Receiving and transmitting combined robust adaptive beam forming method

Abstract

The invention discloses a receiving and transmitting joint robust adaptive beam forming method, which aims at eliminating expected signal components in a sampling covariance matrix of a receiving end, and comprises the following steps of: firstly, estimating interference power through eigenvalue decomposition of the sampling covariance matrix; secondly, power spectrum integration is carried out on an interference signal arrival angle neighborhood to reconstruct an interference and noise covariance matrix, then matrix construction is carried out based on a channel matrix and the covariance matrix, and a main feature vector is extracted; and finally, based on the channel matrix, the main feature vector and the reconstructed covariance matrix, giving a most suitable design scheme for sending a precoding vector and receiving a beam forming vector. According to the method, the model is complete, the design algorithm is reasonable and effective, the optimal beam forming vector can be provided for the transmitting end and the receiving end in a multi-input multi-output communication scene with serious multipath fading, and specific implementation steps are given.

Description

Technical Field

[0001] The present invention belongs to the technical field of wireless communication, and particularly relates to a transceiver joint robust adaptive beamforming method based on covariance matrix reconstruction. Background Art

[0002] In the field of wireless communication, the propagation of electromagnetic waves in space has natural openness and broadcast characteristics, making it extremely easy for the communication receiver to be interfered with unintentionally or maliciously by a third party. Robust adaptive beamforming technology is an important array antenna technology, which is widely used in military or civilian wireless communication fields and can effectively improve the signal-to-interference-plus-noise ratio (SINR) at the receiving end and improve communication quality. Robust adaptive beamforming technology usually refers to using an adaptive beamformer to perform adaptive weighting processing on the signal vector sampled by the receiving end array, so that interference signals from different directions are suppressed, which is a means of anti-interference in the spatial domain. Currently, beamformers can be classified into various technologies according to the design principle, such as diagonal loading, subspace projection, uncertainty set technology, covariance matrix reconstruction, etc.

[0003] Traditional robust adaptive beamforming algorithms usually exhibit excellent interference suppression performance in non-scattering and weak scattering environments. However, in scenarios with severe multipath fading caused by scatterers, the spatial domain anti-interference performance of traditional robust adaptive beamforming algorithms will deteriorate sharply. In this case, the data sampled by the receiving end array at different snapshots contains mixed signals with different amplitudes and time delays, which makes it difficult for the adaptive beamformer to effectively suppress interference and extract the desired signal. Summary of the Invention

[0004] The purpose of the present invention is to solve the problems proposed in the background art, and propose a transceiver joint robust adaptive beamforming method, which can provide the optimal beamforming vectors for both the transmitting and receiving ends in a multiple-input multiple-output (MIMO) communication scenario with severe multipath fading, and specific implementation steps are given.

[0005] To achieve the purpose of the present invention, the present invention discloses a transceiver joint robust adaptive beamforming method, including the following steps:

[0006] Step 1: Establish a joint optimization problem for the transceiver beamforming vectors;

[0007] Step 2: Solve the joint optimization problem to obtain the theoretical optimal solutions of the transmit precoding vector and the receive beamforming vector;

[0008] Step 3: Reconstruct the interference plus noise covariance matrix;

[0009] Step 4: Give the best choices of the transmit precoding vector and the receive beamforming vector in practice.

[0010] Furthermore, in step 1, considering the multi-input multi-output communication scenario, assume that the transmitter is equipped with a linear uniform array with N array elements, and the receiver array has M array elements; the receiver array receives L + 1 far-field narrowband signals, the desired signal is s0(t), and the L interference signals are s1(t), …, s L (t). The sampling vector of the received signal at the k-th (k = 1, 2, …, K) snapshot is

[0011]

[0012] where represent the desired signal, interference signal, and noise respectively, and the three are statistically independent; x m (k) represents the received signal of the m-th antenna, represents the precoding vector at the transmitting end, represents the fading channel matrix, and θ l represents the angle of arrival of the l-th interference signal s l (t), and a(θ l ) is defined as the steering vector and is expressed as follows

[0013]

[0014] where d is the spacing between adjacent antennas and λ is the wavelength;

[0015] If the weighting vector at the receiving end is w, then the received signal at the receiving end becomes

[0016]

[0017] To suppress interference and noise, the following optimization problem is to be solved.

