Details

Autor(en) / Beteiligte

• Li, Zenghui
• Wang, Wei
• Yuan, Saijie
• Pei, Junpeng
• Yang, Qianqian
• Wang, Yousong

Titel

Research on SSVEP‐EEG feature enhancement algorithm based on fractional differentiation

Ist Teil von

• Journal of engineering (Stevenage, England), 2024-03, Vol.2024 (3), p.n/a

Ort / Verlag

Wiley

Erscheinungsjahr

2024

Quelle

Wiley Online Library

Beschreibungen/Notizen

• Steady‐state visual evoked potentials (SSVEP), significant in brain‐computer interfaces (BCI) and medical diagnostics, benefit from enhanced signal processing for improved analysis and interpretation. This study introduces a novel enhancement algorithm for SSVEP electroencephalogram (EEG) signals, employing fractional‐order differentiation operators combined with image processing techniques. Utilizing fractional‐order differentiation within a Laplace pyramid framework, the algorithm achieves hierarchical signal enhancement, facilitating detailed feature extraction and emphasizing SSVEP signal characteristics. This innovative approach merges the precision of fractional calculus with the structural benefits of the Laplace pyramid, leading to enhanced signal clarity and feature discrimination. The efficacy of this method was validated using canonical correlation analysis (CCA), filter bank CCA (FBCCA), and task‐related component analysis (TRCA) on a public dataset. Compared to conventional methods, our algorithm not only mitigates trend components in SSVEP signals but also significantly boosts the recognition accuracy of CCA, FBCCA, and TRCA algorithms. Experimental results indicate a marked improvement in recognition precision, underscoring the algorithm's potential to advance SSVEP‐based BCI research. This study introduces a novel SSVEP‐EEG signal enhancement algorithm using fractional‐order differentiation and the Laplace pyramid, enhancing signal clarity and feature distinctiveness. Tested with CCA, FBCCA, and TRCA on a public dataset, the algorithm significantly improves recognition accuracy, showcasing a notable advancement in SSVEP‐based BCI research.