A radar signal sorting method and apparatus

Abstract

The application discloses a radar signal sorting method and device, comprising: obtaining a historical radar signal pulse description word matrix comprising multiple column pulse description word vectors; compressing the historical radar signal pulse description word matrix; obtaining a characteristic subspace of the pulse description word; obtaining a characteristic subspace library by using the characteristic subspace of the pulse description word; processing a real-time extracted pulse description word vector by using a preset model; obtaining a main characteristic subspace of the radar signal pulse description word; obtaining a load matrix according to the converged main characteristic subspace of the radar signal pulse description word; and obtaining a target radar signal type sorting result according to an orthogonal result of the load matrix and the characteristic subspace library. The method realizes fast and accurate sorting of the radar signal under the premise of reserving main characteristics of original signal description words, provides a theoretical basis and technical support for performance improvement of weapon equipment, and has good engineering application value.

Description

Technical Field

[0001] This invention relates to the field of radar signal processing technology, and more specifically to a radar signal sorting method and apparatus. Background Technology

[0002] Radar countermeasures are an important means of modern information warfare, and radar signal sorting based on parameter processing is a crucial aspect of radar countermeasures. With the development of radar technology, the emergence of new radar systems, improvements in operating modes, the surge in data volume, and the interleaving of signals in the time and frequency domains, the requirements for data processing speed and sorting accuracy in radar signal sorting are becoming increasingly stringent.

[0003] Different types of radars, operating under different systems, have at least one distinct pulse descriptor among their various pulse descriptors. Therefore, radar signal sorting is often based on these pulse descriptors. Traditional signal sorting methods rely on one or a few pulse descriptors; however, when dealing with complex radar sorting applications, they often suffer from low sorting accuracy. On the other hand, using all available pulse descriptors for signal sorting can effectively improve sorting accuracy, but this presents challenges in data processing. Summary of the Invention

[0004] This invention provides a radar signal sorting method, comprising:

[0005] Obtain a historical radar signal pulse description word matrix that includes multiple columns of pulse description word vectors;

[0006] The pulse descriptor matrix of historical radar signals is compressed to obtain the feature subspace of the pulse descriptors, and the feature subspace library is obtained by using the feature subspace of the pulse descriptors.

[0007] The pulse descriptor vectors extracted in real time are processed using a preset model to obtain the main feature subspace of the radar signal pulse descriptor, and the load matrix is ​​obtained based on the converged main feature subspace of the radar signal pulse descriptor.

[0008] Based on the orthogonality between the load matrix and the feature subspace library, the target radar signal type sorting results are obtained.

[0009] Furthermore, the pulse descriptor matrix of historical radar signals is compressed to obtain the feature subspace of the pulse descriptors, and a pulse descriptor feature subspace library is constructed using the feature subspace of the pulse descriptors, including:

[0010] The acquired historical radar signal pulse descriptor matrix is ​​normalized to obtain a data table matrix;

[0011] Eigenvalue decomposition is performed on the covariance matrix of the data table matrix to obtain the eigenspace of the covariance matrix and multiple eigenvalues ​​arranged in descending order;

[0012] Obtain the feature vectors corresponding to multiple feature values ​​that are sorted first in the feature space and whose sum is greater than a preset value, and obtain multiple load vector matrices;

[0013] The feature subspace of the pulse descriptor is obtained by multiplying multiple load vector matrices and historical radar signal pulse descriptor matrices.

[0014] A feature subspace library is constructed based on the feature subspaces of multiple diagonally arranged pulse descriptors.

[0015] Furthermore, the pulse descriptor vectors extracted in real time are processed using a pre-defined model to obtain the main feature subspace of the radar signal pulse descriptor. Based on the converged main feature subspace of the radar signal pulse descriptor, the load matrix is ​​obtained, including:

[0016] Obtain the pulse description word vector d(k) extracted in real time;

[0017] The main feature subspace is updated using the model preset in formula (1) to obtain the main feature subspace of the radar signal pulse descriptor;

[0018] W(k+1)=W(k)+η[R(k)W(k)-W(k)DW T (k)W(k)D -1 (1)

[0019] Where R = E{dd T Let} be the expectation matrix of d(k), W be the principal subspace of R, η be the convergence factor, and D be a pre-defined diagonal matrix whose diagonal elements are constants and arranged in descending order. The expectation matrix R is updated using formula (2):

[0020] R(k+1)=αR(k)+(1-α)[d(k+1)d T (k+1)] (2)

[0021] Where α is the forgetting factor;

[0022] Once formulas (1) and (2) converge, the load matrix is ​​obtained, and its expression includes formula (3).