[0018]

[0019] The optimization objective is to maximize the output signal-to-interference-plus-noise ratio, and the constraint is that the maximum precoding power at the transmitting end is A0, where is the desired signal power, and R i+n is the interference-plus-noise covariance matrix (IPNCM), which is expressed as follows:

[0020]

[0021] where is the power of the l-th interference signal, is the noise power.

[0022] Furthermore, in step 2, from the observation, the optimal solution of the optimization problem (4) must satisfy The optimization problem (4) is equivalent to

[0023]

[0024] Since only the molecule contains f in the optimization objective, w and f are optimized separately in sequence;

[0025] First, assume that f is fixed and optimize and solve for w; at this time, transform Equation (6) into

[0026]

[0027] Using the Lagrange multiplier method, let the Lagrange multiplier be λ, then the objective function is

[0028]

[0029] Take the derivative of g(w) and make g'(w) = 0, that is

[0030] R i+n w = λHf (9)

[0031] Furthermore, we get

[0032]

[0033] From Equation (7), we know that w H Hf = 1, that is (Hf) H w = 1, multiply both sides of Equation (10) by (Hf) H We get

[0034]

[0035] That is, we get

[0036]

[0037] Substitute Equation (12) into Equation (10) to get

[0038]

[0039] Substitute Equations (10) and (12) into the optimization objective w of Equation (7) H R i+n w, and the received power is

[0040]

[0041] Substitute Equations (10) and (12) into the SINR optimization objective of Equation (6) We get

[0042]

[0043] Since the in Equation (15) is a constant, reduce the result of Equation (15) After multiplying by a factor, it is used as the optimization objective to optimize and solve for the variable f, that is

[0044]

[0045] Solve it using the Lagrange method, and the objective function is

[0046]

[0047] Take the derivative of g(f) with respect to f and set it to 0, that is

[0048]

[0049] It is found that λ is the eigenvalue of the matrix and f is times the unit eigenvector of the matrix Multiply both sides of equation (18) by f on the left H to get That is, the optimization objective of equation (16) is equal to λA0, that is, the larger λ is, the better; The eigenvalue decomposition of

[0050]

[0051] is expressed as follows, where Λ = diag(λ1, λ2, …, λ N )(λ1 ≥ λ2 ≥ … ≥ λ NM ≥ 0) is the eigenvalue matrix, and U = [u1, u2, …, u N is a unitary matrix; when the precoding vector f takes times the unit eigenvector u1 corresponding to the largest eigenvalue, the optimization objective of equation (16) is the largest, that is, the theoretically optimal transmit precoding vector is

[0052]

[0053] Substitute equation (20) into equation (13) to get the theoretically optimal receive beamforming vector as

[0054]

[0055] In the subsequent steps, design the transmit precoding and receive beamforming vectors for practical applications according to the theoretically optimal transmit precoding vector and the theoretically optimal receive beamforming vector.

[0056] Furthermore, in step 3, the sampled signal vector x(k) at the receiver in practical applications contains both the desired signal and interference signal components. Therefore, use the sampled covariance matrix to replace the interference plus noise covariance matrix, and the sampled covariance matrix is expressed as

[0057]

[0058] However, the sampling covariance matrix And the interference plus noise covariance matrix R in equation (21) i+n There is a significant difference, and the sampling covariance matrix is needed to reconstruct the interference plus noise covariance matrix; Note that formula (14) is similar to Capon spatial power spectrum estimation, and R in formula (14) i+n Replace with The following spatial power spectrum estimation strategy is obtained

[0059]

[0060] Where H0(θ) is the channel matrix for the arrival angle θ. In practice, since the true values of the complex scattering coefficient and the departure angle of the non-line-of-sight path are difficult to obtain, H0(θ) is simplified to in represents the departure angle of the desired signal transmitter; combined with the power spectrum estimation strategy of formula (23) The interference plus noise covariance matrix is reconstructed by integration and expressed as

[0061]

[0062] where Θ i is the fan-shaped area where the estimated value of the interference signal arrival angle is located, is the minimum eigenvalue of the sampling covariance matrix; assuming that the D sampling points of the receiver in the interference signal sector interval are θ1, θ2, …, θ D , then the integral operation shown in formula (24) can be approximately expressed as a summation operation, as shown below

[0063]

[0064] In subsequent steps, It will be used to replace R in formula (21) i+n To obtain the proposed receive beamforming vector.

[0065] Furthermore, in step 4, when the transmitter sends a signal with precoding f0, the receiver first estimates the noise power by eigenvalue decomposition of the sampling covariance matrix after receiving the signal, and then estimates the interference plus noise covariance matrix by integration. Then use the reconstructed matrix And the channel matrix H constructs the matrix right Perform eigenvalue decomposition to obtain the unit eigenvector corresponding to the maximum eigenvalue From equations (20) and (25), the optimal design of the actual transmission precoding vector is

[0066]

[0067] wherein denotes the unit eigenvector corresponding to the maximum eigenvalue of; furthermore, from equations (21) and (25), the optimal design scheme of the actual received beamforming vector is

[0068]

[0069] After the receiver obtains the most suitable precoding and received beamforming vectors, it feeds back the optimal precoding vector to the transmitter, and the transmitter then transmits the signal with the optimal precoding vector. The receiving end uses the optimal received beamforming vector to receive the signal and suppress interference and noise.

[0070] Compared with the prior art, the significant progress of the present invention lies in: 1) establishing a transceiver joint robust adaptive beamforming optimization problem and giving the theoretical optimal solutions of the transmit precoding and received beamforming vectors, providing a theoretical reference for the design of transceiver joint beamforming vectors in multiple-input multiple-output (MIMO) communication scenarios; 2) the proposed transceiver joint robust adaptive beamforming method based on covariance matrix reconstruction has a complete model and clear physical meaning, providing a specific implementation scheme for the design of actual transceiver joint beamforming vectors in MIMO communication scenarios with severe multipath fading and interference.

[0071] To more clearly illustrate the functional characteristics and structural parameters of the present invention, the following further explains in conjunction with the accompanying drawings and specific embodiments. Description of the Drawings

[0072] The accompanying drawings described herein are used to provide a further understanding of the present invention and constitute a part of this application. The schematic embodiments of the present invention and their descriptions are used to explain the present invention and do not constitute an improper limitation of the present invention. In the drawings:

[0073] Figure 1 is a schematic diagram of a multiple-input multiple-output (MIMO) communication scenario with interference signals in the present invention;

[0074] Figure 2 is a schematic diagram of the actual signal processing flow in the present invention;

[0075] Figure 3 is a schematic diagram of the curve of the signal-to-interference-plus-noise ratio (SINR) output by the receiver varying with the input signal-to-noise ratio (SNR);

[0076] Figure 4 is a schematic diagram of the curve of the SINR output by the receiver varying with the magnitude of the steering vector mismatch;

[0077] Figure 5 is a schematic diagram of the curve of the SINR output by the receiver varying with the number of non-line-of-sight (NLOS) paths. Detailed implementation manners

[0078] Next, the technical solutions in the embodiments of the present invention will be clearly and completely described in conjunction with the accompanying drawings in the embodiments of the present invention. Obviously, the described embodiments are only a part of the embodiments of the present invention, rather than all the embodiments; based on the embodiments of the present invention, all other embodiments obtained by those of ordinary skill in the art without creative efforts belong to the scope of protection of the present invention.

[0079] A transceiver joint robust adaptive beamforming method can provide the best beamforming vectors for both the transceiver ends in a multiple-input multiple-output (MIMO) communication scenario with severe multipath fading, and specific implementation steps are given:

[0080] Step 1: Establish a joint optimization problem for transceiver beamforming vectors.

[0081] Step 2: Solve the joint optimization problem to obtain the theoretical optimal solutions of the transmit precoding vector and the receive beamforming vector.

[0082] Step 3: Reconstruct the interference plus noise covariance matrix.

[0083] Step 4: Give the best choices of the transmit precoding vector and the receive beamforming vector in practice.

[0084] Specifically, in Step 1, a joint optimization problem for transceiver beamforming vectors is established.

[0085] Considering a multiple-input multiple-output (MIMO) communication scenario, assume that the transmitter is equipped with a linear uniform array with N array elements, and the receiver array has M array elements. The receiver array receives L + 1 far-field narrowband signals, where the desired signal is s0(t), and the L interference signals are s1(t), …, s L (t). The sampling vector of the received signal at the k-th (k = 1, 2, …, K) snapshot is

[0086]

[0087] where represent the desired signal, interference signals, and noise respectively, and the three are statistically independent. x m (k) represents the received signal of the m-th antenna, represents the transmit precoding vector, represents the fading channel matrix, θ l represents the angle of arrival of the l-th interference signal s l (t), and a(θ l ) is defined as the steering vector and can be expressed as follows

[0088]

[0089] where d is the spacing between adjacent antennas and λ is the wavelength.

[0090] If the receiving end weighting vector is w, the signal at the receiving end becomes

[0091]

[0092] To suppress interference and noise, the following optimization problem is to be solved.

[0093]

[0094] The optimization objective is to maximize the output signal-to-interference-plus-noise ratio, and the constraint is that the maximum precoding power at the transmitting end is A0, where is the desired signal power, and R i+n is the interference-plus-noise covariance matrix (IPNCM), which is expressed as follows:

[0095]

[0096] where is the power of the l-th interference signal, and is the noise power.

[0097] Specifically, in step 2, the joint optimization problem is solved to obtain the theoretical optimal solutions of the transmit precoding vector and the receive beamforming vector.

[0098] It can be observed that the optimal solution of the optimization problem (4) must satisfy The optimization problem (4) is equivalent to

[0099]

[0100] Since only the numerator contains f in the optimization objective, w and f can be optimized sequentially.

[0101] First, assume that f is fixed and optimize w. At this time, equation (6) can be transformed into

[0102]

[0103] Using the Lagrange multiplier method, let the Lagrange multiplier be λ, then the objective function is

[0104]

[0105] Take the derivative of g(w) and set g'(w) = 0, that is

[0106] R i+n w = λHf (9)

[0107] Furthermore, we get

[0108]

[0109] From equation (7), we know that w H Hf = 1, that is, (Hf) H w = 1. Multiply both sides of equation (10) by (Hf) H to get

[0110]

[0111] That is, we get

[0112]

[0113] Substitute equation (12) into equation (10) to get

[0114]

[0115] Substitute equations (10) and (12) into the optimization objective w of equation (7) H R i+n w, and the received power can be obtained as

[0116]

[0117] Substitute equations (10) and (12) into the SINR optimization objective of equation (6) to get

[0118]

[0119] Since in equation (15) is a constant, divide the result of equation (15) by times and use it as the optimization objective to optimize and solve for the variable f, that is

[0120]

[0121] Solve it using the Lagrange method. The objective function is

[0122]

[0123] Let the derivative of g(f) with respect to f be 0, that is

[0124]

[0125] It can be found that λ is the eigenvalue of the matrix , and f is times the unit eigenvector of the matrix Multiply both sides of equation (18) by f on the left H to get That is, the optimization objective of equation (16) is equal to λA0, that is, the larger λ is, the better The eigenvalue decomposition of

[0126]

[0127] where $\Lambda=\text{diag}(\lambda_1,\lambda_2,\ldots,\lambda$ N $)(\lambda_1\geq\lambda_2\geq\ldots\geq\lambda$ NM $\geq0)$ is the eigenvalue matrix, and $U = [u_1,u_2,\ldots,u$ N is a unitary matrix. When the precoding vector $f$ takes the multiple of the unit eigenvector $u_1$ corresponding to the largest eigenvalue, the optimization objective of Equation (16) is maximized, that is, the theoretically optimal transmit precoding vector is

[0128]

[0129] Substituting Equation (20) into Equation (13), the theoretically optimal receive beamforming vector is obtained as

[0130]

[0131] Specifically, in step 3, the interference plus noise covariance matrix is reconstructed.

[0132] In practical applications, the sampled signal vector $x(k)$ at the receiver contains both the desired signal and interference signal components. Therefore, the sampling covariance matrix is usually used to replace the interference plus noise covariance matrix. The sampling covariance matrix is expressed as

[0133]

[0134] However, the sampling covariance matrix and the interference plus noise covariance matrix $R$ in Equation (21) i+n have obvious differences, and the sampling covariance matrix is needed to reconstruct the interference plus noise covariance matrix. Note that Equation (14) is similar to the Capon spatial power spectrum estimation. Replacing $R$ in Equation (14) i+n with the following spatial power spectrum estimation strategy can be obtained

[0135]

[0136] where $H_0(\theta)$ is the channel matrix with respect to the angle of arrival $\theta$. In practice, since it is difficult to obtain the complex scattering coefficient of the non-line-of-sight path and the true value of the angle of departure, $H_0(\theta)$ is simplified to where represents the angle of departure of the desired signal transmitter. Combining the power spectrum estimation strategy of Equation (23) the interference plus noise covariance matrix is reconstructed through integration, which is expressed as

[0137]

[0138] where Θ i is the sector region where the estimated value of the arrival angle of the interference signal is located, is the minimum eigenvalue of the sampling covariance matrix. Assume that the D sampling points of the receiver in the interference signal sector interval are θ1, θ2, …, θ D , then the integral operation shown in Equation (24) can be approximately expressed as a summation operation, as follows

[0139]

[0140] Specifically, Step 4 gives the optimal design scheme of the actual transmitted precoding vector and the received beamforming vector.

[0141] When the transmitter sends a signal with precoding f0, after receiving the signal, the receiver first estimates the noise power by performing eigenvalue decomposition on the sampling covariance matrix, and then estimates the interference-plus-noise covariance matrix through integration Furthermore, use the reconstructed matrix and the channel matrix H to construct the matrix For perform eigenvalue decomposition to obtain the unit eigenvector corresponding to the maximum eigenvalue From Equation (20) and Equation (25), the optimal design scheme of the actual transmitted precoding vector is

[0142]

[0143] where represents the unit eigenvector corresponding to the maximum eigenvalue of; furthermore, from Equation (21) and Equation (25), the optimal design scheme of the actual received beamforming vector is

[0144]

[0145] After the receiver obtains the most suitable precoding and received beamforming vectors, it feeds back the optimal precoding vector to the transmitter, and the transmitter then sends a signal with the optimal precoding vector, and the receiving end uses the optimal received beamforming vector to receive the signal and suppress interference and noise.

[0146] Embodiment

[0147] A joint transmitting and receiving robust adaptive beamforming method, which considers a multiple-input multiple-output (MIMO) communication scenario with severe multipath fading and interfering signals from different directions of arrival. In order to improve the output signal-to-interference-plus-noise ratio (SINR) at the receiving end, adaptive beamforming is performed at both the transmitting and receiving ends to achieve robust spatial anti-interference. To eliminate the desired signal component in the sampling covariance matrix at the receiving end, first, the interference power is estimated through the eigenvalue decomposition of the sampling covariance matrix. Then, power spectral integration is performed in the neighborhood of the arrival angle of the interfering signals to reconstruct the interference-plus-noise covariance matrix. Next, matrix construction is carried out based on the channel matrix and the covariance matrix, and the principal eigenvector is extracted. Finally, based on the channel matrix, the principal eigenvector, and the reconstructed covariance matrix, the most suitable transmit precoding vector and receive beamforming vector design schemes are given. The model of the present invention is complete, and the designed algorithm is reasonable and effective. It can provide the optimal beamforming vectors for both the transmitting and receiving ends in a MIMO communication scenario with severe multipath fading, and specific implementation steps are given. The embodiments of the present invention are specifically described as follows:

[0148] Figure 1 This is the MIMO communication scenario with interfering signals in the present invention. System simulation is carried out using MATLAB software, and the setting of parameters does not affect generality. During simulation, it is assumed that there is always a random error in the steering vector, that is, the actual steering vector is the sum of the nominal vector e and the error steering vector e, expressed as The error steering vector is where ρ reflects the mismatch amplitude of the error steering vector and ρ follows a uniform distribution [0, ρ max ), and φ m (m = 1, 2,..., M) represents phases that are independent of each other and uniformly distributed in [0, 2π). It is assumed that the power of the non-line-of-sight path signal of the desired signal at the receiver is half of the power of the line-of-sight path of the desired signal. The remaining simulation parameters are set as follows: Both the transmitter and the receiver use uniform linear arrays to transmit and receive signals. The size of the transmitter array is N = 8, the size of the receiver array is M = 10, and the element spacing is half a wavelength. The noise is additive white Gaussian noise. The departure angle of the desired signal at the transmitter array is 50°, and the arrival angle at the receiver array is 10°. The arrival angles of the two interfering signals at the receiver array are -30° and 40° respectively. The interference-to-noise ratio is set to 20 dB, the number of snapshots is 30, and the experimental results are the average of 1000 Monte Carlo runs.

[0149] Figure 2It is a schematic diagram of the actual signal processing flow in the present invention. The signal processing flow of the transceiver joint robust adaptive beamforming method based on covariance matrix reconstruction in the present invention can be divided into three stages. In the first stage, the transmitter uses random precoding to precode the symbols to be transmitted and then transmits them. The receiver reconstructs the interference plus noise covariance matrix using the sampled covariance matrix and integral operation, and then designs the transmit precoding vector and receive beamforming vector based on eigenvalue decomposition. In the second stage, the receiver feeds back the designed transmit precoding vector to the transmitter. In the third stage, the transmitter precodes the symbols to be transmitted using the designed precoding vector, and the receiver processes the received signal using the designed receive beamforming vector.

[0150] Figure 3 It is a curve of the output signal-to-interference-plus-noise ratio (SINR) of the receiver varying with the input signal-to-noise ratio (SNR). The simulation parameters are set as follows: the number of non-line-of-sight paths of the desired signal is 5, and the magnitude of the steering vector error is ρ max = 0.3, and the other parameters remain unchanged. The comparison method is the adaptive beamforming method using Gaussian-Legendre integration (T. Luo, P. Chen, Z. Cao, L. Zheng and Z. Wang. URGLQ: An efficient covariance matrix reconstruction method for robust adaptive beamforming [J]. IEEE Transactions on Aerospace Electronic Systems, vol. 59, no. 5, pp. 5634-5645, Oct. 2023.). Figure 3 It shows that the theoretical optimal output SINR of the adaptive beamforming method proposed in the present invention has an advantage of approximately 3 dB compared with the theoretical optimal output SINR of the comparison method. In addition Figure 3 it also shows that in a relatively large range of input SNR variations, the adaptive beamforming method proposed in the present invention exhibits better output SINR performance than the comparison method. The above simulation results show that the adaptive beamforming method proposed in the present invention can still maintain excellent interference suppression performance in a strong scattering environment with steering vector mismatch.

[0151] Figure 4 It is a curve of the output SINR of the receiver varying with the magnitude of the steering vector mismatch. The simulation parameters are set as follows: the input SNR of the receiver is 0 dB, the number of non-line-of-sight paths of the desired signal is 5, and the other parameters remain unchanged. The comparison method is the same as above. Figure 4 It shows that the theoretical optimal output SINR of the adaptive beamforming method proposed in the present invention is significantly greater than the theoretical optimal output SINR of the comparison method. Figure 4It is also shown that under different steering vector mismatch magnitudes, the adaptive beamforming method proposed by the present invention exhibits an output signal-to-interference-plus-noise ratio (SINR) performance 3 dB higher than that of the comparative method. The above simulation results indicate that the adaptive beamforming method proposed by the present invention can still effectively suppress interference in a strong scattering environment and has high robustness to steering vector mismatch.

[0152] Figure 5 It is a curve of the receiver output SINR varying with the number of non-line-of-sight (NLOS) paths. The simulation parameters are set as follows: the receiver input signal-to-noise ratio is 0 dB, the steering vector error magnitude is ρ max = 0.3, and the other parameter settings remain unchanged. The comparative method is the same as above. Figure 5 It shows that when the number of multi-NLOS paths increases, the theoretically optimal output SINR of the comparative method decreases, while the theoretically optimal output SINR of the adaptive beamforming method proposed by the present invention remains stable. Figure 5 It also shows that the output SINR of the adaptive beamforming method proposed by the present invention decreases slowly as the number of NLOS paths increases, while the output SINR of the comparative method decays rapidly as the number of NLOS paths increases. The above simulation results indicate that the adaptive beamforming method proposed by the present invention can still exhibit good interference suppression performance and strong robustness in environments with different multipath fading degrees.

