现代防御技术 ›› 2025, Vol. 53 ›› Issue (2): 91-98.DOI: 10.3969/j.issn.1009-086x.2025.02.010

• 军事智能 • 上一篇    下一篇

基于改进轻量化神经网络的干扰识别方法

付亦凡, 阮航, 周东平, 穆贺强   

  1. 北京无线电测量研究所,北京 100854
  • 收稿日期:2023-12-29 修回日期:2024-05-08 出版日期:2025-04-28 发布日期:2025-04-30
  • 作者简介:付亦凡(1999-),男,内蒙古东胜人。硕士生,研究方向为电子对抗技术。

Radar Interference Recognition Based on Improved Lightweight Convolutional Neural Networks

Yifan FU, Hang RUAN, Dongping ZHOU, Heqiang MU   

  1. Beijing Institute of Radio Measurement,Beijing 100854,China
  • Received:2023-12-29 Revised:2024-05-08 Online:2025-04-28 Published:2025-04-30

摘要:

针对战场实战电磁对抗作战中,大量雷达干扰信号可以被简单迅速地生成,使用传统卷积神经网络对雷达干扰进行识别存在规模大,难以在小型化装备上搭载的问题。提出一种改进的轻量化卷积神经网络,通过在传统神经网络中使用动态卷积核尺寸技术并添加批量归一化层技术,提高网络的识别效能。通过提取干扰信号时频特征,构建训练集与测试集对网络进行训练。仿真实验表明,该网络对6种干扰信号在-8 dB干噪比条件下识别准确率达到96 %以上,对比其他网络具有更好的识别准确效能。

关键词: 雷达有源干扰, 卷积神经网络, 轻量化, 动态卷积核, 特征提取

Abstract:

Considering a large number of radar interference signals could be built within a short period on the battlefield. Traditional convolutional neural networks face issues due to their large scale. It’s a challenge to deploy jamming recognition system on small-scale equipment. This paper proposes an improved lightweight convolutional neural network to solve the problem by adopting adaptive kernel and batch normalization technology to improve recognition efficiency. By extracting time-frequency character to construct training and testing database for neural network training. Experiment shows that the network achieves over 96% identification accuracy for six kinds of interference signals under -8dB JNR. Compared with other networks, it has a superior accuracy efficiency ratio.

Key words: radar active jamming, convolutional neural network, lightweight, adaptive kernel, feature extraction

中图分类号: