CN-122017737-A - Radar model sorting method, device and storage medium

CN122017737ACN 122017737 ACN122017737 ACN 122017737ACN-122017737-A

Abstract

The invention provides a radar model sorting method, a radar model sorting device and a storage medium, which belong to the technical field of radar sorting, wherein the method comprises the steps of importing a radar signal PDW data sequence to be sorted, and performing matrix analysis on the radar signal PDW data sequence to obtain a radar signal PDW matrix to be processed; the method comprises the steps of constructing a sorting analysis model, sorting analysis of a radar signal PDW matrix to be processed through the sorting analysis model to obtain a plurality of original radar signal PDW vectors, constructing a batch analysis model, and carrying out signal batch analysis on each original radar signal PDW vector through the batch analysis model to obtain batch analysis results. The invention effectively reduces error sorting and omission, improves sorting accuracy, robustness and sorting processing speed, can adapt to more diversified radar signal characteristics, reduces dependence on manual experience and priori knowledge, and overcomes the problem of misjudgment possibly occurring in a single model.

Inventors

ZHU LIWEI
WEN RONGHUI
Pan Minrong
LU SHUSHU
YAO CONG

Assignees

桂林长海发展有限责任公司

Dates

Publication Date: 20260512
Application Date: 20251223

Claims (10)

1. The radar model sorting method is characterized by comprising the following steps of: importing a radar signal PDW data sequence to be sorted, and performing matrix analysis on the radar signal PDW data sequence to obtain a radar signal PDW matrix to be processed; Constructing a sorting analysis model, and sorting and analyzing the radar signal PDW matrix to be processed through the sorting analysis model to obtain a plurality of original radar signal PDW vectors; Establishing a batch analysis model, and respectively carrying out signal batch analysis on each original radar signal PDW vector through the batch analysis model to obtain batch analysis results corresponding to each original radar signal PDW vector; And constructing a screening analysis model, carrying out screening analysis on all the original radar signal PDW vectors and all the batch analysis results through the screening analysis model, and taking the analysis results as radar model sorting results.
2. The radar model sorting method according to claim 1, wherein the radar signal PDW data sequence includes a first radar signal carrier frequency, a first radar signal pulse width, and a first radar signal arrival time, and the process of performing matrix analysis on the radar signal PDW data sequence to obtain a to-be-processed radar signal PDW matrix includes: Vectorizing the first radar signal carrier frequency, the first radar signal pulse width and the first radar signal arrival time respectively to obtain a first radar signal carrier frequency vector corresponding to the first radar signal carrier frequency, a first radar signal pulse width vector corresponding to the first radar signal pulse width and a first radar signal arrival time vector corresponding to the first radar signal arrival time; Constructing an original radar signal PDW matrix through the first radar signal carrier frequency vector, the first radar signal pulse width vector and the first radar signal arrival time vector; judging whether the number of lines in the original radar signal PDW matrix is a preset number of lines, if so, taking the original radar signal PDW matrix as a radar signal PDW matrix to be processed, and if not, filling the original radar signal PDW matrix according to a preset zero vector, and taking a filling result as the radar signal PDW matrix to be processed.
3. The radar model sorting method according to claim 1, wherein the sorting analysis model includes a first long-short-term memory network and a second long-short-term memory network, and the process of sorting analysis of the radar signal PDW matrix to be processed by the sorting analysis model to obtain a plurality of original radar signal PDW vectors includes: Performing feature capture on the radar signal PDW matrix to be processed through the first long-term and short-term memory network to obtain a radar signal PDW hidden state sequence; And extracting features of the radar signal PDW sequence to be processed through the second long-short-term memory network to obtain a plurality of original radar signal PDW vectors.
4. The radar model sorting method according to claim 1, wherein the batch analysis model includes a third long-short-term memory network and a first full-connection layer, and the process of performing signal batch analysis on each of the original radar signal PDW vectors by the batch analysis model to obtain batch analysis results corresponding to each of the original radar signal PDW vectors includes: combining each original radar signal PDW vector with any one of the remaining original radar signal PDW vectors to obtain a plurality of original radar signal PDW vector pairs corresponding to each original radar signal PDW vector; Extracting a plurality of second radar signal carrier frequencies corresponding to the original radar signal PDW vector pairs and a plurality of second radar signal pulse widths corresponding to the original radar signal PDW vector pairs from the original radar signal PDW vector pairs respectively; Respectively splicing a plurality of second radar signal carrier frequencies corresponding to the original radar signal PDW vector pairs and a plurality of second radar signal pulse widths corresponding to the original radar signal PDW vector pairs to obtain a spliced radar signal PDW matrix corresponding to the original radar signal PDW vector pairs; judging whether the number of lines in the spliced radar signal PDW matrix is a preset number of lines, if so, taking the spliced radar signal PDW matrix as a processed radar signal PDW matrix, if not, filling the spliced radar signal PDW matrix according to a preset zero vector, and taking a filling result as the processed radar signal PDW matrix, thereby obtaining the processed radar signal PDW matrix corresponding to each original radar signal PDW vector pair; respectively extracting features of the processed radar signal PDW matrixes through the third long-term and short-term memory network to obtain first feature extracted radar signal PDW matrixes corresponding to the original radar signal PDW vector pairs; Mapping the radar signal PDW matrixes after the first feature extraction through the first full-connection layer respectively to obtain first radar signal PDW matrixes after mapping corresponding to the original radar signal PDW vector pairs; Extracting first confidence coefficients corresponding to the original radar signal PDW vector pairs from the first mapped radar signal PDW matrixes respectively; And judging whether the first confidence coefficient is larger than or equal to a preset first confidence coefficient threshold value, if so, taking the first radar source batch result as a batch analysis result, and if not, taking the second radar source batch result as a batch analysis result, thereby obtaining batch analysis results corresponding to the PDW vectors of the original radar signals.
5. The radar model sorting method according to claim 1, wherein the screening analysis model includes a plurality of convolution layers and a second full connection layer, and the process of screening all the raw radar signal PDW vectors and all the batch analysis results by the screening analysis model and taking the analysis results as radar model sorting results includes: Extracting a third radar signal carrier frequency corresponding to each original radar signal PDW vector, a third radar signal pulse width corresponding to each original radar signal PDW vector and a second radar signal arrival time corresponding to each original radar signal PDW vector from each original radar signal PDW vector; Constructing a radar signal PDW matrix to be judged corresponding to each original radar signal PDW vector according to a third radar signal carrier frequency corresponding to each original radar signal PDW vector, a third radar signal pulse width corresponding to each original radar signal PDW vector and a second radar signal arrival time corresponding to each original radar signal PDW vector; judging whether the number of lines in the radar signal PDW matrix to be judged is a preset number of lines, if so, taking the radar signal PDW matrix to be judged as a radar signal PDW matrix after judgment, if not, filling the radar signal PDW matrix to be judged according to a preset zero vector, and taking a filling result as a radar signal PDW matrix after judgment, thereby obtaining a radar signal PDW matrix after judgment corresponding to each original radar signal PDW vector; Performing feature extraction on each judged radar signal PDW matrix through a plurality of convolution layers to obtain a second feature extracted radar signal PDW matrix corresponding to each original radar signal PDW vector; Mapping the radar signal PDW matrixes after the second feature extraction through the second full-connection layer respectively to obtain second radar signal PDW matrixes after mapping corresponding to the original radar signal PDW vectors; Extracting second confidence coefficients corresponding to the original radar signal PDW vectors from the second mapped radar signal PDW matrixes respectively; And if the second confidence coefficient is greater than or equal to a preset second confidence coefficient threshold value, taking the batch analysis result as a screening analysis result, so as to obtain a screening analysis result corresponding to each original radar signal PDW vector, and taking all the screening analysis results as radar model sorting results.
6. A radar model sorting apparatus, comprising: the importing module is used for importing a radar signal PDW data sequence to be sorted; The matrix analysis module is used for carrying out matrix analysis on the radar signal PDW data sequence to obtain a radar signal PDW matrix to be processed; the sorting analysis module is used for constructing a sorting analysis model, sorting analysis is carried out on the radar signal PDW matrix to be processed through the sorting analysis model, and a plurality of original radar signal PDW vectors are obtained; The signal batch analysis module is used for constructing a batch analysis model, and carrying out signal batch analysis on each original radar signal PDW vector through the batch analysis model to obtain batch analysis results corresponding to each original radar signal PDW vector; The sorting result obtaining module is used for constructing a screening analysis model, carrying out screening analysis on all the PDW vectors of the original radar signals and all the batch analysis results through the screening analysis model, and taking the analysis results as radar model sorting results.
7. The radar model sorting apparatus of claim 6, wherein the radar signal PDW data sequence includes a first radar signal carrier frequency, a first radar signal pulse width, and a first radar signal arrival time, the matrix analysis module being specifically configured to: Vectorizing the first radar signal carrier frequency, the first radar signal pulse width and the first radar signal arrival time respectively to obtain a first radar signal carrier frequency vector corresponding to the first radar signal carrier frequency, a first radar signal pulse width vector corresponding to the first radar signal pulse width and a first radar signal arrival time vector corresponding to the first radar signal arrival time; Constructing an original radar signal PDW matrix through the first radar signal carrier frequency vector, the first radar signal pulse width vector and the first radar signal arrival time vector; judging whether the number of lines in the original radar signal PDW matrix is a preset number of lines, if so, taking the original radar signal PDW matrix as a radar signal PDW matrix to be processed, and if not, filling the original radar signal PDW matrix according to a preset zero vector, and taking a filling result as the radar signal PDW matrix to be processed.
8. The radar model sorting apparatus of claim 6, wherein the sorting analysis model includes a first long-short-term memory network and a second long-short-term memory network, the sorting analysis module being specifically configured to: Performing feature capture on the radar signal PDW matrix to be processed through the first long-term and short-term memory network to obtain a radar signal PDW hidden state sequence; And extracting features of the radar signal PDW sequence to be processed through the second long-short-term memory network to obtain a plurality of original radar signal PDW vectors.
9. A radar model sorting apparatus comprising a memory, a processor and a computer program stored in the memory and executable on the processor, characterized in that the radar model sorting method according to any one of claims 1 to 5 is implemented when the processor executes the computer program.
10. A computer readable storage medium storing a computer program, characterized in that the radar model sorting method according to any one of claims 1 to 5 is implemented when the computer program is executed by a processor.

