Scholarlink

<p>Noise extraction from electroencephalogram (EEG) signals has become indispensable in the clinical field. This paper mainly focuses on designing an efficient Hammerstein spline adaptive filter (HSAF) for the ocular noise extraction from the EEG signals. The paper claims the novelty of applying a metaheuristic algorithm for optimizing the filter taps and control points (knots) of HSAF. A recently developed evolutionary algorithm, namely, the slime mould optimizer (SMO) algorithm, has been adapted to optimize the filter taps of HSAF, and the outcomes are compared with other reported methods on same dataset. The proposed HSAF has performed better in imaginal and precision analysis for removing the ocular noise even in the low-frequency band (10–12&#xa0;Hz) without any alteration in the α-component with improved signal-to-noise-ratio (SNR) of 99.108&#xa0;dB, lesser mean-squared-error (MSE) value of −36.1083&#xa0;dB, and mean-absolute-error (MAE) of 0.001692&#xa0;dB. Further, the real-world feasibility of the proposed method is verified in hardware TMS320C6713 digital signal processor kit.</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

Efficiently Designed Hammerstein Spline Adaptive Filter for Ocular Noise Extraction from EEG Signals

  • 期刊论文

  • 发表日期：

    2025年1月9日

查看全文
  • Shubham Yadav,
  • Suman Kumar Saha,
  • Rajib Kar

摘要

Noise extraction from electroencephalogram (EEG) signals has become indispensable in the clinical field. This paper mainly focuses on designing an efficient Hammerstein spline adaptive filter (HSAF) for the ocular noise extraction from the EEG signals. The paper claims the novelty of applying a metaheuristic algorithm for optimizing the filter taps and control points (knots) of HSAF. A recently developed evolutionary algorithm, namely, the slime mould optimizer (SMO) algorithm, has been adapted to optimize the filter taps of HSAF, and the outcomes are compared with other reported methods on same dataset. The proposed HSAF has performed better in imaginal and precision analysis for removing the ocular noise even in the low-frequency band (10–12 Hz) without any alteration in the α-component with improved signal-to-noise-ratio (SNR) of 99.108 dB, lesser mean-squared-error (MSE) value of −36.1083 dB, and mean-absolute-error (MAE) of 0.001692 dB. Further, the real-world feasibility of the proposed method is verified in hardware TMS320C6713 digital signal processor kit.