Research on SSVEP feature extraction based on HHT

作者：Li Zhao, Pengxian Yuan, Longteng Xiao, Qingguo Meng, Daofu Hu, Hui Shen · 发表于：2010 Seventh International Conference on Fuzzy Systems and Knowledge Discovery · 年份：2010 · DOI：10.1109/fskd.2010.5569537 · 被引用次数：23 · 研究领域：EEG and Brain-Computer Interfaces、Neuroscience and Neural Engineering、Advanced Memory and Neural Computing

Considering of high transmission rate and short training time, Steady State Visual Evoked Potential (SSVEP) rapidly becomes a practical signal in Brain-Computer Interface(BCI) system. This paper study the extraction method of SSVEP based on the Hilbert-Huang Transformation. The SSVEP was processed by a time-frequency processing system. after empirical mode decomposition and Hilbert-Huang Transform(HHT), an eigenvector detected from the result of HHT was viewed as the characteristics of the SSVEP signal that contains different frequency component. Then the eigenvector is classified in a Fisher classifier. Compared with the (Fast Fourier Transform)FFT, the classification accuracy of a one-minute data can reach more than 85 percent.