A matrix reconstruction distributed rotating array anti-jamming method

Abstract

The application discloses a kind of matrix reconstruction distributed rotating array anti-interference method, it includes: establishing the signal model received by isomorphic subarray distributed rotating array;According to signal model, the signal response steering vector of isomorphic subarray distributed rotating array is constructed and sampling covariance matrix is calculated;Based on signal response steering vector and sampling covariance matrix, obtain interference noise covariance matrix;According to signal response steering vector, to obtain the accurate estimation value of signal response steering vector of expected signal;Based on interference noise covariance matrix and the accurate estimation value of signal response steering vector, obtain the weight of digital beam forming of isomorphic subarray distributed rotating array.The application effectively eliminates the information of expected signal in sampling covariance matrix, solves the model mismatch problem, and discrete summation greatly reduces the calculation complexity and application cost, effectively improves the anti-interference ability of distributed rotating array.

Description

TECHNICAL FIELD

[0001] The present application relates to the field of anti-interference technology, in particular to a matrix reconstruction distributed rotating array anti-interference method. BACKGROUND

[0002] With the improvement of the guidance precision requirement of guided cannonball, it is of great significance to study small, low-cost and highly integrated anti-interference satellite navigation system. For the platform of cannonball with a larger caliber, a small distributed array rotating guidance platform can be considered on the cannonball body to ensure the reception of satellite signals and effectively suppress interference signals under the rotating condition of the carrier.

[0003] The main feature of the distributed array is that the distance between the array elements is large, and there is a large error in the robustness of the snapshot and the statistical accuracy of the data. The adaptive beam forming algorithm is very sensitive to the model mismatch when the expected signal appears in the training of the sampling snapshot or the training of the data sample in the application of the distributed array signal processing, and the algorithm performance decreases sharply, and the suppression effect of the interference is poor, which leads to many difficulties in the application of the adaptive beam forming algorithm in the distributed array signal processing. The traditional solution is to use the diagonal loading technology, but the optimal diagonal loading factor is difficult to determine, and the output signal-to-noise ratio obtained by the diagonal loading method is still poor under the condition of large input signal-to-interference-and-noise ratio. In order to eliminate the component of the expected signal, the traditional matrix reconstruction is proposed, but because it needs to perform integral operation on the angle interval of the signal estimation, the calculation complexity is very large, and it is difficult to realize in practical application, and the cost is high. SUMMARY

[0004] In view of this, the present application provides a matrix reconstruction distributed rotating array anti-interference method, which ensures good algorithm performance under high input signal-to-interference-and-noise ratio, reduces the calculation complexity of the matrix reconstruction algorithm, reduces the application cost, and effectively improves the anti-interference ability of the distributed array.

[0005] The present application discloses a matrix reconstruction distributed rotating array anti-interference method, which comprises the following steps:

[0006] Establishing a signal model received by the isomorphic subarray distributed rotating array;

[0007] According to the signal model, the signal response steering vector of the isomorphic subarray distributed rotating array is constructed, and the sampling covariance matrix is calculated;

[0008] Based on the signal response steering vector and the sampling covariance matrix, the interference noise covariance matrix is obtained;

[0009] According to the signal response steering vector, an accurate estimation value of the signal response steering vector of the desired signal is obtained;

[0010] Based on the accurate estimation value of the signal response steering vector and the interference noise covariance matrix, a weight value of digital beamforming of the homogeneous subarray distributed rotating array is obtained.

[0011] Further, the signal model received by the homogeneous subarray distributed array comprises:

[0012] The array model adopted is a homogeneous subarray distributed array, the number of subarrays of which is N, and each subarray is a uniform linear array with M array elements;

[0013] The array element spacing of each subarray is d = λ / 2, i.e. half wavelength, and the baseline length between subarrays is D; assuming that K signals in space are incident to the homogeneous subarray distributed array, taking the first array element of the first subarray as a reference point, the signal model received by the homogeneous subarray distributed array is represented as

[0014] X(t) = Z sv S(t) + N(t)

[0015] Wherein, X(t) is the signal received by the homogeneous subarray distributed array, S(t) is the signal vector of the K signals, N(t) is Gaussian white noise, S(t) and N(t) are statistically independent and irrelevant, and Z sv is the signal response steering vector of the defined homogeneous subarray distributed array.

[0016] Further, the signal response steering vector of the homogeneous subarray distributed array is constructed according to the signal model, comprising:

[0017] Z sv is defined as the signal response steering vector of the homogeneous subarray distributed rotating array, and the expression is:

[0018] Z sv = [Z1, Z2, …, Z N ] T

[0019] Wherein, Z n = [z n (θ1), z n (θ2), …, z n (θ K )]n = 1, 2, …, N is the signal response steering vector matrix of each subarray array, and is the signal response steering vector of the signal θ k , d m-1= (m-1)d, m = 1, 2, ..., M is the distance of each subarray element relative to its first element, where n = 1, 2, ..., N, m = 1, 2, ..., M.

[0020] Further, calculating the sampling covariance matrix based on the signal model includes:

[0021] Based on the algorithm for inverting the sampling covariance matrix in adaptive beamforming, the sampling covariance matrix obtained from maximum likelihood estimation is calculated as follows:

[0022]

[0023] Where L is the number of sampling snapshots.

[0024] Furthermore, based on the signal response steering vector and the sampling covariance matrix, the interference noise covariance matrix is ​​obtained;

[0025] Based on the signal response steering vector, an interference steering vector loop uncertainty set is constructed. Capon spectral integration is performed on this set to obtain the interference covariance matrix C of the l-th interference signal. i-l ;

[0026] By replacing the complex calculations of integration with discrete summation, C can be re-obtained. i-l ;

[0027] For the reacquired C i-l Eigenvalue decomposition is performed, and based on the eigenvalues ​​obtained from the decomposition, a more accurate steering vector estimate and power of the l-th interference signal are obtained.

[0028] Based on a more accurate estimation of the steering vector and power of the l-th interference signal, the interference noise covariance matrix is ​​obtained.

[0029] Furthermore, based on the signal response steering vector, an interference steering vector loop uncertainty set is constructed. Capon spectral integration is performed on this set to obtain the interference covariance matrix C of the l-th interference signal. i-l ,include:

[0030] The angle range Θ of the incoming directions of all K-1 interference signals is obtained using a low-resolution DOA estimation algorithm. int ; ; the angle interval Θ containing all K-1 interference signals iint Divide into K-1 sub-intervals Θ l l = 2, 3, ..., K, each interference signal is located in its corresponding angular interval Θ l Above, all subintervals Θ l Connecting them forms the angle interval Θ int , that is, Θ int =Θ2∪Θ3∪…∪Θl ;;

[0031] The angle interval Θ where the l-th, l=2,3,…,L interference signal is located l Construct the cyclic uncertainty set of the signal response steering vector:

[0032]

[0033] In the uncertain set U z-l (θ∈Θ l The interference covariance matrix of the l-th interference signal is obtained by performing Capon spectral integration on the signal.

[0034]

[0035] Furthermore, the method of replacing complex integration calculations with discrete summation is used to re-obtain C. i-l ,include:

[0036] By approximating the volume integral on the ring uncertainty set surface with the surface integral, then approximating the surface integral with the double integral, and finally replacing the surface integral with discrete summation, the approximate expression for the interference covariance matrix of the l-th interference signal is obtained as follows:

[0037]

[0038] In the formula, P represents the transformation of a continuous angular interval into a discrete angular interval Θ. l ={θ1, θ2, ..., θ P The number of angle values ​​contained in}, where Q represents the angle θ. p The uncertainty set U corresponding to (p = 1, 2, ..., P) z-l (θ p The surface is converted into the number of points contained in a discrete interval.

