电子学报 ›› 2021, Vol. 49 ›› Issue (9): 1840-1851.DOI: 10.12263/DZXB.20200492

• 综述评论 • 上一篇    下一篇

水声信号处理中的稀疏表示理论及应用

冯西安1, 寇思玮1, 谭伟杰2, 毕杨1   

  1. 1.西北工业大学航海学院,陕西 西安 710072
    2.贵州大学计算机科学与技术学院,贵州 贵阳 550025
  • 收稿日期:2020-05-22 修回日期:2021-03-17 出版日期:2021-09-25 发布日期:2021-09-25
  • 作者简介:冯西安 男,1962年6月出生,陕西户县人,教授,博士生导师,中国电子学会高级会员.研究方向为水声信号处理,阵列信号处理,目标跟踪与信息融合. E-mail:fengxa@nwpu.edu.cn
    寇思玮 女,1989年1月出生,陕西西安人,博士研究生. 研究方向为稀疏信号处理,水下目标识别与声成像.
    谭伟杰 男,1981年8月出生, 陕西合阳人,副教授. 2019年获西北工业大学信息与通信工程学科博士学位. 现在贵州大学计算机科学与技术学院工作. 研究方向为稀疏信号处理、阵列信号处理和通信网络安全.
    毕 杨 女,1981年5月出生,陕西汉中人. 2015年获西北工业大学水声工程学科博士学位. 现为西北工业大学信息与通信工程学科在站博士后,研究方向为阵列信号处理,水声信号处理和稀疏信号处理.
  • 基金资助:
    国家自然科学基金(61671378);国家自然科学基金青年基金(12004293)

Sparse Representation Theory and Application in Underwater Acoustic Signal Processing

Xi-an FENG1, Si-wei KOU1, Wei-jie TAN2, Yang BI1   

  1. 1.School of Marine Science and Technology,Northwestern Polytechnical University,Xi’an,Shaanxi 710072,China
    2.College of Computer Science and Technology,Guizhou University,Guiyang,Guizhou 550025,China
  • Received:2020-05-22 Revised:2021-03-17 Online:2021-09-25 Published:2021-09-25

摘要:

稀疏表示研究信号简洁表示与重构的本质问题,能够更好地揭示、分辨和提取信号中所蕴含的信息特征,在水声信号处理的许多应用方面都显示了巨大的优势和潜力.本文综述了水声信号处理中的稀疏表示理论及有关应用问题.首先介绍了稀疏表示模型和典型的稀疏分解算法;然后,研究了自适应过完备字典设计、离网格处理等稀疏表示的关键问题;接着,探索了稀疏表示理论在水下信号处理中的一些重要应用,包括高分辨波达方向(Direction Of Arrival, DOA)估计、水下体目标微多普勒特征提取、运动目标角度-多普勒声成像、水声信号压缩感知与重构;最后,指出稀疏表示理论在水声信号处理中的发展趋势.进行了必要的计算机仿真,提取了水下目标时、频、空域多维度信息特征,并实现了两类典型通信信号的有效压缩和精确重构.

关键词: 水声信号, 稀疏表示, 特征提取, 压缩感知

Abstract:

Sparse representation is a theory to study the essential problem of signal concise representation and precise recovery. It can better reveal, distinguish and extract the characteristic information contained in underwater acoustic signals,so that it has great advantages and potential in many applications of underwater acoustic signal processing. In this paper, the sparse representation theory and its application in underwater acoustic signal processing are reviewed. Firstly, the sparse representation model and typical sparse decomposition algorithms are introduced. Then, the key problems of sparse representation, such as adaptive over-complete dictionary design and off-grid processing and so on, are studied. Thirdly, some important applications of sparse representation theory in underwater signal processing are explored, which include high-resolution DOA estimation, micro-Doppler feature extraction of underwater target, angle-Doppler acoustic imaging of moving target, compressed sensing and reconstruction of underwater acoustic signals. Finally, the development trend of sparse representation theory in underwater acoustic signal processing is pointed out. Some necessary computer simulations are carried out to extract the multi-dimensional information features of underwater target in time, frequency and spatial domain are successfully extracted, and two kinds of typical communication signals are effectively compressed and accurately reconstructed.

Key words: underwater acoustic signal, sparse representation, feature extraction, compressed sensing

中图分类号: