Identification method of radar emitter signal based on one-dimensional convolutional neural network

Abstract

The invention belongs to the technical field of radiation source signal identification, and in particular relates to a radar radiation source signal identification method based on a one-dimensional convolutional neural network. The steps are as follows: (1) generating a radar radiation source signal data set; (2) data pre-processing Processing; (3) constructing convolutional neural network; (4) setting parameters and training convolutional neural network; (5) predicting classification; (6) calculating accuracy; (7) outputting results. The recognition method of the radar radiation source signal based on the one-dimensional convolutional neural network disclosed by the present invention has the following advantages: 1. The feature extraction of the signal is carried out through the network structure in the convolutional neural network, which avoids the need for manual design of features in traditional algorithms. Process; 2. It can correctly identify the intra-pulse modulation methods of various radar emitter signals when the signal-to-noise ratio is as low as -10dB; 3. The implementation is simple.

Description

technical field [0001] The invention belongs to the technical field of radiation source signal identification, and in particular relates to a radar radiation source signal identification method based on a one-dimensional convolutional neural network. Background technique [0002] Radar emitter signal identification is an important link in electronic countermeasures, and plays a key role in electronic intelligence reconnaissance, electronic support reconnaissance and threat warning systems. [0003] With the rapid development of electronic information technology, the confrontation in the modern electronic battlefield is becoming more and more fierce, and the new complex system radar is gradually occupying a dominant position. The electromagnetic environment is increasingly complex and dense, and the traditional pulse descriptors (carrier frequency, pulse arrival time, pulse arrival angle, pulse amplitude, pulse width) are no longer suitable for radar radiation source signals ...

AI Technical Summary

Problems solved by technology

However, these methods have many shortcomings. On the one hand, many methods have a general effect on the recognition of signals under low signal-to-noise conditions. On the other hand, these methods often spend a lot of time on signal feature extraction, and some features extracted are not It is not universal, and if you want to use combined features, you may have to face the disaster of dimensionality or the difficulty of feature selection

Images