[0039] Furthermore, the reacquired C i-l Eigenvalue decomposition is performed, and based on the eigenvalues ​​obtained from the decomposition, a more accurate estimate of the steering vector and power of the l-th interference signal is obtained, including:

[0040] For the interference covariance matrix C i-l Perform eigenvalue decomposition:

[0041]

[0042] In the formula, γ l-b For matrix C i-l The eigenvalues ​​of γ are arranged in descending order, i.e., γ l-1 ≥γ l-2 ≥…≥γ l-M×N bl-b is the eigenvector corresponding to the eigenvalue γ l-b ; where b = 1, 2, …, M x N; and rewrite the eigenvalue decomposition of the sample covariance matrix as:

[0043]

[0044] where λ1≥ λ2≥ … ≥ λ K ≥ λ K+1 … ≥ λ M×N are the M eigenvalues of R in descending order, e m is the eigenvector corresponding to the eigenvalue λ ; E s+i = [e1, e2, …, e K ] is the (M x N) x K dimensional signal-interference subspace matrix, whose column vectors span the signal-interference subspace, Δ s+i is the diagonal matrix composed of the eigenvalues corresponding to the signal-interference subspace, E n is the noise subspace, Δ n is the diagonal matrix composed of the eigenvalues corresponding to the noise-interference subspace;

[0045] The coarse estimate of the steering vector of the lth interference signal is projected onto the signal-interference subspace to obtain a more accurate estimate of the steering vector:

[0046]

[0047] At the same time, the Capon spectrum estimation algorithm is used to estimate the power of the interference signal, i.e.

[0048]

[0049] Based on the more accurate estimate of the steering vector of the lth interference signal and its power, the interference noise covariance matrix is obtained, including:

[0050] The power estimation of the Gaussian white noise is:

[0051]

[0052] where λ b is the eigenvalue of the sample covariance matrix corresponding to the noise, b = K + 1, K + 2, …, M x N; and the interference noise covariance matrix can be reconstructed as:

[0053]

[0054] ​Further, the accurate estimation of the signal response steering vector based on the signal response steering vector and the signal is obtained by:

[0055] The preliminary estimation of the signal response steering vector is

[0056]

[0057] Wherein, E s is the signal subspace, and b1 is the eigenvalue corresponding to the eigenvalue of the signal;

[0058] The minimum value of the signal response steering vector error of the expected signal is solved by using the solution of the convex optimization:

[0059]

[0060]

[0061]

[0062] Wherein, e ⊥ is the vertical component of the signal response steering vector error, and the accurate estimation of the signal response steering vector is finally obtained as:

[0063]

[0064] Further, the weight value of the digital beam forming of the isomorphic subarray distributed array is obtained based on the accurate estimation of the signal response steering vector and the interference noise covariance matrix:

[0065] According to the minimum variance distortionless response algorithm of the spatial adaptive beam forming, the weight value of the digital beam forming of the isomorphic subarray distributed rotating array is obtained as:

[0066]

[0067] The weight value is the weight value of the signal steering vector in all directions when the signal received by the final distributed array antenna is subjected to digital beam forming.

[0068] Due to the adoption of the above technical scheme, the present application has the following advantages:

[0069] According to the application, the stable main beam pointing on the desired signal and the effective interference suppression of the rotating distributed array platform can be obtained, the mismatch error sensitivity caused by the platform rotation to the model is prevented, the effective anti-interference is realized, the calculation amount of the matrix reconstruction operation is greatly reduced, the performance requirement of the hardware device is reduced, and the device cost is reduced. While the beam is stable, the slight vibration or the instability of the interference signal source existing in the actual distributed rotating platform can still have good interference signal suppression ability. BRIEF DESCRIPTION OF DRAWINGS

[0070] In order to more clearly illustrate the technical solutions in the embodiments of the present application, the drawings needed to be used in the embodiment description will be briefly introduced. Obviously, the drawings in the following description are only some embodiments of the present application, and other drawings can be obtained by those skilled in the art according to these drawings.

[0071] Figure 1 A flowchart of a matrix reconstruction distributed rotating array anti-interference method of an embodiment of the present application is shown in the figure.

[0072] Figure 2 A homogeneous subarray distributed array model of an embodiment of the present application is shown in the figure. DETAILED DESCRIPTION

[0073] The present application will be further described in combination with the drawings and embodiments. Obviously, the described embodiments are only some of the embodiments of the present application, but not all the embodiments. All other embodiments obtained by those skilled in the art should belong to the scope of protection of the embodiments of the present application.

[0074] In order to realize the application of the adaptive beam forming algorithm in the distributed array, effectively eliminate the mismatch problem caused by the desired signal, ensure that the algorithm performance is still good under high input interference signal noise ratio, reduce the calculation complexity of the matrix reconstruction algorithm and reduce the application cost, and effectively improve the anti-interference ability of the distributed array.

[0075] Referring to Figure 1 , the present application provides an embodiment of a matrix reconstruction distributed rotating array anti-interference method, which includes the following steps:

[0076] Step 1: The array model used first is a homogeneous subarray distributed array, the number of subarrays is N, and the number of array elements of each subarray is M uniform linear array. The schematic diagram of the homogeneous subarray distributed array is shown in Figure 2 .

[0077] The element spacing of the subarray is d = λ / 2, i.e. half wavelength, and the baseline length between the subarrays is D. For K signals incident on the array in space, with the first element of the first subarray as the reference point, the signal model received by the distributed array can be expressed as

[0078] X(t) = Z sv S(t) + N(t)

[0079] where X(t) is the signal received by the homogenous subarray distributed array, S(t) is the signal vector of the K signals, N(t) is the Gaussian white noise, S(t) and N(t) are statistically independent and uncorrelated. Z sv is the signal response steering vector of the distributed array defined by the method.

[0080] Step 2: define Z sv as the signal response steering vector of the homogenous subarray distributed array, the expression can be written as

[0081] Z sv = [Z1, Z2, …, Z N ] T

[0082] where Z n = [z n (θ1), z n (θ2), …, z n (θ K )]n = 1, 2, …, N is the signal response steering vector matrix of each subarray, and is the signal response steering vector of the signal θ k , d m-1 = (m-1)d, m = 1, 2, …, M is the distance of each subarray element relative to its first element, where n = 1, 2, …, N, m = 1, 2, …, M.

[0083] Step 3: according to the sample covariance matrix inversion algorithm in adaptive beamforming, the sample covariance matrix obtained by maximum likelihood estimation can be calculated as:

[0084]

[0085] where L is the number of sample snaps.

[0086] Step 4: apply the modeling method of the ring uncertainty set to the determination of the signal response steering vector of the distributed array, and use the low-resolution DOA estimation algorithm to obtain the angle interval Θ int in which all K-1 interference signals come from. Divide the angle interval Θ int in which all K-1 interference signals come from into K-1 subintervals Θ l(l = 2, 3,..., K), each interference signal is in its corresponding angle interval Θ l All sub-intervals Θ l are connected together to form a large angle interval Θ int , i.e. Θ int = Θ2∪ Θ3∪... ∪ Θ l . The angle interval Θ l can also be obtained by using DOA estimation algorithm.

