CN-121984554-A - Robust self-adaptive wave beam forming method based on interference power estimation

CN121984554ACN 121984554 ACN121984554 ACN 121984554ACN-121984554-A

Abstract

The invention discloses a robust self-adaptive wave beam forming method based on interference power estimation, which considers the scene of airspace anti-interference by utilizing a multi-antenna array receiver in the environment with interference signals in different incoming wave directions. Firstly, a feature subspace is constructed by using a sampling signal, and the pattern nulls of a sampling matrix inverse beam former are screened to obtain an estimated value of an arrival angle of an interference signal. The desired signal steering vector is then estimated by principal eigenvector extraction. And then estimating the power of the interference signal by constructing a bias projection operator, so as to obtain a reconstructed interference plus noise covariance matrix. And finally, obtaining the beam forming device by using the reconstructed interference plus noise covariance matrix and the estimated expected signal steering vector. The invention has complete model and reasonable and effective design method, can effectively improve the interference suppression capability of the receiving end under the scene that the signal arrival angles are mismatched, and provides specific implementation steps.

Inventors

ZHANG XIAOKAI
LIANG KEXIN
ZHOU XIN
LIU YIYUAN
XU YUHUA
XU YIFAN
FENG ZHIBIN
CHEN RUNFENG
LIU SONGYI
HAN HAO
Xiong Yanan

Assignees

中国人民解放军陆军工程大学

Dates

Publication Date: 20260505
Application Date: 20250715

Claims (5)

