and particularly relates to a full-directional reconfigurable metasurface assisted array radar transmitting beam shaping method for non-line-of-sight target perception. BACKGROUND

[0002] In a complex urban environment, the line-of-sight path between the radar transmitter and the target is blocked, which poses a serious challenge to target perception. Since the reconfigurable intelligent surface (RIS) can intelligently adjust the incident signal, it is used to create a non-line-of-sight path to expand the detectable area. However, RIS requires the radar transmitter and the target to be located on the same side, which limits the detectable area to a hemispherical range. In order to achieve full spatial coverage, multiple RISs need to be deployed, which will increase the hardware cost. The simultaneously transmitting and reflecting surface (STARS) is the key to solving this problem. Unlike traditional RIS, STARS divides the entire area into a transmissive region and a reflective region, and can transmit or reflect incident signals in the two regions respectively, thereby obtaining 360° full-directional coverage.

[0003] STARS was originally applied in wireless communications to expand coverage, improve spectral efficiency, and achieve secure communications. Due to its significant advantages, STARS has also been applied in integrated sensing and communication (ISAC) systems to simultaneously improve the sensing and communication performance of non-line-of-sight paths. For example, Wang et al. utilized STARS to improve parameter estimation performance by reducing the Cramer-Rao bound of radar target parameters, see Z. Wang, X. Mu, and Y. Liu, “STARS enabled integrated sensing and communications,” IEEE Transactions on Wireless Communications, vol. 22, no. 10, pp. 6750-6765, 2023. Wang et al. maximized the average illumination power of multiple radar targets under the constraints of communication user quality-of-service (QoS) and eavesdropping user secrecy rate, see C. Wang, C.-C. Wang, Z. Li, D. W. Kwan Ng, K.-K. Wong, N. Al-Dhahir, and D. Niyato, “STAR-RIS-enabled secure dual-functional radar-communications: Joint waveform and reflective beamforming optimization,” IEEE Transactions on Information Forensics and Security, vol. 18, pp. 4577-4592, 2023. Zhang et al. achieved multi-target detection by maximizing the minimum SINR of all radar targets, see Z. Zhang, W. Chen, Q. Wu, Z. Li, X. Zhu, and J. Yuan, “Intelligent omnisurfaces assisted integrated multi-target sensing and multi-user MIMO communications,” IEEE Transactions on Communications, vol. 13, pp. 4591-4606, 2024. However, all of these studies detect non-line-of-sight targets within a hemispherical space and require knowledge of the direction in which the targets are located. SUMMARY

[0004] The application is a method for omnidirectional reconfigurable metasurface assisted array radar transmit beamforming for non-line-of-sight target perception.

[0005] The technical scheme adopted by the application is: an omnidirectional reconfigurable metasurface assisted array radar transmit beamforming method for non-line-of-sight target perception, comprising:

[0006] For the omnidirectional reconfigurable metasurface assisted array radar, a maximum main lobe gain of a transmission region and a reflection region is taken as an objective function, and an array transmission energy and omnidirectional reconfigurable metasurface transmission coefficients and reflection coefficients are taken as constraints to construct an optimization model, and equivalent transformation is performed on the optimization model;

[0007] A scaled form augmented Lagrangian function of the equivalent transformed optimization model is constructed, the equivalent transformed optimization model is decomposed into a plurality of sub-problems based on an ADMM criterion, a non-convex quadratic constraint quadratic programming and a semi-positive definite relaxation technique are used to solve each sub-problem, and the optimized transmission weight and the omnidirectional reconfigurable metasurface transmission coefficient and reflection coefficient are obtained through iteration.

[0008] Preferably, a maximum main lobe gain of a transmission region and a reflection region is taken as an objective function, and an array transmission energy and omnidirectional reconfigurable metasurface transmission coefficients and reflection coefficients are taken as constraints to construct an optimization model.

[0009]

[0010] In the formula, w is a transmission weight vector of the array radar, V t and V r are respectively a transmission coefficient matrix and a reflection coefficient matrix of the omnidirectional reconfigurable metasurface STARS, θ s and are respectively a wave off angle of the STARS in the transmission region and the reflection region, the STARS has N units in total, v t,n and v r,n are respectively a transmission coefficient and a reflection coefficient of the nth unit of the omnidirectional reconfigurable metasurface STARS, and P is a maximum available transmission power.

[0011] Preferably, a specific method for equivalent transformation of the optimization model is:

[0012] Define Define auxiliary variables and Convert the optimization model into:

[0013]

[0014] In the formula, a H (θ s ) and are the steering vectors of STARS in the transmission and reflection regions, respectively, and G is the channel matrix between the transmit array and STARS.

[0015] Preferably, the scaled augmented Lagrangian function of the constructed equivalent transformed optimization model is specifically:

[0016]

[0017] where ρ>0 is a penalty factor, and are the dual variables, S and Q are the number of discrete main lobe grid points in the transmission and reflection regions, respectively.

[0018] Preferably, based on the ADMM criterion, the equivalent transformed optimization model is decomposed into several sub-problems, which are respectively:

[0019] Sub-problem 1:

[0020]

[0021] Sub-problem 2:

[0022]

[0023] Sub-problem 3:

[0024]

[0025]

[0026] where the superscript k is used to represent the iteration number.

[0027] Preferably, the specific method for solving sub-problem 1 is:

[0028] Define z 1,s = h s + δ s , z 2,q = y q + λ q , z = [z 1,1 ,..., z 1,S , z 2,1 ,..., z 2,Q ] T , a 1,s = a H (θ s ) V t G, A = [a 1,1 ,..., a 1,S , a 2,1 ,..., a 2,Q ]H Ignoring irrelevant terms, subproblem 1 is transformed into

[0029]

[0030] Solve using the Lagrange multiplier method Specifically:

[0031] Define the Lagrange function as:

[0032]

[0033] Where λ>0 are Lagrange multipliers, and the first-order optimality condition of the Lagrange function is:

[0034] z H A(A H A+λI L ) -2 A H z = P

[0035] definition and eigenvalue decomposition into Where D = diag([r1,...,r) L ]) is an eigenvalue matrix with eigenvalues ​​arranged in descending order, and V is the corresponding eigenvector matrix;

[0036] definition The first-order optimal condition is equivalently expressed as:

[0037]

[0038] The optimal λ is obtained using a linear search method. ＊ The optimal solution to subproblem 1 is obtained as follows:

[0039] w ＊ =(A H A+λ ＊ I L ) -1 A H z.

[0040] Preferably, the specific method for solving subproblem 2 is as follows:

[0041] According to the properties of diagonal matrices, V i q = diag(q)v i , i∈{t,r}, v i It is V i The corresponding column vector v i =diag(V i ), where q = Gw; define z t =[z 1,1..., z 1,S ] T , z r = [z 2,1 ..., z 2,Q ] T , ignoring irrelevant terms, subproblem 2 is transformed into:

[0042]

[0043] where B t = A t diag(q), B r = A r diag(q), and A t = [a H (θ1),..., a H (θ S )].

[0044] The above equation is a non-convex non-homogeneous QCQP problem, which is equivalent to the following homogeneous QCQP form:

[0045]

[0046] where,

[0047] SDR is used to solve the homogeneous QCQP form of subproblem 2, define satisfying Y i ≥ 0 and rank(Y i ) = 1, then the SDR form of the QCQP form of subproblem 2 is:

[0048]

[0049] The SDR form of subproblem 2 is a convex problem, which is solved by a standard convex optimization solver, define as the approximate rank-one solution, then the corresponding suboptimal solution is:

[0050]

[0051] Preferably, the specific method for solving subproblem 3 is:

[0052] Define g s = a H (θ s )V t Gw-δ s , Ignoring constant terms, subproblem 3 is simplified to:

[0053]

[0054] In the simplified subproblem 3, h s , y q and ε are coupled, and the optimal h s and y q are expressed as:

[0055]

[0056] The optimal h s and y q are brought into the simplified subproblem 3, and an optimization problem only about variable ε is obtained:

[0057]

[0058] wherein, when |g s |>ε, otherwise when |g q |>ε, otherwise

[0059] Define [σ1,...,σ P ] and [η1,...,η I ] as and After the sequences are de-duplicated and sorted in descending order, [σ1,...,σ P ,η1,...,η I ] are de-duplicated and sorted in descending order to obtain [u1,...,u K ];

[0060] Let u0=0 and u K+1 =∞, and [u1,...,u K ] is changed into a piecewise function N(ε)

[0061] N(ε)={N k (ε)|u k-1 ≤ε≤u k ,k=1,...,K},

[0062] wherein, P' and I' here satisfy and is the kth sub-function, and N(ε) is changed into a quadratic function form N k (ε)=a k ε 2 +b k ε+c k , wherein:

[0063]

[0064] Since a k > 0, The minimum value is determined by ;

[0065] By selecting the minimum value in the K+1 segment function, the optimal ε is obtained:

[0066]

[0067] According to the optimal h s and y q , the optimal and are calculated by the formula

[0068] Preferably, the specific process of obtaining the optimized transmission weight and the omnidirectional reconfigurable metasurface transmission coefficient and reflection coefficient by iteration is as follows:

[0069] S1: Obtain the transmission weight of the k+1 iteration by w ＊ = (A H A+λ ＊ I L ) -1 A H z;

[0070] S2: Obtain the omnidirectional reconfigurable metasurface transmission coefficient and reflection coefficient of the k+1 iteration by ;

[0071] S3: Obtain the ε k+1 of the k+1 iteration according to the formula and

[0072]

[0073] ;

[0074] S4: Obtain the of the k+1 iteration according to the formula:

[0075]

[0076] ;

[0077] S5: Determine whether the iteration stopping condition is met: and If yes, the transmission weight, the transmission coefficient and the reflection coefficient of the STARS obtained in the current iteration are taken as the final results, otherwise, return to step S1.

[0078] Compared with the prior art, the present application has the following advantages: compared with the traditional RIS-assisted array radar, the coverage range of which is only in one hemisphere, the STARS-assisted array radar can achieve 360° full-space detection. In addition, experimental results show that the STARS-assisted array radar can significantly improve the performance in terms of main lobe gain and side lobe level. BRIEF DESCRIPTION OF DRAWINGS

[0079] In order to more clearly illustrate the technical solutions, the drawings used in the prior art description are briefly introduced as follows.

[0080] Figure 1 Figure 1 is a model diagram of the STARS-assisted array radar.

[0081] Figure 2 Figure 2 is a convergence curve of the objective function and the dual variable, Figure 2 (a) is the convergence curve of the objective function, Figure 2 (b) is the convergence curve of the dual variable.

[0082] Figure 3 Figure 3 is a comparison of the performance of the transmitted beam in the transmission area.

[0083] Figure 4 Figure 4 is a comparison of the performance of the transmitted beam in the reflection area. DETAILED DESCRIPTION

[0084] The present application is further illustrated below in combination with the drawings and examples, i.e. a method for transmitting beamforming of an omnidirectional reconfigurable metasurface-assisted array radar for non-line-of-sight target perception.

[0085] A method for transmitting beamforming of an omnidirectional reconfigurable metasurface-assisted array radar for non-line-of-sight target perception, which uses STARS to assist the traditional array radar, realizes 360° full-space detection of non-line-of-sight targets.

[0086] Simultaneously transmitting and reflecting surface (STARS) assisted digital transmit array is used to realize 360° omnidirectional detection of non-line-of-sight targets. Under the constraints of array transmit energy and STARS transmission / reflection coefficient, a transmit beamforming method based on alternating direction methods of multipliers (ADMM) is proposed. Under the constraints of array transmit energy and STARS transmission / reflection coefficient, the algorithm aims to maximize the main lobe gain of the transmission / reflection area at the same time. The final problem is defined as a maxmin problem under non-convex constraints. Based on the ADMM framework, the original problem is decomposed into several subproblems that are easy to solve. Then, non-convex quadratically constrained quadratic program (QCQP) and semi-definite relaxation (SDR) techniques are used to solve it. The specific implementation steps are as follows:

[0087] Step 1: For the array radar assisted by the omnidirectional reconfigurable metasurface STARS, the beam transmitted by the array radar first hits the omnidirectional reconfigurable metasurface, and then is reflected or projected onto the object by the omnidirectional reconfigurable metasurface. The illumination range of the non-line-of-sight signal is divided into two regions, transmission and reflection. The STARS in the present application uses an energy splitting model to simultaneously transmit and reflect signals, and has N unit numbers.

[0088] Define V t and V r as the transmission and reflection coefficient matrices of STARS, respectively, and they are specifically represented as where β i,n ∈[0,1], is the amplitude and phase shift of the nth unit. Since each unit of STARS must comply with the law of conservation of energy, its transmission coefficient and reflection coefficient satisfy |v t,n | 2 +|v r,n | 2 =1. Define the transmit weight vector of the array radar as w=[w1,w2,…,w L ] T , which satisfies where P is the maximum available transmit power.

[0089] Discretize the main lobe directions in the array radar transmission and reflection regions, i.e. and where S and Q are the number of discrete main lobe grid points in the transmission and reflection regions, respectively.

[0090] f(θ s ,w,V t )=a H (θ s )V t Gw and are the radiation beam patterns of the transmit and receive regions, respectively, where is the steering vector of the STARS, θ s and are the wave departure angles of the STARS in the transmit and receive regions, respectively, k is the wave number, and d is the inter-element spacing of the STARS. G is the channel matrix between the transmit array and the STARS.

[0091] To maximize the main lobe gain of both the transmit and receive regions simultaneously, we model this as the following optimization problem:

[0092]

[0093] Step 2: Define and define auxiliary variables and Problem (1) is equivalently transformed as

[0094]

[0095] Step 3: Define the scaled form of the augmented Lagrangian function of problem (26) as

[0096]

[0097] where ρ > 0 is the penalty factor, and are the dual variables.

[0098] Based on the ADMM criterion, solving problem (2) can be reduced to the following subproblems:

[0099]

[0100]

[0101] Step 4: Define z 1,s = h s + δ s , z 2,q = y q + λ q , z = [z 1,1 ,..., z 1,S , z 2,1 ,..., z 2,Q ] T , a 1,s = aH (θ s )V t G, A=[a 1,1 ,...,a 1,S ,a 2,1 ,...,a 2,Q ] H , ignoring the irrelevant terms, problem (4) is transformed into

[0102]

[0103] Problem (9) is solved by using the Lagrange multiplier method. Define the Lagrange function as

[0104]

[0105] where λ > 0 is the Lagrange multiplier. The first-order optimal condition of equation (10) is

[0106] z H A(A H A+λI L ) -2 A H z=P. (11)

[0107] Define and the eigenvalue decomposition of where D = diag([r1,...,r L ]) is the eigenvalue matrix and the eigenvalues are arranged in descending order, and V is the corresponding eigenvector matrix. Define Equation (11) can be equivalently expressed as

[0108]

[0109] The optimal λ ＊ is obtained by using the typical linear search method, so that the optimal solution of problem (9) is

[0110] w ＊ =(A H A+λ ＊ I L ) -1 A H z. (13)

[0111] Step 5: According to the properties of diagonal matrices, V i q=diag(q)v i , i ∈ {t, r}, where q = Gw. Define z t =[z 1,1 ,…,z 1,S ] T , zr = [z 2,1 ,…,z 2,Q ] T , ignoring irrelevant terms, problem (5) is transformed into

[0112]

[0113] where B t = A t diag(q), B r = A r diag(q), and A t = [a H (θ1),...,a H (θ S )].

[0114] Problem (14) is a non-convex non-homogeneous QCQP problem, which can be equivalently represented in the homogeneous QCQP form as

[0115]

[0116] where

[0117] Using SDR to solve problem (15), define which satisfies Y i ≥ 0 and rank(Y i ) = 1, then the SDR form of problem (15) is

[0118]

[0119] Problem (16) is a convex problem, which can be solved by standard convex optimization solvers. Define as the approximate rank-one solution of problem (14), then the suboptimal solution of problem (14) is

[0120]

[0121] Step 6: Define g s = a H (θ s )V t Gw-δ s , ignoring constant terms, problem (6) is simplified as

[0122]

[0123] In problem (18), h s , y qThere is a coupling between h, g and ε. However, when ε is fixed, the optimal h s and y q The optimal ε can be shown to be:

[0124]

[0125] Substituting (19) into (18), we get an optimization problem only in terms of ε

[0126]

[0127] where, when |g s | < ε, otherwise Similarly, when |g q | > ε, otherwise

[0128] Define [σ1,...,σ P ] and [η1,...,η I ] as and After removing duplicates and sorting in descending order, we get [u1,...,u P ] by removing duplicates and sorting [σ1,...,σ I ,η1,...,η K in descending order. Let u0=0 and u K+1 =∞, and we get a piecewise function N(ε)

[0129] N(ε)={N k (ε)|u k-1 ≤ε≤u k ,k=1,...,K}, (21)

[0130] where, is the kth sub-function. We change N(ε) to a quadratic function N k (ε)=a k ε 2 +b k ε+c k , where

[0131]

[0132] Since a k >0, the minimum value of N(ε) is determined by .

[0133] By choosing the minimum value among the K+1 piecewise functions, we get the optimal ε ＊ as

[0134]

[0135] Once ε ＊ is determined, the optimal and

[0136] In summary, the algorithmic flow of the optimization problem is as follows:

[0137] 1) Input parameters: Ω t , Ω r , ρ, K max , τ

[0138] 2) Initialization settings:

[0139] 3) Start iteration:

[0140] i. Obtain w k+1 by equation (13);

[0141] ii. Obtain by equation (17);

[0142] iii. Obtain ε k+1 by equations (22) and (19),

[0143] iv. Determine and

[0144] v. k = k + 1

[0145] 4) If and , stop the loop.

[0146] 5) Output: transmit weight w, transmission and reflection coefficients V t and V r of STARS.

[0147] The present application studies the omni-directional transmit beamforming problem of array radar assisted by STARS under the constraints of radar transmit power and STARS transmission / reflection coefficients.

[0148] Embodiment

[0149] The omni-directional reconfigurable metasurface assisted array radar transmit beamforming method for non-line-of-sight target perception of the present application is further illustrated by matlab simulation.

[0150] 1) Simulation system parameter settings

[0151] In the experiment, the distance between STARS and the transmit array is 15m, and the channel G obeys the Rician distribution. Each simulation is a uniform linear array with L = 16, the element spacing is d = λ / 2, the maximum transmit power is P = 16W, the maximum iteration number is K max = 2000, and the ADMM tolerance is τ = 10 -5 . The scaled dual variables and auxiliary variables are randomly initialized.

[0152] 2) Beam pattern drawing

[0153] In order to intuitively show the effect of the transmit beamforming method, the obtained weight vector w and the transmission and reflection coefficients V t , V r of STARS are weighted and combined, and finally the transmit beam pattern of the STARS-assisted array radar is drawn, and compared with the RIS-assisted transmit beam pattern. The abscissa of the beam pattern is the transmission region of [-180°, 0°] and the reflection region of [0°, 180°], and the ordinate is the minimum value of the main lobe gain in the two regions after being normalized by the path loss into dB.

[0154] 3) Measurement index

[0155] In the present application, the effect of the final beamforming needs to be measured, and in addition to drawing the beam pattern, the minimum main lobe gain index also needs to be used.

[0156]

[0157] Wherein: f (θ s , w, V t ) and are the radiation patterns of the transmission / reflection region. Under the same initial conditions, the larger the minimum main lobe gain, the better the effect of beamforming.

[0158] 4) Result analysis

[0159] The present application has carried out three example simulations.

[0160] Figure 1 is the model diagram of the STARS-assisted array radar; Figure 2 is the convergence curve of the objective function and the dual variable of the algorithm; Figure 3 and Figure 4 are the performance comparison of the STARS-assisted transmit array and the standard RIS-assisted transmit array under the conditions of N = 64 and N = 128.

[0161] In Figure 1In the simulation, a STARS-assisted radar transmitter with a uniform linear array of L elements and a STARS composed of N transmissive-refractive metasurfaces is considered. Assuming that the line-of-sight path between the transmitting array and the target is blocked, the target can only be detected through the non-line-of-sight path generated by the STARS.

[0162] In Figure 2 In (a) and (b), it can be observed that the objective function gradually converges to a stable value as the number of iterations increases, while the dual variable gradually decreases and tends to zero. These phenomena prove the convergence of the algorithm proposed in the present application.

[0163] In Figure 3 and Figure 4 In and, the transmissive / reflective regions are set to [-120°, -100°] and [100°, 120°], respectively, and two standard RISs are used to independently form the transmissive and reflective beams as a comparison object, and the number of metasurface units of them is N RIS =N / 2. It is equivalent to the STARS in the mode switching state, where half of the units work in the transmissive mode and the other half work in the reflective mode. In Figure 3 In, it can be observed that in the transmissive region, the main lobe gain of the STARS-assisted array is 1.85 dB and 1.19 dB higher than that of the standard RIS-assisted array when N=64 and N=128, respectively. In addition, compared with the standard RIS, the STARS significantly reduces the sidelobe level. Specifically, the integrated sidelobe level is 10.47 dB and 13.63 dB lower than that of the standard RIS, respectively. Similarly, in Figure 4 In, it can be observed that in the reflective region, the main lobe gain of the STARS-assisted array is 1.85 dB and 1.19 dB higher than that of the standard RIS-assisted array when N=64 and N=128, respectively, and the integrated sidelobe level is 10.35 dB and 13.29 dB lower than that of the standard RIS, respectively.

[0164] In summary, the STARS-assisted array radar transmitting beamforming method proposed in the present application has good comprehensive performance. Compared with the beamforming method of the standard RIS-assisted array, the proposed method has higher main lobe gain and lower sidelobe level. This indicates that the STARS is an effective method that can realize full-space non-line-of-sight target perception.

Claims

1. A method for omni-directional reconfigurable metasurface assisted array radar transmit beamforming for non-line-of-sight target perception, characterized in that, Comprise: For the omni-directional reconfigurable metasurface assisted array radar, an optimization model is constructed with the main lobe gain of the transmission area and the reflection area as the objective function, and the array transmission energy and the transmission coefficient and the reflection coefficient of the omni-directional reconfigurable metasurface as the constraints, and the optimization model is equivalently transformed, wherein the constructed optimization model is: In the formula, w is the transmitting weight vector of the array radar, V t and V r are respectively the transmission coefficient matrix and the reflection coefficient matrix of the omnidirectional reconfigurable metasurface STARS, θ s and are respectively the wave off angle of the STARS in the transmission and reflection regions, the STARS has N units in total, v t,n and v r,n are respectively the transmission coefficient and the reflection coefficient of the nth unit of the omnidirectional reconfigurable metasurface STARS, and P is the maximum available transmitting power; Wherein, the scaled form augmented Lagrangian function of the optimization model after the equivalent transformation is: Based on the ADMM criterion, the optimization model after the equivalent transformation is decomposed into a plurality of sub-problems, the non-convex quadratic constraint quadratic programming and the semi-positive definite relaxation technology are used to solve each sub-problem, and the optimized transmission weight and the transmission coefficient and the reflection coefficient of the omni-directional reconfigurable metasurface are obtained through iteration; Based on the ADMM criterion, the optimization model after the equivalent transformation is decomposed into a plurality of sub-problems, the non-convex quadratic constraint quadratic programming and the semi-positive definite relaxation technology are used to solve each sub-problem, and the optimized transmission weight and the transmission coefficient and the reflection coefficient of the omni-directional reconfigurable metasurface are obtained through iteration; The specific method for equivalently transforming the optimization model is: The scaled form augmented Lagrangian function of the optimization model after the equivalent transformation is constructed specifically as: where the superscript k is used to denote the iteration number, δ s and λ q are dual variables, h s , y q are auxiliary variables, and ε is a variable parameter.

2. The omni-directional reconfigurable metasurface assisted array radar transmit beamforming method for non-line-of-sight target awareness according to claim 1, wherein, The specific method for solving the sub-problem 1 is: Definitions Defining auxiliary variables and Converting the optimization model to: where a H (θ s ) and are the steering vectors of STARS in the transmission and reflection regions, respectively, and G is the channel matrix between the transmit array and STARS.

3. The omni-directional reconfigurable metasurface assisted array radar transmit beamforming method for non-line-of-sight target awareness according to claim 2, characterized in that, The Lagrangian function is defined as: where p > 0 is a penalty factor, and are dual variables, S and Q are the number of discrete main-lobe grid points in the transmission and reflection regions, respectively.

