Complex signal sorting method based on visual graphic features

Abstract

The invention relates to a complex signal sorting method based on visual graphic features. Directed at the current situation that complex system radar signals with carrier frequency and repetition frequency changing rapidly are difficult to sort, priori knowledge is provided for signal sorting based on recognition signal parameters of graphic visualization features. And according to the preprocessed multi-dimensional visual features of the radiation source description word (PDW) information in the time domain and the frequency domain, Hough transformation is utilized to quickly and automatically extract graphic parameters, and multi-dimensional feature parameter association processing in the time domain is adopted to realize complex signal sorting of specified parameters. According to themethod, complex signal sorting based on visual graphic feature information parameters overcomes the difficulty that frequency domains in conventional sorting are difficult to cluster, and hidden features, categories, relations and trends can be effectively found from graphic visualization feature information. The method has good universality and practicability.

Description

technical field [0001] The invention belongs to the field of radar signal sorting and recognition. Background technique [0002] With the rapid development of radar technology, the use of conventional pulse radar signals in the radar signal environment has become less and less, and the sorting and identification of complex radar signals has always been a key technology and problem in electronic countermeasures. Complex signal features are mainly extracted from multidimensional features such as time domain, frequency domain, intrapulse features, and airspace. The intrapulse modulation features of radar signals include linear frequency modulation, nonlinear frequency modulation, two-phase encoding, four-phase encoding, and frequency encoding. The inter-modulation features include frequency-domain modulation and time-domain modulation. The frequency-domain modulation features include pulse-to-pulse frequency agility, pulse group frequency agility, pulse-to-pulse frequency modul...

AI Technical Summary

Problems solved by technology

The characteristics of the modern electronic countermeasure signal environment are high signal density, complex and changeable waveforms, wide operating frequency bands and partial overlaps, dense signals in the time domain and increasingly serious overlaps, and the signals arriving at the input of the radar reconnaissance system are random pulse streams , conventional radar signal sorting leads to an increase in false targets, and complex signals are difficult to sort

[0003] At present, the methods and technical means of key researches on complex signal sorting and identification can be summarized into six categories: template matching method, PRI sorting, multi-parameter correlation comparison and multi-parameter sorting, cluster sorting, artificial intelligence and neural network method, signal sorting based on intrapulse features, etc., the radar emitter signal has a lot of noise in the process of propagation and reception, and the SNR changes greatly. In the case of requiring prior data for training, this is very important in many electronic reconnaissance situations. Difficult to satisfy, also greatly increased the difficulty of sorting

Images