电子学报 ›› 2019, Vol. 47 ›› Issue (1): 1-8.DOI: 10.3969/j.issn.0372-2112.2019.01.001

• 学术论文 •    下一篇

基于半监督矩形网络的通信电台个体识别

黄健航, 雷迎科   

  1. 国防科技大学电子对抗学院, 安徽合肥 230037
  • 收稿日期:2017-07-03 修回日期:2018-02-16 出版日期:2019-01-25
    • 作者简介:
    • 黄健航 男,1994年2月出生于广东深圳.国防科技大学电子对抗学院硕士研究生,研究方向为通信信号处理;雷迎科 男,1975年2月出生于安徽安庆.国防科技大学电子对抗学院指挥与控制系副教授,研究方向为通信信号处理、模式识别.
    • 基金资助:
    • 国家自然科学基金 (No.61272333); 国防科技重点实验室基金 (No.9140C130502140C13068); 总装预研项目基金 (No.9140A33030114JB39470)

Communication Radio Individual Recognition Based on Semi-Supervised Rectangular Network

HUANG Jian-hang, LEI Ying-ke   

  1. Electronic Countermeasures Institution of National University of Defense Technology, Hefei, Anhui 230037, China
  • Received:2017-07-03 Revised:2018-02-16 Online:2019-01-25 Published:2019-01-25
    • Supported by:
    • National Natural Science Foundation of China (No.61272333); Key Laboratory of national defense science and Technology Foundation (No.9140C130502140C13068); Pre-research Program Fund of the General Armaments Department (No.9140A33030114JB39470)

摘要: 通信电台信号的小样本条件造成了电台个体识别准确性欠佳的问题,本文首次提出基于半监督矩形网络进行通信电台个体识别,克服小样本条件对电台个体识别效果的影响.首先提取电台信号的矩形积分双谱特征,人为注入噪声构成污染样本,在半监督矩形网络编码器中有监督训练,其训练结果通过网络径向连接传给解码器,解码器再无监督学习,重构未污染的原始样本,从网络顶层提取电台个体特征,输入softmax分类器实现分类识别.在实际采集的电台数据集上的实验结果说明,本算法在小样本条件下相比现有算法能更准确识别同型号的电台个体.

关键词: 小样本条件, 电台个体识别, 半监督学习, 矩形积分双谱, 自编码器

Abstract: Small sample condition of communication radio signal caused poor individual recognition on radios.To solve this problem,a method about communication radio individual recognition based on semi-supervised rectangular network was proposed innovatively.Firstly,the square integral bispectrum feature was extracted from radio signal and then was corrupted by Gaussian noise.The corrupted sample was passed to the encoder of semi-supervised rectangular network for supervised training.The trained parameterization was then mirrored to decoder through the lateral connection across the model.And the output was forced by decoder through unsupervised learning to be close to the clean input.Then the essential feature extracted was referred as the individual feature of radio signals.Individual recognition was finally accomplished by a softmax classifier.The experiment results on several radio datasets collected in actual environment indicated that the method had superior performance on identifying radio individuals with the same types under small sample condition.

Key words: small sample condition, radio individual recognition, semi-supervised learning, square integral bispectra, auto-encoder

中图分类号: