基于复值卷积神经网络的微震干扰识别。

Recognition of Micro-Motion Jamming Based on Complex-Valued Convolutional Neural Network.

机构信息

Information and Navigation School, Air Force Engineering University, Xi'an 710077, China.

Equipment Management and Unmanned Aerial Vehicle Engineering School, Air Force Engineering University, Xi'an 710051, China.

出版信息

Sensors (Basel). 2023 Jan 18;23(3):1118. doi: 10.3390/s23031118.

DOI:10.3390/s23031118

PMID:36772157

原文链接:https://pmc.ncbi.nlm.nih.gov/articles/PMC9919671/

Abstract

Micro-motion jamming is a new jamming method to inverse synthetic aperture radar (ISAR) in recent years. Compared with traditional jamming methods, it is more flexible and controllable, and is a great threat to ISAR. The prerequisite of taking relevant anti-jamming measures is to recognize the patterns of micro-motion jamming. In this paper, a method of micro-motion jamming pattern recognition based on complex-valued convolutional neural network (CV-CNN) is proposed. The micro-motion jamming echo signals are serialized and input to the network, and the result of recognition is output. Compared with real-valued convolutional neural network (RV-CNN), it can be found that the proposed method has a higher recognition accuracy rate. Additionally, the recognition accuracy rate is analyzed with different signal-to-noise ratio (SNR) and number of training samples. Simulation results prove the effectiveness of the proposed recognition method.

摘要

微多普勒干扰是近年来针对逆合成孔径雷达（ISAR）的一种新的干扰方法。与传统的干扰方法相比，它更加灵活可控，对 ISAR 构成了极大的威胁。采取相关抗干扰措施的前提是识别微多普勒干扰的模式。本文提出了一种基于复值卷积神经网络（CV-CNN）的微多普勒干扰模式识别方法。将微多普勒干扰回波信号进行序列化并输入到网络中，输出识别结果。与实值卷积神经网络（RV-CNN）相比，发现所提出的方法具有更高的识别准确率。此外，还分析了不同信噪比（SNR）和训练样本数量对识别准确率的影响。仿真结果证明了所提出的识别方法的有效性。