CN-116359867-B - Signal sorting method, device, storage medium and equipment based on feature tree clustering

CN116359867BCN 116359867 BCN116359867 BCN 116359867BCN-116359867-B

Abstract

The invention relates to the field of radar signal sorting, and particularly discloses a signal sorting method, a device, a storage medium and equipment based on feature tree clustering, wherein the signal sorting method based on feature tree clustering comprises the steps of converting standard data of each radar pulse signal into corresponding feature classes after processing target feature parameters of each radar pulse signal into standard data, wherein the feature classes are nodes of a feature tree; and when the target feature class is determined on the feature tree based on the radius of the target feature class and/or the target sample threshold value, determining a signal sorting result according to the target feature class. The method solves the technical problems that the traditional radar signal sorting method is poor in accuracy and cannot cope with complex electromagnetic environments.

Inventors

HE XIANGCHEN
LI XIN
XIE LI

Assignees

北京遥感设备研究所

Dates

Publication Date: 20260508
Application Date: 20221230

Claims (7)

1. The signal sorting method based on the feature tree clustering is characterized by comprising the following steps of: after processing the target characteristic parameters of each radar pulse signal into standard data, converting the standard data of each radar pulse signal into corresponding characteristic classes, wherein the characteristic classes are nodes of a characteristic tree; Performing layer-by-layer clustering on the current feature class according to the feature class merging condition until all the feature classes are merged into the same feature class to obtain the feature tree, wherein the top layer of the feature tree is the same feature class; determining a signal sorting result according to the target feature class when determining the target feature class on the feature tree based on the target feature class radius and/or the target sample threshold; The method comprises the steps of determining a target feature class, updating a feature tree based on a received new radar pulse signal and a feature class exceeding a time window, sequentially deleting the expiration feature class from three or more upper classes including the feature class, connecting the feature class closest to the expiration feature class with a three-level class corresponding to a two-level class of the expiration feature class when the distance between the candidate feature class and the new feature class is smaller than the radius of the target feature class, replacing the candidate feature class with the new feature class, merging upper classes of the candidate feature class to update the feature tree, searching the expiration feature class exceeding the time window based on the input time of the corresponding radar pulse signal, and sequentially deleting the expiration feature class from the three or more upper classes including the feature class, and deleting the expiration feature class and the two-level class of the expiration feature class when the distance between the candidate feature class and the new feature class is smaller than the radius of the target feature class.
2. The method of claim 1, further comprising, prior to processing the target characteristic parameter of each radar pulse signal into standard data: converting the received radar pulse signals into pulse description words; Determining the information quantity of each feature in the pulse description by using information entropy; and determining target feature parameters from the features according to the information quantity of the features.
3. The method according to claim 1, wherein the step of layer-by-layer clustering the current feature class according to the merging condition of the feature class includes: calculating the class distance of every two feature classes in all feature classes, and determining two feature classes closest to the class distance; combining the two nearest feature classes to obtain a two-layer class; and replacing the two feature classes by the two-layer class, and continuing to perform merging and clustering according to class distances.
4. The method of claim 1, further comprising, after deriving the feature tree: and continuing to search for radar pulse signals if the target feature class is not determined on the feature tree based on the target feature class radius and/or the target sample threshold.
5. A signal sorting apparatus based on feature tree clustering, applying the method of any one of claims 1 to 4, comprising: the conversion unit is used for converting the standard data of each radar pulse signal into corresponding feature classes after processing the target feature parameters of each radar pulse signal into standard data, wherein the feature classes are nodes of a feature tree; A clustering unit, configured to perform layer-by-layer clustering on the current feature class according to the merging condition of the feature classes until all feature classes are merged into the same feature class, so as to obtain the feature tree, where a top layer of the feature tree is the same feature class; And the determining unit is used for determining a signal sorting result according to the target feature class when the target feature class is determined on the feature tree based on the target feature class radius and/or the target sample threshold value.
6. A computer readable storage medium, characterized in that the computer readable storage medium comprises a stored program, wherein the program when run performs the method of any one of claims 1 to 4.
7. An electronic device comprising a memory and a processor, characterized in that the memory has stored therein a computer program, the processor being arranged to execute the method according to any of the claims 1-4 by means of the computer program.

Description

Signal sorting method, device, storage medium and equipment based on feature tree clustering Technical Field The invention relates to the field of radar signal sorting, in particular to a signal sorting method, device, storage medium and equipment based on feature tree clustering. Background In order to acquire useful radar signals, various electromagnetic mixed pulse signals received into the system must be classified according to the corresponding radiation sources so as to be capable of subsequent signal processing. With the wide application and rapid development of electronic equipment, the space electromagnetic environment is more complex and variable, and various dense electromagnetic signals become electronic interference signals for detecting radar, so that the signal sorting efficiency of a radiation source is affected. The working frequency band is continuously widened, signal parameters are overlapped in multiple dimensions, and various unconventional radar radiation source signals have new technologies such as frequency agility, pulse compression and the like, so that the characteristic parameter variation amplitude is large, and the accuracy of the traditional radiation source signal sorting method is difficult to guarantee. The radar radiation source signal sorting is a technology for separating each radar pulse sequence under the condition that a plurality of radar pulses are mutually staggered, and estimating and identifying parameters of each radar. The radar signal sorting can obtain a plurality of real radar pulse sequences, so that the processing such as interference, positioning or tracking can be further performed. Radar signal sorting is a basis for radar detection signal processing, the importance of which is self-evident. However, with the continuous development of radar technology, the rapidly increased pulse current density, wider and wider frequency spectrum, complex and changeable modulation modes and continuously-changed radar parameters all make the electromagnetic environment which needs to be dealt with by radar signal sorting become worse, the sorting difficulty is also increased, and the traditional radar signal sorting method has poor accuracy and cannot cope with the complex electromagnetic environment. In view of the above problems, no effective solution has been proposed at present. Disclosure of Invention The embodiment of the invention provides a signal sorting method, a device, a storage medium and equipment based on feature tree clustering, which at least solve the technical problems that the traditional radar signal sorting method is poor in accuracy and cannot cope with complex electromagnetic environments. According to one aspect of the embodiment of the invention, a signal sorting method based on feature tree clustering is provided, which comprises the following steps of. After processing target feature parameters of each radar pulse signal into standard data, converting the standard data of each radar pulse signal into corresponding feature classes, wherein the feature classes are nodes of a feature tree, clustering the current feature classes layer by layer according to the combination condition of the feature classes until all the feature classes are combined into the same feature class, obtaining the feature tree, wherein the top layer of the feature tree is the same feature class, and determining a signal sorting result according to the target feature class when determining the target feature class on the feature tree based on the radius of the target feature class and/or a target sample threshold value. Preferably, before processing the target characteristic parameter of each radar pulse signal into standard data, the method further comprises the steps of converting the received radar pulse signal into a pulse description word, determining information quantity of each characteristic in the pulse description by using information entropy, and determining the target characteristic parameter from each characteristic according to the information quantity of each characteristic. Preferably, the step of performing layer-by-layer clustering on the current feature class according to the feature class merging condition comprises the steps of calculating the class distance of every two feature classes in all feature classes, determining two feature classes closest to each other, merging the two feature classes closest to each other to obtain two-layer classes, and using the two-layer classes to replace the two feature classes to continue merging and clustering according to the class distance. Preferably, after the feature tree is obtained, the method further comprises continuing to search for radar pulse signals if the target feature class is not determined on the feature tree based on the target feature class radius and/or the target sample threshold. Preferably, after continuing to search for radar pulse signals or after determining target feature classes, the met