Description

Radar model sorting method, device and storage medium Technical Field The invention mainly relates to the technical field of radar sorting, in particular to a radar model sorting method, a radar model sorting device and a storage medium. Background In the fields of electronic warfare and radar reconnaissance, radar signal sorting is a vital link. Because of the large number of different types of radar systems of different functions that may operate simultaneously or alternately in modern battlefield environments, transmitting pulsed signals with similar parameters, such as frequency, pulse width, and Pulse Repetition Interval (PRI), these dense and complex signal environments present a significant challenge to radar signal sorting. Traditional radar signal sorting methods rely mainly on techniques based on rules and template matching, such as time difference of arrival (TDOA) methods, pulse repetition interval analysis (e.g. PRI-fixed, PRI-parametric, PRI-jittered, etc.), clustering algorithms (e.g. K-means, DBSCAN), etc. These methods are able to handle a simple signal environment to some extent, but tend to be poorly effective when faced with: The signal parameters are highly similar, and the signals emitted by different radars can be very close in frequency, pulse width, pulse repetition interval and the like, so that the traditional method is difficult to distinguish. The radar may adopt anti-interference technologies such as frequency hopping, pulse width jitter, PRI jitter and the like, so that signal parameters change rapidly along with time, and the traditional method based on the static template fails. And in a dense signal environment, a large number of signals arrive at the same time, so that the time relation is fuzzy, and TDOA is difficult to calculate accurately or pulse repetition interval analysis is difficult to perform. Disclosure of Invention The invention aims to solve the technical problem of providing a radar model sorting method, a radar model sorting device and a storage medium aiming at the defects of the prior art. The technical scheme for solving the technical problems is as follows, the radar model sorting method comprises the following steps: importing a radar signal PDW data sequence to be sorted, and performing matrix analysis on the radar signal PDW data sequence to obtain a radar signal PDW matrix to be processed; Constructing a sorting analysis model, and sorting and analyzing the radar signal PDW matrix to be processed through the sorting analysis model to obtain a plurality of original radar signal PDW vectors; Establishing a batch analysis model, and respectively carrying out signal batch analysis on each original radar signal PDW vector through the batch analysis model to obtain batch analysis results corresponding to each original radar signal PDW vector; And constructing a screening analysis model, carrying out screening analysis on all the original radar signal PDW vectors and all the batch analysis results through the screening analysis model, and taking the analysis results as radar model sorting results. The other technical scheme for solving the technical problems is as follows, namely a radar model sorting device comprises: the importing module is used for importing a radar signal PDW data sequence to be sorted; The matrix analysis module is used for carrying out matrix analysis on the radar signal PDW data sequence to obtain a radar signal PDW matrix to be processed; the sorting analysis module is used for constructing a sorting analysis model, sorting analysis is carried out on the radar signal PDW matrix to be processed through the sorting analysis model, and a plurality of original radar signal PDW vectors are obtained; The signal batch analysis module is used for constructing a batch analysis model, and carrying out signal batch analysis on each original radar signal PDW vector through the batch analysis model to obtain batch analysis results corresponding to each original radar signal PDW vector; The sorting result obtaining module is used for constructing a screening analysis model, carrying out screening analysis on all the PDW vectors of the original radar signals and all the batch analysis results through the screening analysis model, and taking the analysis results as radar model sorting results. Based on the radar model sorting method, the invention further provides a radar model sorting system. Another technical scheme of the invention for solving the technical problems is that the radar model sorting system comprises a memory, a processor and a computer program which is stored in the memory and can run on the processor, and the radar model sorting method is realized when the processor executes the computer program. Based on the radar model sorting method, the invention further provides a computer readable storage medium. Another technical solution of the present invention to solve the above technical problem is a computer-readable storage medium st