Intra-pulse modulation recognition method based on time frequency image features

Abstract

The invention belongs to the field of radar radiation source signal recognition technology and particularly relates to an intra-pulse modulation recognition method based on time frequency image features. The method comprises the steps that (1) radar signals are subjected to Wigner-Vill distribution time frequency transformation to obtain a time frequency distribution image reflecting signal energy along with time and frequency; (2) the time frequency image obtained in the step (1) is subjected to image preprocessing; (3) the image obtained after preprocessing is subjected to interpolation to scatter an original track; (4) the track scattered into single points in the step (3) is analogized into a target location in target tracking, and joint probabilistic data association is introduced to recognize tracks of different radar signals; and (5) a central moment is utilized to extract shape features, etc. of the time frequency image. According to the method, the radar signals are converted into the time frequency image by means of time frequency distribution transformation, and modulation modes of the radar signals which arrive at the same moment under the condition that frequencies are overlapped are recognized respectively through digital image processing.

Description

technical field [0001] The invention belongs to the technical field of radar radiation source signal identification, and in particular relates to an intrapulse modulation identification method based on time-frequency image features. Background technique [0002] Radar emitter signal recognition is a key process in electronic intelligence reconnaissance, electronic support reconnaissance and threat warning systems, and its recognition level is directly related to the technical advancement of radar countermeasure equipment. With the fierce confrontation of modern electronic warfare, new complex system radars are continuously put into use and gradually occupy a dominant position. The signal density of the electromagnetic threat environment is as high as 1.2 million pulses per second, and the radar operating frequency coverage has reached 0.1-20GHz, and is being Extending to 0.05-140GHz, the radar signal waveform changes simultaneously in multiple domains such as time and freque...

AI Technical Summary

Problems solved by technology

These methods are mainly aimed at a few types of signals, and the recognition effect of other signals needs to be further studied. At the same time, these methods only consider the situation of only a single signal at the same time, and do not discuss the overlapping of multiple signals at the same time.

Images