[0023]

[0024] Where P is called the load matrix, Up is the principal subspace of the load matrix with dimension p, Λ p It is a diagonal matrix, and its diagonal elements are the p principal eigenvalues ​​corresponding to the principal eigenspace.

[0025] Furthermore, based on the orthogonality between the load matrix and the pulse descriptor feature subspace library, the target radar signal type sorting results are obtained, including:

[0026] Based on the historical radar signal pulse descriptor matrix, the feature subspace of the pulse descriptor is obtained;

[0027] Based on the feature subspaces of the diagonally arranged descriptive characters, a feature subspace library is obtained;

[0028] Obtain the orthogonal matrix between the load matrix and the feature subspace library;

[0029] Using preset vectors to process orthogonal matrices, a diagonal matrix is ​​obtained;

[0030] By using the historical radar signal pulse descriptor matrix corresponding to the maximum value in the diagonal array, the radar signal type corresponding to the historical radar signal pulse descriptor matrix can be obtained.

[0031] The present invention also provides a radar signal sorting device, comprising:

[0032] The acquisition module is used to acquire a historical radar signal pulse description word matrix that includes multiple columns of pulse description word vectors;

[0033] The first processing module is used to compress the pulse descriptor matrix of historical radar signals to obtain the feature subspace of the pulse descriptor, and to obtain a feature subspace library using the feature subspace of the pulse descriptor.

[0034] The second processing module is used to process the pulse descriptor vector extracted in real time using a preset model, obtain the main feature subspace of the radar signal pulse descriptor, and obtain the load matrix based on the converged main feature subspace of the radar signal pulse descriptor.

[0035] The output module is used to obtain the target radar signal type sorting results based on the orthogonality between the load matrix and the feature subspace library.

[0036] The present invention also provides an electronic device, which includes a processor and a memory. The memory stores at least one instruction, at least one program, code set or instruction set. The processor loads and executes the at least one instruction, at least one program, code set or instruction set to implement a radar signal sorting method.

[0037] The present invention also provides a computer-readable storage medium storing at least one instruction, at least one program, code set or instruction set, wherein the at least one instruction, at least one program, code set or instruction set is loaded and executed by a processor to implement a radar signal sorting method.

[0038] This invention provides a radar signal sorting method, which has the following advantages compared with the prior art:

[0039] This invention addresses the issue of low signal rate or accuracy in traditional radar signal sorting methods by providing a fast radar signal sorting method based on feature compression. While preserving the main features of the original signal descriptors, it achieves rapid and accurate radar signal sorting, providing a theoretical basis and technical support for improving weapon system performance, and possesses significant engineering application value. Attached Figure Description

[0040] Figure 1 A flowchart of a radar signal sorting method provided in an embodiment of the present invention;

[0041] Figure 2 This is a schematic diagram of a radar signal sorting device provided in an embodiment of the present invention. Detailed Implementation

[0042] The technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. 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 are within the scope of protection of the present invention.

[0043] See Figure 1 This invention provides a radar signal sorting method, which includes:

[0044] S101. Obtain the historical radar signal pulse description word matrix, which includes multiple columns of pulse description word vectors;

[0045] Select various typical descriptors that can characterize the features of radar pulse signals, and construct the pulse descriptor vector described in step S101. To achieve accurate sorting of radar signals, it is necessary to incorporate as many signal feature descriptors as possible. In practical applications, the following typical radar signal feature descriptors can be selected:

[0046] ①Pulse arrival time (TOA)

[0047] The Time of Arrival (TOA) refers to the time interval between the transmission of a radar pulse signal and its reception by the receiver. Clearly, TOA can be used to calculate target range.

[0048] ②Pulse Repetition Interval (PRI)

[0049] The pulse repetition frequency (PRF) refers to the time interval between two adjacent pulse signals emitted by the same radar. Based on the variation pattern of the PRF, it can be classified into fixed PRF signals, staggered PRF signals, jittered PRF signals, slip-modulated PRF signals, and sinusoidal modulated PRF signals, etc.

[0050] ③ Pulse arrival direction angle (DOA)

[0051] The DOA refers to the relative angle between the radar receiver and the target radiation source. Clearly, the DOA changes with the relative motion between the receiver and the target.

[0052] ④ Pulse Width (PW)

[0053] The time interval between the arrival and end times of a pulse signal, i.e., the duration of a single pulse.

