CN105204008A - Adaptive antenna wave beam forming nulling widening method based on covariance matrix extension - Google Patents

Abstract

The invention discloses an adaptive antenna wave beam forming nulling widening method based on covariance matrix extension. The directions of interference signals reaching array antennas are estimated in advance; then a covariance expansive matrix is obtained and used for replacing a sampling covariance matrix; a nulling widening region is determined, and a steering vector correlation matrix in the widening region is established; linear constraint is formed in the rough direction of interference through feature vectors corresponding to large feature values; the minimum variance of the linear constraint is obtained, and an optimal weighting vector is obtained; data received by an antenna array are weighted through the optimal weighting vector, and output data of the antenna array are obtained. Virtual array elements are established, wider and deeper nulling is achieved under the condition that the number of the array elements is definite, the purpose of restraining interference of position disturbance is achieved, and the anti-interference capacity of an adaptive antenna wave beam forming device is enhanced.

Description

A kind of adaptive antenna Wave beam forming zero based on covariance matrix expansion falls into method for widening

Technical field

The invention belongs to the control field of adaptive array antenna, particularly relate to a kind of adaptive antenna Wave beam forming zero based on covariance matrix expansion and fall into method for widening.

Background technology

Adaptive antenna beam forming technique is weighted by the reception data of pair array antenna, adaptively forms constant gain in wanted signal direction, forms zero and falls into, be widely used in radar, sonar, the fields such as mobile communication at jamming bearing.When disturbing the vibration of rapid movement or carrier platform, or when adaptive weight value updating speed is relatively slow, weight vector training data and application data possibility mismatch, interference likely can not get effectively suppressing, and this can cause the performance of adaptive beamforming device obviously to decline.

Cause the problem of weight vector training data and application data mismatch for interference rapid movement, zero falls into broadening technology falls into by forming wider zero near interference, the effectively undesired signal of suppression rapid movement.Mailloux proposes and utilizes a matrix to carry out to covariance matrix the method (list of references: Covariancematrixaugmentationtoproduceadaptivearraypatter nroughs [J] .ElectronicsLetters that sharpening carrys out null-broadening, 1995,31 (10): 771-772), the method makes zero to fall into depth shallower, and array gain declines.AlonAmar proposes and falls into method for widening (list of references: Alinearlyconstrainedminimumvariancebeamformerwithapre-sp ecifiedsuppressionleveloverapre-definedbroadnullsector [J] based on zero of linear restriction (LCSS), SignalProcessing, 2015,109 (1): 165-171), zero can be deepened fall into while null-broadening, the method that its array gain proposes higher than Mailloux, but the method needs to consume more array freedom.

In practical application, due to the factor such as size, cost, the array number of array antenna is usually very limited, now, the LCSS zero that AlonAmar proposes falls into method for widening, and can reach zero sunken width and the degree of depth all can be restricted because of array number, and cannot reach ideal situation.

Summary of the invention

The object of this invention is to provide a kind of higher output Signal to Interference plus Noise Ratio of can obtaining, the adaptive antenna Wave beam forming zero based on covariance matrix expansion falls into method for widening.

Adaptive antenna Wave beam forming zero based on covariance matrix expansion falls into a method for widening, comprises the following steps,

Step one: pre-estimation is carried out to the undesired signal orientation arriving array antenna;

Step 2: build expansion covariance matrix

Step 3: be by steering vector respective extension

Step 4: determine that zero falls into broadening region Θ, build the steering vector correlation matrix C in broadening region _Θ, feature decomposition is carried out to steering vector correlation matrix, N after expansion ²take out in individual eigenwert individual large eigenwert characteristic of correspondence vector

Step 5: by form linear restriction $\stackrel{\‾}{C}=\[b\left({\mathrm{\θ}}_d\right),{\stackrel{\‾}{U}}_{\stackrel{\‾}{r}}\],$ Its binding occurrence ${\stackrel{\‾}{e}}_1={\left[\begin{array}{cc}1& 0^T\end{array}\right]}^T;$

Step 6: solve linear constraint minimal variance, obtains optimal weighting vector

Step 7: utilize optimal weighting vector W to be weighted antenna array receiver data, obtains aerial array and exports data.

Beneficial effect:

The present invention is directed to adaptive antenna when there is interference position disturbance, the problem that output performance declines.The inventive method, is expanded by covariance matrix, adds the degree of freedom of adaptive antenna, can form more linear restriction, and then obtains wider and darker zero and fall into.

