系统工程与电子技术 ›› 2022, Vol. 44 ›› Issue (7): 2319-2328.doi: 10.12305/j.issn.1001-506X.2022.07.29

• 通信与网络 • 上一篇    下一篇

基于Deep SVDD的通信信号异常检测方法

康颖1,2, 赵治华1,2, 吴灏1,2,*, 李亚星1,2, 孟进1,2   

  1. 1. 海军工程大学军用电气科学与技术研究所, 湖北 武汉 430033
    2. 海军工程大学舰船综合电力技术国防科技重点实验室, 湖北 武汉 430033
  • 收稿日期:2021-05-21 出版日期:2022-06-22 发布日期:2022-06-28
  • 通讯作者: 吴灏
  • 作者简介:康颖 (1994—), 男, 博士研究生, 主要研究方向为通信信号处理|赵治华 (1962—), 男, 教授, 博士, 主要研究方向为电磁兼容、电磁领域的分析与计算、自适应干扰对消|吴灏 (1988—), 男, 讲师, 博士, 主要研究方向为通信信号处理、波形设计|李亚星 (1988—), 男, 讲师, 博士, 主要研究方向为通信信号处理|孟进 (1977—), 男, 研究员, 博士, 主要研究方向为电磁干扰与防护
  • 基金资助:
    国家杰出青年科学基金(52025072);国家自然科学基金(61801502);国家自然科学基金(61801501);国家自然科学基金(62001497);国防科技重点实验室稳定支持项目(6142217200105);湖北省自然科学基金(ZRMS202001331);海军工程大学发展基金(425317S126)

Deep SVDD-based anomaly detection method for communication signals

Ying KANG1,2, Zhihua ZHAO1,2, Hao WU1,2,*, Yaxing LI1,2, Jin MENG1,2   

  1. 1. Institute of Military Electrical Science and Technology, Naval University of Engineering, Wuhan 430033, China
    2. National Key Laboratory of Science and Technology on Vessel Integrated Power System, Naval University of Engineering, Wuhan 430033, China
  • Received:2021-05-21 Online:2022-06-22 Published:2022-06-28
  • Contact: Hao WU

摘要:

针对复杂电子对抗场景中的非理想信道环境, 该文提出了一种基于深度学习的异常检测(anomaly detection, AD)方法。首先, 分析了利用时频同相/正交(in-phase/quadrature, I/Q)采样数据进行AD的可行性; 然后, 设计了深度学习网络架构, 并提出基于深度支持向量描述(deep support vector data description, Deep SVDD)和调制识别的AD方法。仿真及实验结果表明: 相比于经典的单分类检测算法, 该方法检测性能和实时性明显提升, 且在非理想信道环境下表现鲁棒。该方法已在某型号项目原理样机上得到验证, 具有很高应用价值。

关键词: 异常检测, Deep SVDD, 调制识别, 干扰预警

Abstract:

To solve the problem of anomaly detection (AD) of the non-ideal channel in complex electronic countermeasures, a deep learning based method is presented. First, the feasibility of using time-frequency in-phase/quadrature (I/Q) sampling data for anomaly detection (AD) is analyzed. Then, a deep learning network architecture is designed and an AD method based on deep support vector data description (Deep SVDD) and modulation classification is proposed. Simulation and experimental results show that the detection performance and real-time performance of the method are significantly improved compared with the classical algorithms of one-class classification, and the performance is robust in non-ideal channel environment. The method is validated on a sample machine and is of high application value.

Key words: anomaly detection (AD), deep support vector data description (Deep SVDD), modulation classification, interference warning

中图分类号: