A full-spatial coverage array direction of arrival estimation method and apparatus

Abstract

The application discloses a full-space coverage array wave direction estimation method and equipment, the method comprises the following steps: using a single polarization dipole antenna or a double cone antenna to form a three-layer uniform cylindrical array; using a darkroom to collect measured data to construct a calibration data set; obtaining a to-be-measured vector of an unknown signal; operating the to-be-measured vector and the calibration data set to form a spatial spectrum; finding the position of the strongest energy point in the spatial spectrum to obtain wave direction estimation; finding the double polarization calibration data of the strongest energy direction in the calibration data set; and obtaining polarization information estimation through the corresponding double polarization calibration data and the to-be-measured vector. The application can effectively solve the technical problems of insufficient full-space coverage and slow direction finding speed, improve the full-space coverage and timeliness of radar signal direction finding in a complex electromagnetic environment, and has the advantages of less required antennas, low cost and low operation complexity.

Description

Technical Field

[0001] This invention relates to the field of radar signal processing technology, specifically to a method and device for estimating the direction of arrival (DOA) across the entire airspace in complex electromagnetic environments, which can be applied to scenarios such as electromagnetic reconnaissance and radiation source localization. Background Technology

[0002] In the electromagnetic environment of modern battlefields, radar signals are characterized by full-space distribution and high-density interweaving. Traditional DOA direction finding equipment has defects such as limited airspace coverage (unable to achieve full-space direction finding without blind spots), slow direction finding speed, and insufficient direction finding success rate in complex environments. This results in incomplete battlefield electromagnetic situational awareness and delayed response to radiation source location, making it difficult to meet the needs of electronic warfare in actual combat.

[0003] Existing full-space DOA (Direction-of-Area) direction-finding technologies suffer from high requirements for array aperture, a large number of array elements, and high computational complexity. Current full-space DOA arrays often employ a multi-array splicing and spatial partitioning design, dividing the 360° full space into multiple independent sectors, with different sub-arrays responsible for signal reception and direction finding in their respective sectors. Ultimately, full-space coverage direction finding is achieved through signal fusion. Furthermore, traditional full-space DOA algorithms typically rely on feature decomposition as their core computational step, whose high computational complexity leads to low algorithm efficiency, failing to meet the dual requirements of timely direction finding and convenient hardware deployment in practical applications. Summary of the Invention

[0004] Purpose of the invention: The present invention aims to provide a method and device for estimating the direction of arrival of a full-space coverage array, which solves the technical problems of high aperture requirements, large number of array elements, and high computational complexity of existing full-space direction finding arrays, and improves the globality and timeliness of radar signal direction finding in complex electromagnetic environments.

[0005] Technical solution: A method for estimating the direction of arrival (DOA) of a full-space coverage array, comprising the following steps:

[0006] Three layers of uniform cylindrical arrays are arranged using single-polarized dipole antennas or biconical antennas to form a full-space coverage array.

[0007] In an anechoic chamber, measured data of the antenna array were collected. The main signal components were extracted using the feature decomposition method. The denoised and power-normalized dual-polarization steering vector was obtained as a calibration dataset.

[0008] The antenna array is used to acquire unknown signals to obtain corresponding amplitude and phase data. The amplitude and phase data are denoised and the power is normalized to obtain the vector to be measured. The vector to be measured and the calibration dataset are processed to form a spatial spectrum. The location of the point with the strongest energy in the spatial spectrum is found to obtain the direction of arrival estimation. The dual-polarization steering vector with the strongest energy direction is found in the calibration dataset. The polarization information is estimated by using the corresponding dual-polarization steering vector and the vector to be measured.

[0009] Furthermore, a three-layer uniform cylindrical array is arranged using single-polarized dipole antennas or biconical antennas to form a full-space coverage array, including:

[0010] Five single-polarized dipole antennas or biconical antennas are arranged at equal intervals along the same circumference to form a five-element uniform circular ring subarray, with one end of each antenna raised upwards at a specified angle to the Z-axis. Three such five-element uniform circular ring subarrays are coaxially superimposed to form a cylindrical array configuration. The spatial attitude and spacing of each layer of the subarray are adjusted, with the second layer rotated 24° clockwise relative to the first layer and the third layer rotated 24° relative to the second layer. The interlayer spacing is set to 40mm and 60mm respectively from top to bottom.

[0011] Furthermore, measured data of the antenna array were collected in an anechoic chamber, and the main signal components were extracted using the eigenvalue decomposition method to obtain the denoised and power-normalized dual-polarization steering vector, including:

[0012] In an anechoic chamber, a horizontally polarized radiation source is fixed and positioned directly in front of the antenna array. The array's servo motor is controlled to rotate and collect array response data covering the entire airspace. The radiation source is then changed to vertical polarization, and the array response data is collected repeatedly.

[0013] Calculate the covariance matrix of the response data for the two arrays respectively;

[0014] Eigenvalue decomposition of the covariance matrix yields the eigenvector corresponding to the largest eigenvalue;

[0015] The array eigenvectors that preserve the dual-polarization response of the full-coverage spatial domain are dual-polarization steering vectors.

[0016] Furthermore, the covariance matrix of the array response data is calculated using the following formula:

[0017] ;

[0018] in, Let covariance matrix be the variance matrix. This indicates that the array is receiving data. , The total number of snapshots taken. It is the conjugate transpose of the i-th received data.

[0019] Furthermore, eigenvalue decomposition is performed on the covariance matrix to obtain the eigenvector corresponding to the largest eigenvalue, as shown in the following formula:

[0020] ;

[0021] in and That is, the largest eigenvalue and its corresponding eigenvector. and These are the j-th eigenvalue and its corresponding eigenvector, respectively.

[0022] Furthermore, the measured vector and the calibration dataset are processed to form a spatial spectrum, as shown in the following formula:

[0023] ;

[0024] in , , Let be the vector to be measured. This represents the array feature vector corresponding to the horizontally polarized radiation source. This represents the array feature vector corresponding to the vertically polarized radiation source. Indicates the azimuth and elevation angles of the signal.

[0025] Furthermore, the location of the point with the strongest energy in the spatial spectrum is used to obtain the direction of arrival estimate, using any of the following formulas:

[0026] ;

[0027] ;

[0028] in, , The first The azimuth and elevation angles of each signal and This is the parameter space for azimuth and elevation angles.

[0029] Furthermore, polarization information is estimated by using the corresponding dual-polarization steering vector and the vector to be measured, as shown in the following formula:

[0030] ;

[0031] ;

[0032] in, and The first The polarization auxiliary angle and polarization phase difference of each signal and At the angle of incidence The norm of the original bipolar steering vector, This indicates taking the phase angle.

[0033] An electronic device includes: one or more processors; a memory; and one or more programs, wherein the one or more programs are stored in the memory and configured to be executed by the one or more processors, and the programs, when executed by the processors, implement the following methods:

[0034] The array response data of the unknown signal is obtained to obtain the corresponding amplitude and phase data. The amplitude and phase data are denoised and the power is normalized to obtain the vector to be measured. The vector to be measured and the calibration dataset are processed to form a spatial spectrum. The position of the point with the strongest energy in the spatial spectrum is found to obtain the direction of arrival estimation. The dual polarization steering vector with the strongest energy direction is found in the calibration dataset. The polarization information is estimated by the corresponding dual polarization steering vector and the vector to be measured.

[0035] The array is a full-space coverage array, which is a three-layer uniform cylindrical array composed of single-polarized dipole antennas or biconical antennas.

[0036] The calibration dataset consists of measured data of the antenna array collected in an anechoic chamber, the main signal components extracted using the feature decomposition method, and the resulting denoised and power-normalized dual-polarization steering vector.

[0037] A computer-readable storage medium having a computer program stored thereon, the computer program implementing the following method when executed by a processor:

[0038] The array response data of the unknown signal is obtained to obtain the corresponding amplitude and phase data. The amplitude and phase data are denoised and the power is normalized to obtain the vector to be measured. The vector to be measured and the calibration dataset are processed to form a spatial spectrum. The position of the point with the strongest energy in the spatial spectrum is found to obtain the direction of arrival estimation. The dual polarization steering vector with the strongest energy direction is found in the calibration dataset. The polarization information is estimated by the corresponding dual polarization steering vector and the vector to be measured.

[0039] The array is a full-space coverage array, which is a three-layer uniform cylindrical array composed of single-polarized dipole antennas or biconical antennas.

[0040] The calibration dataset consists of measured data of the antenna array collected in an anechoic chamber, the main signal components extracted using the feature decomposition method, and the resulting denoised and power-normalized dual-polarization steering vector.

[0041] Compared with the prior art, the present invention has the following beneficial effects:

[0042] (1) Make full use of the wide coverage of the polarized antenna radiation pattern to achieve DOA direction finding with full airspace coverage. Use 15 antennas with different polarizations evenly distributed on a 360° ring to eliminate azimuth ambiguity. The spacing between layers is 40mm and 60mm respectively, which also constitutes a longitudinal virtual array element to eliminate elevation ambiguity. This can improve the accuracy and correctness of incident signals near the elevation angle of 90° and enable direction finding of the complete spherical airspace.

[0043] (2) The measured data were collected in an anechoic chamber, and the main signal components were extracted using the feature decomposition method. The denoised and power-normalized dual-polarization steering vector was obtained as a calibration dataset. This dataset has high accuracy, which makes subsequent calculations simpler.

[0044] (3) The direction of arrival estimation calculation process does not involve complex matrix decomposition, has low computational complexity, and fits the proposed antenna array configuration. In the spatial spectrum construction, a single point only requires two vector product operations. This invention achieves accurate direction finding of the entire spatial domain while reducing the array hardware scale (small aperture, few array elements) and computational complexity, and simultaneously completes polarization information estimation. It has the technical advantages of low hardware cost, simple calculation process, fast direction finding speed and high direction finding accuracy. Attached Figure Description

[0045] Figure 1 This is a schematic diagram of the dipole array arrangement method in a specific implementation of the present invention.

[0046] Figure 2 This is a schematic diagram of a five-uniform circular array according to a specific embodiment of the present invention.

[0047] Figure 3 This is a schematic diagram of a tilted five-uniform circular array, which is a specific embodiment of the present invention.

[0048] Figure 4 This is a schematic diagram of a three-layer torsion uniform cylindrical array, which is a specific embodiment of the present invention.

[0049] Figure 5 This is a schematic diagram (front view) of a three-layer torsion deblurring uniform cylindrical array, which is a specific embodiment of the present invention.

[0050] Figure 6 This is a schematic diagram (top view) of a three-layer torsion deblurring uniform cylindrical array, which is a specific implementation of the present invention.

[0051] Figure 7 This is a schematic diagram of the antenna array anechoic chamber calibration process in a specific implementation of the present invention.

[0052] Figure 8 This is a schematic diagram of the fast DOA estimation process for antenna arrays in a specific implementation of the present invention.

[0053] Figure 9The azimuth RMS diagram is a simulation result of a specific implementation of the present invention.

[0054] Figure 10 The image shows the pitch angle RMS plot, which represents the simulation results of a specific implementation of this invention.

[0055] Figure 11 The image shows the polarization tilt angle RMS plot, which is a simulation result of a specific implementation of the present invention.

[0056] Figure 12 The simulation results of the polarization ellipse angle RMS diagram are shown for a specific implementation of the present invention. Detailed Implementation

[0057] The technical solution of the present invention will be further described below with reference to the accompanying drawings and embodiments.

[0058] This invention provides a method for estimating the direction of arrival (DOA) of a full-space coverage array, comprising the following steps:

[0059] S1 uses a single-polarized dipole antenna or a biconical antenna to arrange three layers of uniform cylindrical arrays to form a full-space coverage array;

[0060] S2. Collect measured data of the antenna array in an anechoic chamber, extract the main signal components using the feature decomposition method, and obtain the denoised and power-normalized dual-polarization steering vector as the calibration dataset.

[0061] S3. Use an antenna array to collect unknown signals to obtain corresponding amplitude and phase data. Denoise the amplitude and phase data and normalize the power to obtain the vector to be measured. Calculate the vector to be measured and the calibration dataset to form a spatial spectrum. Find the position of the point with the strongest energy in the spatial spectrum to obtain the direction of arrival estimation. Find the dual-polarization steering vector with the strongest energy direction in the calibration dataset. Obtain the polarization information estimation by using the corresponding dual-polarization steering vector and the vector to be measured.

[0062] In step S1 of this invention, increasing the polarization difference between antennas in the array arrangement effectively expands the array's spatial detection range, increases the deambiguity probability, and improves direction-finding accuracy. Specifically, the embodiment uses a dipole array method, such as... Figure 1 As shown, it includes:

[0063] Step S11: As Figure 2 As shown, five dipole antennas are arranged in a five-uniform circular array with a radius of 100 mm.

[0064] Step S12: As Figure 3 As shown, one end of each dipole antenna is raised to form an angle of 54.74° with the Z-axis.

[0065] Step S13: Repeat steps S11 and S12 to generate two more five-uniform circular arrays.

[0066] Step S14: Stack the three circular arrays into a cylinder, and implement a clockwise twist design between the layers at a fixed angle of 24°. For example... Figure 4 As shown, with the first layer of blue circular array as the orientation reference, the orientation of the second layer of red circular array is rotated 24° clockwise relative to the first layer of blue circular array; the orientation of the third layer of yellow circular array is rotated 48° clockwise relative to the first layer of blue circular array (that is, it continues to rotate 24° clockwise relative to the second layer of red circular array), thus forming a three-layer torsional superimposed cylindrical array structure.

[0067] Step S15: As Figure 5 and Figure 6 As shown, the spacing between the layers is adjusted to 40mm and 60mm respectively. At this time, the 15 array elements are evenly distributed on the 360° ring to eliminate azimuth ambiguity. They also form a longitudinal virtual array element to eliminate pitch ambiguity, which can improve the accuracy and correctness of the incident signal near the pitch angle of 90°.

[0068] For such an array composed of polarized antennas, if there exists in space A narrowband electromagnetic signal is incident on In an array of short dipole antennas, let the first one be... The azimuth angle, elevation angle, polarization auxiliary angle, polarization phase difference, and wavelength of each signal are respectively , , , and , record The spatial rectangular coordinates of each array element are: , No. The azimuth and elevation angles pointed to by the dipoles are respectively , The data received by the polarization-sensitive array can be represented as

[0069] (1)

[0070] in , The total number of snapshots taken. For signal sampling matrix

[0071] (2)

[0072] For the first An incident signal, It is a stationary Gaussian zero-mean random white noise that is independent of signal statistics.

[0073] The guiding vector matrix of the array can then be represented as:

[0074] (3)

[0075] in For the first The steering vector of each signal:

[0076] (4)

[0077] (5)

[0078] (6)

[0079] (7)

[0080] (8)

[0081] By separating the incident signal angle from the polarization information, we can obtain:

[0082] (9)

[0083] in The element spatial phase matrix describes the phase difference of the received signal generated by the element position. The polarization sensitivity matrix describes the selective reception of polarization components by the pointing of the dipole antenna. The polarization-angle domain steering vector describes the three-dimensional decomposition of the polarization of the electric field components of the incident signal along the propagation direction, yielding the three polarization components of the incident signal's electric field, which are ultimately controlled by the polarization-sensitive matrix. Selective weighted reception.

[0084] In step S2 of this invention, the radiation pattern of the proposed array is calibrated in an anechoic chamber. Since the radiation pattern of the antenna differs from the ideal, it is necessary to collect the actual response data of the incident signal in the target spatial domain on the array to form a calibration dataset, which can effectively improve the final direction finding accuracy and correctness.

[0085] Specifically, the antenna array anechoic chamber calibration method, such as... Figure 7 As shown, it includes:

[0086] Step S21: Fix a horizontally polarized radiation source in the dark room, facing the antenna array.

[0087] Step S22: Control the servo motors of the control array to rotate and collect array response data covering the entire airspace.

[0088] Step S23: Change the radiation source to vertical polarization and repeat step S22.

[0089] Step S24: Calculate the array response data covariance matrix As shown in equation (10):

[0090] (10)

[0091] This indicates taking the conjugate transpose of the i-th response data, according to... The covariance matrix R can be obtained.

[0092] Step S25: Perform eigenvalue decomposition on the covariance matrix to obtain the eigenvector corresponding to the largest eigenvalue, as shown in equation (11):

[0093] (11)

[0094] in and That is, the largest eigenvalue and its corresponding eigenvector. and These are the j-th eigenvalue and its corresponding eigenvector, respectively. This eigenvector is equivalent to a denoised and power-normalized array steering vector.

[0095] Step S26: Save the array feature vector of the dual-polarization response of the full-coverage spatial domain. The array feature vector corresponding to the horizontally polarized radiation source is called... The array eigenvector corresponding to the vertically polarized radiation source is called the eigenvector. .

[0096] In step S3 of this invention, rapid DOA estimation of the antenna array is performed. The proposed array is used to estimate the DOA and polarization information of the unknown signal. The proposed array is used to acquire the amplitude and phase data of the unknown signal, which are then processed and calculated with the calibration dataset in A2 to obtain the estimated DOA and polarization information of the corresponding unknown signal.

[0097] Specifically, such as Figure 8 As shown, the fast DOA estimation algorithm for the antenna array includes:

[0098] Step S31: Use an antenna array to collect unknown signals to obtain the corresponding amplitude and phase data, as shown in equation (1).

[0099] Step S32: Denoise and normalize the power of the amplitude and phase data to obtain the vector to be measured, i.e., adjust the covariance matrix. The mean of each column is calculated as shown in equation (12):

[0100] (12)

[0101] Will L2 norm normalization is performed, as shown in equation (13):

[0102] (13)

[0103] Step S33: Calculate the spatial spectrum by combining the measured vector and the calibration dataset, as shown below:

[0104] make , The spatial spectrum can then be calculated using equation (14):

[0105] (14)

[0106] Represents the spatial spectrum.

[0107] Step S34: Locate the location of the point with the strongest energy in the spatial spectrum to obtain the DOA estimate, as shown in equations (15) and (16):

[0108] (15)

[0109] (16)

[0110] Either method can be chosen to obtain the same result; formula 16 is simpler to calculate, while formula 15 provides a more intuitive spatial spectrum. and This is the parameter space for azimuth and elevation angles.

[0111] Step S35: Locate the dual-polarization calibration data with the strongest energy direction in the calibration dataset. and .

[0112] Step S36: Obtain polarization information estimation through the corresponding dual polarization calibration data and the vector to be measured, as shown in equations (17) and (18):

[0113] Assuming at the angle of incidence The norms of the original dual-polarization steering vectors are respectively and ,So

[0114] (17)

[0115] (18)

[0116] This indicates taking the phase angle.

[0117] The correctness of the algorithm will be verified through simulation experiments.

[0118] Assuming the incident signal is incident from different angles, with azimuth angles of 0°~360° (40 points) and elevation angles of 50°~105° (20 points), traversing 800 points evenly, the signal frequency is 6GHz, with 45° oblique polarization, 128 snapshots, and a signal-to-noise ratio of 13dB (±1dB fluctuation). The maximum random channel phase inconsistency is ±5°. The spectral peak search is performed at 1° intervals between azimuth and elevation angles, using 100 Monte Carlo runs. The statistical results are as follows. Figure 9 , 10 As shown in Figures 11 and 12. Among them, Figure 9 and Figure 10 Simulation results demonstrate that the array and algorithm can accurately determine the direction of signals incident in the entire airspace, with root mean square errors of azimuth and elevation angles both below 0.5°. Figure 11 and Figure 12 Simulation results demonstrate that the array and algorithm can accurately measure polarization parameters of signals incident throughout the entire spatial domain, with the root mean square error of the polarization tilt angle being less than 0.5° and the root mean square error of the polarization ellipse angle being less than 1.5°.

[0119] The method of this invention effectively solves the technical problems of insufficient full-space coverage and slow direction finding speed in existing technologies, improving the global coverage and timeliness of radar signal direction finding in complex electromagnetic environments. It also has the following advantages: fewer antennas and channels are required, the equipment occupies less space, and the cost is lower; it utilizes measured steering vectors, further improving estimation accuracy and precision, and simplifying subsequent calculations; the computational complexity is low, the proposed algorithm does not involve complex matrix decomposition, has low computational load, and is compatible with the antenna array proposed in this invention, requiring only two vector product operations for a single point in spatial spectrum construction.

[0120] The present invention also provides an electronic device, comprising: one or more processors; a memory; and one or more programs, wherein the one or more programs are stored in the memory and configured to be executed by the one or more processors, and the programs, when executed by the processors, implement the following method:

[0121] The array response data of the unknown signal is obtained to obtain the corresponding amplitude and phase data. The amplitude and phase data are denoised and the power is normalized to obtain the vector to be measured. The vector to be measured and the calibration dataset are processed to form a spatial spectrum. The position of the point with the strongest energy in the spatial spectrum is found to obtain the direction of arrival estimation. The dual polarization steering vector with the strongest energy direction is found in the calibration dataset. The polarization information is estimated by the corresponding dual polarization steering vector and the vector to be measured.

[0122] The array is a full-space coverage array, which is a three-layer uniform cylindrical array composed of single-polarized dipole antennas or biconical antennas.

[0123] The calibration dataset consists of measured data of the antenna array collected in an anechoic chamber, the main signal components extracted using the feature decomposition method, and the resulting denoised and power-normalized dual-polarization steering vector.

[0124] The present invention also provides a computer-readable storage medium having a computer program stored thereon, wherein the computer program, when executed by a processor, implements the following method:

[0125] The array response data of the unknown signal is obtained to obtain the corresponding amplitude and phase data. The amplitude and phase data are denoised and the power is normalized to obtain the vector to be measured. The vector to be measured and the calibration dataset are processed to form a spatial spectrum. The position of the point with the strongest energy in the spatial spectrum is found to obtain the direction of arrival estimation. The dual polarization steering vector with the strongest energy direction is found in the calibration dataset. The polarization information is estimated by the corresponding dual polarization steering vector and the vector to be measured.

[0126] The array is a full-space coverage array, which is a three-layer uniform cylindrical array composed of single-polarized dipole antennas or biconical antennas.

[0127] The calibration dataset consists of measured data of the antenna array collected in an anechoic chamber, the main signal components extracted using the feature decomposition method, and the resulting denoised and power-normalized dual-polarization steering vector.

[0128] The specific calculation methods for antenna array arrangement, radiation pattern calibration of the proposed array in the anechoic chamber, and fast DOA estimation of the antenna array can be found in the description of the aforementioned method embodiments, and will not be repeated here.

Claims

1. A full-sky coverage method for array direction of arrival estimation, characterized in that, Includes the following steps: A three-layer uniform cylindrical array is arranged using single-polarized dipole antennas or biconical antennas to form a full-space coverage array. This includes: using five single-polarized dipole antennas or biconical antennas, arranged at equal intervals along the same circumference to form a five-element uniform circular ring subarray, with one end of each antenna raised upwards at a specified angle to the Z-axis; coaxially stacking three such five-element uniform circular ring subarrays to form a cylindrical array configuration; and adjusting the spatial attitude and spacing of each layer of subarray, wherein the second layer is rotated 24° clockwise relative to the first layer and the third layer is rotated 24° relative to the second layer, and the interlayer spacing is set to 40mm and 60mm respectively from top to bottom. In an anechoic chamber, measured data of the antenna array were collected. The main signal components were extracted using eigenvalue decomposition to obtain the denoised and power-normalized dual-polarization steering vector, which served as the calibration dataset. The process included: fixing a horizontally polarized radiation source directly in front of the antenna array in the anechoic chamber; controlling the array's servo motor to rotate and collecting array response data across the entire airspace; changing the radiation source to vertical polarization and repeating the array response data collection; calculating the covariance matrix of the two array response data; performing eigenvalue decomposition on the covariance matrix to obtain the eigenvector corresponding to the largest eigenvalue; and saving the array eigenvector of the dual-polarization response across the entire airspace as the dual-polarization steering vector. The antenna array is used to acquire unknown signals to obtain corresponding amplitude and phase data. The amplitude and phase data are denoised and the power is normalized to obtain the vector to be measured. The vector to be measured and the calibration dataset are processed to form a spatial spectrum. The location of the point with the strongest energy in the spatial spectrum is found to obtain the direction of arrival estimation. The dual-polarization steering vector with the strongest energy direction is found in the calibration dataset. The polarization information is estimated by using the corresponding dual-polarization steering vector and the vector to be measured.

2. The method according to claim 1, characterized in that, The covariance matrix of the array response data is calculated using the following formula: ； in, Let covariance matrix be the variance matrix. This indicates that the array is receiving data. , This represents the total number of snapshots taken. It is the conjugate transpose of the i-th received data.

3. The method according to claim 1, characterized in that, For 15 array elements uniformly distributed on a circular ring, the covariance matrix is ​​decomposed into eigenvalues ​​to obtain the eigenvector corresponding to the largest eigenvalue, as shown in the following formula: ； in and That is, the largest eigenvalue and its corresponding eigenvector. and These are the j-th eigenvalue and its corresponding eigenvector, respectively.

4. The method according to claim 1, characterized in that, The spatial spectrum is generated by operating on the vector to be measured and the calibration dataset, as shown in the following formula: ； in , , Let be the vector to be measured. This represents the array feature vector corresponding to the horizontally polarized radiation source. This represents the array feature vector corresponding to the vertically polarized radiation source. Indicates the azimuth and elevation angles of the signal.

5. The method according to claim 4, characterized in that, The direction of arrival (DOA) is estimated by locating the point of highest energy in the spatial spectrum using any of the following formulas: ； ； in, , The first The azimuth and elevation angles of each signal and This is the parameter space for azimuth and elevation angles.

6. The method according to claim 4, characterized in that, The polarization information is estimated by using the corresponding dual-polarization steering vector and the vector to be measured, as shown in the following formula: ； ； in, and The first The polarization auxiliary angle and polarization phase difference of each signal and At the angle of incidence The norm of the original bipolar steering vector, This indicates taking the phase angle.

7. An electronic device, characterized in that, include: One or more processors; Memory; And one or more programs, wherein the one or more programs are stored in the memory and configured to be executed by the one or more processors, wherein when the programs are executed by the processors, they implement the full-space coverage array direction-of-arrival estimation method according to any one of claims 1-6.

8. A computer-readable storage medium having a computer program stored thereon, characterized in that, When the computer program is executed by the processor, it implements the direction-of-arrival estimation method for full-space coverage arrays as described in any one of claims 1-6.

Images