[0087] Therefore, the ring uncertainty set of signal response steering vector can be constructed on the angle interval Θ l where the lth (l = 2, 3,..., L) interference signal is located:

[0088]

[0089] Obviously, the ring uncertainty set U z (θ ∈ Θ l ) of the lth (l = 2, 3,..., L) interference signal is smaller than the ring uncertainty set U z (θ ∈ Θ l ) of the (l-1)th interference signal. Meanwhile, the steering vector of the lth interference will be located on the ring uncertainty set U z-l (θ ∈ Θ l ), and similarly, the interference covariance matrix of the lth interference signal can be obtained by performing Capon spectral integration on U z-l (θ ∈ Θ l ):

[0090]

[0091] Step 5: In order to reduce the huge calculation amount of integral operation, the surface integral on the ring uncertainty set is first replaced by the volume integral on the surface of the ring uncertainty set, then the double integral is used to approximate the area integral on the ring uncertainty set, and finally the discrete summation is used to replace the area integral, so that the interference covariance matrix of the lth interference signal can be approximately obtained as follows:

[0092]

[0093] In the formula, P represents the number of angle values contained in the discrete angle interval Θ l = {θ1, θ2,..., θ P} converted from the continuous angle interval, and Q represents the number of points contained in the discrete interval converted from the surface of the uncertainty set U z-l (θ p ) corresponding to the angle θ p (p = 1, 2,..., P).

[0094] Step 6: In order to effectively reduce the error caused by the above approximate processing, the matrix Ci-l Eigenvalue decomposition is performed:

[0095]

[0096] where γ l-b are the eigenvalues of the matrix C i-l and are arranged in descending order, i.e., γ l-1 ≥ γ l-2 ≥... ≥ γ l-M×N , and b l-b is the eigenvector corresponding to the eigenvalue γ l-b . The eigenvalue decomposition of the sample covariance matrix is rewritten as

[0097]

[0098] where λ1≥ λ2≥... ≥ λ K ≥ λ K+1 ... ≥ λ M×N are the M eigenvalues of C arranged in descending order, e m is the eigenvector of C corresponding to the eigenvalue λ s+i . E K = [e1, e2,..., e s+i ] is the (M x N) x K dimensional signal-plus-interference subspace matrix, whose column vectors span the signal-plus-interference subspace, Δ n is the diagonal matrix whose entries are the eigenvalues corresponding to the signal-plus-interference subspace. E n is the noise subspace, and Δ l-b is the diagonal matrix whose entries are the eigenvalues corresponding to the noise-plus-interference subspace.

[0099] The coarse estimate of the steering vector of the lthinterfering signal is projected onto the signal-plus-interference subspace to obtain a more accurate estimate of the steering vector

[0100]

[0101] The power of the interfering signal can also be easily estimated using the Capon spectral estimation algorithm, i.e.,

[0102]

[0103] Step 7: The power of the Gaussian white noise can be estimated as:

[0104]

[0105] where γ l-b (b = 1, 2,..., M x N) are the eigenvalues of the sample covariance matrix The eigenvalue corresponding to the noise. In summary, the interference noise covariance matrix can be reconstructed as:

[0106]

[0107] Step 8: calculate the signal response steering vector of the expected signal Where θ1 is the direction of arrival of the expected signal. Through the method of step 6, we can obtain that the preliminary estimation value of the signal response steering vector is

[0108]

[0109] Where, E s Is the signal subspace, and b1 is the eigenvector corresponding to the eigenvalue of the signal. The minimum value of the signal steering vector error of the expected signal is solved by using the solution of convex optimization:

[0110]

[0111]

[0112]

[0113] Where, e ⊥ Is the vertical component of the signal response steering vector error, and the parallel component does not affect its size and can be ignored. Finally, the accurate estimation value of the signal response steering vector is:

[0114]

[0115] Step 9: according to the minimum variance distortionless response algorithm of the spatially adaptive beam forming, the weight value of the digital beam forming of the distributed array is:

[0116]

[0117] The weight value is the weight value of the signal steering vector in all directions when the signal received by the final distributed array antenna is subjected to digital beam forming, and finally the receiver end of the rotating platform can realize a robust adaptive beam former, which can effectively suppress interference signals and improve the anti-interference ability of the rotating distributed array.

