系统工程与电子技术 ›› 2020, Vol. 42 ›› Issue (1): 30-36.doi: 10.3969/j.issn.1001-506X.2020.01.05

• 电子技术 • 上一篇    下一篇

稀疏字典学习海面微弱动目标检测

董自巍1,2(), 孙俊1,2(), 孙晶明1,2(), 潘美艳1,2()   

  1. 1. 南京电子技术研究所, 江苏 南京 210039
    2. 中国电子科技集团公司智能感知技术重点实验室, 江苏 南京 210039
  • 收稿日期:2019-04-28 出版日期:2020-01-01 发布日期:2019-12-23
  • 作者简介:董自巍(1994-),女,硕士研究生,主要研究方向为通信与电子信息工程。E-mail:1551653725@qq.com|孙俊(1974-),男,研究员,博士,主要研究方向为雷达信号处理、目标检测。E-mail:sunjun@ustc.edu|孙晶明(1984-),男,高级工程师,博士,主要研究方向为雷达信号处理。E-mail:sjm@alumni.hust.edu.cn|潘美艳(1993-),女,硕士研究生,主要研究方向为通信与电子信息工程。E-mail:1151079873@qq.com
  • 基金资助:
    “十三五”装备预研领域基金

Marine weak moving target detection based on sparse dictionary learning

Ziwei DONG1,2(), Jun SUN1,2(), Jingming SUN1,2(), Meiyan PAN1,2()   

  1. 1. Nanjing Research Institute of Electronics Technology, Nanjing 210039, China
    2. Key Laboratory of IntelliSense Technology, China Electronics Technology Group Corporation, Nanjing 210039, China
  • Received:2019-04-28 Online:2020-01-01 Published:2019-12-23
  • Supported by:
    “十三五”装备预研领域基金

摘要:

针对强海杂波背景下微弱动目标信号提取困难、雷达检测性能差的问题,在稀疏表示理论的基础上,提出利用字典学习算法抑制海杂波、重构目标信号。该算法通过K类奇异值分解(K-singular value decomposition,K-SVD)算法学习海杂波和目标的稀疏域主成分特征,得到相应的学习字典,抑制海杂波并对目标信号稀疏重建,解决了以往固定字典与高海况下雷达回波匹配度低、目标信号提取效果差的问题;并通过算法参数的分析和优化,进一步提高了算法性能和工程实用性。基于实测数据的实验结果表明,相比传统检测方法,所提算法能够有效检测高海况下微弱动目标,显著提升检测性能。

关键词: 稀疏字典学习, 海杂波抑制, 信号重构, 微弱动目标检测

Abstract:

Aiming at the difficulty of weak moving target signals extraction and the poor radar detection performance under strong sea clutter background, a dictionary learning algorithm is proposed based on the sparse representation theory to suppress sea clutter and reconstruct the target signals. The K-singular value decomposition (K-SVD) algorithm is used in this method to learn the characteristics of principal components of sea clutter and targets in sparse domain, obtaining corresponding learning dictionaries. Based on the dictionaries, the sea clutter is first suppressed and then the target signals are reconstructed, overcoming the disadvantage of signals' low-matching degree and poor effects in signal extraction under fixed dictionaries. And then the algorithm parameters are analyzed and optimized to further promote the performance and engineering practicability. The experimental results based on the measured data show that compared with the traditional detection methods, the proposed algorithm can effectively detect the weak moving target under high sea conditions and significantly improve the detection performance.

Key words: sparse dictionary learning, sea clutter suppression, signal reconstruction, weak moving target detection

中图分类号: