Direction of arrival estimation method and device, electronic equipment, medium and program product
By acquiring the spectral characteristics of the noise subspace matrix and pseudospectral function, the angle estimation interval of the target signal is determined, and the final incident angle is determined within this interval. This solves the problems of low angle estimation accuracy and large computational load in the multiple signal classification method, and achieves more efficient direction of arrival estimation.
Patent Information
- Application Number
- CN202511485608.2
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- Filing Date
- 2025-10-17
- Publication Date
- 2026-01-23
AI Technical Summary
Existing multiple signal classification methods suffer from low angle estimation accuracy and high computational cost in direction-of-arrival estimation. Especially when the signal coherence is strong, the signal-to-noise ratio is low, or the angles of multiple targets are close, the pseudo-spectral peaks are prone to broadening, splitting, or shifting, making it difficult to meet the requirements of high processing speed.
By obtaining the noise subspace matrix and pseudospectral function of the target signal received by the antenna array, the angle estimation interval of the target signal is determined based on the spectral characteristics of the noise subspace matrix, and the final incident angle is determined based on the extreme points of the pseudospectral function within this interval, thus narrowing the search range and reducing the amount of computation.
It improves the accuracy of direction-of-arrival estimation and reduces the computational load, avoids searching across the entire angular range, and improves estimation efficiency.
Smart Images

Figure CN121385782A_ABST
Abstract
Description
Technical Field
[0001] This invention relates to the field of data processing technology, and in particular to a direction-of-arrival estimation method, apparatus, electronic device, medium, and program product. Background Technology
[0002] In modern wireless communication, radar detection, sonar positioning and other fields, direction of arrival (DOA) estimation is a key technology for acquiring spatial information of signals. Its performance directly affects the positioning accuracy and signal processing efficiency of the system, and it is a research hotspot in the field of signal processing.
[0003] The traditional approach used in the field of DOA estimation is the multiple signal classification method. This method separates the signal subspace and the noise subspace by performing eigenvalue decomposition on the covariance matrix of the signal received by the antenna array, then uses the orthogonality of the two to construct a pseudo-spectral function, and finally searches for the peak value of the pseudo-spectral function to determine the incident angle of the target signal.
[0004] However, the multiple signal classification method has obvious drawbacks: on the one hand, the angle estimation accuracy is low, as it relies on the global peak search of the pseudospectral function. When the signal coherence is strong, the signal-to-noise ratio is low, or the angles of multiple targets are close, the pseudospectral peaks are prone to broadening, splitting, or shifting, making it impossible to accurately capture the true incident angle. On the other hand, the angle estimation computation is large, requiring a full traversal search of peaks across the entire angle range. When the search step size is reduced to increase the resolution, the computational load increases exponentially, making it difficult to adapt to scenarios requiring high processing speeds. Summary of the Invention
[0005] This invention provides a direction-of-arrival (DOA) estimation method, apparatus, electronic device, medium, and program product to improve the estimation accuracy of DOA and reduce the computational load of DOA estimation.
[0006] In one aspect of this invention, a direction-of-arrival estimation method is provided, comprising:
[0007] Obtain the noise subspace matrix and pseudospectral function when the antenna array receives the target signal;
[0008] The angle estimation interval of the target signal is determined based on the spectral characteristics of the noise subspace matrix;
[0009] Based on the extreme points of the pseudospectral function, the final incident angle of the target signal is determined within the angle estimation interval.
[0010] In one aspect of this invention, a direction-of-arrival (DOA) estimation device is provided, comprising:
[0011] The matrix acquisition module is used to acquire the noise subspace matrix and pseudospectral function when the antenna array receives the target signal;
[0012] The interval determination module is used to determine the angle estimation interval of the target signal based on the spectral characteristics of the noise subspace matrix;
[0013] The angle determination module is used to determine the final incident angle of the target signal within the angle estimation interval based on the extreme points of the pseudospectral function.
[0014] In another aspect of the present invention, an electronic device is provided, comprising:
[0015] At least one processor; and
[0016] Memory that is communicatively connected to at least one processor;
[0017] The memory stores a computer program that can be executed by at least one processor, and the computer program is executed by at least one processor so that at least one processor can execute the direction-of-arrival estimation method of any embodiment of the present invention.
[0018] In another aspect of the present invention, a computer-readable medium is provided, comprising: computer instructions that enable a processor to execute the direction-of-arrival estimation method of any embodiment of the present invention.
[0019] In another aspect of the present invention, a computer program product is provided, comprising: a computer program configured to enable a processor to execute the direction-of-arrival estimation method of any embodiment of the present invention when executed.
[0020] This invention obtains the noise subspace matrix and pseudospectral function of an antenna array when receiving a target signal. Based on the spectral characteristics of the noise subspace matrix, it determines the angle estimation interval of the target signal. Based on the extreme points of the pseudospectral function, it determines the final incident angle of the target signal within the angle estimation interval. By determining the final incident angle within the angle estimation interval, this invention narrows the search range of the target signal angle, avoids searching across the entire angle range, reduces computational load, and improves the accuracy of the final incident angle determination.
[0021] It should be understood that the description in this section is not intended to identify key or essential features of the embodiments of the present invention, nor is it intended to limit the scope of the invention. Other features of the invention will become readily apparent from the following description. Attached Figure Description
[0022] To more clearly illustrate the technical solutions in the embodiments of the present invention, the accompanying drawings used in the description of the embodiments will be briefly introduced below. Obviously, the accompanying drawings described below are only some embodiments of the present invention. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.
[0023] Figure 1 This is a flowchart of a direction-of-arrival estimation method provided in Embodiment 1 of the present invention;
[0024] Figure 2 This is a flowchart of another direction-of-arrival estimation method provided in Embodiment 2 of the present invention;
[0025] Figure 3 This is a flowchart of another direction-of-arrival estimation method provided in Embodiment 3 of the present invention;
[0026] Figure 4 This is a flowchart illustrating an iterative solution for the final incident angle using a quadratic interpolation method according to Embodiment 3 of the present invention;
[0027] Figure 5 This is a schematic diagram of a direction of arrival estimation result provided in Embodiment 3 of the present invention;
[0028] Figure 6 This is a schematic diagram of the root mean square error curves of MUSIC and Q-MUSIC algorithms as a function of signal-to-noise ratio, provided in Embodiment 3 of the present invention.
[0029] Figure 7 This is a schematic diagram of the root mean square error curve of MUSIC and Q-MUSIC algorithms as a function of the number of array elements, provided in Embodiment 3 of the present invention.
[0030] Figure 8 This is a schematic diagram of the root mean square error curve of MUSIC and Q-MUSIC algorithms as a function of the number of snapshots, provided in Embodiment 3 of the present invention.
[0031] Figure 9 This is a schematic diagram of the performance variation curves of Rife, Rife-MUSIC and Q-MUSIC algorithms when the number of snapshots is 10, according to Embodiment 3 of the present invention.
[0032] Figure 10 This is a schematic diagram of the performance variation curves of Rife, Rife-MUSIC and Q-MUSIC algorithms when the number of snapshots is 100, according to Embodiment 3 of the present invention.
[0033] Figure 11 This is a structural diagram of another direction-of-arrival estimation device provided in Embodiment 4 of the present invention;
[0034] Figure 12 This is a block diagram of an electronic device for performing a direction-of-arrival estimation method according to Embodiment 5 of the present invention. Detailed Implementation
[0035] To enable those skilled in the art to better understand the present invention, the technical solutions of the present invention will be clearly and completely described below with reference to the accompanying drawings of the embodiments of the present invention. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort should fall within the scope of protection of the present invention.
[0036] It should be noted that the terms "first," "second," etc., in the specification, claims, and accompanying drawings of this invention are used to distinguish similar objects and are not necessarily used to describe a specific order or sequence. It should be understood that such data can be interchanged where appropriate so that the embodiments of the invention described herein can be implemented in orders other than those illustrated or described herein. Furthermore, the terms "comprising" and "having," and any variations thereof, are intended to cover a non-exclusive inclusion; for example, a process, method, apparatus, product, or device that comprises a series of steps or units is not necessarily limited to those steps or units explicitly listed, but may include other steps or units not explicitly listed or inherent to such processes, methods, products, or devices.
[0037] Figure 1 This invention provides a flowchart of a direction-of-arrival (DOA) estimation method, applicable to target localization and / or target tracking scenarios. The method can be executed by a DOA estimation device, which can be implemented in hardware and / or software. This device can be configured in a semiconductor testing server or a distributed computing cluster composed of multiple servers. Figure 1 As shown, the method includes:
[0038] S110, Obtain the noise subspace matrix and pseudospectral function when the antenna array receives the target signal.
[0039] In this context, the target signal refers to the signal whose incident angle needs to be estimated in an antenna array reception scenario. By estimating the incident angle of the target signal, target localization and / or target tracking can be achieved. In actual reception scenarios, due to noise effects such as antenna array hardware noise or environmental interference, the target signal received by the antenna array is actually composed of a mixture of useful and noise signals. For example, the noise subspace matrix can be obtained by performing eigenvalue decomposition on the target signal composed of the mixture of useful and noise signals.
[0040] The noise subspace matrix is a matrix obtained by decomposing the covariance matrix of the target signal received from the antenna array. It reflects the characteristics of the noise signal in the target signal and is used to determine the angle estimation interval of the target signal.
[0041] A pseudospectral function can be understood as a function used to estimate the final incident angle of a target signal, obtained through mathematical operations on the target signal. The extreme points of the pseudospectral function are related to the final incident angle of the target signal, and can be used to determine the final incident angle. For example, a pseudospectral function can be constructed using the orthogonality of the signal subspace matrix and the noise subspace matrix. The formula for calculating this pseudospectral function includes at least the following: ,in, Let be the signal subspace matrix. The noise subspace matrix, for The conjugate transpose of the matrix. for The conjugate transpose of the matrix. The signal subspace matrix and noise subspace matrix are obtained by eigenvalue decomposition of the covariance matrix of the target signal.
[0042] Specifically, the target signal received by the antenna array is processed to obtain the covariance matrix of the target signal. Eigenvalue decomposition is performed on the covariance matrix, and the obtained eigenvalues are sorted by size. Eigenvectors corresponding to eigenvalues with noise power less than or equal to the noise signal are selected. The eigenvectors corresponding to eigenvalues with noise power less than or equal to the noise power are arranged into a noise subspace matrix, and the eigenvectors corresponding to eigenvalues with noise power greater than or equal to the noise power are arranged into a signal subspace matrix. Combining the calculation formula of the pseudospectral function, the pseudospectral function of the target signal is obtained based on the signal subspace matrix and the noise subspace matrix.
[0043] S120. Determine the angle estimation interval of the target signal based on the spectral characteristics of the noise subspace matrix.
[0044] Spectral characteristics can be understood as the distribution characteristics exhibited by the spectrum of a series of target signals. For example, spectral characteristics may include: the magnitude of spectral values or the location of spectral extrema. The spectral characteristics of the noise subspace matrix reflect the distribution and characteristics of the noise signal in the frequency domain and can be used to determine the angle estimation interval of the target signal.
[0045] An angle estimation interval can be understood as a range of initially determined incident angles of a target signal, which includes the final incident angle of the target signal and is used to narrow down the range of determination of the final incident angle of the target signal.
[0046] Specifically, signal processing operations such as fast Fourier transform or wavelet transform can be performed on the noise subspace matrix to obtain the spectral characteristics of the noise subspace matrix. Based on the spectral characteristics, the spectral range of the final incident angle of a target signal can be determined, and the angle boundary corresponding to the angle estimation interval can be determined based on the spectral range.
[0047] S130. Based on the extreme points of the pseudospectral function, determine the final incident angle of the target signal within the angle estimation interval.
[0048] The final incident angle refers to the angle at which the target signal actually arrives at the antenna array, and is used to estimate the incident angle of the target signal.
[0049] Specifically, the extreme points of the pseudospectral function can be solved by using gradient descent or quadratic interpolation, and these extreme points can be used as the final incident angle of the target signal.
[0050] This invention acquires the noise subspace matrix and pseudospectral function when an antenna array receives a target signal. Based on the spectral characteristics of the noise subspace matrix, it determines the angle estimation interval of the target signal. Based on the extreme points of the pseudospectral function, it determines the final incident angle of the target signal within the angle estimation interval. By determining the final incident angle within the angle estimation interval, this invention narrows the search range of the target signal angle, avoids searching across the entire angle range, reduces computational load, and improves the accuracy of the final incident angle determination.
[0051] Furthermore, this embodiment of the invention refines step S110, specifically detailing how to obtain the noise subspace matrix of the target signal, including:
[0052] Extract the noise signal from the target signal and obtain the noise power of the noise signal; perform eigenvalue decomposition on the covariance matrix to obtain the eigenvalues of the covariance matrix and the corresponding eigenvectors. The number of eigenvalues is equal to the number of rows or columns of the covariance matrix; arrange the eigenvectors corresponding to the eigenvalues that are less than or equal to the noise power in descending order of the eigenvalues to form the noise subspace matrix.
[0053] Specifically, to acquire the target signal received by the antenna array, noise signals can be extracted from the target signal using methods such as filtering or principal component analysis. The noise power, which measures the magnitude of the noise signal energy, can then be obtained using methods such as time-domain average power or frequency-domain integral power. Covariance calculation is performed on the target signal to obtain its covariance matrix. Eigenvalue decomposition is then performed on this covariance matrix, decomposing it into multiple eigenvalues and a unique eigenvector corresponding to each eigenvalue. The number of eigenvalues is equal to the number of rows or columns of the covariance matrix. The decomposed eigenvalues are sorted in descending order. Each eigenvalue is compared with the previously calculated noise power, and eigenvalues less than or equal to the noise power are selected. The corresponding eigenvectors are then determined, and the selected eigenvectors are arranged sequentially as column vectors to form the noise subspace matrix in descending order of eigenvalues.
[0054] Example 2
[0055] Figure 2 This invention provides a flowchart of another direction-of-arrival estimation method for Embodiment 2 of the present invention. This embodiment of the present invention is a refinement of the above embodiment, specifically refining the specific steps of how to determine the angle estimation interval and how to determine the final incident angle.
[0056] like Figure 2 As shown, another method for estimating direction of arrival may include the following specific steps:
[0057] S210. Perform a Fourier transform on the noise subspace matrix to obtain the spectrum matrix of the noise subspace matrix.
[0058] The spectrum matrix is the matrix obtained by performing a Fourier transform on the noise subspace matrix. It is used to describe the characteristics of the noise subspace matrix from the perspective of the frequency domain. For example, the spectrum matrix can describe the energy, amplitude, or phase distribution of the noise subspace matrix at different frequencies.
[0059] Specifically, the Fourier transform algorithm is used to transform the noise subspace matrix from the time domain to the frequency domain, obtaining the spectrum matrix. Examples of Fourier transform algorithms include: Fast Fourier Transform (FFT) or Discrete Fourier Transform (DFT), etc.
[0060] S220. Obtain the mean of all column vector elements in each column vector of the spectrum matrix, and arrange the mean values into an average noise subspace vector according to the order of each column vector in the spectrum matrix.
[0061] The average noise subspace vector is a vector obtained by arranging the mean values of all column vector elements in each column vector of the spectrum matrix according to the order of the column vectors in the spectrum matrix. The dimension of this vector is equal to the number of columns in the spectrum matrix.
[0062] Specifically, for each column vector of the spectrum matrix, all elements in each column vector are traversed sequentially, the sum of the elements of each column vector is calculated, and then divided by the number of elements in that column to obtain the mean of the elements of each column vector. The mean values of each column vector of the spectrum matrix are then arranged in the original order of the column vectors in the spectrum matrix to form the average noise subspace vector.
[0063] S230. Identify the minimum spectral value in the average noise subspace vector, and obtain the left neighbor spectral value to the left of the minimum spectral value and the right neighbor spectral value to the right of the minimum spectral value.
[0064] The minimum spectral value refers to the smallest element value in the average noise subspace vector. The left neighbor spectral value refers to the element value to the left of the minimum spectral value in the average noise subspace vector, and the right neighbor spectral value refers to the element value to the right of the minimum spectral value in the average noise subspace vector. The spectral index of the left neighbor spectral value plus 1 equals the spectral index of the minimum spectral value, and the spectral index of the left neighbor spectral value minus 1 equals the spectral index of the minimum spectral value.
[0065] Specifically, a traversal comparison method can be used, starting from the first element of the vector and taking it as the current minimum value. This is then compared with subsequent elements in turn. If a smaller element is encountered, the current minimum value is updated until all elements have been traversed, and the minimum spectral value is finally obtained. Alternatively, a library function search can be used, utilizing the minimum value search function provided by the mathematical library to directly obtain the minimum spectral value in the average noise subspace vector. After determining the minimum spectral value, the adjacent element value to the left of the minimum spectral value is taken as the left neighbor spectral value, and the adjacent element value to the right of the minimum spectral value is taken as the right neighbor spectral value.
[0066] S240. Determine the angle estimation interval based on the left neighbor spectrum value, the right neighbor spectrum value, the signal wavelength of the target signal, the number of elements in the antenna array, and the spacing between the elements in the antenna array.
[0067] Specifically, the left neighbor spectrum value, the right neighbor spectrum value, the signal wavelength of the target signal, the number of elements in the antenna array, and the element spacing of the antenna array are obtained. The angle estimation interval is determined based on the left neighbor spectrum value, the right neighbor spectrum value, the signal wavelength of the target signal, the number of elements in the antenna array, and the element spacing of the antenna array.
[0068] S250: Obtain the average angle after accumulating the left boundary angle and the right boundary angle.
[0069] The left boundary angle can be understood as an angle value that forms the left endpoint of the angle estimation interval, used to limit the left side of the interval; the right boundary angle can be understood as an angle value that forms the right endpoint of the interval, used to limit the right side of the interval. The left and right boundary angles form a complete angle estimation interval, providing a clear angle termination boundary for the subsequent determination of the final incident angle.
[0070] The average angle can be understood as an angle value obtained by averaging the left boundary angle and the right boundary angle. It is used as another angular reference point between the left boundary angle and the right boundary angle to determine the final incident angle.
[0071] Specifically, the values corresponding to the left and right boundary angles of the determined angle estimation interval are added together to obtain the cumulative value of the left and right boundary angles. This cumulative value is then divided by 2 to obtain the average angle of the left and right boundary angles.
[0072] S260. Substitute the left boundary angle, right boundary angle, and average angle into the pseudospectral function to obtain the first function value, the second function value, and the third function value, respectively.
[0073] The first function value can be understood as a function value calculated by substituting the left boundary angle into the pseudospectral function; the second function value can be understood as a function value calculated by substituting the right boundary angle into the pseudospectral function; and the third function value can be understood as a function value calculated by substituting the average angle into the pseudospectral function. The first, second, and third function values provide equation conditions for solving the coefficients of the preset quadratic function and reflect the values of the pseudospectral function at different angles.
[0074] Specifically, the left boundary angle, right boundary angle, and average angle are substituted into the pseudospectral function in sequence, and the corresponding function values are obtained through function calculation, namely the first function value, the second function value, and the third function.
[0075] S270. Obtain a preset quadratic function. Substitute the left boundary angle, right boundary angle, average angle, first function value, second function value, and third function value into the preset quadratic function to obtain the coefficients of the quadratic term and the linear term of the preset quadratic function.
