Scholarlink

<p>The classical filtered-x least mean square (FxLMS) algorithm has been widely applied in active noise control (ANC) systems. However, its efficacy in attenuating noise is frequently inadequate in the presence of impulsive noise. To address this deficiency, this paper proposes a filtered-x quaternion-valued least mean square (FxtanhQLMS) algorithm with a modified hyperbolic tangent function, which uses quaternion-valued LMS as the core filter to update the weight coefficients and nonlinear transformation to have an adjustable compression level for error signal. Moreover, the error information is reused by the weight-based <i>k</i> nearest neighbors method to further diminish the gap between the anti-noise signal and the primary noise. The simulation results demonstrate that the proposed algorithms exhibit superior convergence performance and tracking capability in comparison to other algorithms.</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

Active control of impulsive noise based on a robust quaternion-valued adaptive algorithm with hyperbolic tangent function

  • 期刊论文

  • 发表日期：

    2025年7月13日

查看全文
  • Bingyan Zhang,
  • Xiaomei Chen

摘要

The classical filtered-x least mean square (FxLMS) algorithm has been widely applied in active noise control (ANC) systems. However, its efficacy in attenuating noise is frequently inadequate in the presence of impulsive noise. To address this deficiency, this paper proposes a filtered-x quaternion-valued least mean square (FxtanhQLMS) algorithm with a modified hyperbolic tangent function, which uses quaternion-valued LMS as the core filter to update the weight coefficients and nonlinear transformation to have an adjustable compression level for error signal. Moreover, the error information is reused by the weight-based k nearest neighbors method to further diminish the gap between the anti-noise signal and the primary noise. The simulation results demonstrate that the proposed algorithms exhibit superior convergence performance and tracking capability in comparison to other algorithms.