Sparse low-rank decomposition DOA estimation method based on virtual array interpolation

By inserting virtual sensors at the locations of holes in a virtual array, a Toeplitz matrix is ​​constructed for sparse low-rank decomposition, overcoming the shortcomings of sparse reconstruction and subspace methods, and achieving higher accuracy and resolution DOA estimation.

CN119001593AActive Publication Date: 2024-11-22Chinese People's Liberation Army Cyberspace Force Information Engineering University

Patent Information

Application Number
CN202411086088.3
Authority / Receiving Office
CN · China
Patent Type
Applications(China)
Current Assignee / Owner
Filing Date
2024-08-08
Publication Date
2024-11-22
Estimated Expiration
2044-08-08

AI Technical Summary

Technical Problem

The basis mismatch problem caused by sparse reconstruction-based DOA estimation methods, and the insufficient utilization of virtual matrix metadata by subspace-based methods, lead to inaccurate DOA estimation.

Method used

A sparse low-rank decomposition method based on virtual array interpolation is adopted. Virtual sensors are inserted at the hole positions of the virtual array to form a virtual uniform array. The Toeplitz matrix is ​​constructed for low-rank and sparse matrix decomposition. The MUSIC algorithm is used for eigenvalue decomposition to estimate DOA.

Benefits of technology

It improves the accuracy and resolution of DOA estimation, especially under low signal-to-noise ratio conditions, and can better utilize all virtual array element information, reduce information loss, and improve estimation accuracy.

✦ Generated by Eureka AI based on patent content.

Smart Images

  • Figure CN119001593A_ABST
    Figure CN119001593A_ABST
Patent Text Reader

Abstract

The invention relates to the technical field of signal processing, in particular to a sparse low-rank decomposition DOA estimation method based on virtual array interpolation, and the method comprises the steps: building a receiving signal model; calculating a sampling covariance matrix of the received signal model, performing vectorization operation on the sampling covariance matrix, inserting a virtual sensor at a hole position of a virtual array # imgabs1 # due to the fact that the virtual array # imgabs0 # has a hole which is a non-uniform virtual array, forming a virtual uniform array # imgabs2 # to reconstruct the received signal model of the virtual uniform array # imgabs3 # into a Toeplitz matrix, and calculating the sampling covariance matrix of the received signal model and performing vectorization operation on the sampling covariance matrix. Carrying out low-rank and sparse matrix decomposition to obtain a signal covariance matrix and a noise covariance matrix; and reconstructing a signal covariance matrix by using a low-rank matrix recovery theory, recovering to obtain an expected signal covariance matrix, carrying out characteristic decomposition on the expected signal covariance matrix by using a MUSIC algorithm, and estimating a target DOA. According to the method, all virtual array element information is fully utilized, and inaccurate DOA estimation caused by information loss is avoided.
Need to check novelty before this filing date? Find Prior Art

Description

Technical Field

[0001] The present invention relates to the field of signal processing technology, and in particular to a sparse low-rank decomposition DOA estimation method based on virtual array interpolation. Background Art

[0002] Direction of Arrival (DOA) estimation is a major research technology in the field of array signal processing. It is mainly used to obtain the spatial positions of multiple desired signals in the airspace and has broad application prospects in both military and civilian fields. Currently, the methods for DOA estimation of signals on coprime arrays can be divided into two categories: sparse reconstruction methods based on compressed sensing and subspace methods. Sparse reconstruction methods mainly use the sparsity of the signal source relative to the entire space to design and solve optimization problems. One solution is to over-completely represent the spatial power spectrum with predefined grid points, build a data fitting model based on sparse theory, and use the Least Absolute Shrinkage and Selection Operator (Lasso) method for sparse signal recovery. However, this will lead to basis mismatch problems, resulting in estimation errors when estimating the direction of arrival. One of the main methods in the subspace class is the DOA estimation method based on the virtual domain. The core idea of ​​this method is to vectorize the virtual array data model, obtain a full-rank covariance matrix through spatial smoothing, and then introduce the MUSIC algorithm for DOA estimation. This method utilizes the structural advantages of sparse arrays and can effectively increase the degrees of freedom, but it only utilizes the continuous part of the virtual array, resulting in partial information loss. Summary of the invention

[0003] The purpose of the present invention is to solve the basis mismatch problem caused by sparse reconstruction type DOA estimation methods and the problem of insufficient utilization of virtual array data by subspace type methods. A sparse low-rank decomposition DOA estimation method based on virtual array interpolation is proposed, which makes full use of all virtual array element information and avoids inaccurate DOA estimation caused by information missing.

[0004] To achieve the above purpose, the technical solution adopted is:

[0005] A sparse low-rank decomposition DOA estimation method based on virtual array interpolation, comprising:

[0006] Establish a received signal model;

[0007] Calculate the sampling covariance matrix of the received signal model and perform vector operations on the sampling covariance matrix. The existence of holes is a non-uniform virtual array. Insert virtual sensors at the hole positions to form a virtual uniform array

[0008] Virtual uniform array The received signal model is reconstructed into a Toeplitz matrix, and low-rank and sparse matrix decomposition is performed into a signal covariance matrix and a noise covariance matrix;

[0009] The signal covariance matrix is ​​reconstructed using the low-rank matrix recovery theory, and the expected signal covariance matrix is ​​recovered. The MUSIC algorithm is used to perform eigendecomposition and estimate the target DOA.

[0010] According to the sparse low-rank decomposition DOA estimation method based on virtual array interpolation of the present invention, further, assuming that K far-field narrowband independent signals come from different wave directions θ k (k=1,2,...,K) is incident on a coprime array, then The received signal of an array element at time t is expressed as:

[0011]

[0012] Among them, s k (t) is the original reference signal, The direction is θ k The array manifold of the signal, For the first i Zero-mean Gaussian white noise on the array elements;

[0013] The received signal model expression is:

[0014]

[0015] in, is the array flow matrix, s(t) is the original reference signal, is zero-mean Gaussian white noise.

[0016] According to the sparse low-rank decomposition DOA estimation method based on virtual array interpolation of the present invention, further, the sampling covariance matrix calculation formula of the received signal model is as follows:

[0017]

[0018] Where T represents the number of snapshots, (·) H represents the conjugate transpose of a matrix;

[0019] Vectorize the above formula to get:

[0020]

[0021] Among them, the virtual array flow matrix is the orientation vector, is the conjugate of the steering vector, is the incident signal power vector, is the noise power and I is the unit matrix.

[0022] According to the sparse low-rank decomposition DOA estimation method based on virtual array interpolation of the present invention, further, the virtual uniform array The received signal model is expressed as:

[0023]

[0024] Among them, <·> l is the received signal at the lth virtual array element, Indicates that the lth virtual array element is in the set But not in the collection middle.

[0025] According to the sparse low-rank decomposition DOA estimation method based on virtual array interpolation of the present invention, further, the reconstructed Toeplitz matrix is ​​as follows:

[0026]

[0027] in,

[0028] According to the sparse low-rank decomposition DOA estimation method based on virtual array interpolation of the present invention, the Toeplitz matrix is ​​further decomposed into a signal covariance matrix and a noise covariance matrix:

[0029]

[0030] Where I is L sub ×L sub The unit matrix of the array. Usually, the number of signal sources is much smaller than the number of array elements, so we can get rank(R s )=K≤L sub , that is, the signal covariance matrix R s Has low rank characteristics; the noise covariance matrix R n It is a full rank matrix whose elements are all zero except the diagonal elements, that is, the noise covariance matrix R n It has sparse characteristics, based on which the signal covariance matrix is ​​reconstructed using low-rank matrix recovery theory.

[0031] According to the sparse low-rank decomposition DOA estimation method based on virtual array interpolation of the present invention, further, reconstructing the signal covariance matrix using low-rank matrix recovery theory includes:

[0032] Based on the expected signal covariance matrix Ts and the expected noise covariance matrix T n The low rank and sparse characteristics of compose the following optimization problem:

[0033]

[0034] subject to R T =C○T s +T n

[0035] T s ≥0,T n ∈D +

[0036] Where rank(·) is the rank of the matrix, ||·||0 is the l0 norm, λ is the regularization parameter that balances the l0 norm and the error, and C is a matrix R T Binary matrices of the same dimension, which in R T The value of non-zero positions is 1, and the values ​​of other positions are 0. + is a semi-positive definite diagonal matrix; T s is a matrix satisfying the Hermitian Toeplitz structure;

[0037] Since both the rank minimization problem and the l0 norm are difficult NP problems to solve, the above equation is convexly relaxed to the following form:

[0038]

[0039] subject to T s ≥0,T n ∈D +

[0040] Where tr(·) is the trace of the matrix, and the expected signal covariance matrix T can be restored. s .

[0041] According to the sparse low-rank decomposition DOA estimation method based on virtual array interpolation of the present invention, further, the desired signal covariance matrix T is converted into s The eigendecomposition is further divided into signal subspace and noise subspace, and the expression is:

[0042]

[0043] Among them, U s , U n are the signal subspace and the noise subspace respectively, Σ s ,Σ n are K large eigenvalues ​​and L sub-A diagonal matrix composed of K small eigenvalues; using the orthogonality of the noise eigenvector and the signal eigenvector, the spatial spectrum function is obtained:

[0044]

[0045] Wherein, a(θ) is the steering vector. As the angle θ changes, the value of the spatial spectrum peak is the estimated value of the direction of arrival.

[0046] The beneficial effects achieved by adopting the above technical solution are:

[0047] The sparse low-rank decomposition DOA estimation method based on virtual array interpolation proposed in the present invention adopts virtual array element interpolation technology to insert virtual sensors at the hole positions of the virtual array to form a virtual uniform array. This method does not need to discard the virtual array elements containing holes, so that all virtual array element information can be fully utilized. By constructing a Toeplitz matrix for the interpolated virtual uniform array receiving signal model, and constructing a sparse and low-rank decomposition optimization problem to recover the missing data in the Toeplitz matrix, the DOA estimation of the signal on the coprime array is realized. The present invention can obtain higher estimation accuracy, and can obtain higher estimation accuracy and resolution at low signal-to-noise ratio. BRIEF DESCRIPTION OF THE DRAWINGS

[0048] In order to more clearly illustrate the technical solutions of the embodiments of the present invention, the following briefly introduces the drawings of the embodiments of the present invention, wherein the drawings are only used to illustrate some embodiments of the present invention, but not to limit all embodiments of the present invention thereto.

[0049] Figure 1 4 is a flow chart of a sparse low-rank decomposition DOA estimation method based on virtual array interpolation according to an embodiment of the present invention;

[0050] Figure 2 is a spatial spectrum distribution diagram of the proposed method according to an embodiment of the present invention;

[0051] Figure 3 is the resolution of the proposed method in the embodiment of the present invention;

[0052] Figure 4 is the root mean square error of each method in the embodiment of the present invention changes with the signal-to-noise ratio;

[0053] Figure 5 The root mean square error of each method in the embodiments of the present invention varies with the number of snapshots. DETAILED DESCRIPTION

[0054] The following will be combined with the drawings of specific embodiments of the present invention to clearly and completely describe the exemplary scheme of the embodiment of the present invention. Unless otherwise defined, the technical terms or scientific terms used in the present invention should be the common meanings understood by people with ordinary skills in the field.

[0055] like Figure 1 As shown, this embodiment discloses a sparse low-rank decomposition DOA estimation method based on virtual array interpolation, comprising the following steps:

[0056] Step S101: Establish a received signal model.

[0057] Assume that K far-field narrowband independent signals come from different wave directions θ k (k=1,2,...,K) is incident on a coprime array, then The received signal of an array element at time t is expressed as:

[0058]

[0059] Among them, s k (t) is the original reference signal, The direction is θ k The array manifold of the signal, For the first i Zero mean Gaussian white noise on the array elements. Therefore, as shown in formula (1), the received signal of a single array element is expanded to the received signal on the coprime array, and the expression is:

[0060]

[0061] Among them, the received signal can be expressed as:

[0062]

[0063] The original reference signal can be expressed as:

[0064] s(t)=[s1(t),s2(t),…,s K (t)] T (4)

[0065] The noise data can be expressed as:

[0066]

[0067] The array flow matrix can be expressed as:

[0068]

[0069] The steering vector It can be expressed as:

[0070]

[0071] Step S102: Calculate the sampling covariance matrix of the received signal model and perform vectorization operation on the sampling covariance matrix. The existence of holes is a non-uniform virtual array. Insert virtual sensors at the hole positions to form a virtual uniform array

[0072] The covariance matrix of the received signal model is calculated by formula (2) as follows:

[0073]

[0074] in, is the noise power, I is the (M+N-1)×(M+N-1)-dimensional identity matrix, R s is the signal covariance matrix.

[0075] In practical applications, the received signal model is T snapshots of sampled data. Since the number of snapshots is limited, formula (8) can be written as:

[0076]

[0077] Where T represents the number of snapshots, (·) H Represents the conjugate transpose of a matrix.

[0078] Performing vector operations on formula (9), we get:

[0079]

[0080] Among them, the virtual array flow matrix is the orientation vector, is the conjugate of the steering vector, is the incident signal power vector, is the noise power, and I is the unit matrix. The values ​​of each element in can be regarded as Then the virtual array element position set That is, the virtual array element position is composed of the difference between the positions of each element in the coprime array. Since there are holes after the difference between the positions of each element, the virtual array It is a non-uniform virtual array. The common method of processing virtual arrays is usually to extract the continuous part of the non-uniform array as the processing object and discard the part containing holes. However, this actually discards some virtual array elements, resulting in information loss, which in turn affects the subsequent DOA estimation performance.

[0081] This embodiment adopts virtual array element interpolation technology. Virtual sensors are inserted at the hole positions to form a virtual uniform array It should be noted that the virtual sensor inserted into the virtual array does not really exist, so the virtual sensor can be regarded as an inoperative antenna array element, and the received signal of the array element can be initialized to zero. The remaining array elements in the array are virtual arrays. The array elements in the virtual array receive signals. The received signal of the virtual uniform array The received signal model can be expressed as:

[0082]

[0083] Among them, <·> l is the received signal at the lth virtual array element, Indicates that the lth virtual array element is in the set But not in the collection middle.

[0084] Step S103: virtual uniform array The received signal model is reconstructed into a Toeplitz matrix, and low-rank and sparse matrix decomposition is performed into a signal covariance matrix and a noise covariance matrix.

[0085] Formula (11) can be regarded as a virtual uniform array The single-snap sampling data of the received signal can be used to solve the rank deficiency problem caused by the single-snap sampling data of the received signal by constructing a Toeplitz matrix. The matrix form is as follows:

[0086]

[0087] in,

[0088] At this time, the Toeplitz matrix can also be regarded as a length of L sub The covariance matrix constructed from the uniform array received data can be decomposed into the signal covariance matrix and the noise covariance matrix, namely:

[0089]

[0090] Where I is L sub ×L sub The unit matrix of the array. Usually, the number of signal sources is much smaller than the number of array elements, so we can get rank(R s )=K≤L sub , that is, the signal covariance matrix R sHas low rank characteristics; the noise covariance matrix R n It is a full rank matrix whose elements are all zero except the diagonal elements, that is, the noise covariance matrix R n It has sparse characteristics. Based on this, the signal covariance matrix can be reconstructed using low-rank matrix recovery theory.

[0091] Step S104: reconstruct the signal covariance matrix using low-rank matrix recovery theory to recover the desired signal covariance matrix, perform eigendecomposition on it using the MUSIC algorithm, and estimate the target DOA.

[0092] Based on the expected signal covariance matrix T s and the expected noise covariance matrix T n The low rank and sparse characteristics of compose the following optimization problem:

[0093]

[0094] Where rank(·) is the rank of the matrix, ||·||0 is the l0 norm, λ is the regularization parameter that balances the l0 norm and the error, and C is a matrix R T Binary matrices of the same dimension, which in R T The value of non-zero positions is 1, and the values ​​of other positions are 0. + is a semi-positive definite diagonal matrix; T s is a matrix with Hermitian Toeplitz structure.

[0095] Since the number of signal samples in practical applications is limited, the matrix R T There is usually an estimation error, namely R T ≠C○T s +T n , the following convex error model is constructed based on the Frobenius norm:

[0096]

[0097] Where δ is the constrained expected signal covariance matrix T during the optimization process s The error parameter.

[0098] Based on the above convex error model, formula (14) can be expressed as:

[0099]

[0100] Since the rank minimization problem and l0 norm in formula (16) are both NP problems that are difficult to solve, the above problem can be convexly relaxed as follows:

[0101]

[0102] Since the matrix Rs is a semi-positive definite matrix, formula (17) can be further expressed as:

[0103]

[0104] where tr(·) is the trace of the matrix.

[0105] By solving the above optimization problem, the desired signal covariance matrix T can be recovered s In practical applications, due to the influence of errors, the recovered expected signal covariance matrix T s The influence of noise cannot be completely eliminated, so the MUSIC algorithm can be used to perform feature decomposition and further divide it into signal subspace and noise subspace, namely:

[0106]

[0107] Among them, U s , U n are the signal subspace and the noise subspace respectively, Σ s ,Σ n are K large eigenvalues ​​and L sub -A diagonal matrix composed of K small eigenvalues; using the orthogonality of the noise eigenvector and the signal eigenvector, the spatial spectrum function is obtained:

[0108]

[0109] Wherein, a(θ) is the steering vector. As the angle θ changes, the value of the spatial spectrum peak can be obtained, which is the estimated value of the direction of arrival.

[0110] The present invention simulates and analyzes the proposed method from three aspects: spatial spectrum distribution, resolution, and wave direction estimation accuracy. Set a coprime array of M=5, N=6. It can be considered that the number of physical array elements of the coprime array is 10, the virtual hole positions of the coprime array are {-23,-22,-21,-16,-11,11,16,21,22,23}, and the noise environment is Gaussian white noise with zero mean and unit variance. It should be noted that since the present invention decomposes the covariance matrix into the signal covariance matrix and the noise covariance matrix, the regularization parameter λ related to the noise covariance matrix should not be too large. In the actual simulation, the parameter λ in formula (18) is set to 0.1, and the grid accuracy in the spectrum peak search is set to 0.1°.

[0111] The present invention first simulates the spatial spectrum of the DOA estimation method. The simulation sets 11 signals with incident angles uniformly distributed in the interval [-57.33°, 55.44°], the signal-to-noise ratio is 5dB, and the number of snapshots is 500. The spatial spectrum of the present invention is as follows Figure 2 As shown, from Figure 2 It can be seen that the present invention can accurately estimate 11 incident signal sources, and the spectrum peak is sharp.

[0112] The present invention simulates the resolution of the DOA estimation method. The simulation sets two incident signals from different directions [31.5°, 34.6°], the signal-to-noise ratio is 5dB, and the number of snapshots is 500. The resolution of the present invention is as follows Figure 3 As shown, it is verified by the spatial spectrum distribution diagram. Figure 3 It can be seen from the figure that the present invention can accurately estimate the information sources whose incident directions are close to each other, and has a high resolution.

[0113] Finally, simulations are performed to verify the effects of different snapshot numbers and different signal-to-noise ratios on the estimation accuracy of each method. The root mean square error (RMSE) is used to verify the DOA estimation performance of all methods. The RMSE result is obtained by averaging L=500 Monte Carlo experiments. The specific formula is as follows:

[0114]

[0115] in is the jth source estimated by the lth Monte Carlo experiment, θ j is the real j-th information source.

[0116] Assume that there are 10 signals uniformly distributed in the interval [-57.33°, 55.44°] incident on the coprime array. Figure 4 To show the influence of different signal-to-noise ratios on the estimation accuracy under the same number of snapshots, the number of snapshots is set to 500, the signal-to-noise ratio range is -10 to 20 dB, and the step size is 5 dB. In general, the RMSE value of the method proposed in the present invention is smaller than that of other methods, and the estimation performance is better than that of other methods. At the same time, under high signal-to-noise ratio, the performance of the proposed method is similar to that of the MC algorithm. This is because the proposed method decomposes the matrix into the signal covariance matrix and the noise covariance matrix when performing low-rank matrix recovery, thereby reducing the influence of noise, making the proposed method better than other methods.

[0117] Figure 5 To investigate the effect of different snapshot numbers on the estimation accuracy under the same signal-to-noise ratio, the signal-to-noise ratio is set to 5 dB, the snapshot number range is 10 to 3010, and the step size is 300. Figure 5 It can be seen that the RMSE curves of the proposed method and the comparison method gradually tend to be flat as the number of snapshots increases. Due to the ability of the proposed method to suppress noise, its estimation performance is better than other methods at a low number of snapshots.

[0118] Those skilled in the art will appreciate that all or part of the steps in the above method can be completed by instructing related hardware through a program, and the program can be stored in a computer-readable storage medium, such as a read-only memory, a disk or an optical disk. Optionally, all or part of the steps in the above embodiment can also be implemented using one or more integrated circuits, and accordingly, each module / unit in the above embodiment can be implemented in the form of hardware or in the form of software function modules. The present invention is not limited to any specific form of combination of hardware and software.

[0119] Finally, it should be noted that the above-described embodiments are only specific implementations of the present invention, which are used to illustrate the technical solutions of the present invention, rather than to limit them. The protection scope of the present invention is not limited thereto. Although the present invention is described in detail with reference to the above-described embodiments, ordinary technicians in the field should understand that any technician familiar with the technical field can still modify the technical solutions recorded in the above-described embodiments within the technical scope disclosed by the present invention, or can easily think of changes, or make equivalent replacements for some of the technical features therein; and these modifications, changes or replacements do not make the essence of the corresponding technical solutions deviate from the spirit and scope of the technical solutions of the embodiments of the present invention, and should be included in the protection scope of the present invention. Therefore, the protection scope of the present invention shall be subject to the protection scope of the claims.

Claims

1. A sparse low-rank decomposition DOA estimation method based on virtual array interpolation, characterized in that: Include: Establish a received signal model; Calculate the sampling covariance matrix of the received signal model and perform vector operations on the sampling covariance matrix. The existence of holes is a non-uniform virtual array. Insert virtual sensors at the hole positions to form a virtual uniform array Virtual uniform array The received signal model is reconstructed into a Toeplitz matrix, and low-rank and sparse matrix decomposition is performed into a signal covariance matrix and a noise covariance matrix; The signal covariance matrix is ​​reconstructed using the low-rank matrix recovery theory, and the expected signal covariance matrix is ​​recovered. The MUSIC algorithm is used to perform eigendecomposition and estimate the target DOA.

2. The sparse low-rank decomposition DOA estimation method based on virtual array interpolation according to claim 1 is characterized in that: Assume that K far-field narrowband independent signals come from different wave directions θ k (k=1,2,…,K) is incident on a coprime array, then The received signal of an array element at time t is expressed as: Among them, s k (t) is the original reference signal, The direction is θ k The array flow pattern of the signal, n li (t) is the first i Zero-mean Gaussian white noise on the array elements; The received signal model expression is: in, is the array flow matrix, s(t) is the original reference signal, is zero-mean Gaussian white noise.

