Radar communication integrated beamforming method based on linearly constrained power allocation

Abstract

The present invention belongs to the field of signal processing technology and relates to a radar communication integrated beamforming method based on linear constrained power allocation. The present invention first constructs a MU-MIMO radar communication integrated system model and directly designs the transmit covariance matrices of multiple users. These transmit covariance matrices must satisfy radar pattern constraints, signal-to-leakage-to-noise ratio maximization constraints, and transmit power constraints. All transmit covariance matrices are rank-one decomposed to obtain beamforming weight vectors. The beamforming weight vectors obtained by the present invention have no power loss before and after rank-one decomposition. Through beamforming technology, an effective radar communication integrated spatial power allocation effect is achieved in a multi-user scenario.

Description

Technical Field

[0001] The present invention belongs to the technical field of signal processing and relates to a radar communication integrated beamforming method based on linear constrained power allocation. Background Art

[0002] The number of radar and communication systems is rapidly increasing, wireless spectrum resources are becoming increasingly scarce, and equipment hardware costs are rising. To address these issues, radar-communication integration has become a mainstream research direction. As an interdisciplinary topic combining radar and communications, radar-communication integration aims to achieve the coexistence of radar detection and wireless communications through unified design at the signal and channel levels, while sharing spectrum, energy, space, and time resources. This ultimately improves spectrum utilization efficiency and enhances system integration.

[0003] Beamforming technology is a core technology in the field of array signal processing. It can realize the spatial power allocation of the transmitted signal. When it is applied in the field of radar communication integration, the radar target can be regarded as a unit in the line-of-sight channel, and the communication receiver is a unit in the non-line-of-sight channel. The radar communication integration problem can be regarded as designing the beamforming weight vector of the transmitting array and allocating spatial power to each unit in the line-of-sight channel and the non-line-of-sight channel to simultaneously meet the radar's expected radiation pattern requirements and the communication performance requirements of the communication user.

[0004] Currently, there are methods for spatially allocating power between radar and communications using beamforming technology. A representative method is the multi-user multiple-input multiple-output (MU-MIMO) radar communication joint beamforming method. This method is targeted at radar target detection and multi-user downlink communication scenarios, using the communication signal as the radar detection signal. The MU-MIMO radar communication joint beamforming method first designs a radar transmit covariance matrix that satisfies power constraints and radar pattern constraints, and then designs a multi-user transmit covariance matrix that satisfies signal-to-interference-and-noise ratio constraints. The sum of all user transmit covariance matrices approximates the radar transmit covariance matrix. However, the performance of the beamforming weight vector obtained after the rank-one decomposition of the covariance matrix in this method is significantly attenuated, making it unsuitable for practical applications and unable to achieve efficient radar communication integration. Summary of the Invention

[0005] In response to the above-mentioned existing problems, the purpose of the present invention is to provide an integrated beamforming method for MU-MIMO radar communication based on linear constrained power allocation. For radar target detection and multi-user downlink communication scenarios, a multi-user transmit covariance matrix that simultaneously satisfies radar pattern constraints, signal-to-leakage-and-noise ratio (SLNR) constraints, and transmit power constraints is designed. The covariance matrix designed by the present invention has no power loss before and after rank-one decomposition. The beamforming weight vector obtained after rank-one decomposition can be applied to actual scenarios and can achieve effective MU-MIMO radar communication integrated spatial power allocation effect.

[0006] In order to achieve the above object, the technical solution adopted by the present invention is as follows:

[0007] A radar communication integrated beamforming method based on linear constrained power allocation is proposed. Considering a MU-MIMO radar communication integrated system, the system includes N t uniform linear array transmitting array of array elements, N r The uniform linear array radar receiving array has K elements and K communication users equipped with single antennas. The expected number of radar main lobes is Q and the expected main lobe pointing direction is θ. q ,q=1,...,Q. The channel information between the communication user and the transmitting array is accurately estimated and known by the transmitting end, including the following steps:

[0008] S1, using the communication signal as the radar detection signal, the received signal y at the i-th communication user i (t) is:

[0009]

[0010] Where t is the time parameter. Without loss of generality, the channel between the transmitting array and the single-antenna communication user is considered to be a flat Rayleigh fading channel, and H is K×N t dimensional channel matrix, h i is the i-th row of H. k is the transmit beamforming weight vector of the kth user signal, s k (t) is the communication signal of the kth user, n i (t) is the receiving noise of the i-th communication user.

[0011] The transmission covariance matrix of the kth communication user Then the radar communication combined transmission power P t for:

[0012]

[0013] Here, ||·|| represents the bi-norm of the vector, and tr(·) represents the trace of the matrix.

[0014] S2. Design multi-user transmission covariance matrix

[0015] For K communication users in a non-line-of-sight channel, it is necessary to solve the K user transmission covariance matrix W k ,k=1,...,K, the transmission covariance matrix W of the kth communication user k The SLNR maximization constraint for the kth communication user must be satisfied, and W k The corresponding radar transmit pattern must satisfy the expected radar pattern constraint, and the sum of all transmit covariance matrices must satisfy the overall transmit power constraint. Considering the transmit power is P0, the number of main lobes of the radar expected pattern is Q, and the expected main lobe pointing is θ q ,q=1,...,Q, the pattern is in the sidelobe area The internal demand is less than the preset sidelobe level ρ. Then the following optimization problem needs to be solved:

[0016]

[0017] stdiag(G H W k G)=f k

[0018]

[0019]

[0020]

[0021] rank(W k )=1

[0022] The above problem is a non-convex optimization problem and can be solved by ignoring the rank-one constraint. H is the conjugate transpose, (·) T is the transpose, and diag(·) means taking the diagonal elements of the matrix and arranging them into column vectors. α and ε are the power coefficients of the radar target and the communication user respectively. The matrix G = [CH H ], where C=[a(θ1) … a(θ Q )],a(θ q ),q=1,...,Q is the desired main lobe direction θ q The corresponding steering vector, H is the channel matrix. k =[f r,k f c,k ] T is the power allocation vector, where f r,k Control radar target power, which is a 1×Q-dimensional element. The row vector of f c,kControls the communication user power, which is a 1×K-dimensional row vector with the kth element being ε and the rest being 0. Radar pattern sidelobe area Existence L s Angle sampling, a(θ l ) is the sampling angle θ l The corresponding steering vector. ≥ indicates positive semidefinite, and rank(·) indicates the rank of the matrix.

[0023] S3. Decompose the multi-user transmit covariance matrix rank one to obtain the multi-user beamforming weight vector

[0024] The transmission covariance matrix W of K users k ,k=1,...,K performs rank-one decomposition to obtain the beamforming weight vector w of K users k ,k=1,...,K, the transmission covariance matrix W of the kth user k For example, first calculate the matrix W k Perform singular value decomposition to obtain:

[0025]

[0026] Where U is the matrix of singular vectors, Λ is N t Singular values The real diagonal matrix is ​​formed by taking the first column of the matrix Λ and taking the square root of the first element to get the vector σ:

[0027]

[0028] Then the beamforming weight vector w of the kth user is k It can be expressed as:

[0029] w k =Uσ

[0030] After K-order rank-one decomposition, the beamforming weight vectors of K users are obtained. They contribute the same weight vector w within the radar main lobe range. k The received SLNR of the kth user can be maximized while satisfying the transmit power constraint.

[0031] The present invention provides an integrated beamforming method for MU-MIMO radar communications based on linearly constrained power allocation. The method directly designs the transmit covariance matrices of multiple users, which must satisfy radar pattern constraints, signal-to-leakage-and-noise ratio maximization constraints, and transmit power constraints. All transmit covariance matrices are then rank-one decomposed to obtain beamforming weight vectors. The resulting beamforming weight vectors exhibit no power loss before and after the rank-one decomposition, achieving the spatial power allocation required for effective radar target detection and multi-user downlink communications. BRIEF DESCRIPTION OF THE DRAWINGS

[0032] Figure 1 A flowchart of the process for implementing the present invention;

[0033] Figure 2 Schematic diagram of the MU-MIMO radar communication integrated system model of the present invention;

[0034] Figure 3 The radar transmission pattern obtained by the method proposed in the present invention and the sub-patterns corresponding to each user;

[0035] Figure 4 Comparison of radar transmission patterns obtained by the method proposed in the present invention and the MU-MIMO radar communication joint beamforming method. (a) is the radar transmission pattern obtained by the method proposed in the present invention, and (b) is the radar transmission pattern obtained by the MU-MIMO radar communication joint beamforming method. DETAILED DESCRIPTION

[0036] The technical solution of the present invention will be further described below with reference to the accompanying drawings and specific implementation methods.

[0037] The purpose of this embodiment is to simulate and verify the feasibility and effectiveness of the MU-MIMO radar communication integrated beamforming method based on linear constrained power allocation. Figure 1 As shown, the schematic diagram of the MU-MIMO radar communication integrated system model constructed in the embodiment is shown in the attached Figure 2 In the embodiment, the simulation conditions are set as the transmission power P0 = 20dBm, the noise power N0 = 0dBm, the uniform linear array with the array element spacing of half a wavelength is used as the transmission array, and the number of transmission array elements N t =20. The expected radiation pattern of the radar is a multi-main lobe radiation pattern, the expected number of main lobes is Q=3, and the expected main lobe points to θ q =[-45° 0° 45°], preset sidelobe level ρ = -10dB.