[0153] In summary, the transceiver joint robust adaptive beamforming method based on covariance matrix reconstruction proposed by the present invention establishes a transceiver joint robust beamforming optimization problem, gives the theoretically optimal solutions of the transmit precoding vector and the receive beamforming vector, and gives the most suitable design schemes of the transmit precoding vector and the receive beamforming vector in practical applications based on covariance matrix reconstruction, providing a reference for the actual transceiver joint beamforming vector design in multi-input multi-output communication scenarios with severe multipath fading and interference, and can achieve efficient and robust interference suppression and improve the output SINR at the receiver.

[0154] It should be noted that in this article, relational terms such as first and second are only used to distinguish one entity or operation from another entity or operation, and do not necessarily require or imply any actual relationship or order between these entities or operations. Moreover, the term "comprising", "including" or any other variant thereof is intended to cover non-exclusive inclusion, so that a process, method, article or device comprising a series of elements not only includes those elements, but also includes other elements not expressly listed, or further includes elements inherent to such process, method, article or device.

[0155] Although embodiments of the present invention have been shown and described, those of ordinary skill in the art will appreciate that various changes, modifications, substitutions and variations can be made to these embodiments without departing from the principles and spirit of the present invention. The scope of the present invention is defined by the appended claims and their equivalents.

Claims

1. A joint transmitting and receiving robust adaptive beamforming method, characterized in that It includes the following steps: Step 1: Establish a joint optimization problem for the transmit and receive beamforming vectors; Step 2: Solve the joint optimization problem to obtain the theoretical optimal solutions for the transmit precoding vector and the receive beamforming vector; Step 3: Reconstruct the interference plus noise covariance matrix; Step 4: Give the best choices for the transmit precoding vector and the receive beamforming vector in practice.

2. The method for transceiver joint robust adaptive beamforming according to claim 1, wherein In step 1, considering the multi-input multi-output communication scenario, assume that the transmitter is equipped with a linear uniform array with N array elements, and the receiver array has M array elements; the receiver array receives L + 1 far-field narrowband signals, the desired signal is s0(t), and the L interference signals are s1(t), …, s L (t), and the sampling vector of the received signal at the k-th (k = 1, 2, …, K) snapshot is where represent the desired signal, interference signal, and noise, respectively, and the three are statistically independent; x m (k) represents the received signal of the m-th antenna, represents the precoding vector at the transmitter, represents the fading channel matrix, θ l represents the angle of arrival of the l-th interference signal s l (t), and a(θ l ) is defined as the steering vector and is expressed as follows Where d is the spacing between adjacent antennas and λ is the wavelength; If the receive-end weighting vector is w, the receive-end signal becomes after weighted processing To suppress interference and noise, the following optimization problem is to be solved. The optimization objective is to maximize the output signal-to-interference-plus-noise ratio, and the constraint is that the maximum precoding power at the transmitter is A0, where is the desired signal power, and R i+n is the interference-plus-noise covariance matrix (IPNCM), which is expressed as follows: wherein is the power of the l-th interference signal, is the noise power.