[0076] The preset quadratic function can be understood as a pre-defined quadratic polynomial function used to fit the variation trend of the pseudospectral function within the angle estimation interval. For example, the preset quadratic function can take the form of: etc., among which, For a predefined quadratic function, This refers to the left boundary angle, right boundary angle, or average angle. The coefficient of the quadratic term, The coefficient of the linear term, The constant term is represented by the coefficient of the quadratic term. The coefficient of the quadratic term can be understood as a coefficient multiplied by the square of the angle in a predefined quadratic function, determining the opening direction and size of the graph of the predefined quadratic function. The coefficient of the linear term can be understood as a coefficient multiplied by the linear term of the angle in a predefined quadratic function, used to influence the translation direction of the graph of the predefined quadratic function.
[0077] Specifically, a quadratic polynomial function is pre-defined, i.e., a preset quadratic function. The left boundary angle, right boundary angle, average angle, and the corresponding first function value, second function value, and third function value are substituted into the preset quadratic function. The coefficients of the quadratic term and the coefficients of the linear term of the preset quadratic function are obtained by solving the system of equations.
[0078] S280. Obtain twice the value of the quadratic coefficient, take the opposite of the ratio of the linear coefficient to the twice value as the extreme point of the pseudospectral function, and take the extreme point as the final incident angle.
[0079] Specifically, the coefficient of the quadratic term is multiplied by 2 to obtain twice the value of the quadratic term coefficient. The ratio of the coefficient of the linear term to this twice value is calculated, and the negative of this ratio is taken as the extreme point of the pseudospectral function. This extreme point is taken as the final incident angle of the target signal.
[0080] In this embodiment of the invention, a Fourier transform is performed on the noise subspace matrix to obtain its spectrum matrix. The mean value of all column vector elements in each column vector of the spectrum matrix is obtained, and the mean values are arranged into an average noise subspace vector according to the order of the column vectors in the spectrum matrix. The minimum spectral value in the average noise subspace vector is identified, and the left and right neighboring spectral values are obtained. Based on the left and right neighboring spectral values, the signal wavelength of the target signal, the number of elements in the antenna array, and the element spacing of the antenna array, the angle estimation interval is determined, and the left boundary is obtained. The average angle obtained by accumulating the angles of the left and right boundaries is used to calculate the pseudospectral function. Substituting the left and right boundary angles, the right boundary angle, and the average angle into the pseudospectral function yields the first, second, and third function values, respectively. A preset quadratic function is then obtained. Substituting the left and right boundary angles, the average angle, the first, second, and third function values into the preset quadratic function yields the coefficients of the quadratic and linear terms. The quadratic coefficient is doubled, and the negative of the ratio of the linear coefficient to the doubled value is taken as the extreme point of the pseudospectral function, which is then used as the final incident angle. This embodiment of the invention simplifies the data representation of the spectrum matrix by using the average noise subspace vector, improving computational efficiency. By combining the frequency domain characteristics of the target signal and the physical parameters of the antenna array, the angle estimation interval of the target signal can be determined more accurately, improving the accuracy of angle estimation. The extreme point of the pseudospectral function is calculated based on the coefficients of the preset quadratic function and used as the final incident angle, reducing computational load and further improving estimation efficiency.
[0081] Furthermore, this embodiment of the invention further refines step S240, specifically detailing how to determine the angle estimation interval based on the left neighbor spectral value, the right neighbor spectral value, the signal wavelength of the target signal, the number of elements in the antenna array, and the element spacing of the antenna array, including:
[0082] S2401. Obtain the first product of the frequency index corresponding to the left neighbor's spectrum value and the signal wavelength, the second product of the frequency index corresponding to the right neighbor's spectrum value and the signal wavelength, and the third product of the number of array elements and the array element spacing.
[0083] The frequency index can be understood as a number that identifies the position of a spectral value, used to locate the specific position of the spectral value in the spectral sequence.
[0084] The first product value is obtained by multiplying the frequency index corresponding to the left neighboring spectral value by the wavelength of the target signal. The second product value is obtained by multiplying the frequency index corresponding to the right neighboring spectral value by the wavelength of the target signal. The third product value is obtained by multiplying the number of elements in the antenna array by the element spacing. The first, second, and third product values are used together to determine the angle estimation interval.
[0085] Specifically, the frequency index corresponding to the left neighbor spectrum value is identified, and the frequency index corresponding to the left neighbor spectrum value is multiplied by the signal wavelength to obtain the first product value. The frequency index corresponding to the right neighbor spectrum value is identified, and the frequency index corresponding to the right neighbor spectrum value is multiplied by the signal wavelength to obtain the second product value. The number of array elements and the element spacing of the antenna array are obtained, and the number of array elements and the element spacing of the antenna array are multiplied to obtain the third product value.
[0086] S2402. Perform an arcsine operation on the ratio of the first product value to the third product value to obtain the left boundary angle of the angle estimation interval.
[0087] Specifically, the first product value is used as the numerator, the third product value is used as the denominator, the ratio of the first product value to the third product value is obtained, the arcsine operation is performed on the ratio, the result of the arcsine operation is obtained, and the result is used as the left boundary angle of the angle estimation interval.
[0088] S2403. Perform an arcsine operation on the ratio of the second product value to the third product value to obtain the right boundary angle of the angle estimation interval.
[0089] Specifically, the second product value is used as the numerator and the third product value is used as the denominator to obtain the ratio of the second product value to the third product value. The arcsine operation is performed on this ratio to obtain the result of the arcsine operation. This result is used as the left boundary angle of the angle estimation interval.
[0090] Furthermore, the embodiments of the present invention also include: using a quadratic interpolation method to iterate the pseudospectral function to solve for the extreme points of the pseudospectral function; for each iteration, obtaining the absolute value of the difference between the extreme point and the average angle; comparing the absolute value with a preset angle estimation threshold, stopping the iteration when the absolute value is less than the preset angle estimation threshold, and taking the extreme point corresponding to the absolute value less than the preset angle estimation threshold as the final incident angle.
[0091] Specifically, using a quadratic interpolation method, the pseudospectral function is iterated. In the first iteration, the left boundary angle, right boundary angle, and average angle within the angle estimation interval are selected as three sampling points. These are substituted into the pseudospectral function to obtain the corresponding function values, and a quadratic polynomial is fitted to solve for the extreme points. In the second iteration, using the extreme points obtained in the first iteration as the center, two angles with a preset step size are reselected, and these angles, along with the extreme points obtained in the first iteration, form new three sampling points. Quadratic interpolation is then performed again to solve for the new extreme points. This process is repeated until the absolute value of the difference between the extreme point and the average angle is less than a preset angle estimation threshold.
[0092] Example 3
[0093] Figure 3 Flowchart of another direction-of-arrival estimation method provided in Embodiment 3 of the present invention; Figure 4 This is a flowchart illustrating an iterative solution for the final incident angle using a quadratic interpolation method, as provided in Embodiment 3 of the present invention. This embodiment of the invention is an optimization of the above embodiment, specifically supplementing the steps for iteratively solving the final incident angle using the quadratic interpolation method.
[0094] like Figure 3 As shown, another method for estimating direction of arrival (DOA) includes: preprocessing the noise subspace matrix of the antenna array when receiving the target signal using a Fast Fourier Transform (FFT) to obtain the angle prediction interval; and constructing the objective function. The predicted angle range is used as the objective function. The initial iteration value is used to iteratively find the minimum point within the prediction interval using the quadratic interpolation method, which is the DOA estimate.
[0095] The process of preprocessing the noise subspace matrix using a Fast Fourier Transform (FFT) to obtain the angle prediction interval includes the following steps:
[0096] For the noise subspace matrix No. Liede FFT of each element yields:
[0097]
[0098] in, For the noise subspace of the first Liede The element corresponding to the row, where N represents the number of FFT points.
[0099] The average noise subspace vector is obtained by summing the element values of each column vector in the noise subspace matrix and then averaging them. ,in, This represents the total number of columns in the noise subspace.
[0100] By comparing the magnitudes of the spectral values in the average noise subspace vector, if a certain spectral value is simultaneously smaller than the spectral values on both sides, and is the minimum value, then the angles corresponding to the spectral values on both sides are considered to be the angle prediction interval. ,in .
[0101] Wherein, construct the objective function The steps may include: using the pseudospectral function P(θ) of the antenna array when receiving the target signal as the objective function:
[0102]
[0103] If the angle prediction range is Given three points The corresponding function values are respectively Construct a quadratic function p(x) such that its values at the three points are equal to... Corresponding function value:
[0104]
[0105] Will , Substituting into the above system of equations, we can obtain... .
[0106] Let the derivative of p(x) The extreme point can be obtained. , ,Will Substituting the value into the above formula, we get:
[0107]
[0108] Based on the results The size of the point determines which of the three points to replace, until... When ε is reached, the iteration stops, where ε is the required precision. The detailed steps for solving the minimum value using quadratic interpolation iteration are summarized below (where...). (Indicates the leftward magnitude), flowchart as follows: Figure 4 As shown:
[0109] (1) Comparison , Size, if If yes, proceed to step (2); otherwise, proceed to step (3).
[0110] (2) If ,but Proceed to step (4); otherwise, Proceed to step (4);
[0111] (3) If ,but Proceed to step (4); otherwise, Proceed to step (4);
[0112] (4) In the new search interval [ , Calculate the minimum point of the interpolation function on [the graph]. Proceed to step (5);
[0113] (5) If Proceed to step (1); otherwise, the iteration ends, and the current value is recorded. This is the minimum value, which yields an accurate angle estimate.
[0114] Furthermore, in this embodiment of the invention, a complexity analysis was also performed on another direction-of-arrival estimation method provided by this embodiment of the invention. In this embodiment of the invention, the direction-of-arrival estimation method can be defined as the Q-MUSIC algorithm.
[0115] Table 1 shows the complexity of the MUSIC algorithm and the Q-MUSIC algorithm for direction-of-arrival estimation. In Table 1, N is the number of array elements and M is the number of signal sources. L is the number of FFT points, and L is the number of snapshots. The number of peak searches performed by the MUSIC algorithm. Let represent the number of iterations for the quadratic interpolation. Assuming the number of array elements is 8, the number of sources is 2, the number of FFT points is 32, the number of snapshots is 100, the spectral peak search range of the MUSIC algorithm is [-90°, 90°], and the search step size is set to 0.01°, then the number of searches is 18000. Through multiple simulation experiments, the number of iterations for the quadratic interpolation is generally 5. Therefore, we can obtain... This represents the computational complexity of the MUSIC algorithm. This represents the computational complexity of the Q-MUSIC algorithm. It can also be compared to the Rife-MUSIC algorithm, whose computational complexity is... It can be seen that the Q-MUSIC algorithm proposed in this embodiment reduces computational complexity.
[0116]
[0117] Furthermore, in this embodiment of the invention, the effectiveness of another direction-of-arrival estimation method provided by this embodiment of the invention was also verified.
[0118] In this embodiment of the invention, to verify the effectiveness of the proposed Q-MUSIC algorithm in measuring angles, a uniform linear array consisting of 8 elements with an element spacing of half a wavelength was set. The incident signal angles were [-22°, 15.5°, 42°], the number of snapshots was 100, and the signal-to-noise ratio was 10dB. Simultaneously, a preset angle estimation threshold ε = 0.01° was set. The simulation results of the MUSIC algorithm are compared as follows... Figure 5 As shown.
[0119] Depend on Figure 5 It can be seen that the angle estimated by the proposed Q-MUSIC algorithm is consistent with the peak position of the MUSIC algorithm and is very close to the true angle. However, the Q-MUSIC algorithm does not require spectral peak search; it only needs to use a quadratic interpolation method to find the extreme value within the predicted angle range, thus reducing the computational load. In addition, with ε=0.01°, the number of iterations is only 5, indicating that high-precision estimation is maintained while reducing the computational load.
[0120] The influence of signal-to-noise ratio (SNR) variation on the root mean square (RMS) error of the MUSIC and Q-MUSIC algorithms was investigated. A uniform linear array of 8 elements was used, with an element spacing of half a wavelength. The signal was incident at 15.5° and 42°, with a snapshot number of 100. The SNR was varied from -10dB to 20dB in 2dB steps. Simultaneously, ε = 0.01° was set, and the peak search step size of the MUSIC algorithm was 0.01°. 500 Monte Carlo simulations were performed, and the simulation results comparing the two algorithms are shown below. Figure 6 As shown.
[0121] Depend on Figure 6 It can be seen that when the signal-to-noise ratio (SNR) is less than -4dB, the estimation performance of both algorithms is poor, and neither can estimate the incident direction of the signal. However, under the same SNR, the proposed Q-MUSIC algorithm has better accuracy than the MUSIC algorithm. When the SNR is greater than -2dB, the RMSE performance of the two algorithms is basically the same, and as the SNR increases, the RMSE of both gradually decreases and approaches 0. Under the same conditions, the proposed Q-MUSIC algorithm has better estimation performance than the MUSIC algorithm or is basically the same as the MUSIC algorithm, but the computational complexity of the Q-MUSIC algorithm is much lower than that of the traditional MUSIC algorithm. Therefore, the Q-MUSIC algorithm outperforms the MUSIC algorithm.
