Radar information source number detection method and system based on density clustering, and terminal

Abstract

The invention discloses a radar information source number detection method and system based on density clustering and a terminal, and the method comprises the steps: generating an antenna array receiving data representation according to target radar antenna array information, obtaining a target covariance matrix according to the antenna array receiving data representation, carrying out the unitary transformation of the target covariance matrix, and obtaining the number of radar information sources according to the unitary transformation. A target transformation matrix is obtained, and a radius vector corresponding to the target transformation matrix is generated according to the Gerschner circle theorem; obtaining a noise background, and when the noise background is a Gaussian color noise background, compressing and standardizing the radius vector to obtain a target radius matrix; clustering the target radius matrix according to a target clustering algorithm to obtain a clustering result; and obtaining a target cluster according to the clustering result, and obtaining the number of radar information sources according to the target cluster. According to the invention, the accuracy of signal source number estimation under low signal-to-noise ratio and color noise is improved.

Description

Technical Field

[0001] The present invention relates to the field of array signal processing, and in particular to a radar source number detection method, system and terminal based on density clustering. Background Art

[0002] In array radar signal processing technology, signal source angle estimation, that is, radar echo signal arrival angle (Direction of Arrival, DOA) estimation, is a key issue. Many spatial angle estimation algorithms, such as Multiple Signal Classification (MUSIC) and Estimation of Signal Parameters via Rotational Invariance Techniques (ESPRIT), rely on the accurate number of radar signal sources (radar echo signals) as a priori input. However, in practical applications, the number of signal sources is often unknown. Therefore, it is necessary to accurately estimate the number of signal sources.

[0003] At present, the performance of various signal source number estimation methods proposed is poor under colored noise conditions. Among them, although the Gay circle method can estimate the number of signal sources under spatial colored noise, its accuracy is not high under low signal-to-noise ratio conditions, which will affect the estimation of the number of signal sources and lead to inaccurate estimation of the arrival angle of the radar echo signal.

[0004] Therefore, the prior art still needs to be improved and developed. Summary of the invention

[0005] The main purpose of the present invention is to provide a radar source number detection method, system, terminal and computer-readable storage medium based on density clustering, aiming to solve the problem that the various signal source number estimation methods proposed in the prior art have poor performance under color noise conditions. Among them, although the Gay circle method can estimate the number of signal sources under spatial color noise, its accuracy is not high under low signal-to-noise ratio conditions, which will affect the estimation of the number of signal sources, thereby leading to the problem of inaccurate estimation of the arrival angle of the radar echo signal.

[0006] To achieve the above object, the present invention provides a radar source number detection method based on density clustering, and the radar source number detection method based on density clustering comprises the following steps: Generate an antenna array receiving data representation according to the target radar antenna array information, obtain a target covariance matrix according to the antenna array receiving data representation, perform a unitary transformation on the target covariance matrix to obtain a target transformation matrix, and generate a radius vector corresponding to the target transformation matrix according to the Gauss circle theorem; Acquire a noise background, and when the noise background is a Gaussian color noise background, compress and standardize the radius vector to obtain a target radius matrix; Clustering the target radius matrix according to a target clustering algorithm to obtain a clustering result; A target cluster is obtained according to the clustering result, and the number of radar sources is obtained according to the target cluster.

[0007] Optionally, generating an antenna array received data representation according to the target radar antenna array information, obtaining a target covariance matrix according to the antenna array received data representation, performing a unitary transformation on the target covariance matrix to obtain a target transformation matrix, and generating a radius vector corresponding to the target transformation matrix according to the Gauss circle theorem specifically includes: Constructing an array steering vector corresponding to each signal source according to the target radar antenna array information, generating a steering vector matrix according to all the array steering vectors, and generating the antenna array reception data representation according to the steering vector matrix; Obtaining the target covariance matrix and the radar echo signal covariance matrix according to the antenna array received data representation; Performing a unitary transformation on the target covariance matrix according to the radar echo signal covariance matrix to obtain the target transformation matrix; According to the Gagnes circle theorem, the Gagnes circle radius of each Gagnes circle is obtained from the target transformation matrix, and the radius vector is generated according to all the Gagnes circle radii.

[0008] Optionally, performing a unitary transformation on the target covariance matrix according to the radar echo signal covariance matrix to obtain the target transformation matrix specifically includes: Divide the target covariance matrix into blocks according to the radar echo signal covariance matrix, and obtain a main sub-matrix according to the result of the block division; Performing eigenvalue decomposition on the principal submatrix to obtain an eigenspace vector of the principal submatrix; A unitary transformation matrix is ​​constructed according to the eigenspace vector, and a unitary transformation operation is performed on the eigenspace vector according to the unitary transformation matrix to obtain the target transformation matrix.

[0009] Optionally, the acquiring of the noise background, when the noise background is a Gaussian color noise background, compressing and normalizing the radius vector to obtain a target radius matrix, specifically includes: Acquire a noise background, and determine the type of the noise background; When the noise background is a Gaussian color noise background, compressing each Gaussian circle radius in the radius vector according to the unitary transformation matrix and the center of each Gaussian circle to obtain a plurality of compressed radii; Performing standardization on each of the compression radii to obtain a target radius of each Geiger circle; The target radius matrix is ​​generated according to all the target radii.