3. A transceiver joint robust adaptive beamforming method according to claim 2, wherein In step 2, it can be observed that the optimal solution of the optimization problem (4) must satisfy The optimization problem (4) is equivalent to Since only the numerator in the optimization objective contains f, w and f are optimized separately in turn; First, assume that f is fixed and optimize w; at this time, equation (6) is transformed into Using the Lagrange multiplier method, let the Lagrange multiplier be λ, then the objective function is Take the derivative of g(w) and make g'(w) = 0, that is R i+n w = λHf (9) Furthermore, we get From Equation (7), we know that w H Hf = 1, that is, (Hf) H w = 1. Multiply both sides of Equation (10) by (Hf) H to obtain That is, we get Substitute equation (12) into equation (10) to get Substitute equations (10) and (12) into the optimization objective \(w\) of equation (7) H R i+n \(w\), and the received power is Substitute equations (10) and (12) into the SINR optimization objective in equation (6). We get Since in Equation (15) is a constant, the result of Equation (15) is scaled down by times and used as the optimization objective to optimize and solve for the variable f, that is Since in Equation (15) is a constant, the result of Equation (15) is scaled down by times and used as the optimization objective to optimize and solve for the variable f, that is Since in Equation (15) is a constant, the result of Equation (15) is scaled down by times and used as the optimization objective to optimize and solve for the variable f, that is Solve it using the Lagrange method, and the objective function is Make the derivative of g(f) with respect to f equal to 0, that is It is found that λ is the eigenvalue of the matrix , and f is the -fold of the unit eigenvector of the matrix ; Multiply both the left and right sides of Equation (18) by f H to obtain That is, the optimization objective of Equation (16) is equal to λA0, that is, the larger λ is, the better; The eigenvalue decomposition of where $\Lambda=\text{diag}(\lambda_1,\lambda_2,\ldots,\lambda$ N $)(\lambda_1\geq\lambda_2\geq\ldots\geq\lambda$ NM $\geq0)$ is the eigenvalue matrix, and $U = [u_1,u_2,\ldots,u$ N is a unitary matrix; when the precoding vector $f$ takes times the unit eigenvector $u_1$ corresponding to the largest eigenvalue, the optimization objective of Equation (16) is maximized, that is, the theoretically optimal transmit precoding vector is times, the optimization objective of Equation (16) is maximized, that is, the theoretically optimal transmit precoding vector is Substitute equation (20) into equation (13) to get the theoretical optimal receive beamforming vector as In the subsequent steps, based on the theoretical optimal transmit precoding vector and the theoretical optimal receive beamforming vector, design the transmit precoding and receive beamforming vectors for practical applications.

4. A joint transceiver robust adaptive beamforming method according to claim 3, wherein In step 3, the sampled signal vector x(k) at the receive end in practical applications contains both the desired signal and interference signal components. Therefore, use the sampling covariance matrix to replace the interference plus noise covariance matrix, and the sampling covariance matrix is expressed as However, the sampled covariance matrix and the interference plus noise covariance matrix \(R\) in Equation (21) i+n have significant differences, and it is necessary to use the sampled covariance matrix to reconstruct the interference plus noise covariance matrix; noting that Equation (14) is similar to the Capon spatial power spectrum estimation, replacing \(R\) in Equation (14) i+n with yields the following spatial power spectrum estimation strategy where \(H_0(\theta)\) is the channel matrix with respect to the angle of arrival \(\theta\); in practice, since it is difficult to obtain the complex scattering coefficient of the non-line-of-sight path and the true value of the angle of departure, \(H_0(\theta)\) is simplified during integration to where represents the angle of departure of the desired signal transmitter; combining the power spectrum estimation strategy of Equation (23) The interference plus noise covariance matrix is reconstructed by integration, denoted as where Θ i is the sector area where the estimated value of the angle of arrival of the interference signal is located, is the minimum eigenvalue of the sampling covariance matrix; Assume that the D sampling points of the receiver in the interference signal sector interval are θ1, θ2, …, θ D , then the integral operation shown in Equation (24) is approximately represented as a summation operation, as follows In subsequent steps, which is used to replace R in Equation (21) i+n to obtain the proposed receive beamforming vector.

5. A joint transceiver robust adaptive beamforming method according to claim 4, wherein In step 4, when the transmitter sends a signal with precoding f0, after receiving the signal, the receiver first estimates the noise power by performing eigenvalue decomposition on the sampled covariance matrix, and then estimates the interference plus noise covariance matrix through integration. Furthermore, use the reconstructed matrix and the channel matrix H to construct the matrix For perform eigenvalue decomposition to obtain the unit eigenvector corresponding to the maximum eigenvalue. From equations (20) and (25), the optimal design scheme for the actual transmitted precoding vector is where denotes the unit eigenvector corresponding to the maximum eigenvalue; furthermore, from equations (21) and (25), the optimal design scheme of the actual received beamforming vector is After the receiver obtains the most suitable precoding and receive beamforming vectors, it feeds back the optimal precoding vector to the transmitter, and the transmitter then transmits the signal with the optimal precoding vector, and the receiver uses the optimal receive beamforming vector to receive the signal and suppress interference and noise.

Images