Scholarlink

<p>Normal mode extraction has attracted extensive attention over the past few decades due to its practical value in enhancing the performance of underwater acoustic signal processing. Singular value decomposition (SVD) is an effective method to extract modal depth functions using vertical line arrays (VLA), particularly in scenarios when no prior environment information is available. However, the SVD method requires rigorous orthogonality conditions, and its performance severely degenerates in the presence of mode degeneracy. Consequently, the SVD approach is often not feasible in practical scenarios. This paper proposes a full rank decomposition (FRD) method to address these issues. Compared to the SVD method, the FRD method has three distinct advantages: 1) the conditions that the FRD method requires are much easier to be fulfilled in practical scenarios; 2) both modal depth functions and wavenumbers can be simultaneously extracted <i>via</i> the FRD method; 3) the FRD method is not affected by the phenomenon of mode degeneracy. Numerical simulations are conducted in two types of waveguides to verify the FRD method. The impacts of environment configurations and noise levels on the precision of the extracted modal depth functions and wavenumbers are also investigated through simulation.</p>

错误：搜索内容不能为空，请输入英文关键词
错误：关键词超出字数限制，请精简
高级检索

Extraction of Modal Depth Functions and Wavenumbers Using Full Rank Decomposition Method with a Vertical Line Array

  • 期刊论文

  • 发表日期：

    2025年5月8日

查看全文
  • Shuang Zhang,
  • Yinquan Zhang,
  • Jinxing Qin,
  • Zhenglin Li,
  • Yonggang Guo,
  • Shuanglin Wu

摘要

Normal mode extraction has attracted extensive attention over the past few decades due to its practical value in enhancing the performance of underwater acoustic signal processing. Singular value decomposition (SVD) is an effective method to extract modal depth functions using vertical line arrays (VLA), particularly in scenarios when no prior environment information is available. However, the SVD method requires rigorous orthogonality conditions, and its performance severely degenerates in the presence of mode degeneracy. Consequently, the SVD approach is often not feasible in practical scenarios. This paper proposes a full rank decomposition (FRD) method to address these issues. Compared to the SVD method, the FRD method has three distinct advantages: 1) the conditions that the FRD method requires are much easier to be fulfilled in practical scenarios; 2) both modal depth functions and wavenumbers can be simultaneously extracted via the FRD method; 3) the FRD method is not affected by the phenomenon of mode degeneracy. Numerical simulations are conducted in two types of waveguides to verify the FRD method. The impacts of environment configurations and noise levels on the precision of the extracted modal depth functions and wavenumbers are also investigated through simulation.