3. The sparse low-rank decomposition DOA estimation method based on virtual array interpolation according to claim 2 is characterized in that: The sampling covariance matrix calculation formula of the received signal model is as follows: Where T represents the number of snapshots, (·) H represents the conjugate transpose of a matrix; Vectorizing the above equation, we get: Among them, the virtual array flow matrix is the orientation vector, is the conjugate of the steering vector, is the incident signal power vector, is the noise power and I is the unit matrix.

4. The sparse low-rank decomposition DOA estimation method based on virtual array interpolation according to claim 3 is characterized in that: Virtual Uniform Array The received signal model is expressed as: Among them, <·> l is the received signal at the lth virtual array element, Indicates that the lth virtual array element is in the set But not in the collection middle.

5. The sparse low-rank decomposition DOA estimation method based on virtual array interpolation according to claim 4 is characterized in that: The reconstructed Toeplitz matrix is ​​as follows: in, 6. The sparse low-rank decomposition DOA estimation method based on virtual array interpolation according to claim 5 is characterized in that: Decompose the Toeplitz matrix into the signal covariance matrix and the noise covariance matrix: Where I is L sub ×L sub The unit matrix of the array. Usually, the number of signal sources is much smaller than the number of array elements, so we can get rank(R s )=K≤L sub , that is, the signal covariance matrix R s Has low rank characteristics; the noise covariance matrix R n It is a full rank matrix whose elements are all zero except the diagonal elements, that is, the noise covariance matrix R n It has sparse characteristics, based on which the signal covariance matrix is ​​reconstructed using low-rank matrix recovery theory.

7. The sparse low-rank decomposition DOA estimation method based on virtual array interpolation according to claim 6 is characterized in that: Reconstructing the signal covariance matrix using low-rank matrix recovery theory includes: Based on the expected signal covariance matrix T s and the expected noise covariance matrix T n The low rank and sparse characteristics of compose the following optimization problem: subject to R T =CοT s +T n T s ≥0,T n ∈D + Where rank(·) is the rank of the matrix, ||·||0 is the l0 norm, λ is the regularization parameter that balances the l0 norm and the error, and C is a matrix R T Binary matrices of the same dimension, which in R T The value of non-zero positions is 1, and the values ​​of other positions are 0. + is a semi-positive definite diagonal matrix; T s is a matrix satisfying the Hermitian Toeplitz structure; Since both the rank minimization problem and the l0 norm are difficult NP problems to solve, the above equation is convexly relaxed to the following form: subject to T s ≥0,T n ∈D + Where tr(·) is the trace of the matrix, and the expected signal covariance matrix T can be restored. s .

8. The sparse low-rank decomposition DOA estimation method based on virtual array interpolation according to claim 7 is characterized in that: The expected signal covariance matrix T is converted into s The eigendecomposition is further divided into signal subspace and noise subspace, and the expression is: Among them, U s , U n are the signal subspace and the noise subspace respectively, Σ s ,Σ n are K large eigenvalues ​​and L sub -A diagonal matrix composed of K small eigenvalues; using the orthogonality of the noise eigenvector and the signal eigenvector, the spatial spectrum function is obtained: Where a(θ) is the steering vector. As the angle θ changes, the value of the spatial spectrum peak is the estimated value of the direction of arrival.

Citation Information

Patent Citations

  • Direction-of-arrival estimation method based on sparse array direct interpolation

    CN107329110A

  • Co-prime array direction or arrival estimation method based on interpolation virtual array covariance matrix subspace rotation invariance

    CN108872929A

  • Co-prime array coherent signal arrival angle estimation method based on electromagnetic vector sensor

    CN109901101A

  • Robust direction of arrival (DOA) estimation method based on sparse and low-rank recovery

    CN110045321A

  • DOA (Direction of Arrival) estimation method and device under array element failure based on matrix filled nested array

    CN110308417A

Cited By

  • SAR (Synthetic Aperture Radar) anti-radio frequency interference method based on low-rank decomposition and matrix filling

    CN119716851A

  • Sparse array virtualization and degree of freedom improvement method based on polynomial interpolation

    CN121578232A

  • A method for sparse matrix virtualization and degree of freedom promotion based on polynomial interpolation

    CN121578232B