Scholarlink

<p>When applying compressed sensing orthogonal matching pursuit algorithm (CS-OMP) to localize multiple sound sources, the number of sound sources must be given in advance, although it is often unknown. To address this issue, firstly, source number estimation methods were investigated, and the performance of the estimation methods under varying number of snapshots was compared by numerical simulations and experiments. Secondly, the revised Akaike Information Criterion (RAIC) sound source estimation algorithm was selected to improve the orthogonal matching pursuit algorithm (OMP), which obtained the orthogonal matching pursuit based on sound source estimation algorithm (SNE-OMP). Finally, the proposed SNE-OMP algorithm was tested and compared with Conventional Beamforming (CBF) algorithm to verify the feasibility and advantages in multiple sound source localization, thereby addressing the problem of sound source number determination in the compressed sensing source localization algorithm.</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

Research on Sound Source Number Estimation Method Under Compressed Sensing Sound Source Localization

  • 期刊论文

  • 发表日期：

    2025年7月5日

查看全文
  • Xiaohui Liu,
  • Qiannian Cao,
  • Shuai Wang,
  • Lei Guo,
  • Leilei Cao

摘要

When applying compressed sensing orthogonal matching pursuit algorithm (CS-OMP) to localize multiple sound sources, the number of sound sources must be given in advance, although it is often unknown. To address this issue, firstly, source number estimation methods were investigated, and the performance of the estimation methods under varying number of snapshots was compared by numerical simulations and experiments. Secondly, the revised Akaike Information Criterion (RAIC) sound source estimation algorithm was selected to improve the orthogonal matching pursuit algorithm (OMP), which obtained the orthogonal matching pursuit based on sound source estimation algorithm (SNE-OMP). Finally, the proposed SNE-OMP algorithm was tested and compared with Conventional Beamforming (CBF) algorithm to verify the feasibility and advantages in multiple sound source localization, thereby addressing the problem of sound source number determination in the compressed sensing source localization algorithm.