CSP-Ph-PS: Learning CSP-phase space and Poincare sections based on evolutionary algorithm for EEG signals recognition
作者：
Highlights：
• We proposed a new feature extraction method.
• Features are extracted based on optimal Poincare sections that fitted on phase space.
• Optimal phase space reconstructed based on filtered data using CSP.
• BCI Competition III and BCI Competition IV datasets is utilized to evaluation.
• Presented model reached the accuracy of 89.76% and 71.87% for datasets, respectively.
摘要
•We proposed a new feature extraction method.•Features are extracted based on optimal Poincare sections that fitted on phase space.•Optimal phase space reconstructed based on filtered data using CSP.•BCI Competition III and BCI Competition IV datasets is utilized to evaluation.•Presented model reached the accuracy of 89.76% and 71.87% for datasets, respectively.
论文关键词：Phase space,Poincare section,Feature extraction,Electroencephalogram signals,Motor imagery
论文评审过程：Received 19 April 2022, Revised 29 July 2022, Accepted 16 August 2022, Available online 19 August 2022, Version of Record 30 August 2022.
论文官网地址：https://doi.org/10.1016/j.eswa.2022.118621