The present invention constructs Virtual array, is equivalent to the array number adding antenna array, there is extra processing gain, thus obtains higher output Signal to Interference plus Noise Ratio.

Accompanying drawing explanation

Fig. 1 is performing step schematic diagram of the present invention;

Fig. 2 is the adaptive beam figure after zero sunken broadening;

Output Signal to Interference plus Noise Ratio curve when Fig. 3 is the change of fast umber of beats;

When Fig. 4 is the change of array input signal-to-noise ratio, corresponding array exports Signal to Interference plus Noise Ratio curve.

Embodiment

Below in conjunction with accompanying drawing, the present invention is described in further details.

The object of the present invention is to provide a kind of method, improve adaptive antenna Wave beam forming zero falls into broadening performance, thus improves adaptive array antenna performance.

The object of the present invention is achieved like this:

The present invention includes following steps:

(1) pre-estimation is carried out to the general orientation of the interference arriving array antenna;

(2) expansion covariance matrix is built, and become new covariance matrix with its alternative sample covariance matrix R.

(3) by steering vector respective extension be $b\left(\mathrm{\θ}\right)=a\left(\mathrm{\θ}\right)\⊗a^{*}\left(\mathrm{\θ}\right)$

(4) determine that zero falls into broadening region Θ and the steering vector correlation matrix built in broadening region, feature decomposition is carried out to steering vector correlation matrix, N after expansion ²take out in individual eigenwert individual larger eigenwert characteristic of correspondence vector

(5) again by form linear restriction, its binding occurrence corresponding

$\begin{array}{cc}\stackrel{\‾}{C}=\[b\left({\mathrm{\θ}}_d\right),{\stackrel{\‾}{U}}_{\stackrel{\‾}{r}}\]& {\stackrel{\‾}{e}}_1={\left[\begin{array}{cc}1& 0^T\end{array}\right]}^T\end{array}$

(6) solve the optimum weight vector of linearly constrained minimum variance beamformer to have

$W\left(\stackrel{\‾}{r}\right)={\tilde{R}}^{-1}\stackrel{\‾}{C}{\left({\stackrel{\‾}{C}}^H{\tilde{R}}^{-1}\stackrel{\‾}{C}\right)}^{-1}{\stackrel{\‾}{e}}_1$

(7) beam forming, utilizes optimal weighting vector W to be weighted antenna array receiver data, obtains aerial array and exports data.

Wherein:

() ^*: represent conjugation;

Θ: represent that wave beam zero falls into broadening region;

R: the second order covariance matrix representing array antenna received data;

representing matrix enlargement oprator, the covariance matrix namely after expansion;

I: representation unit matrix.

represent Kronecker product;

A (θ): the steering vector representing actual array;

B (θ): the virtual array steering vector that representing matrix enlargement oprator is corresponding;

N: represent array antenna array element number;

represent that zero falls into broadening region, the eigenvector of steering vector correlation matrix.

B (θ _d): represent the virtual array steering vector that wanted signal is corresponding.

In step 2, expansion covariance matrix by the conjugation R of a unit matrix I and sample covariance matrix ^*do Kronecker product to obtain.

In step 2, expansion covariance matrix can be loaded by the diagonal angle of 10dB and obtain better performance.

The present invention is not only applicable to single user adaptive antenna beam forming, is also applicable to Multi User Adaptive antenna beam and is shaped.

What the present invention relates to is a kind of control method of adaptive antenna, specifically for adaptive antenna beam-shaper when there is disturbance in interference position, the problem of hydraulic performance decline, propose a kind of adaptive antenna Wave beam forming zero based on covariance matrix expansion and fall into method for widening, the method can pass through constructing virtual array element, obtain darker wider zero to fall into when array number is certain, improve the interference free performance of adaptive antenna.

The present invention includes following steps: first do Kronecker product with unit matrix and sample covariance matrix, thus obtain covariance extended matrix, and with its alternative sample covariance matrix, adopt feature decomposition technology again, by large eigenwert characteristic of correspondence vector, form linear restriction in the roughly direction of interference, null-broadening, reach the object suppressing position to occur the interference of disturbance.The present invention can pass through constructing virtual array element, obtains wider darker zero and falls into, reach the object suppressing position to occur the interference of disturbance, enhance the antijamming capability of adaptive antenna Beam-former when array number is certain.

With reference to Fig. 1, specific embodiment of the invention step is as follows:

Embodiment:

Step 1: pre-estimation is carried out to the general orientation of the interference arriving array antenna;

Step 2: build expansion covariance matrix,

$\tilde{R}=I\⊗R^{*},---\left(1\right)$

Step 3: replace with corresponding for steering vector

Step 4: determine that zero falls into broadening region Θ and the steering vector correlation matrix C built in broadening region _Θ, feature decomposition is carried out to it, obtains

$C_{\mathrm{\Θ}}=\stackrel{\‾}{U}\stackrel{\‾}{\Σ}{\stackrel{\‾}{U}}^H$

Wherein representation feature value diagonal matrix, represent and eigenwert diagonal matrix corresponding eigenvectors matrix.

Step 5: N after expansion ²take out in individual eigenwert individual larger eigenwert characteristic of correspondence vector form linear restriction, have

$\stackrel{\‾}{C}=\[b\left({\mathrm{\θ}}_d\right),{\stackrel{\‾}{U}}_{\stackrel{\‾}{r}}\]---\left(3\right)$

${\stackrel{\‾}{e}}_1={\left[\begin{array}{cc}1& 0^T\end{array}\right]}^T---\left(4\right)$

Step 6: according to linearly constrained minimum variance, solves the optimum weight vector meeting formula (3) (4) and form, has

$W\left(\stackrel{\‾}{r}\right)={\tilde{R}}^{-1}\stackrel{\‾}{C}{\left({\stackrel{\‾}{C}}^H{\tilde{R}}^{-1}\stackrel{\‾}{C}\right)}^{-1}{\stackrel{\‾}{e}}_1---\left(5\right)$

Step 7: beam forming, utilizes optimal weighting vector W to be weighted antenna array receiver data, obtains aerial array and exports data.

Method of the present invention adopts following emulation to verify:

Simulated conditions: suppose that 1 wanted signal incident angle is 0 °, signal to noise ratio snr=0dB, 3 independent interference incident angles are respectively-50 °, 20 ° and 60 °, dry making an uproar compares INR=30dB, also separate between wanted signal and interference, optionally null-broadening centered by-50 ° and 60 °, adaptive antenna is 10 array element uniform straight line arrays, fast umber of beats is 200, the LCSS zero that the method propose the present invention and AlonAmar propose falls into method for widening (list of references: Alinearlyconstrainedminimumvariancebeamformerwithapre-sp ecifiedsuppressionleveloverapre-definedbroadnullsector [J], SignalProcessing, 2015, 109 (1): 165-171) contrast.

In Fig. 2, it is 20 ° that the method zero that the present invention proposes falls into width, and it is 10 ° that the LCSS method zero that AlonAmar proposes falls into width, compares two method directional diagrams.Output Signal to Interference plus Noise Ratio curve when Fig. 3 is the change of fast umber of beats.Output Signal to Interference plus Noise Ratio curve when Fig. 4 is input signal-to-noise ratio change.Can see, under the prerequisite that main lobe width is close, LCSS method is when zero sunken width is 10 °, zero falls into the degree of depth about-70dB, and the method that the present invention proposes is when zero sunken width is 20 °, zero falls into the degree of depth about-140dB, no matter is that zero sunken width or the zero sunken degree of depth all improve one times.This proves that method in this paper is compared with LCSS method, can obtain wider darker zero and fall into.From Fig. 3 and Fig. 4, no matter be change fast umber of beats or change input signal-to-noise ratio, the output Signal to Interference plus Noise Ratio of institute of the present invention extracting method is all the time higher than LCSS method.It can thus be appreciated that the present invention, compared with LCSS method, owing to constructing Virtual array, is equivalent to the array number adding antenna array, can builds more constraint condition, obtain darker wider zero and fall into, higher output Signal to Interference plus Noise Ratio can be obtained again.

Claims (1)

1. the adaptive antenna Wave beam forming zero based on covariance matrix expansion falls into a method for widening, it is characterized in that: comprise the following steps,

Step one: pre-estimation is carried out to the undesired signal orientation arriving array antenna;

Step 2: build expansion covariance matrix

Step 3: be by steering vector respective extension

Step 4: determine that zero falls into broadening region Θ, build the steering vector correlation matrix C in broadening region _Θ, feature decomposition is carried out to steering vector correlation matrix, N after expansion ²the individual large eigenwert characteristic of correspondence vector of r is taken out in individual eigenwert

Step 5: by form linear restriction its binding occurrence ${\stackrel{\‾}{e}}_1={\left[\begin{array}{cc}1& 0^T\end{array}\right]}^T;$

Step 6: solve linear constraint minimal variance, obtains optimal weighting vector

Step 7: utilize optimal weighting vector W to be weighted antenna array receiver data, obtains aerial array and exports data.