[0038] The radar emission pattern and the sub-patterns corresponding to each user obtained by the method proposed in the present invention are as follows: Figure 3 As shown. It can be found that the radar transmission pattern obtained by the present invention achieves the desired purpose of multiple mainlobe patterns, successfully realizing the control of the mainlobe pointing and the suppression of the sidelobe level. The present invention adopts the equal contribution constraint for the sub-patterns of each user, and the sub-patterns of each user are effectively controlled to the same mainlobe level, while the sidelobe levels are suppressed. Therefore, the radar transmission pattern obtained by the present invention and the sub-patterns of each user are consistent with expectations.

[0039] The comparison of radar emission patterns obtained by the proposed method and the MU-MIMO radar communication joint beamforming method is shown in the figure below. Figure 4 As shown. It can be found that the performance of the radar pattern obtained by the method proposed in the present invention does not decrease before and after the rank-one decomposition, and the final beamforming weight vector power is 20dBm, which still meets the transmission power constraint. The performance of the radar pattern obtained by the MU-MIMO radar communication joint beamforming method is greatly attenuated before and after the rank-one decomposition, and the final beamforming weight vector power is 7.8032dBm, which does not meet the transmission power constraint. The received SLNR of each user achieved by the method proposed in the present invention reached 22.9668dB, which meets the communication requirements. In summary, the beamforming weight vector obtained by the present invention can effectively achieve the spatial power allocation effect required for radar target detection and multi-user downlink communication, and always meets the transmission power constraint, which is more conducive to application in actual scenarios.

Claims

1. Radar communication integrated beamforming method based on linear constrained power allocation for MU-MIMO radar communication integrated system. The system is defined as N t uniform linear array transmitting array of array elements, N r The uniform linear array radar receiving array has K elements and K communication users equipped with single antennas. The expected number of radar main lobes is Q and the expected main lobe pointing direction is θ. q ,q=1,...,Q, the channel information between the communication user and the transmitting array is accurately estimated and known by the transmitting end, which is characterized by, The beamforming method comprises the following steps: S1. Use the communication signal as the radar detection signal and define the received signal y at the i-th communication user i (t) is: Where t is the time parameter, h i is the i-th row of H, where H is K×N t dimensional channel matrix, w k is the transmit beamforming weight vector of the kth user signal, s k (t) is the communication signal of the kth user, n i (t) is the receiving noise of the i-th communication user; The transmission covariance matrix of the kth communication user Then the radar communication combined transmission power P t for: Where ||·|| represents the bi-norm of the vector, and tr(·) represents finding the trace of the matrix; S2. Obtain the multi-user transmit covariance matrix by solving the following optimization problem: s.t.diag(G H W k G)=f k rank(W k )=1 in,(·) H is the conjugate transpose, (·) T is transposed, diag(·) means taking the diagonal elements of the matrix side by side as column vectors, α and ε are the power coefficients of the radar target and the communication user respectively, the matrix G=[CH H ], where C=[a(θ1) … a(θ Q )],a(θ q ) is the desired main lobe direction θ q The corresponding steering vector, q = 1, ..., Q, H is the channel matrix, f k =[f r,k f c,k ] T is the power allocation vector, where f r,k Control radar target power, the elements of 1×Q dimension are The row vector of f ck Control communication user power, which is a 1×K-dimensional row vector with the kth element being ε and the rest being 0; P0 is the transmit power, ρ is the preset sidelobe level, and the radar pattern sidelobe area Existence L s Angle sampling, a(θ l ) is the sampling angle θ l The corresponding steering vector; ≥ indicates semi-positive definite, rank(·) indicates the rank of the matrix; S3, the transmission covariance matrix W of K users k Perform rank-one decomposition to obtain the beamforming weight vector w of K users k ,k=1,...,K, the transmission covariance matrix W for the kth user k , first the matrix W k Perform singular value decomposition to obtain: Where U is the matrix of singular vectors, Λ is N t Singular values The real diagonal matrix is ​​formed; take the first column of the matrix Λ and take the square root of the first element to get the vector σ: Then the beamforming weight vector w of the kth user is k Expressed as: In k =Uσ After K-order rank-one decomposition, the beamforming weight vectors of K users are obtained. They contribute the same weight vector w within the radar main lobe range. k Maximize the received SLNR of the kth user while satisfying the transmit power constraint.

Images