Scholarlink

<p>To improve the accuracy of sparse system identification with impulsive measurement interference, a scaled least mean square (LMS) algorithm with block sparsity constraint is proposed. The proposed method adopts a scalar for scaling and filtering the impulsive interference. The proposed scalar avoids drastic changes in channel estimation due to the wrong information from the abrupt impulsive interference. On the other hand, by exploiting the sparsity of the system, the proposed block sparsity constraint searches sparse solution by using an approximated block <InlineEquation ID="IEq1"> <EquationSource Format="TEX">\(\ell _0\)</EquationSource> </InlineEquation> norm constraint. Additionally, the step-size of the proposed method is gradually adjusted based on the gradient factors combining block sparse constraint and impulsive interference. Simulations, including sparse system identification and tracking, are conducted to confirm the effectiveness of the proposed algorithm for sparse system identification with impulsive interference. Moreover, this paper includes a comparison and analysis of underwater acoustic channel estimation results, demonstrating the superiority and effectiveness of the proposed 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

Adaptive Block Sparse System Identification in an Impulsive Interference Environment

  • 期刊论文

  • 发表日期：

    2026年1月12日

查看全文
  • Fei-Yun Wu,
  • Guanquan Dai,
  • Dan Song

摘要

To improve the accuracy of sparse system identification with impulsive measurement interference, a scaled least mean square (LMS) algorithm with block sparsity constraint is proposed. The proposed method adopts a scalar for scaling and filtering the impulsive interference. The proposed scalar avoids drastic changes in channel estimation due to the wrong information from the abrupt impulsive interference. On the other hand, by exploiting the sparsity of the system, the proposed block sparsity constraint searches sparse solution by using an approximated block \(\ell _0\) norm constraint. Additionally, the step-size of the proposed method is gradually adjusted based on the gradient factors combining block sparse constraint and impulsive interference. Simulations, including sparse system identification and tracking, are conducted to confirm the effectiveness of the proposed algorithm for sparse system identification with impulsive interference. Moreover, this paper includes a comparison and analysis of underwater acoustic channel estimation results, demonstrating the superiority and effectiveness of the proposed algorithm.