1. A robust adaptive beamforming method based on interference power estimation, comprising the steps of: Step 1, constructing a characteristic subspace, and further estimating an arrival angle of an interference signal; Step2, estimating a desired signal steering vector through main feature vector extraction; step 3, estimating the power of the interference signal, and further reconstructing an interference plus noise covariance matrix; and 4, obtaining the robust self-adaptive beam forming device.
2. The robust adaptive beamforming method according to claim 1, wherein step 1 is specifically: the receiver array is set to be a non-uniform linear array containing M array elements, L+1 far-field narrowband signals are emitted to the array in space, wherein the expected signals are expressed as s 0 (t), the L interference signals are expressed as s 1 (t),…,s L (t), and the sampling data of the receiving array at the kth (k=1, 2, the..K) sampling point are expressed as Wherein the method comprises the steps of Respectively representing the desired signal component, the interference signal component and the noise, and satisfying the statistical independence of the three, wherein θ 0 represents the desired signal arrival angle DOA, θ l represents the arrival angle of the first interference signal, a (θ) is a guide vector related to the arrival angle θ, and is expressed as Wherein (-) T denotes the transpose operation, x m (m=1, 2,., M) denotes the position coordinates of the mth element, λ denotes the signal wavelength, and the receiver filtering weight is set to be The result of processing the sampled data x (k) with the filter weight w is Wherein (-) H represents the conjugate transpose operation, and the output SINR is expressed as Wherein the method comprises the steps of Representing the desired signal power, R i+n represents the interference plus noise covariance matrix, as follows Wherein the method comprises the steps of Representing the power of the first interfering signal, Representing the noise power, I representing an M x M-dimensional unit array, and a minimum variance distortion-free response MVDR beamformer capable of maximizing the output SINR shown in (4), the MVDR beamformer being represented as Using a sampling covariance matrix Instead of R i+n in equation (6), a sample matrix inversion SMI beamformer is obtained, denoted as Wherein the covariance matrix is sampled Represented as The SMI beamformer can cause the pattern to null in the direction of the interfering signal when the interfering signal power is much greater than the noise power, and the normalized array response in the theta direction corresponding to the SMI beamformer is expressed as Assuming that the complement of the angle sector where the expected signal is positioned is omega s ,Ω s and is an interference angle sector omega j , omega s +Ω j forms the whole airspace, determining the potential arrival angle of the interference signal according to (9) the null position of the pattern B SMI (theta) of the SMI beam former on the interference angle sector omega j , and representing the Q angles obtained by the null position of B SMI (theta) on the interference angle sector omega j as a set For sampling covariance matrix The eigenvalue decomposition was performed as follows Wherein the diagonal matrix Λ=diag ([ lambda 1 ,λ 2 ,…,λ M ) is a eigenvalue matrix, u= [ U 1 ,u 2 ,…,u M ] is a corresponding eigenvector matrix, lambda 1 ≥λ 2 ≥…≥λ M is set without loss of generality, and an eigenvalue subspace matrix is constructed as follows Wherein the method comprises the steps of Namely, a feature vector corresponding to a larger feature value is selected to construct a feature subspace matrix, and a function g (theta) is defined as Then for two angles α, β within the interference sector Ω j the following holds g(α)>>g(β),α∈Θ j , (14) Where Θ j represents the set Θ j ＝{θ 1 ,θ 2 ,…,θ L containing the actual angle of arrival of the interfering signal, screening the Q angles within Φ using equation (14) to estimate the actual angle of arrival of the interfering signal, and The angle in (3) is substituted into the function corresponding to the formula (13) to obtain a vector b The elements in vector b are arranged in descending order by permutation matrix G 0 , denoted as The Q angles in Φ are then reordered with a permutation matrix G 0 to yield a vector d, denoted as The first Q 0 elements in vector d are then selected as estimates of DOA for the true interference signal, i.e., Q 0 interference signals DOA selected from Φ are Wherein the method comprises the steps of I.e. Q 0 angles are screened out from Q angles of Φ to be used as the estimation of the interference signal DOA, and the interference power corresponding to Q 0 interference signals is estimated one by one in the subsequent step to reconstruct the interference plus noise covariance matrix.
3. The robust adaptive beamforming method according to claim 1, wherein step 2 is specifically: in order to align the main lobe peak value of the direction diagram with the actual expected signal direction when the expected signal DOA is in mismatch, the expected signal steering vector is estimated, and the space power spectrum is firstly integrated linearly in the expected signal sector omega s to obtain a matrix Then to matrix Performing eigenvalue decomposition Wherein D=Diag ([ ζ 1 ,ξ 2 ,…,ξ M ]) is a eigenvalue matrix, V= [ V 1 ,v 2 ,…,v M ] is a corresponding eigenvector matrix, and ζ 1 ≥ξ 2 ≥…≥ξ M is not lost in generality, and because the space power spectrum value of the expected signal sector omega s at the actual expected signal DOA is larger than the space power spectrum values corresponding to other angles, the space power spectrum value of the expected signal sector omega s is larger than the space power spectrum value corresponding to other angles There will be a significantly larger eigenvalue among the eigenvalues of (a) through the matrix Is expressed as a main feature vector v 1 to estimate a desired signal steering vector Desired signal steering vector estimated in subsequent steps Will be used to estimate the interfering signal power.
4. The robust adaptive beamforming method according to claim 1, wherein step 3 is specifically: first using the expected signal steering vector estimated in step 2 Q 0 interference signal arrival angles estimated in step 1 Construction of Q 0 matrices Further obtain Q 0 matrices Corresponding Q 0 orthogonal projection matrices And then Q 0 oblique projection operators are obtained by the oblique projection theory Represented as In this case, the oblique projection operator is utilized The power of the Q (q=1, 2,., Q 0 ) interfering signal is estimated, and the estimated value of the interfering signal power corresponding to the signal arrival angle gamma q is obtained as follows Wherein the method comprises the steps of For estimating noise power, a sampling covariance matrix is generally taken Reconstructing the interference plus noise covariance matrix according to equations (5), (18), (25), expressed as Covariance matrix reconstructed in subsequent step Will be used to generate the beamformer proposed by the present invention.
5. The robust adaptive beamforming method based on interference power estimation according to claim 1, wherein step 4 obtains a robust adaptive beamformer, specifically: based on the reconstructed interference plus noise covariance matrix Estimated desired signal steering vector The beam former w prop proposed by the present invention is obtained by combining (6).

Description

Robust self-adaptive wave beam forming method based on interference power estimation Technical Field The invention belongs to the technical field of wireless communication, and particularly relates to a robust self-adaptive beam forming method based on interference power estimation. Background In recent years, the rapid development of the array signal processing field makes new interference technologies such as directional interference, intelligent interference and the like increasingly diversified, and makes a serious challenge on the safety and stability of a wireless communication system. In this context, enhancing the spatial domain interference immunity of the receiving device has become a critical issue to be addressed. In order to effectively cope with multi-directional interference, a modern receiving array generally adopts a self-adaptive wave beam forming strategy, and the communication performance is ensured by improving the output signal-to-interference-and-noise ratio of a receiving end. The core of the technology is that the adaptive weighting algorithm is utilized to process the array received signals in real time, and interference signals are restrained to the greatest extent while target signals are enhanced, so that effective spatial filtering is realized. The current mainstream self-adaptive beam forming scheme mainly comprises key technologies such as diagonal loading, characteristic subspace projection, covariance matrix reconstruction and the like. The conventional adaptive beamforming method generally performs well in terms of spatial interference suppression at the communication receiving end. However, when the observed value of the angle of arrival of the desired signal or the interfering signal deviates from the actual value, the interference resistance of the conventional method is significantly reduced. This is mainly due to the fact that the null of the beam pattern is offset from the actual arrival direction of the interfering signal, thereby reducing the anti-interference effect. In addition, if the arrival angles of the desired signals are mismatched, the main lobe peaks of the beam pattern may not be aligned with the incident direction of the desired signals, resulting in partial suppression of the useful signals and finally resulting in a decrease in the signal-to-interference-and-noise ratio of the system output. Disclosure of Invention The invention aims to solve the problems in the background art, and provides a robust self-adaptive wave beam forming method based on interference power estimation, which can improve the accuracy of interference plus noise covariance matrix reconstruction through the interference power estimation, so that a main lobe peak value of a directional diagram is aligned with an expected signal incoming wave direction, and a directional diagram null is aligned with the interference signal incoming wave direction, thereby reducing the sensitivity of a wave beam forming device to signal arrival angle mismatch, and providing specific implementation steps. In order to achieve the purpose of the invention, the invention discloses a robust self-adaptive wave beam forming method based on interference power estimation, which comprises the following steps: Step 1, constructing a characteristic subspace, and further estimating an arrival angle of an interference signal; Step2, estimating a desired signal steering vector through main feature vector extraction; step 3, estimating the power of the interference signal, and further reconstructing an interference plus noise covariance matrix; and 4, obtaining the robust self-adaptive beam forming device. Further, the step 1 specifically comprises the following steps: the receiver array is set to be a non-uniform linear array containing M array elements, L+1 far-field narrowband signals are emitted to the array in space, wherein the expected signals are expressed as s 0 (t), the L interference signals are expressed as s 1(t),…,sL (t), and the sampling data of the receiving array at the kth (k=1, 2, the..K) sampling point are expressed as Wherein the method comprises the steps ofRespectively representing the desired signal component, the interference signal component and the noise, and satisfying the statistical independence of the three, wherein θ 0 represents the desired signal arrival angle DOA, θ l represents the arrival angle of the first interference signal, a (θ) is a guide vector related to the arrival angle θ, and is expressed as Wherein (-) T denotes the transpose operation, x m (m=1, 2,., M) denotes the position coordinates of the mth element, λ denotes the signal wavelength, and the receiver filtering weight is set to beThe result of processing the sampled data x (k) with the filter weight w is Wherein (-) H represents the conjugate transpose operation, and the output SINR is expressed as Wherein the method comprises the steps ofRepresenting the desired signal power, R i+n represents the interference plu