4. The omni-directional reconfigurable metasurface assisted array radar transmit beamforming method for non-line-of-sight target awareness according to claim 3, characterized in that, Wherein, λ>0 is the Lagrange multiplier, and the first-order optimal condition of the Lagrangian function is: Definition z 1,s = h s + δ s , z 2,q = y q + λ q , z = [z 1,1 ,...,z 1,S , z 2,1 ,...,z 2,Q ] T , a 1,s = a H (θ s ) V t G, A = [a 1,1 ,...,a 1,S , a 2,1 ,...,a 2,Q ] H , neglecting irrelevant terms, subproblem 1 is transformed into The Lagrange multiplier method is used to solve Specifically: The specific method for solving the sub-problem 2 is: The above formula is a non-convex non-homogeneous QCQP problem, which is equivalently represented in the following homogeneous QCQP form: z H A(A H A+λI L ) -2 A H z=P Definition and the eigenvalue decomposition of where D = diag([r1,...,r L ]) is the eigenvalue matrix and the eigenvalues are arranged in descending order, V is the corresponding eigenvector matrix; Definitions The first order optimality condition is equivalent to: The optimal λ is obtained using a linear search method ★ The optimal solution of subproblem 1 is obtained as w ★ = (A H A+λ ★ I L ) -1 A H z。 5. The omni-directional reconfigurable metasurface assisted array radar transmit beamforming method for non-line-of-sight target awareness according to claim 4, characterized in that, The specific method for solving the sub-problem 3 is: According to the properties of diagonal matrices, V i q = diag(q)v i , i∈{t,r}, v i It is V i The corresponding column vector v i =diag(V i ), where q = Gw; define z t =[z 1,1 ,…,z 1,S ] T , z r =[z 2,1 ,…,z 2,Q ] T Ignoring irrelevant terms, subproblem 2 is transformed into: where B t = A t diag(q), B r = A r diag(q), and A t = [a H (θ1),...,a H (θ S )]. By selecting the minimum value in K+1 segments of functions, the optimal ε is obtained: where The SDR is used to solve the homogeneous QCQP form of subproblem 2, defined as satisfying and rank(Y i ) = 1, the SDR form of the QCQP form of subproblem 2 is: Y t (n+1,n+1) = 1, Y r (n+1,n+1) = 1, Subproblem 2 in SDR form is a convex problem, which can be solved using standard convex optimization solvers, defined as The corresponding suboptimal solution for the approximate rank-one solution is 6. The omni-directional reconfigurable metasurface assisted array radar transmit beamforming method for non-line-of-sight target awareness according to claim 5, characterized in that, The specific process for obtaining the optimized transmission weight and the transmission coefficient and the reflection coefficient of the omni-directional reconfigurable metasurface through iteration is: Definition g s = a H (θ s V t Gw-δ s , Neglecting the constant term, the subproblem 3 is simplified as: In the simplified subproblem 3, h s , y q and ε are coupled, and when ε is fixed, the optimal h s and y q are expressed as: The optimal h s and y q into the simplified subproblem 3, we obtain an optimization problem only about variable ε: wherein, when |g s | > ε, else when |g q | > ε, else define [σ1,...,σ P ] and [η1,...,η I ] as and After de-duplication and descending order, the sequence is obtained again [σ1,...,σ P ,η1,...,η I ] de-duplicate and descending order, and then [u1,...,u K ] is obtained. Let u0= 0 and u K+1 = ∞, [u1,..., u K ] becomes a piecewise function N(ε) N(ε) = {N k (ε)|u k-1 ≤ε≤u k ,k = 1,..., K}, wherein, Here P' and I' satisfy and is the kth sub-function, N(ε) is changed into a quadratic function form N k (ε) = a k ε 2 + b k ε + c k wherein: Since a k > 0, the minimum of a is determined by S4: According to the formula: According to the optimal h s and y q The calculation formula calculates the optimal and 7. The omni-directional reconfigurable metasurface assisted array radar transmit beamforming method for non-line-of-sight target awareness according to claim 6, characterized in that, ​ S1: by w ★ = (A H A+λ ★ I L ) -1 A H z, obtain the transmit weight of the k+1th iteration; S2: by obtaining the omni-directional reconfigurable metasurface transmission coefficient and reflection coefficient of the k+1th iteration; S3: According to the formula and ε is obtained for the (k+1)th iteration k+1 , ​ obtaining the (k+1)th iteration of S5: judge whether the iteration stop condition is satisfied or not: and If yes, the transmission weight obtained in the current iteration, the transmission coefficient and the reflection coefficient of STARS are taken as the final result, otherwise, return to step S1.

Images