[0122] The influence of the number of array elements on the root mean square error (RMSE) of the MUSIC and Q-MUSIC algorithms was investigated. A uniform linear array with an element spacing of half a wavelength was used, with signals incident at 15.5° and 42°, a snapshot number of 100, a signal-to-noise ratio of 10 dB, and element numbers of 8, 16, 32, 64, and 128. Simultaneously, ε = 0.01° was set, and the peak search step size of the MUSIC algorithm was 0.01°. 500 Monte Carlo experiments were conducted, and the simulation results comparing the two algorithms are shown below. Figure 7 As shown
[0123] Depend on Figure 7 It can be seen that when the number of array elements is 8, the root mean square error (RMSE) of both algorithms is only 0.18°. When the number of array elements is 128, the RMSSE of both algorithms is basically the same, and both can accurately estimate the target angle. With the same number of array elements, the proposed Q-MUSIC algorithm maintains the same estimation accuracy as the MUSIC algorithm, exhibiting the same estimation performance, but the Q-MUSIC algorithm has a very low computational cost. Experimental results show that the proposed Q-MUSIC algorithm can reduce the computational cost while maintaining the same estimation accuracy as the MUSIC algorithm. Therefore, the Q-MUSIC algorithm has better estimation performance.
[0124] The effect of varying snapshot number on the root mean square error (RMSE) of the MUSIC and Q-MUSIC algorithms was investigated. A uniform linear array of 8 elements was used, with an element spacing of half a wavelength. The signal was incident at 15.5° and 42°, and the snapshot number varied from 10 to 150 in step sizes of 20, with a signal-to-weight ratio (SBR) of 10 dB. Simultaneously, ε was set to 0.01, and the peak search step size of the MUSIC algorithm was 0.01. 500 Monte Carlo experiments were conducted, and the simulation results of the RMSE of the two algorithms for DOA estimation are shown below. Figure 8 As shown.
[0125] Depend on Figure 8 It can be seen that when the number of snapshots is 10, the root mean square error (RMSE) of both algorithms reaches 0.14°; when the number of snapshots reaches 150, the RMSE of both algorithms is 0.035°. At the same number of snapshots, the proposed Q-MUSIC algorithm consistently maintains the same estimation accuracy as the MUSIC algorithm. With the increase of the number of snapshots, the RMSE of both algorithms shows a consistent decreasing trend. Experimental results indicate that with the increase of the number of snapshots, both algorithms perform better, but the Q-MUSIC algorithm has lower computational complexity.
[0126] The performance of the Rife algorithm and the Rife-MUSIC algorithm is compared with that of the Q-MUSIC algorithm proposed in this embodiment. A uniform linear array consisting of 8 antenna elements is set up with an element spacing of half a wavelength. The incident signals are 15.5° and 42°. The signal-to-noise ratio is varied from -5dB to 20dB in steps of 1dB. ε is set to 0.01 and the spectral peak search step size is 0.01°. 500 Monte Carlo experiments are performed.
[0127] When the number of snapshots is 10, the changes in the root mean square error of DOA estimation for the three algorithms are as follows: Figure 9 As shown in the figure. When the number of snapshots is 100, the changes in the root mean square error of DOA estimation for the three algorithms are as follows. Figure 10 As shown.
[0128] Depend on Figure 9 It can be seen that when the number of snapshots is small and the signal-to-noise ratio is low (less than 0dB), the estimation performance of the three algorithms has a large error. However, the Rife algorithm and the Rife-MUSIC algorithm are both better than the Q-MUSIC algorithm. When the signal-to-noise ratio is greater than 0dB, the estimation accuracy of the Q-MUSIC algorithm is consistent with that of the Rife-MUSIC algorithm, and both are better than the Rife algorithm.
[0129] Depend on Figure 10 It can be observed that, under multiple snapshot conditions, the angle measurement accuracy of the Rife-MUSIC and O-MUSIC algorithms is basically the same when the signal-to-noise ratio (SNR) is between -5dB and 5dB; when the SNR is greater than 5dB, the Q-MUSIC algorithm's angle measurement performance is slightly better than that of the Rife-MUSIC algorithm. Furthermore, under the same SNR, the RMSE of both the Rife-MUSIC and Q-MUSIC algorithms is much lower than that of the Rife algorithm. Extensive experiments show that as the number of snapshots increases, the difference in angle measurement accuracy between the Rife algorithm and the other two algorithms becomes increasingly significant.
[0130] Example 4
[0131] Figure 11 This is a schematic diagram of another direction-of-arrival estimation device provided in Embodiment 4 of the present invention. Figure 11 As shown, the device includes: a matrix acquisition module 310, an interval determination module 320, and an angle determination module 330.
[0132] The matrix acquisition module 310 is used to acquire the noise subspace matrix and pseudospectral function when the antenna array receives the target signal;
[0133] The interval determination module 320 is used to determine the angle estimation interval of the target signal based on the spectral characteristics of the noise subspace matrix;
[0134] Angle determination module 330 is used to determine the final incident angle of the target signal within the angle estimation interval based on the extreme points of the pseudospectral function.
[0135] Optionally, the matrix acquisition module 310 is specifically used to extract noise signals from the target signal and obtain the noise power of the noise signals; perform eigenvalue decomposition on the covariance matrix to obtain the eigenvalues of the covariance matrix and the eigenvectors corresponding to the eigenvalues, the number of eigenvalues being equal to the number of rows or columns of the covariance matrix; and arrange the eigenvectors corresponding to the eigenvalues that are less than or equal to the noise power into a noise subspace matrix according to the descending order of the eigenvalues.
[0136] Optionally, the interval determination module 320 includes: a matrix acquisition unit, used to perform a Fourier transform on the noise subspace matrix to obtain the spectrum matrix of the noise subspace matrix; a vector acquisition unit, used to acquire the mean value of all column vector elements in each column vector of the spectrum matrix, and arrange the mean values into an average noise subspace vector according to the order of each column vector in the spectrum matrix; a spectrum value acquisition unit, used to identify the minimum spectrum value in the average noise subspace vector, and acquire the left neighbor spectrum value to the left of the minimum spectrum value and the right neighbor spectrum value to the right of the minimum spectrum value; and an interval determination unit, used to determine the angle estimation interval based on the left neighbor spectrum value, the right neighbor spectrum value, the signal wavelength of the target signal, the number of elements of the antenna array, and the element spacing of the antenna array.
[0137] Optionally, the interval determination unit is specifically used to obtain the first product value of the frequency index corresponding to the left neighbor spectrum value and the signal wavelength, the second product value of the frequency index corresponding to the right neighbor spectrum value and the signal wavelength, and the third product value of the number of array elements and the spacing between array elements; perform an arcsine operation on the ratio of the first product value and the third product value to obtain the left boundary angle of the angle estimation interval; and perform an arcsine operation on the ratio of the second product value and the third product value to obtain the right boundary angle of the angle estimation interval.
[0138] Optionally, the angle determination module 330 is specifically used to obtain the average angle after accumulating the left boundary angle and the right boundary angle; substitute the left boundary angle, the right boundary angle, and the average angle into the pseudo-spectral function to obtain the first function value, the second function value, and the third function value, respectively; obtain a preset quadratic function, substitute the left boundary angle, the right boundary angle, the average angle, the first function value, the second function value, and the third function value into the preset quadratic function to obtain the quadratic coefficient and the linear coefficient of the preset quadratic function; obtain twice the value of the quadratic coefficient, take the opposite of the ratio of the linear coefficient to the twice value as the extreme point of the pseudo-spectral function, and take the extreme point as the final incident angle.
[0139] Optionally, the direction-of-arrival estimation device further includes: an extreme point solving module, used to iterate the pseudospectral function using a quadratic interpolation method to solve for the extreme points of the pseudospectral function; an absolute value acquisition module, used to acquire the absolute value of the difference between the extreme point and the average angle for each iteration; and an angle comparison module, used to compare the absolute value with a preset angle estimation threshold, and stop the iteration when the absolute value is less than the preset angle estimation threshold, and take the extreme point corresponding to the absolute value less than the preset angle estimation threshold as the final incident angle.
[0140] The direction-of-arrival estimation device provided in the embodiments of the present invention can execute the direction-of-arrival estimation method provided in any embodiment of the present invention, and has the corresponding beneficial effects of executing the method.
[0141] Example 5
[0142] Embodiment 5 of the present invention provides an electronic device for performing a direction-of-arrival estimation method, a computer-readable medium, and a computer program product.
[0143] Figure 12 A schematic diagram of an electronic device is shown that can be used to implement the direction-of-arrival estimation method of any embodiment of the present invention. The electronic device is intended to represent various forms of digital computers, such as laptop computers, desktop computers, workstations, personal digital assistants, servers, blade servers, mainframe computers, and other suitable computers. The electronic device can also represent various forms of mobile devices, such as personal digital processors, cellular phones, smartphones, wearable devices (such as helmets, glasses, watches, etc.), and other similar computing devices. The components shown in the embodiments of the present invention, their connections and relationships, and their functions are merely illustrative and are not intended to limit the implementation of the embodiments of the present invention described and / or claimed herein.
[0144] like Figure 12 As shown, the electronic device includes at least one processor 11 and a memory, such as a read-only memory (ROM) 12 or a random access memory (RAM) 13, communicatively connected to the at least one processor 11. The memory stores computer programs executable by the at least one processor. The processor 11 can perform various appropriate actions and processes based on the computer program stored in the ROM 12 or loaded into the RAM 13 from the storage unit 18. The RAM 13 can also store various programs and data required for device operation. The processor 11, ROM 12, and RAM 13 are interconnected via a bus 14. An input / output (I / O) interface 15 is also connected to the bus 14.
[0145] Multiple components in the electronic device are connected to the I / O interface 15, including: an input unit 16, such as a keyboard, mouse, etc.; an output unit 17, such as various types of displays, speakers, etc.; a storage unit 18, such as a disk, optical disk, etc.; and a communication unit 19, such as a network card, modem, wireless transceiver, etc. The communication unit 19 allows the electronic device to exchange information / data with other devices through computer networks such as the Internet and / or various telecommunications networks.
[0146] Processor 11 can be a variety of general-purpose and / or special-purpose processing components with processing and computing capabilities. Some examples of processor 11 include, but are not limited to, central processing units, graphics processing units, various special-purpose artificial intelligence computing chips, various processors running machine learning model algorithms, digital signal processors, and any suitable processor, controller, microcontroller, etc. Processor 11 performs the various methods and processes described above, such as direction-of-arrival estimation methods.
[0147] In some embodiments, the direction-of-arrival (DOA) estimation method may be implemented as a computer program tangibly contained in a computer-readable medium, such as storage unit 18. In some embodiments, part or all of the computer program may be loaded into and / or installed on an electronic device via ROM 12 and / or communication unit 19. When the computer program is loaded into RAM 13 and executed by processor 11, one or more steps of the DOA estimation method may be performed. Alternatively, in other embodiments, processor 11 may be configured as the DOA estimation method by any other suitable means (e.g., by means of firmware).
[0148] Various embodiments of the systems and technologies described above in these embodiments of the present invention can be implemented in digital electronic circuit systems, integrated circuit systems, field-programmable gate arrays, application-specific integrated circuits (ASICs), application-specific standard products (ASICs), systems-on-a-chip (SoCs), payload programmable logic devices, computer hardware, firmware, software, and / or combinations thereof. These various embodiments may include implementations in one or more computer programs that can be executed and / or interpreted on a programmable system including at least one programmable processor, which may be a dedicated or general-purpose programmable processor, capable of receiving data and instructions from a storage system, at least one input system, and at least one output system, and transmitting data and instructions to the storage system, the at least one input system, and the at least one output system.
[0149] Computer programs for implementing the methods of embodiments of the present invention may be written in any combination of one or more programming languages. These computer programs may be provided to a processor of a general-purpose computer, a special-purpose computer, or other programmable data processing system, such that when executed by the processor, the functions / operations specified in the flowcharts and / or block diagrams are implemented. The computer programs may be executed entirely on a machine, partially on a machine, or as a standalone software package, partially on a machine and partially on a remote machine, or entirely on a remote machine or server.
[0150] In the context of embodiments of the present invention, a computer-readable medium may be a tangible medium that may contain or store a computer program for use by or in conjunction with an instruction execution system, system, or device. A computer-readable medium may include, but is not limited to, electronic, magnetic, optical, electromagnetic, infrared, or semiconductor systems, systems, or devices, or any suitable combination of the foregoing. Alternatively, a computer-readable medium may be a machine-readable signal medium. More specific examples of machine-readable media include electrical connections based on one or more wires, portable computer disks, hard disks, RAM, ROM, erasable programmable read-only memory (EPROM or flash memory), optical fibers, compact disc read-only memory (CD-ROM), optical storage devices, magnetic storage devices, or any suitable combination of the foregoing.
[0151] To provide interaction with a user, the systems and techniques described herein can be implemented on an electronic device having: a display system (e.g., a cathode ray tube or liquid crystal display monitor) for displaying information to the user; and a keyboard and pointing system (e.g., a mouse or trackball) through which the user provides input to the electronic device. Other types of systems can also be used to provide interaction with the user; for example, feedback provided to the user can be any form of sensory feedback (e.g., visual feedback, auditory feedback, or tactile feedback); and input from the user can be received in any form (including: sound input, voice input, or tactile input).
[0152] The systems and technologies described herein can be implemented in computing systems that include backend components (e.g., as data servers), or computing systems that include middleware components (e.g., application servers), or computing systems that include frontend components (e.g., user computers with graphical user interfaces or web browsers through which users can interact with implementations of the systems and technologies described herein), or any combination of such backend, middleware, or frontend components. The components of the system can be interconnected via digital data communication of any form or medium (e.g., communication networks). Examples of communication networks include local area networks (LANs), wide area networks (WANs), blockchain networks, and the Internet.
[0153] A computing system can include clients and servers. Clients and servers are generally located far apart and typically interact through communication networks. The client-server relationship is created by computer programs running on the respective computers and having a client-server relationship with each other. The server can be a cloud server, also known as a cloud computing server or cloud host, which is a host product within the cloud computing service system. This addresses the shortcomings of traditional physical hosts and virtual private servers, such as high management difficulty and weak business scalability.
[0154] It should be understood that the various forms of procedures shown above can be used, with steps reordered, added, or deleted. For example, the steps described in this invention can be executed in parallel, sequentially, or in different orders, as long as the desired result of the technical solution of this invention can be achieved, and no limitation is imposed herein.
[0155] The specific embodiments described above do not constitute a limitation on the scope of protection of the present invention. Those skilled in the art should understand that various modifications, combinations, sub-combinations, and substitutions can be made according to design requirements and other factors. Any modifications, equivalent substitutions, and improvements made within the spirit and principles of the present invention should be included within the scope of protection of the present invention.
Claims
1. A direction-of-arrival estimation method, characterized in that, The method includes: Obtain the noise subspace matrix and pseudospectral function when the antenna array receives the target signal; The angle estimation interval of the target signal is determined based on the spectral characteristics of the noise subspace matrix; Based on the extreme points of the pseudospectral function, the final incident angle of the target signal is determined within the angle estimation interval.
2. The method according to claim 1, characterized in that, Determining the angle estimation interval of the target signal based on the spectral characteristics of the noise subspace matrix includes: Perform a Fourier transform on the noise subspace matrix to obtain the spectrum matrix of the noise subspace matrix; Obtain the mean of all column vector elements in each column vector of the spectrum matrix, and arrange the mean values into an average noise subspace vector according to the order of each column vector in the spectrum matrix. Identify the minimum spectral value in the average noise subspace vector, and obtain the left neighbor spectral value to the left of the minimum spectral value and the right neighbor spectral value to the right of the minimum spectral value; The angle estimation interval is determined based on the left neighbor spectrum value, the right neighbor spectrum value, the signal wavelength of the target signal, the number of elements in the antenna array, and the element spacing of the antenna array.
3. The method according to claim 2, characterized in that, Determining the angle estimation interval based on the left neighbor spectrum value, the right neighbor spectrum value, the signal wavelength of the target signal, the number of elements in the antenna array, and the element spacing of the antenna array includes: Obtain the first product of the frequency index corresponding to the left neighbor spectrum value and the signal wavelength, the second product of the frequency index corresponding to the right neighbor spectrum value and the signal wavelength, and the third product of the number of array elements and the array element spacing; The left boundary angle of the angle estimation interval is obtained by performing an arcsine operation on the ratio of the first product value to the third product value. The right boundary angle of the angle estimation interval is obtained by performing an arcsine operation on the ratio of the second product value to the third product value.
4. The method according to claim 1, characterized in that, The angle estimation interval includes: the left boundary angle and the right boundary angle. Determining the final incident angle of the target signal within the angle estimation interval based on the extreme points of the pseudospectral function includes: Obtain the average angle after summing the left boundary angle and the right boundary angle; Substituting the left boundary angle, the right boundary angle, and the average angle into the pseudo-spectral function, respectively, yields the first function value, the second function value, and the third function value. Obtain a preset quadratic function, and substitute the left boundary angle, the right boundary angle, the average angle, the first function value, the second function value, and the third function value into the preset quadratic function to obtain the coefficients of the quadratic term and the coefficients of the linear term of the preset quadratic function; Obtain twice the value of the quadratic term coefficient, take the opposite of the ratio of the linear term coefficient to the twice value as the extreme point of the pseudospectral function, and take the extreme point as the final incident angle.
5. The method according to claim 1, characterized in that, The method further includes: The pseudospectral function is iterated using a quadratic interpolation method to solve for its extreme points; For each iteration, obtain the absolute value of the difference between the extreme point and the average angle; The absolute value is compared with a preset angle estimation threshold. When the absolute value is less than the preset angle estimation threshold, the iteration stops, and the extreme point corresponding to the absolute value less than the preset angle estimation threshold is taken as the final incident angle.
6. The method according to claim 1, characterized in that, The acquisition of the noise subspace vector when the antenna array receives the target signal includes: Extract the noise signal from the target signal and obtain the noise power of the noise signal; The covariance matrix is subjected to eigenvalue decomposition to obtain the eigenvalues of the covariance matrix and the eigenvectors corresponding to the eigenvalues. The number of eigenvalues is equal to the number of rows or columns of the covariance matrix. The eigenvectors corresponding to eigenvalues less than or equal to the noise power are arranged in descending order to form the noise subspace matrix.
7. A direction-of-arrival estimation device, characterized in that, The device includes: The matrix acquisition module is used to acquire the noise subspace matrix and pseudospectral function when the antenna array receives the target signal; An interval determination module is used to determine the angle estimation interval of the target signal based on the spectral characteristics of the noise subspace matrix; An angle determination module is used to determine the final incident angle of the target signal within the angle estimation interval based on the extreme points of the pseudospectral function.
8. An electronic device, characterized in that, The electronic device includes: At least one processor; and A memory communicatively connected to the at least one processor; wherein, The memory stores a computer program that can be executed by the at least one processor, the computer program being executed by the at least one processor to enable the at least one processor to perform the direction-of-arrival estimation method according to any one of claims 1-6.
9. A computer-readable medium, characterized in that, The computer-readable medium stores computer instructions that cause a processor to execute the direction-of-arrival estimation method according to any one of claims 1-6.
10. A computer program product, characterized in that, The computer program product includes a computer program that, when executed by a processor, implements the direction-of-arrival estimation method according to any one of claims 1-6.