[0118] The application adopts an interval discrete summation combined with subspace projection matrix reconstruction method for the homogeneous subarray distributed array, so as to effectively eliminate the information of the expected signal in the sampling covariance matrix, solve the model mismatch problem, greatly reduce the calculation complexity through discrete summation, and further improve the anti-interference ability of the algorithm through the controllable null widening method.

[0119] It should be pointed out finally that the above embodiments are only used to illustrate the technical solutions of the present application but not to limit it. Although the present application has been described in detail with reference to the above embodiments, it should be understood by those skilled in the art that the specific embodiments of the present application can be modified or replaced equivalently without departing from the spirit and scope of the present application, and any modification or equivalent replacement should be covered in the protection scope of the claims of the present application.

Claims

1. A distributed rotating array anti-jamming method of matrix reconstruction, characterized in that, The method comprises the following steps: establishing a signal model of a homogenous subarray distributed rotating array; constructing a signal response steering vector of the homogenous subarray distributed rotating array and calculating a sample covariance matrix according to the signal model; obtaining an interference noise covariance matrix based on the signal response steering vector and the sample covariance matrix; obtaining an accurate estimation value of the signal response steering vector of an expected signal according to the signal response steering vector; obtaining a weight value of digital beam forming of the homogenous subarray distributed rotating array based on the accurate estimation value of the signal response steering vector and the interference noise covariance matrix; the step of establishing the signal model of the homogenous subarray distributed array comprises: the array model adopted is a homogenous subarray distributed array, the number of subarrays of the homogenous subarray distributed array is N, and each subarray is a uniform linear array with M array elements; the array element spacing of each subarray is d = λ / 2, i.e. half wavelength, and the baseline length between the subarrays is D; assuming that K signals in space are incident to the homogenous subarray distributed array, the signal model received by the homogenous subarray distributed array is represented as X(t) = Z sv S(t) + N(t) Wherein, X(t) is the signal received by the homogeneous subarray distributed array, S(t) is the signal vector of K signals, N(t) is the Gaussian white noise, S(t) and N(t) are statistically independent and irrelevant, Z sv is the signal response steering vector of the defined homogeneous subarray distributed array; the step of constructing the signal response steering vector of the homogenous subarray distributed array according to the signal model comprises: Definition Z sv The signal response steering vector for a distributed rotating array of identical subarrays is given by Z sv = [Z1, Z2,..., Z N ] T wherein Z n = [z n (θ1), z n (θ2), …, z n (θ K )]n=1,2,…,N is the signal response steering vector matrix of each subarray array, and is the signal response steering vector of the signal θ k , d m-1 = (m - 1) d, m = 1, 2, …, M is the distance of each subarray element relative to its first element, wherein n = 1, 2, …, N, m = 1, 2, …, M. the step of calculating the sample covariance matrix according to the signal model comprises: the sample covariance matrix obtained by maximum likelihood estimation is calculated according to the sample covariance matrix inversion algorithm in adaptive beam forming, and is wherein L is the number of sampling shots; obtaining the interference noise covariance matrix based on the signal response steering vector and the sample covariance matrix; According to the signal response steering vector, an interference steering vector ring uncertainty set is constructed, Capon spectral integration is performed in the set, and an interference covariance matrix of the lth interference signal is obtained as C i-l ; By the method of discrete summation instead of the complex operation of integration, C i-l ; C i-l Eigenvalue decomposition is performed, and based on the eigenvalues obtained from the decomposition, a more accurate steering vector estimate of the 1th interference signal and its power are obtained; obtaining the interference noise covariance matrix based on the more accurate steering vector estimation and the power of the lth interference signal; The interference steering vector ring uncertainty set is constructed according to the signal response steering vector, Capon spectral integration is carried out in the set, and the interference covariance matrix of the lth interference signal is C i-l , comprising: The angle range Θ of the incoming directions of all K-1 interference signals is obtained using a low-resolution DOA estimation algorithm. int ; ; the angle interval Θ containing all K-1 interference signals int Divide into K-1 sub-intervals Θ l l = 2, 3, ..., K, each interference signal is located in its corresponding angular interval Θ l Above, all subintervals Θ l Connecting them forms the angle interval Θ int , that is, Θ int =Θ2∪Θ3∪…∪Θ l ; In the first, l = 2, 3, …, L, the angle interval Θ l The ring uncertainty set of the signal response steering vector is constructed as follows: On the uncertain set U z-l (θ∈Θ l ) on the Capon spectral integral to obtain the interference covariance matrix of the lth interference signal The method through discrete summation instead of complex operation of integral, reacquire C i-l , comprising: the approximate expression of the interference covariance matrix of the lth interference signal is finally obtained by replacing the volume integral on the ring uncertainty set with the area integral on the surface of the ring uncertainty set, replacing the area integral on the ring uncertainty set with double integral, and replacing the area integral with discrete summation: where P denotes the conversion of continuous angle intervals into discrete angle intervals Θ l = {θ1, θ2, …, θ P} contains the number of angle values, Q denotes the angle θ p corresponding to the uncertainty set U z-l (θ p ) surface into a discrete interval contains more points, p = 1, 2, …, P; The pair of re-acquired C i-l Eigenvalue decomposition is performed, and based on the eigenvalues obtained by the decomposition, a more accurate steering vector estimate of the lth interference signal and its power are obtained. The interference covariance matrix C i-l Eigenvalue decomposition is performed: where γ l-b are eigenvalues of the matrix C i-l and are arranged in descending order, i.e. γ l-1 ≥ γ l-2 ≥... ≥ γ l-M×N , b l-b is the eigenvector corresponding to the eigenvalue γ l-b ; where b = 1, 2,..., M x N; and the eigenvalue decomposition of the sampled covariance matrix is rewritten as: where λ1≥ λ2≥... ≥ λ K ≥ λ K+1 ≥ λ M×N is the M eigenvalues of H in descending order, e m is the eigenvector corresponding to the eigenvalue λ ; E s+i = [e1, e2,..., e K ] is an (M x N) x K dimensional signal-interference subspace matrix, whose column vectors span the signal-interference subspace, Δ s+i is a diagonal matrix whose entries are the eigenvalues corresponding to the signal-interference subspace, E n is the noise subspace, Δ n is a diagonal matrix whose entries are the eigenvalues corresponding to the noise-interference subspace. A coarse estimate of the steering vector of the first interfering signal projected onto the signal-interference subspace of to obtain a more accurate estimate of the steering vector: at the same time, the power of the interference signal is estimated by using the Capon spectrum estimation algorithm, i.e. obtaining the interference noise covariance matrix based on the more accurate steering vector estimation and the power of the lth interference signal comprises: the power estimation of the Gaussian white noise is where λ b is the sample covariance matrix corresponding to the noise, b = K + 1, K + 2, …, M x N; the interference noise covariance matrix can be reconstructed as:

2. The method of claim 1, wherein, obtaining the accurate estimation value of the signal response steering vector of the expected signal according to the signal response steering vector comprises: the preliminary estimation value of the signal response steering vector is wherein E s is the signal subspace, and b1is the eigenvector corresponding to the eigenvalue of the signal. the minimum value of the signal steering vector error of the expected signal is solved by using the solution of convex optimization: where e ⊥ is the vertical component of the signal response steering vector error, and the final accurate estimate of the signal response steering vector is 3. The method of claim 2, wherein, obtaining the weight value of digital beam forming of the homogenous subarray distributed array based on the accurate estimation value of the signal response steering vector and the interference noise covariance matrix comprises: the weight value of digital beam forming of the homogenous subarray distributed rotating array is obtained according to the minimum variance distortionless response algorithm of spatial adaptive beam forming, and is the weight value is the weight value of the signal steering vector in all directions when the signal received by the final distributed array antenna is subjected to digital beam forming.

Images