[0010] Optionally, the acquiring of the noise background and determining the type of the noise background further includes: When the noise background is a Gaussian white noise background, performing a logarithmic transformation on the radius vector to obtain a logarithmic transformation matrix; Clustering the logarithmic transformation matrix according to a target clustering algorithm to obtain a white noise clustering result; A white noise target cluster is obtained according to the white noise clustering result, and the corresponding number of radar sources is obtained according to the white noise target cluster.

[0011] Optionally, clustering the target radius matrix according to a target clustering algorithm to obtain a clustering result specifically includes: Obtaining a preset distance threshold and a minimum number of neighbors, taking the target radius matrix, the distance threshold and the minimum number of neighbors as input, and performing clustering; In the clustering process, each element in the target radius matrix is ​​traversed, a core object is selected according to the minimum number of neighbors, the core object is added to the core object set and an ordered arrangement is generated, the core distance and the reachable distance of each core object are calculated, and according to the core distance and the reachable distance of each core object in the ordered arrangement, all the core objects are respectively divided into one of the current cluster cluster, the new cluster cluster and the noise point; A clustering result is generated according to the current cluster, the new cluster and the noise point.

[0012] Optionally, acquiring a target cluster according to the clustering result, and obtaining the number of radar sources according to the target cluster specifically includes: Calculating the average values ​​of the current cluster and the new cluster in the clustering result, and selecting the cluster with the larger average value as the target cluster; The number of elements in the target cluster is counted, and the number is output as the number of radar sources.

[0013] In addition, to achieve the above object, the present invention also provides a radar source number detection system based on density clustering, wherein the radar source number detection system based on density clustering includes: A radius vector generation module is used to generate an antenna array receiving data representation according to the target radar antenna array information, obtain a target covariance matrix according to the antenna array receiving data representation, perform a unitary transformation on the target covariance matrix to obtain a target transformation matrix, and generate a radius vector corresponding to the target transformation matrix according to the Gay circle theorem; A target radius matrix generation module is used to obtain a noise background. When the noise background is a Gaussian color noise background, the radius vector is compressed and standardized to obtain a target radius matrix. A clustering result generating module, used for clustering the target radius matrix according to a target clustering algorithm to obtain a clustering result; The result output module is used to obtain the target cluster according to the clustering result, and obtain the number of radar sources according to the target cluster.

[0014] In addition, to achieve the above-mentioned purpose, the present invention also provides a terminal, wherein the terminal includes: a memory, a processor, and a radar source number detection program based on density clustering stored in the memory and executable on the processor, wherein the radar source number detection program based on density clustering implements the steps of the radar source number detection method based on density clustering as described above when executed by the processor.

[0015] In addition, to achieve the above-mentioned purpose, the present invention also provides a computer-readable storage medium, wherein the computer-readable storage medium stores a radar signal source number detection program based on density clustering, and when the radar signal source number detection program based on density clustering is executed by a processor, the steps of the radar signal source number detection method based on density clustering as described above are implemented.

[0016] In the present invention, an antenna array receiving data representation is generated according to the target radar antenna array information, a target covariance matrix is ​​obtained according to the antenna array receiving data representation, a unitary transformation is performed on the target covariance matrix to obtain a target transformation matrix, and a radius vector corresponding to the target transformation matrix is ​​generated according to the Gaussian circle theorem; a noise background is obtained, and when the noise background is a Gaussian color noise background, the radius vector is compressed and standardized to obtain a target radius matrix; the target radius matrix is ​​clustered according to a target clustering algorithm to obtain a clustering result; a target cluster is obtained according to the clustering result, and the number of radar signal sources is obtained according to the target cluster. The present invention compresses and standardizes the Gaussian circle radius obtained after the unitary transformation of the signal covariance matrix under the Gaussian color noise background, thereby reducing redundant information and improving the discrimination of noise signals, and then clustering the target radius matrix using a target clustering algorithm, and the corresponding radius is divided into a noise cluster and a signal cluster by clustering, thereby effectively distinguishing noise and signals, thereby improving the accuracy of the estimation of the number of signal sources under low signal-to-noise ratio and color noise. BRIEF DESCRIPTION OF THE DRAWINGS

[0017] Figure 1 It is a flow chart of a preferred embodiment of the radar source number detection method based on density clustering of the present invention; Figure 2It is a schematic diagram of the Gay disk of the original covariance matrix in the radar source number detection method based on density clustering of the present invention; Figure 3 It is a schematic diagram of a Gay disk of a target transformation matrix after unitary transformation in a radar source number detection method based on density clustering of the present invention; Figure 4 1. It is a schematic diagram of performance comparison of the method for estimating the number of signal sources under a white noise background in the radar source number detection method based on density clustering of the present invention; Figure 5 1. It is a schematic diagram of performance comparison of the method for estimating the number of signal sources under a color noise background in the radar source number detection method based on density clustering of the present invention; Figure 6 It is a structural diagram of a preferred embodiment of a radar source number detection system based on density clustering of the present invention; Figure 7 It is a structural diagram of a preferred embodiment of the terminal of the present invention. DETAILED DESCRIPTION

[0018] In order to make the purpose, technical solution and advantages of the present invention clearer and more specific, the present invention is further described in detail below with reference to the accompanying drawings and examples. It should be understood that the specific embodiments described herein are only used to explain the present invention and are not used to limit the present invention.

[0019] In array radar signal processing technology, signal source angle estimation, that is, the arrival angle estimation of radar echo signals, is a key issue. Many spatial angle estimation algorithms, such as multiple signal classification and rotation invariant subspace technology, rely on the accurate number of radar signal sources as a priori input. However, in practical applications, the number of signal sources is often unknown. Therefore, it is necessary to accurately estimate the number of signal sources. For the problem of estimating the number of radar signal sources, various current methods for estimating the number of signal sources include information theory methods, such as the Akaike Information Criterion (AIC) criterion, the Minimum Description Length (MDL) criterion, the Effective Detection Criterion (EDC) criterion, the Spatial Smoothing Rank (SSR) method, and the Gerschgorin Disk Estimator (GDE) method. Under the condition of spatial white noise, the MDL criterion of the information theory method is a consistent estimate, but the AIC criterion is not a consistent estimate. Therefore, in the case of large samples, the AIC criterion has the problem of over-estimation. Moreover, since these two methods are derived based on the white noise signal model, their performance is greatly reduced or even completely failed under the condition of colored noise. Although the Gaussian circle method can estimate the number of signal sources under spatial colored noise, its accuracy is low under low signal-to-noise ratio conditions, which will affect the estimation of the number of signal sources, resulting in inaccurate estimation of the arrival angle of the radar echo signal.

[0020] In response to one or more of the above problems, the present invention generates an antenna array received data representation according to target radar antenna array information, obtains a target covariance matrix according to the antenna array received data representation, performs a unitary transformation on the target covariance matrix to obtain a target transformation matrix, and generates a radius vector corresponding to the target transformation matrix according to the Gaussian circle theorem; obtains a noise background, and when the noise background is a Gaussian colored noise background, compresses and standardizes the radius vector to obtain a target radius matrix; clusters the target radius matrix according to a target clustering algorithm to obtain a clustering result; obtains a target cluster according to the clustering result, and obtains the number of radar sources according to the target cluster.

[0021] The radar source number detection method based on density clustering described in the preferred embodiment of the present invention is as follows: Figure 1 As shown, the radar source number detection method based on density clustering includes the following steps: Step S10, generating an antenna array received data representation according to the target radar antenna array information, obtaining a target covariance matrix according to the antenna array received data representation, performing a unitary transformation on the target covariance matrix to obtain a target transformation matrix, and generating a radius vector corresponding to the target transformation matrix according to the Gauss circle theorem.

[0022] It should be noted that, under the condition of low signal-to-noise ratio and spatial color noise, the present invention solves the problem of estimating the number of array radar signal sources. In the present invention, the target radar antenna array information includes the number of array elements, radar echo signal, angle, radar echo wavelength and antenna array element spacing. The antenna array receiving data representation is obtained by corresponding to the target radar antenna array information, so as to obtain the number of radar signal sources by applying the corresponding processing process in the present invention.

[0023] Further, generating an antenna array received data representation according to the target radar antenna array information, obtaining a target covariance matrix according to the antenna array received data representation, performing a unitary transformation on the target covariance matrix to obtain a target transformation matrix, and generating a radius vector corresponding to the target transformation matrix according to the Gay circle theorem, specifically includes: Constructing an array steering vector corresponding to each signal source according to the target radar antenna array information, generating a steering vector matrix according to all the array steering vectors, and generating the antenna array reception data representation according to the steering vector matrix; Obtaining the target covariance matrix and the radar echo signal covariance matrix according to the antenna array received data representation; Performing a unitary transformation on the target covariance matrix according to the radar echo signal covariance matrix to obtain the target transformation matrix; According to the Gagnes circle theorem, the Gagnes circle radius of each Gagnes circle is obtained from the target transformation matrix, and the radius vector is generated according to all the Gagnes circle radii.

[0024] Specifically, for the radar antenna array, The array elements are arranged in a straight line, and the far field has The radar echo signal is from the angle The radar signal echo wavelength is incident on the array. , the antenna array element spacing is , then the uniform linear array The array steering vector corresponding to the signal source (Array flow pattern) can be expressed as the following formula (1): ; (1) in, Indicates exponential operation. is the representation of plural units, , Represents a transpose operation.

[0025] Based on all the array steering vectors, a steering vector matrix can be generated. The steering vector matrix corresponding to the antenna array receiving signal is It is expressed as the following formula (2): ; (2) Therefore, in At this moment, the antenna array receives data (containing radar echo signal and noise) can be represented by the antenna array receiving data representation, as shown in the following formula (3): ; (3) in, yes The radar echo signal incident on the array at any time is dimension column vector, is Gaussian white noise or colored noise with cross-correlation, dimension column vector, and They are expressed as the following formula (4) and formula (5) respectively: ; (4) ; (5) in, express The first radar echo signal incident on the array at time, express The second radar echo signal incident on the array at time, express The first A radar echo signal, express The Gaussian white noise or colored noise with cross-correlation corresponding to the first array element at the moment, express The Gaussian white noise or colored noise with cross-correlation corresponding to the second array element at the moment, express Moment Gaussian white noise or colored noise with cross-correlation corresponding to the array elements.

[0026] Then for the antenna array receiving data representation The covariance matrix of (i.e., the target covariance matrix) and The covariance matrix of (i.e., the radar echo signal covariance matrix), which are respectively expressed by the following formulas (6) and (7): ; (6) ; (7) In the above formula, represents the conjugate transpose, Indicates the number of time domain samples, that is, the number of snapshots.

[0027] Further, performing a unitary transformation on the target covariance matrix according to the radar echo signal covariance matrix to obtain the target transformation matrix specifically includes: Divide the target covariance matrix into blocks according to the radar echo signal covariance matrix, and obtain a main sub-matrix according to the result of the block division; Performing eigenvalue decomposition on the principal submatrix to obtain an eigenspace vector of the principal submatrix; A unitary transformation matrix is ​​constructed according to the eigenspace vector, and a unitary transformation operation is performed on the eigenspace vector according to the unitary transformation matrix to obtain the target transformation matrix.

[0028] It should be noted that the present invention uses the Gay disk method, and does not need to know the exact value of the eigenvalue. Instead, it estimates the number of signal sources by estimating the distribution range of the eigenvalue based on the Gay disk theorem. Specifically, it can be set for dimensional matrix, Row and The elements of the column are , for The line is removed Line Sum the absolute values ​​of the elements in the column to get the radius of the disk , expressed as the following formula (8): ; (8) The corresponding Disc The point on the complex plane is expressed as follows: ; (9) in For the The center of a disk, which is a Geiger disk, symbol represents a complex number on the complex plane that satisfies formula (8).

[0029] And according to the Gay disk theorem, the covariance matrix is ​​drawn When the Geiger disk Figure 2 As shown in Figure 2, it can be found that both the radius of the Gaige disk of the radar echo signal and the radius of the Gaige disk of the noise are significantly larger, and the center positions of all Gaige disks are almost coincident, which shows that using the current covariance matrix Estimating the number of signal sources will lack resolution and cannot obtain accurate results. In view of this situation, the present invention needs to separate the signal disk and the noise disk.

[0030] Specifically, in order to separate the signal disk from the noise disk, the present invention needs to calculate the covariance matrix Perform unitary transformation. First, After the blocks are divided, the result after the blocks is obtained, which is expressed as formula (10): : (10) in, for dimensional principal submatrix, for Middle Front of column elements, For the matrix The last row and last column element of .

[0031] Then the main matrix By performing eigenvalue decomposition, we can obtain formula (11): : (11) in, for The eigenvalue of The diagonal matrix formed is , It means to generate a diagonal matrix by taking the elements of a vector as diagonal elements. express The first eigenvalue of express The second eigenvalue of express The third eigenvalue of express No. eigenvalues, for The eigenvector space of , express The eigenvector space of The first eigenvector in express The eigenvector space of The second eigenvector in , express The eigenvector space of The According to the characteristics of the eigenvector, we can get ,here for dimensional identity matrix.

[0032] Further, according to The eigenvector space of Constructing a unitary transformation matrix , the corresponding unitary transformation matrix is ​​expressed as the following formula (12): ; (12) According to the covariance matrix Perform unitary transformation operation to obtain the unitary transformed matrix, that is, the target transformation matrix , expressed as the following formula (13): ; (13) in, , is the steering vector matrix Before Line and front The matrix composed of columns, represents the complex conjugate, is the first parameter, expressed as the following formula (14): ; (14) in, , , .

[0033] Furthermore, in formula (13), is the radius element, expressed as the following formula (15): ; (15) in, for No. feature vectors. According to the Gay circle theorem, in formula (13), i.e., the target transformation matrix, we can obtain Radius of the Geiger circle It is expressed as the following formula (16): ; (16) in, Represents the absolute value operation of elements or elements in a vector or matrix.

[0034] According to of The radius of the Geiger circle can form the corresponding radius vector , expressed as the following formula (17): ; (17) For each radius of the Geiger circle, according to the inequality principle, the following formula (18) can be obtained: ; (18) in, is a row vector.

[0035] From formula (18), we can get: and It is irrelevant, which means that the radius of the Geiger circle depends only on , when the eigenvector When taking the eigenvector of the noise, it is orthogonal to the array flow pattern, so The elements in are close to zero when the eigenvector When taking the characteristic vector of the radar echo signal, The elements in are significantly greater than zero.

[0036] Based on the Gay disk theorem, the corresponding drawing Geiger disk, such as Figure 3 As shown in the figure, it can be seen that after the unitary transformation In the generated Geiger disk, the disk radius corresponding to the radar echo signal is significantly larger than the disk radius corresponding to the noise, and the separation between the disk centers is high. This result shows that the matrix after unitary transformation is The generated Geiger disks have the distribution characteristics of high contrast, low overlap, and strong separability.

[0037] Step S20: Acquire a noise background. When the noise background is a Gaussian color noise background, compress and standardize the radius vector to obtain a target radius matrix.

[0038] Specifically, the characteristics of the Geis disk generated by the target transformation matrix meet the requirements of the Ordering Points To Identify the Clustering Structure (OPTICS) algorithm for density separability and cluster spacing. Using this data as the input of the OPTICS algorithm can significantly improve the reliability and accuracy of the estimation of the number of radar signal sources. The extracted radius is subjected to OPTICS clustering, wherein the OPTICS algorithm is the target clustering algorithm. Before clustering, in order to eliminate the influence of data scale and provide more standardized data input for the OPTICS clustering algorithm, the present invention corresponds to processing based on the noise background for response.

[0039] Furthermore, the acquiring of the noise background, when the noise background is a Gaussian color noise background, compressing and normalizing the radius vector to obtain a target radius matrix, specifically includes: Acquire a noise background, and determine the type of the noise background; When the noise background is a Gaussian color noise background, compressing each Gaussian circle radius in the radius vector according to the unitary transformation matrix and the center of each Gaussian circle to obtain a plurality of compressed radii; Performing standardization on each of the compression radii to obtain a target radius of each Geiger circle; The target radius matrix is ​​generated according to all the target radii.

[0040] Specifically, when the noise background is a Gaussian color noise background, each Gaussian circle radius in the radius vector is compressed according to the unitary transformation matrix and the center of each Gaussian circle to obtain a plurality of compressed radii: , that is, compressed by the following formula (19): ; (19) in, , is the center of Gay's circle, is the unitary transformation matrix The element in the last row and last column is obtained by and The ratio of can be compressed to effectively reduce the numerical range of the radius, thereby reducing the dynamic range of the data.

[0041] Each compression radius is then normalized to obtain the target radius of each Geiger circle. , that is, it is standardized by the following formula (20): ; (20) in, Returns the maximum value of an element in a collection. It means to select the maximum value from all the compressed radii, normalize the compressed radius, and normalize the radius value to range, which helps to eliminate the impact of data scale.

[0042] Then the target radius is summarized to obtain the target radius matrix, as shown in formula (21): ;(twenty one) in, is the target radius matrix.

[0043] Furthermore, the step of obtaining the noise background and determining the type of the noise background further includes: When the noise background is a Gaussian white noise background, performing a logarithmic transformation on the radius vector to obtain a logarithmic transformation matrix; Clustering the logarithmic transformation matrix according to a target clustering algorithm to obtain a white noise clustering result; A white noise target cluster is obtained according to the white noise clustering result, and the corresponding number of radar sources is obtained according to the white noise target cluster.

[0044] Specifically, when the noise background is a Gaussian white noise background, the radius vector is logarithmically transformed to obtain a logarithmic transformation matrix: , that is, logarithmic transformation is performed through formula (22): ;(twenty two) The obtained logarithmic transformation matrix is ​​then clustered using a target clustering algorithm, and a white noise clustering result is obtained to obtain the number of radar sources at this time. The clustering process and the process of obtaining the number of radar sources according to the corresponding clusters are the same as the processing process of the target radius matrix.

[0045] Step S30: clustering the target radius matrix according to a target clustering algorithm to obtain a clustering result.

[0046] Specifically, after eliminating the influence of the data scale, the target radius matrix is ​​clustered using a target clustering algorithm to distinguish signal components from noise components.

[0047] Furthermore, clustering the target radius matrix according to the target clustering algorithm to obtain a clustering result specifically includes: Obtaining a preset distance threshold and a minimum number of neighbors, taking the target radius matrix, the distance threshold and the minimum number of neighbors as input, and performing clustering; In the clustering process, each element in the target radius matrix is ​​traversed, a core object is selected according to the minimum number of neighbors, the core object is added to the core object set and an ordered arrangement is generated, the core distance and the reachable distance of each core object are calculated, and according to the core distance and the reachable distance of each core object in the ordered arrangement, all the core objects are respectively divided into one of the current cluster cluster, the new cluster cluster and the noise point; A clustering result is generated according to the current cluster, the new cluster and the noise point.

[0048] Specifically, in the present invention, the target clustering algorithm is the OPTICS clustering algorithm. For the OPTICS clustering algorithm, given a data set , sample point , and a distance threshold and the minimum number of neighbors MinPts, then the OPTICS related definitions are as follows: -Neighborhood, for sample points ,That -Neighborhood is defined as Distance is The set of sample points within the range, that is , the number of sample sets is recorded as , Indicates that from the sample point To the sample point distance.

[0049] Core objects, for , The condition for being a core object is that -The sample set corresponding to the neighborhood Contains no less than MinPts sample points; Density direct, if is the core object, and the sample point lie in of -neighborhood, then it is called Depend on Density direct access; The density can reach and , there exists a sample sequence , if satisfied , , and the sample sequence Depend on The density is directly Depend on Density can reach; Density connected, for two points and , if there is a core object , and All by The density can reach and Connected by density; At the current core In the neighborhood of The minimum neighborhood radius of a core point is called Core distance , defined as the following formula (23): ;(twenty three) in, Representing a collection Middle and Node The node of the MinPts nearest neighbor, for the sample point , when its -Number of sample points in the neighborhood Less than MinPts, then Does not meet the conditions for being called a core point, the core distance Cannot be defined, so marked as , express arrive distance.

[0050] for , about The reachable distance is defined as follows: ;(twenty four) For two sample points and ,if of -Number of sample points in the neighborhood Less than MinPts, then It is not the core point. about The reachable distance Cannot be defined, so marked as , express The core distance, express arrive distance.

[0051] Specifically, in the present invention, the radius in the target radius matrix is ​​classified by the OPTICS clustering algorithm, and the target radius matrix obtained under the color noise background is For example, the goal of the OPTICS algorithm is to output an ordered arrangement and the target radius of each Geiger circle. The two properties of are the core distance set and the reachable distance set.

[0052] In the clustering process, the specific steps are as follows: Step 1, input: The present invention will and neighborhood parameters are input into the algorithm as a given sample set, where the neighborhood parameters include the distance threshold and the minimum number of neighbors (MinPts), is infinite, so each sample point is a core object. MinPts is 2, then the core distance of each core object is the distance from the nearest sample point to itself, and the sample point itself is also one of the neighborhood elements.

[0053] Step 2, Initialization: Traversal All elements in and initialize the core object collection Is empty and starts to traverse each element in the sample set.

[0054] Step 3, processing: first traverse all elements in the sample set, calculate the core distance of each element, and if its core distance exists, take it as a core object and put it into the core object set If the core distance does not exist, it is not a core object.

[0055] When all elements in the sample set have been processed, Randomly select a core object that has not been processed, push it into the ordered list, check the neighborhood of the core object, calculate the reachable distance from the unvisited points in the neighborhood to the core object, and put these points into the seed set in turn according to the size of the reachable distance middle.

[0056] If the seed collection is empty, then the core object collection The next unprocessed core object is randomly selected for processing. If the seed object is not empty, the point with the unvisited state and the minimum reachable distance is obtained from the seed set, its state is marked as visited, and pushed into the ordered rehearsal column. If the neighboring point obtained at this time is a core object, the unvisited neighboring points of the point are added to and recalculate the seed set The remaining points in the seed set are re-arranged according to the reachable distance from the point to the largest, and the seed set is judged to be empty and the above steps are repeated; until all points in the seed set are in the state of visited, the points in the core object set are The next core object that has not been processed is randomly selected for processing. After all elements in have been processed, this step ends and the output is The corresponding ordered arrangement , core distance set and reachable distance set.

[0057] Step 4, output: After step 3, the ordered arrangement is obtained , core distance set and reachable distance set, and then enter a reachable distance threshold , in the ordered arrangement obtained after the algorithm is executed, sample points are taken out in order: if the reachable distance of the point is , it belongs to the current cluster; if the reachable distance of the point is greater than , the core distance of the point , is a new cluster; if the reachable distance of the point is greater than , and the core distance of the point , is a noise point.

[0058] Through this clustering, we can finally get the current cluster, new cluster and noise points, that is, the corresponding clustering results.

[0059] Step S40: acquiring a target cluster according to the clustering result, and obtaining the number of radar sources according to the target cluster.

[0060] After the corresponding clustering results are obtained, the corresponding number of radar sources is obtained through the clustering results.

[0061] Further, the acquiring a target cluster according to the clustering result, and obtaining the number of radar sources according to the target cluster, specifically includes: Calculating the average values ​​of the current cluster and the new cluster in the clustering result, and selecting the cluster with the larger average value as the target cluster; The number of elements in the target cluster is counted, and the number is output as the number of radar sources.

[0062] Specifically, the average values ​​of each cluster in the previous clustering and the new clustering are calculated, and the average radius values ​​are compared. The cluster with the largest average value is the cluster corresponding to the signal source. The number of points in the cluster is counted, which is the number of signal sources, and finally the number of signal sources is output.

[0063] Furthermore, the present invention conducts simulation experiments to compare the estimation performance of the technical method of the present invention with AIC, MDL and Geiger circle radius method.

[0064] like Figure 4 As shown, design an array with The antenna spacing is The radar antenna array model is as follows: the radar echo signal is incident on the above array with incident angles of 0, 10 and 30 degrees respectively. The noise background is Gaussian white noise. The signal-to-noise ratio is from -20dB to 20dB with a step size of 2dB. The signal power is calculated and the signal power is calculated according to the change. (Signal-to-Noise Ratio) value adjusts the noise power to ensure that the noise power and signal-to-noise ratio are The number of snapshots is set to 500, and the OPTIC reachable distance threshold is 0.26 (selected based on actual experience). The Monte Carlo experiment was repeated 1000 times, and the number of signal sources estimated in each experiment was recorded. If the estimated value is consistent with the actual number of signal sources, it is counted as a successful estimation. The final accuracy acc It is calculated by dividing the number of successes by the total number of experiments, as shown in the following formula (25): ; (25) in To estimate the number of successes, is the total number of experiments, and finally the accuracy curve is calculated and drawn. The vertical axis represents the accuracy. Figure 4 In the present invention, when the signal-to-noise ratio is greater than -5dB, the signal source estimation accuracy of MDL is always maintained at 1, and it shows high stability. When the signal-to-noise ratio ranges from -5dB to 5dB, the accuracy of the signal source number estimation of the technical method of the present invention is significantly better than that of the Geiger circle radius method. When the signal-to-noise ratio is greater than 5dB, its accuracy is not only higher than AIC, but its stability is also significantly better than AIC.

[0065] like Figure 5 As shown, we also design an array with the same number of elements The antenna spacing is The radar antenna array model, the radar echo signal is incident on the above array, the incident angles are 0 degrees, 10 degrees and 30 degrees respectively, the noise background is colored noise, and the Monte Carlo test is repeated 1000 times with a signal-to-noise ratio of -20dB to 20dB and a step size of 2dB. The number of snapshots is 500, and the OPTIC reachable distance threshold is 0.26 (based on actual experience and inspiration). In order to simulate the actual noise environment, in this experiment, the white noise is passed through a 4th-order Butterworth low-pass filter with a cutoff frequency of 0.6. The noise generated after filtering becomes colored noise. The experiment will also compare different methods according to Experiment 1. Calculate and plot the accuracy curve. Figure 5 As shown in the figure, under the background of colored noise, the performance of MDL and AIC drops significantly, and the accuracy is close to 0, indicating that these two methods are almost completely ineffective under colored noise conditions; in contrast, the Gay circle radius method performs better than MDL and AIC when facing colored noise. However, when the signal-to-noise ratio ranges from 0dB to 12dB, the accuracy of the technical method of the present invention in estimating the number of signal sources is significantly higher than that of the Gay circle radius method, and when the signal-to-noise ratio is greater than 5dB, its accuracy is close to 1 and maintains extremely high stability.

[0066] The present invention generates antenna array receiving data representation according to target radar antenna array information, obtains target covariance matrix according to antenna array receiving data representation, performs unitary transformation on the target covariance matrix to obtain target transformation matrix, and generates radius vector corresponding to the target transformation matrix according to Gaussian circle theorem; obtains noise background, and when the noise background is Gaussian color noise background, compresses and standardizes the radius vector to obtain target radius matrix; clusters the target radius matrix according to target clustering algorithm to obtain clustering result; obtains target cluster according to the clustering result, and obtains the number of radar signal sources according to the target cluster. The present invention compresses and standardizes the Gaussian circle radius obtained after unitary transformation of the signal covariance matrix under Gaussian color noise background, thereby reducing redundant information and improving the discrimination of noise signal, and then clusters the target radius matrix using target clustering algorithm, and divides the corresponding radius into noise cluster and signal cluster by clustering, thereby effectively distinguishing noise and signal, thereby improving the accuracy of signal source number estimation under low signal-to-noise ratio and color noise.

[0067] Furthermore, if Figure 6 As shown, based on the above-mentioned radar source number detection method based on density clustering, the present invention also provides a radar source number detection system based on density clustering, wherein the radar source number detection system based on density clustering includes: A radius vector generating module 61 is used to generate an antenna array receiving data representation according to the target radar antenna array information, obtain a target covariance matrix according to the antenna array receiving data representation, perform a unitary transformation on the target covariance matrix to obtain a target transformation matrix, and generate a radius vector corresponding to the target transformation matrix according to the Gay circle theorem; A target radius matrix generating module 62 is used to obtain a noise background. When the noise background is a Gaussian color noise background, the radius vector is compressed and normalized to obtain a target radius matrix. A clustering result generating module 63 is used to cluster the target radius matrix according to a target clustering algorithm to obtain a clustering result; The result output module 64 is used to obtain a target cluster according to the clustering result, and obtain the number of radar sources according to the target cluster.

[0068] Furthermore, if Figure 7 As shown, based on the above-mentioned radar source number detection method and system based on density clustering, the present invention also provides a terminal accordingly, and the terminal includes a processor 10, a memory 20 and a display 30. Figure 7 Only some components of the terminal are shown, but it should be understood that it is not required to implement all of the components shown, and more or fewer components may be implemented instead.

[0069] In some embodiments, the memory 20 may be an internal storage unit of the terminal, such as a hard disk or memory of the terminal. In other embodiments, the memory 20 may also be an external storage device of the terminal, such as a plug-in hard disk, a smart media card (SMC), a secure digital (SD) card, a flash card, etc. equipped on the terminal. Further, the memory 20 may also include both an internal storage unit of the terminal and an external storage device. The memory 20 is used to store application software and various types of data installed in the terminal, such as the program code of the installation terminal. The memory 20 may also be used to temporarily store data that has been output or is to be output. In one embodiment, a radar source number detection program 40 based on density clustering is stored on the memory 20, and the radar source number detection program 40 based on density clustering can be executed by the processor 10, thereby realizing the radar source number detection method based on density clustering in the present invention.

[0070] In some embodiments, the processor 10 may be a central processing unit (CPU), a microprocessor or other data processing chip, used to run the program code or process data stored in the memory 20, such as executing the radar source number detection method based on density clustering.

[0071] In some embodiments, the display 30 may be an LED display, a liquid crystal display, a touch-sensitive liquid crystal display, an OLED (Organic Light-Emitting Diode) touch device, etc. The display 30 is used to display information on the terminal and to display a visual user interface.

[0072] In one embodiment, when the processor 10 executes the radar signal source number detection program 40 based on density clustering in the memory 20 , the steps of the above radar signal source number detection method based on density clustering are implemented.

[0073] The present invention also provides a computer-readable storage medium, wherein the computer-readable storage medium stores a radar signal source number detection program based on density clustering, and when the radar signal source number detection program based on density clustering is executed by a processor, the steps of the radar signal source number detection method based on density clustering as described above are implemented.

[0074] It should be noted that, in this article, the terms "include", "comprises" or any other variations thereof are intended to cover non-exclusive inclusion, so that a process, method, article or terminal including a series of elements includes not only those elements, but also other elements not explicitly listed, or also includes elements inherent to such process, method, article or terminal. In the absence of further restrictions, an element defined by the sentence "includes a ..." does not exclude the existence of other identical elements in the process, method, article or terminal including the element.

[0075] Of course, those skilled in the art can understand that all or part of the processes in the above-mentioned embodiments can be implemented by instructing related hardware (such as a processor, a controller, etc.) through a computer program, and the program can be stored in a computer-readable storage medium that can be read by a computer, and the program can include the processes of the above-mentioned method embodiments when executed. The computer-readable storage medium can be a memory, a disk, an optical disk, etc.

[0076] It should be understood that the application of the present invention is not limited to the above examples. For ordinary technicians in this field, improvements or changes can be made based on the above description. All these improvements and changes should fall within the scope of protection of the claims attached to the present invention.

Claims

1. A radar source number detection method based on density clustering, characterized in that: The radar source number detection method based on density clustering includes: Generate an antenna array receiving data representation according to the target radar antenna array information, obtain a target covariance matrix according to the antenna array receiving data representation, perform a unitary transformation on the target covariance matrix to obtain a target transformation matrix, and generate a radius vector corresponding to the target transformation matrix according to the Gauss circle theorem; Acquire a noise background, and when the noise background is a Gaussian color noise background, compress and standardize the radius vector to obtain a target radius matrix; Clustering the target radius matrix according to a target clustering algorithm to obtain a clustering result; A target cluster is obtained according to the clustering result, and the number of radar sources is obtained according to the target cluster.

2. The radar source number detection method based on density clustering according to claim 1 is characterized in that: The method of generating an antenna array receiving data representation according to the target radar antenna array information, obtaining a target covariance matrix according to the antenna array receiving data representation, performing a unitary transformation on the target covariance matrix to obtain a target transformation matrix, and generating a radius vector corresponding to the target transformation matrix according to the Gay circle theorem specifically includes: Constructing an array steering vector corresponding to each signal source according to the target radar antenna array information, generating a steering vector matrix according to all the array steering vectors, and generating the antenna array reception data representation according to the steering vector matrix; Obtaining the target covariance matrix and the radar echo signal covariance matrix according to the antenna array received data representation; Performing a unitary transformation on the target covariance matrix according to the radar echo signal covariance matrix to obtain the target transformation matrix; According to the Gagnes circle theorem, the Gagnes circle radius of each Gagnes circle is obtained from the target transformation matrix, and the radius vector is generated according to all the Gagnes circle radii.

3. The radar source number detection method based on density clustering according to claim 2 is characterized in that: The unitary transformation of the target covariance matrix according to the radar echo signal covariance matrix to obtain the target transformation matrix specifically includes: Divide the target covariance matrix into blocks according to the radar echo signal covariance matrix, and obtain a main sub-matrix according to the result of the block division; Performing eigenvalue decomposition on the principal submatrix to obtain an eigenspace vector of the principal submatrix; A unitary transformation matrix is ​​constructed according to the eigenspace vector, and a unitary transformation operation is performed on the eigenspace vector according to the unitary transformation matrix to obtain the target transformation matrix.

4. The radar source number detection method based on density clustering according to claim 3 is characterized in that: The acquiring of the noise background, when the noise background is a Gaussian color noise background, compressing and normalizing the radius vector to obtain a target radius matrix, specifically includes: Acquire a noise background, and determine the type of the noise background; When the noise background is a Gaussian color noise background, compressing each Gaussian circle radius in the radius vector according to the unitary transformation matrix and the center of each Gaussian circle to obtain a plurality of compressed radii; Performing standardization on each of the compression radii to obtain a target radius of each Geiger circle; The target radius matrix is ​​generated according to all the target radii.

5. The radar source number detection method based on density clustering according to claim 4 is characterized in that: The step of obtaining the noise background and determining the type of the noise background further includes: When the noise background is a Gaussian white noise background, performing a logarithmic transformation on the radius vector to obtain a logarithmic transformation matrix; Clustering the logarithmic transformation matrix according to a target clustering algorithm to obtain a white noise clustering result; A white noise target cluster is obtained according to the white noise clustering result, and the corresponding number of radar sources is obtained according to the white noise target cluster.

6. The radar source number detection method based on density clustering according to claim 1, characterized in that: The step of clustering the target radius matrix according to the target clustering algorithm to obtain a clustering result specifically includes: Obtaining a preset distance threshold and a minimum number of neighbors, taking the target radius matrix, the distance threshold and the minimum number of neighbors as input, and performing clustering; In the clustering process, each element in the target radius matrix is ​​traversed, a core object is selected according to the minimum number of neighbors, the core object is added to the core object set and an ordered arrangement is generated, the core distance and the reachable distance of each core object are calculated, and according to the core distance and the reachable distance of each core object in the ordered arrangement, all the core objects are respectively divided into one of the current cluster cluster, the new cluster cluster and the noise point; A clustering result is generated according to the current cluster, the new cluster and the noise point.

7. The radar source number detection method based on density clustering according to claim 6 is characterized in that: The acquiring of the target cluster according to the clustering result and obtaining the number of radar sources according to the target cluster specifically includes: Calculating the average values ​​of the current cluster and the new cluster in the clustering result, and selecting the cluster with the larger average value as the target cluster; The number of elements in the target cluster is counted, and the number is output as the number of radar sources.

8. A radar source number detection system based on density clustering, characterized in that: The radar source number detection system based on density clustering includes: A radius vector generation module is used to generate an antenna array receiving data representation according to the target radar antenna array information, obtain a target covariance matrix according to the antenna array receiving data representation, perform a unitary transformation on the target covariance matrix to obtain a target transformation matrix, and generate a radius vector corresponding to the target transformation matrix according to the Gay circle theorem; A target radius matrix generation module is used to obtain a noise background. When the noise background is a Gaussian color noise background, the radius vector is compressed and standardized to obtain a target radius matrix. A clustering result generating module, used for clustering the target radius matrix according to a target clustering algorithm to obtain a clustering result; The result output module is used to obtain the target cluster according to the clustering result, and obtain the number of radar sources according to the target cluster.

9. A terminal, characterized in that: The terminal includes: a memory, a processor, and a radar source number detection program based on density clustering stored in the memory and executable on the processor. When the radar source number detection program based on density clustering is executed by the processor, the steps of the radar source number detection method based on density clustering as described in any one of claims 1 to 7 are implemented.

10. A computer-readable storage medium, characterized in that: The computer-readable storage medium stores a radar source number detection program based on density clustering, and when the radar source number detection program based on density clustering is executed by a processor, the steps of the radar source number detection method based on density clustering as described in any one of claims 1 to 7 are implemented.

Images