[0054] ⑤ Pulse amplitude (PA).

[0055] This refers to the maximum voltage amplitude of the pulse signal, which is related to the instantaneous power of the radiation source. The power level affects the detection range.

[0056] ⑥ Carrier frequency (RF)

[0057] The frequency of the carrier signal in the modulated pulse signal.

[0058] ⑦ Other pulse description words

[0059] This includes intrapulse parameter modulation methods, antenna scanning parameters, etc.

[0060] After determining the pulse descriptor, in order to facilitate radar signal sorting, they are combined to establish a pulse descriptor vector:

[0061] d = [TOA,PRI,DOA,PW,PA,RF,L] T .

[0062] S102. Compress the pulse descriptor matrix of historical radar signals to obtain the feature subspace of the pulse descriptor, and use the feature subspace of the pulse descriptor to obtain the feature subspace library.

[0063] In step S102, the historical radar signal pulse description word matrix is ​​compressed by extracting its main feature subspace based on existing radar data, and then projecting the pulse description word vector described in step S101 onto this feature subspace to obtain a low-dimensional pulse description word vector.

[0064] To improve the accuracy of radar signal sorting, it is necessary to incorporate as much pulse descriptor information as possible. Therefore, the pulse descriptor vector established in step S102 often has a high dimensionality. On the other hand, there is a certain correlation between the pulse descriptors in the vector, resulting in some redundancy. The high dimensionality and redundancy make this vector unsuitable for direct signal sorting and require dimensionality reduction processing first.

[0065] S103. Process the pulse descriptor vector extracted in real time using a preset model to obtain the main feature subspace of the radar signal pulse descriptor, and obtain the load matrix based on the converged main feature subspace of the radar signal pulse descriptor.

[0066] In step S103, radar signals are sorted by comparing the direction consistency between the extracted real-time radar signal pulse description word vector and the historical radar signal pulse description word feature vector.

[0067] Because different radars have different operating systems and modes, the extracted pulse descriptors vary, resulting in different directions for the compressed radar signal pulse descriptor vectors. The radar signal pulse descriptor vectors extracted based on historical data through the above steps can be considered as vectors representing the radar's characteristic information. Furthermore, by extracting the radar signal pulse descriptor vectors in real time from the radar's operating signals and finding the historical radar signal pulse descriptor feature vectors whose directions are closest to these vectors, radar signal sorting can be achieved. In addition, due to the existence of random disturbances, the extracted real-time radar signal pulse descriptor vectors have a certain random deviation. However, by reducing the dimensionality and removing background noise through principal subspace extraction, the speed and accuracy of signal sorting can be significantly improved.

[0068] S104. Based on the orthogonality between the load matrix and the feature subspace library, obtain the target radar signal type sorting results.

[0069] This invention addresses the need for rapid radar signal sorting by establishing radar signal pulse descriptor vectors and then compressing their features. This allows for maximum feature compression while preserving the main features of the original signal descriptor, thereby reducing the computational complexity of signal sorting and enabling rapid radar signal sorting.

[0070] In one possible implementation, the pulse descriptor matrix of historical radar signals is compressed to obtain the feature subspace of the pulse descriptors, and a pulse descriptor feature subspace library is constructed using the feature subspace of the pulse descriptors, including:

[0071] The acquired historical radar signal pulse descriptor matrix is ​​normalized to obtain a data table matrix;

[0072] Eigenvalue decomposition is performed on the covariance matrix of the data table matrix to obtain the eigenspace of the covariance matrix and multiple eigenvalues ​​arranged in descending order;

[0073] Obtain the feature vectors corresponding to multiple feature values ​​that are sorted first in the feature space and whose sum is greater than a preset value, and obtain multiple load vector matrices;

[0074] The feature subspace of the pulse descriptor is obtained by multiplying multiple load vector matrices and historical radar signal pulse descriptor matrices.

[0075] A feature subspace library is constructed based on the feature subspaces of multiple diagonally arranged pulse descriptors.

[0076] In the embodiments provided by this invention, based on multiple sets of historical radar data of a certain radar, the historical radar signal pulse descriptor matrix of that radar is obtained, i.e.

[0077]

[0078] Where m is the number of selected pulse descriptor words, and n is the number of extracted pulse descriptor word vectors. To eliminate the influence of differences in the metric size of each pulse descriptor word on the signal sorting results, the data needs to be normalized.

[0079]

[0080] Where, d i,j d represents the element in the i-th row and j-th column of the data table matrix D. j,max and d j,min These represent the maximum and minimum values ​​in column j, respectively.

[0081] The covariance matrix of data table matrix D Perform eigenvalue decomposition, i.e.

[0082] R = UΛU T

[0083] Where U is called the eigenspace of the covariance matrix R, and its columns are called eigenvectors. Λ is a diagonal matrix whose diagonal elements λi are arranged in descending order, representing the eigenvalues ​​of the covariance matrix R. In feature compression applications, it is generally believed that the eigenvectors corresponding to larger eigenvalues ​​represent most of the original information, while the eigenvectors corresponding to smaller eigenvalues ​​represent redundant information in the original data. Therefore, the eigenvectors P = U(:,1:r) corresponding to the first r largest eigenvalues ​​of U can be used to compress the data table matrix D, with a compression ratio of:

[0084]

[0085] Generally, η ≥ 85%, and the compression method is...

[0086] T = PD

[0087] In the formula, P is called the load vector of the data table matrix D, T is called the score vector of the data table matrix D. The purpose of feature compression is to obtain the load vector matrix from the original m-dimensional data, and then project the data onto the eigen-subspace of the load vector to obtain the eigen-subspace of the pulse descriptor of the new low-dimensional data with r dimensions (r << m). Each dimension of the eigen-subspace of the pulse descriptor is a linear combination of the variables of each dimension of the original data, but the variables of each dimension of the new data are independent of each other and there is no redundant information.

[0088] In a possible implementation, a preset model is used to process the pulse descriptor vector extracted in real time to obtain the main eigen-subspace of the radar signal pulse descriptor. Obtaining the load matrix according to the converged main eigen-subspace of the radar signal pulse descriptor includes:

[0089] Obtain the pulse descriptor vector d(k) extracted in real time, where k represents the actual number of pulse descriptors extracted;

[0090] Use the preset model in formula (1) to update the main eigen-subspace to obtain the main eigen-subspace of the radar signal pulse descriptor;

[0091] W(k + 1) = W(k) + η[R(k)W(k) - W(k)DW T (k)W(k)D -1 (1)

[0092] where R = E{dd T} is the expected matrix of d(k), W is the principal subspace of the expected matrix R, η is the convergence factor, D is a preset diagonal matrix. Here, D in the preset model is a preset diagonal matrix, which is different from the previous data table matrix D. Its diagonal elements are constants and arranged in descending order. The expected matrix R is updated using formula (2):

[0093] R(k + 1) = αR(k) + (1 - α)[d(k + 1)d T (k + 1)] (2)

[0094] where α is the forgetting factor;

[0095] When formulas (1) and (2) converge, obtain the load matrix, and its expression includes formula (3);

[0096]

[0097] where P is called the load matrix, Up is the principal subspace of the load matrix, whose dimension is p, Λ p is a diagonal matrix, and its diagonal elements are the p main eigenvalues corresponding to the main eigen-subspace.

[0098] In one possible implementation, the target radar signal type sorting result is obtained based on the orthogonality result of the load matrix and the pulse descriptor feature subspace library, including:

[0099] Based on the historical radar signal pulse descriptor matrix, the feature subspace of the pulse descriptor is obtained;

[0100] Based on the feature subspaces of the diagonally arranged descriptive characters, a feature subspace library is obtained;

[0101] Obtain the orthogonal matrix between the load matrix and the feature subspace library;

[0102] Using preset vectors to process orthogonal matrices, a diagonal matrix is ​​obtained;

[0103] By using the historical radar signal pulse descriptor matrix corresponding to the maximum value in the diagonal array, the radar signal type corresponding to the historical radar signal pulse descriptor matrix can be obtained.

[0104] In the embodiments provided by this invention, among the known q-type radars, for each type of radar, the feature subspace of its pulse descriptor can be obtained based on historical data. Assuming the feature subspace of radar i is Ξ i Define matrix

[0105]

[0106] Given a feature subspace library for radar signals, calculate the orthogonality between the load matrix and the feature subspace library:

[0107]

[0108] definition

[0109] L=[l1 l2 L l q ] T

[0110] in,

[0111]

[0112] calculate

[0113] Q = L T YL

[0114] Finally, the calculated matrix Q is a q×q diagonal matrix, and the historical radar signal pulse descriptor matrix corresponding to the maximum value in its diagonal elements is obtained, thus yielding the radar signal type corresponding to the historical radar signal pulse descriptor matrix.

[0115] The present invention also provides a radar signal sorting device 200, comprising:

[0116] The acquisition module 201 is used to acquire a historical radar signal pulse description word matrix including multiple columns of pulse description word vectors;

[0117] The first processing module 202 is used to compress the pulse descriptor matrix of historical radar signals to obtain the feature subspace of the pulse descriptor, and to obtain a feature subspace library using the feature subspace of the pulse descriptor.

[0118] The second processing module 203 is used to process the pulse descriptor vector extracted in real time using a preset model, obtain the main feature subspace of the radar signal pulse descriptor, and obtain the load matrix based on the converged main feature subspace of the radar signal pulse descriptor.

[0119] Output module 204 is used to obtain the target radar signal type sorting result based on the orthogonality result of the load matrix and the feature subspace library.

[0120] This invention also provides an electronic device, which includes a processor and a memory. The memory stores at least one instruction, at least one program, code set, or instruction set. The processor loads and executes the at least one instruction, at least one program, code set, or instruction set to implement a radar signal sorting method.

[0121] This invention also provides a computer-readable storage medium storing at least one instruction, at least one program, code set, or instruction set, wherein the at least one instruction, at least one program, code set, or instruction set is loaded and executed by a processor to implement a radar signal sorting method.

[0122] In the above embodiments, implementation can be achieved entirely or partially through software, hardware, firmware, or any combination thereof. When implemented using software, it can be implemented entirely or partially in the form of a computer program product. The computer program product includes multiple computer instructions. When the computer program instructions are loaded and executed on a computer, all or part of the processes or functions described in the embodiments of the present invention are generated. The computer can be a general-purpose computer, a special-purpose computer, a computer network, or other programmable device. The computer instructions can be stored in a computer-readable storage medium or transmitted from one computer-readable storage medium to another. For example, the computer instructions can be transmitted from one website, computer, server, or data center to another website, computer, server, or data center via wired (e.g., coaxial cable, fiber optic, digital subscriber line (DSL)) or wireless (e.g., infrared, wireless, microwave, etc.) means. The computer-readable storage medium can be any available medium that a computer can access or a data storage device such as a server or data center that integrates multiple available media. The available medium can be a magnetic medium (e.g., floppy disk, hard disk, magnetic tape), an optical medium (e.g., DVD), or a semiconductor medium (e.g., solid-state disk (SSD)).

[0123] It should be noted that, in this document, relational terms such as "first" and "second" are used only to distinguish one entity or operation from another, and do not necessarily require or imply any such actual relationship or order between these entities or operations. Furthermore, the terms "comprising," "including," or any other variations thereof are intended to cover non-exclusive inclusion, such that a process, method, article, or apparatus that comprises a list of elements includes not only those elements but also other elements not expressly listed, or elements inherent to such a process, method, article, or apparatus. Without further limitations, an element defined by the phrase "comprising one..." does not exclude the presence of other identical elements in the process, method, article, or apparatus that includes said element.

[0124] The above-disclosed embodiments are merely a few specific examples of the present invention. Those skilled in the art can make various modifications and variations to the embodiments of the present invention without departing from the spirit and scope of the present invention. However, the embodiments of the present invention are not limited thereto, and any changes that those skilled in the art can conceive of should fall within the protection scope of the present invention.

Claims

1. A radar signal sorting method, characterized in that, include: Obtain a historical radar signal pulse description word matrix that includes multiple columns of pulse description word vectors; The pulse descriptor matrix of historical radar signals is compressed to obtain the feature subspace of the pulse descriptors, and the feature subspace library is obtained by using the feature subspace of the pulse descriptors. The pulse descriptor vectors extracted in real time are processed using a preset model to obtain the main feature subspace of the radar signal pulse descriptor, and the load matrix is ​​obtained based on the converged main feature subspace of the radar signal pulse descriptor. Based on the orthogonality between the load matrix and the feature subspace library, the target radar signal type sorting results are obtained; The process of using a preset model to process the real-time extracted pulse descriptor vectors to obtain the main feature subspace of the radar signal pulse descriptor, and obtaining the load matrix based on the converged main feature subspace of the radar signal pulse descriptor, includes: Obtain the real-time extracted pulse description word vector d(k), where k represents the number of real-time extracted pulse description words; The main feature subspace is updated using the model preset in formula (1) to obtain the main feature subspace of the radar signal pulse descriptor; （1） in, Let d(k) be the expectation matrix, and W be the principal subspace of R. Let D be the convergence factor, and let D be a pre-defined diagonal matrix whose diagonal elements are constants and arranged in descending order. The expected matrix R is updated using formula (2): （2） in, Forgetting factor; Once formulas (1) and (2) converge, the load matrix is ​​obtained, and its expression includes formula (3). （3） Where P is called the load matrix, and Up is the principal subspace of the load matrix with dimension p. It is a diagonal matrix, and its diagonal elements are the p principal eigenvalues ​​corresponding to the principal eigenspace; The process of obtaining target radar signal type sorting results based on the orthogonality of the load matrix and the pulse descriptor feature subspace library includes: Based on the historical radar signal pulse descriptor matrix, the feature subspace of the pulse descriptor is obtained; Based on the feature subspaces of the diagonally arranged descriptive characters, a feature subspace library is obtained; Obtain the orthogonal matrix between the load matrix and the feature subspace library; Using preset vectors to process orthogonal matrices, a diagonal matrix is ​​obtained; By using the historical radar signal pulse descriptor matrix corresponding to the maximum value in the diagonal array, the radar signal type corresponding to the historical radar signal pulse descriptor matrix can be obtained.

2. The radar signal sorting method as described in claim 1, characterized in that, The process of compressing the historical radar signal pulse descriptor matrix to obtain the feature subspace of the pulse descriptors, and constructing a pulse descriptor feature subspace library using the feature subspace of the pulse descriptors, includes: The acquired historical radar signal pulse descriptor matrix is ​​normalized to obtain a data table matrix; Eigenvalue decomposition is performed on the covariance matrix of the data table matrix to obtain the eigenspace of the covariance matrix and multiple eigenvalues ​​arranged in descending order; Obtain the feature vectors corresponding to multiple feature values ​​that are sorted first in the feature space and whose sum is greater than a preset value, and obtain multiple load vector matrices; The feature subspace of the pulse descriptor is obtained by multiplying multiple load vector matrices and historical radar signal pulse descriptor matrices. A feature subspace library is constructed based on the feature subspaces of multiple diagonally arranged pulse descriptors.

3. A radar signal sorting device, characterized in that, include: The acquisition module is used to acquire a historical radar signal pulse description word matrix that includes multiple columns of pulse description word vectors; The first processing module is used to compress the pulse descriptor matrix of historical radar signals to obtain the feature subspace of the pulse descriptor, and to obtain a feature subspace library using the feature subspace of the pulse descriptor. The second processing module is used to process the real-time extracted pulse descriptor vector using a preset model, obtain the main feature subspace of the radar signal pulse descriptor, and obtain the load matrix based on the converged main feature subspace of the radar signal pulse descriptor. This includes: obtaining the real-time extracted pulse descriptor vector d(k), where k represents the number of real-time extracted pulse descriptors. The main feature subspace is updated using the model preset in formula (1) to obtain the main feature subspace of the radar signal pulse descriptor; （1） in, Let d(k) be the expectation matrix, and W be the principal subspace of R. Let D be the convergence factor, and let D be a pre-defined diagonal matrix whose diagonal elements are constants and arranged in descending order. The expected matrix R is updated using formula (2): （2） in, Forgetting factor; Once formulas (1) and (2) converge, the load matrix is ​​obtained, and its expression includes formula (3). （3） Where P is called the load matrix, and Up is the principal subspace of the load matrix with dimension p. It is a diagonal matrix, and its diagonal elements are the p principal eigenvalues ​​corresponding to the principal eigenspace; The output module is used to obtain the target radar signal type sorting results based on the orthogonality between the load matrix and the feature subspace library, including: Based on the historical radar signal pulse descriptor matrix, the feature subspace of the pulse descriptor is obtained; Based on the feature subspaces of the diagonally arranged descriptive characters, a feature subspace library is obtained; Obtain the orthogonal matrix between the load matrix and the feature subspace library; Using preset vectors to process orthogonal matrices, a diagonal matrix is ​​obtained; By using the historical radar signal pulse descriptor matrix corresponding to the maximum value in the diagonal array, the radar signal type corresponding to the historical radar signal pulse descriptor matrix can be obtained.

4. An electronic device, characterized in that, The electronic device includes a processor and a memory, wherein the memory stores at least one instruction, at least one program, a code set, or an instruction set, and the at least one instruction, the at least one program, the code set, or the instruction set is loaded and executed by the processor to implement the radar signal sorting method as described in claim 2.

5. A computer-readable storage medium, characterized in that, The storage medium stores at least one instruction, at least one program, code set, or instruction set, wherein the at least one instruction, the at least one program, the code set, or instruction set is loaded and executed by a processor to implement the radar signal sorting method as described in claim 2.

Images