A Direction Finding Method Based on Projection Coincidence

Through the direction finding method based on projection overlap, the covariance matrix feature decomposition and spatial spectrum function construction, the problem of insufficient spatial resolution in traditional passive detection technology is solved, and high-precision source direction finding is achieved, suitable for various types of information and is suitable for use in the world. Array and coherent source environment.

CN114518557BActive Publication Date: 2024-11-26THE 54TH RESEARCH INSTITUTE OF CHINA ELECTRONICS TECHNOLOGY GROUP CORPORATION
View PDF 0 Cites 0 Cited by

Patent Information

Application Number
CN202210151880.7
Authority / Receiving Office
CN · China
Patent Type
Patents(China)
Current Assignee / Owner
Filing Date
2022-02-18
Publication Date
2024-11-26
Estimated Expiration
2042-02-18

AI Technical Summary

Technical Problem

The limited spatial resolution of traditional passive detection technology leads to insufficient direction finding accuracy, limiting the application of the system in electronic warfare.

Method used

Using a direction finding method based on projection overlap, the spatial spectrum function is constructed to achieve high-precision direction finding by calculating the covariance matrix, feature decomposition, construction of source subspace and noise subspace, and space scanning.

Benefits of technology

It achieves spatial resolution higher than traditional passive direction finding technology, can maintain high direction finding performance under small aperture and few snaps, and is suitable for various array formats and coherent source environments.

✦ Generated by Eureka AI based on patent content.

Smart Images

  • Figure CN114518557B_ABST
    Figure CN114518557B_ABST
Patent Text Reader

Abstract

The present invention provides a direction finding method based on projection coincidence, belonging to the technical field of array signal processing. The present invention utilizes the spatial relationship between a source steering vector and its projection vector in a signal subspace to construct a direction finding spatial spectrum, thereby achieving high-precision solution of the source wave direction.
Need to check novelty before this filing date? Find Prior Art

Description

Technical Field

[0001] The invention relates to a direction finding method based on projection coincidence in the technical field of array signal processing. Background Art

[0002] With the continuous development of electronic reconnaissance technology, technical drawbacks such as low concealment of active detection systems and poor anti-interference ability of the system have gradually become prominent. In order to make up for the shortcomings of active detection in electronic warfare, passive detection technology has received more and more attention, and the most important of them is the direction finding and positioning of the target. Traditional direction finding technologies mainly include interferometer method and beam synthesis method, but due to the Rayleigh limitation, the spatial resolution of the traditional passive direction finding system cannot be further improved, which limits the direction finding accuracy of the system and limits the above algorithm in engineering applications. Summary of the invention

[0003] The technical problem to be solved by the present invention is to provide a new method with high spatial resolution direction finding capability for the passive detection field.

[0004] The technical problem to be solved by the present invention is achieved by the following technical solutions:

[0005] A direction finding method based on projection coincidence comprises the following steps:

[0006] Step 1: Obtain the array receiving signal and calculate the covariance matrix R using the sampled samples containing k source information;

[0007] Step 2: Perform eigendecomposition on the covariance matrix R, and sort the eigenvalue sequence and the corresponding eigenvector in descending order of eigenvalue to obtain the eigenvalue matrix λ and the eigenvector matrix e;

[0008] Step 3: Take the first k vectors in the eigenvector matrix e as the source eigenvectors and the rest as noise eigenvectors, and span them into the source subspace U S With the noise subspace U N ;

[0009] Step 4: Using U S Calculate the source subspace projection matrix

[0010]

[0011] In the above formula, the superscript H is the symbol for the conjugate transpose of the matrix, and the superscript -1 is the symbol for finding the generalized inverse matrix of the matrix;

[0012] Step 5: Perform spatial scanning within the angle search range to obtain the projection vector corresponding to the steering vector at each scanning angle

[0013]

[0014] In the formula, θ is the scanning angle, f is the signal frequency, d j is the distance from the jth array element to the reference array element, M is the number of array elements, c is the speed of light, and the superscript T is the transpose sign of the matrix;

[0015] Step 6: Construct the spatial spectrum function P(θ), then the corresponding angles of the k spectrum peaks in P(θ) are the azimuths of the k signal sources.

[0016] Furthermore, the spatial spectrum function P(θ) is:

[0017]

[0018] Furthermore, the spatial spectrum function P(θ) is:

[0019]

[0020] In the formula, <·> is the symbol for finding the inner product of a vector, and ||·|| is the symbol for calculating the modulus.

[0021] Compared with the prior art, the present invention has the following advantages:

[0022] 1. The present invention has no restrictions on array forms and is applicable to various array forms, and can process signals received by any array.

[0023] 2. The present invention can perform direction finding under the condition that a coherent signal source exists.

[0024] 3. The spatial resolution of direction finding of the present invention is higher than that of the traditional passive direction finding technology.

[0025] 4. The present invention still has a high direction finding performance under the conditions of small aperture and few quick shots. BRIEF DESCRIPTION OF THE DRAWINGS

[0026] Figure 1 It is the overall flow chart of the present invention.

[0027] Figure 2 This is a comparison chart of the direction finding results of the direction finding algorithm based on projection coincidence and the CBF direction finding algorithm when the source wave directions are 64°, 78° and 128° respectively. DETAILED DESCRIPTION

[0028] A direction finding method based on projection coincidence comprises the following steps:

[0029] Step 1: Obtain the array receiving signal and calculate the covariance matrix R using the sampled samples containing k source information;

[0030] Step 2: Perform eigendecomposition on the covariance matrix R, and sort the eigenvalue sequence and the corresponding eigenvector in descending order of eigenvalue to obtain the eigenvalue matrix λ and the eigenvector matrix e;

[0031] Step 3: Take the first k vectors in the eigenvector matrix e as the source eigenvectors and the rest as noise eigenvectors, and span them into the source subspace U S With the noise subspace U N ;

[0032] Step 4: Using U S Calculate the source subspace projection matrix

[0033]

[0034] Step 5: Perform spatial scanning within the angle search range to obtain the projection vector corresponding to the steering vector at each scanning angle

[0035]

[0036] Step 6: Construct the spatial spectrum function P(θ):

[0037]

[0038] Then the angles corresponding to the k spectral peaks in P(θ) are the azimuths of the k signal sources.

[0039] Furthermore, the spatial spectrum function P(θ) in step 6 can also be constructed as follows:

[0040]

[0041] In the above formula <·> is the symbol for finding the inner product of vectors.

[0042] The main working process of this method includes: calculating the covariance matrix of the sampled samples of the source by the receiving array, obtaining the source subspace and noise subspace by decomposing the covariance matrix, generating the direction finding spatial spectrum by using the projection matrix of the source subspace, and finally obtaining the source spectrum peak in the spatial spectrum by spatial domain scanning, thereby obtaining the azimuth information of the source. Based on the array receiving signal, this method constructs the direction finding spatial spectrum by using the spatial relationship between the source steering vector and its projection vector in the signal subspace, and realizes the high-precision estimation of the source wave direction.

[0043] Here is a more specific example:

[0044] Reference Figure 1 , a direction finding method based on projection coincidence, comprising the following steps:

[0045] Step 1: Get the array receiving signal and calculate the covariance matrix R using the sampled samples containing k source information;

[0046] Step 2: Perform eigendecomposition on the covariance matrix R, and sort the eigenvalue sequence and the corresponding eigenvector in descending order of eigenvalue to obtain the eigenvalue matrix λ and the eigenvector matrix e;

[0047] Step 3: Take the first k vectors in the eigenvector matrix e as the source eigenvectors and the rest as noise eigenvectors, and span them into the source subspace U S With the noise subspace U N ;

[0048] Step 4: Using U S Calculate the source subspace projection matrix

[0049]

[0050] Step 5: Perform spatial scanning within the angle search range to obtain the projection vector corresponding to the steering vector at each scanning angle

[0051]

[0052] In the above formula, f is the signal frequency, d j is the distance from the jth array element to the reference array element, M is the number of array elements, and c is the speed of light.

[0053] Step 6: Since the space spanned by the signal feature vector and the space spanned by the source steering vector are the same space, only the space spanned by the source feature (θ d , f d ) generates a steering vector a(θ d ) exists in the signal subspace U S In the example, the subscript d is used to identify the angle θ generated by the “source”. d and frequency f d Therefore, for the source steering vector a(θ d ) is more S The projection vector in The spatial relationship between them can be described as follows:

[0054]

[0055] or:

[0056]

[0057] In the above formula, λ is the coefficient, and st represents the conditions for the corresponding equation to be established.

[0058] According to the above spatial relationship, the spatial spectrum P(θ) can be constructed as follows:

[0059]

[0060] or:

[0061]

[0062] The angles corresponding to the k spectral peaks in P(θ) are the azimuths of the k signal sources.

[0063] Simulation verification:

[0064] Simulation conditions: M=5-element uniform linear array is used, the signal frequency is 800MHz, the bandwidth is B=10MHz, the number of snapshots is N=128, and the directions of the waves from the three information sources are 64°, 78° and 128° respectively.

[0065] The simulation results are as follows Figure 2 shown.

[0066] pass Figure 2 It can be seen that when the source orientation is close (64° and 78°), the two spectrum peaks of the conventional beamforming (CBF) algorithm overlap with each other and the source angle cannot be distinguished; however, the direction finding algorithm based on projection overlap can still accurately calculate the incoming wave directions of the three sources due to its high spatial resolution.

Claims

1. A direction finding method based on projection coincidence, It is characterized in that The following steps are involved: Step 1: Obtain the array receiving signal and calculate the covariance matrix R using the sampled samples containing k source information; Step 2: Perform eigendecomposition on the covariance matrix R, and sort the eigenvalue sequence and the corresponding eigenvector in descending order of eigenvalue to obtain the eigenvalue matrix λ and the eigenvector matrix e; Step 3: Take the first k vectors in the eigenvector matrix e as the source eigenvectors, and the rest As the noise feature vector, they are respectively spanned into the source subspace U S With the noise subspace U N ; Step 4: Using U S Calculate the source subspace projection matrix In the above formula, the superscript H is the symbol for the conjugate transpose of the matrix, and the superscript -1 is the symbol for finding the generalized inverse matrix of the matrix; Step 5: Perform spatial scanning within the angle search range to obtain the projection vector corresponding to the steering vector at each scanning angle In the formula, θ is the scanning angle, f is the signal frequency, d j is the distance from the jth array element to the reference array element, M is the number of array elements, c is the speed of light, and the superscript T is the transpose sign of the matrix; Step 6: Construct a spatial spectrum function P(q), where the corresponding angles of the k spectrum peaks in P(q) are the azimuths of the k information sources; the spatial spectrum function P(q) is: