Scholarlink

<p>This study suggests a novel Logarithmic Hyperbolic Sine Squared (LHSS) cost function and the development of its associated LHSS algorithm for effectiveness in environments affected by impulsive noise. The LHSS cost function is highly resilient to outliers, as its logarithmic growth and saturated gradient effectively restrict their impact. To exploit sparse systems, an improved proportionate term is inserted in the LHSS algorithm, which capitalizes on the sparse nature to speed up adaptation process, while the LHSS cost function is found to be resilient in an environment with impulsive noise. Therefore, the IPLHSS algorithm turns out to be effective and has a faster rate of convergence. Additionally, a filtered-x LHSS (Fx-LHSS) algorithm is suggested to attain room equalization. Simulations were conducted for the sparse system identification and room equalization scenario to demonstrate that the suggested approach performs better and is more robust in an environment with impulsive noise.</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

A Novel Logarithmic Hyperbolic Sine Squared Adaptive Filter for Sparse Systems

  • 期刊论文

  • 发表日期：

    2026年5月29日

查看全文
  • Rosalin,
  • Ansuman Patnaik,
  • Bishnu Prasad Mishra,
  • Deepak Kumar Rout,
  • Sarita Nanda

摘要

This study suggests a novel Logarithmic Hyperbolic Sine Squared (LHSS) cost function and the development of its associated LHSS algorithm for effectiveness in environments affected by impulsive noise. The LHSS cost function is highly resilient to outliers, as its logarithmic growth and saturated gradient effectively restrict their impact. To exploit sparse systems, an improved proportionate term is inserted in the LHSS algorithm, which capitalizes on the sparse nature to speed up adaptation process, while the LHSS cost function is found to be resilient in an environment with impulsive noise. Therefore, the IPLHSS algorithm turns out to be effective and has a faster rate of convergence. Additionally, a filtered-x LHSS (Fx-LHSS) algorithm is suggested to attain room equalization. Simulations were conducted for the sparse system identification and room equalization scenario to demonstrate that the suggested approach performs better and is more robust in an environment with impulsive noise.