Open AccessResearch on Feature Extraction Algorithm Commonly Used in Brain-computer Interface Technology
Author(s) -
Yifan Zhang,
Yaojun Wang
Publication year - 2021
Publication title -
journal of physics. conference series
Language(s) - English
Resource type - Journals
SCImago Journal Rank - 0.21
H-Index - 85
eISSN - 1742-6596
pISSN - 1742-6588
DOI - 10.1088/1742-6596/1861/1/012027
Subject(s) - brain–computer interface , computer science , interface (matter) , feature extraction , computer technology , wavelet transform , algorithm , spectral density , process (computing) , feature (linguistics) , channel (broadcasting) , electroencephalography , wavelet , pattern recognition (psychology) , artificial intelligence , computer engineering , psychology , telecommunications , computer network , linguistics , philosophy , bubble , maximum bubble pressure method , psychiatry , parallel computing , operating system
Brain-computer interface (BCI) is an effective and direct channel of information exchange between human brain and external devices such as computer, which can provide auxiliary information acquisition and treatment means for the medical and other fields in the future. This paper focuses on four kinds of feature extraction algorithms such as power spectral density (PSD), wavelet transform, Hilbert-Huang transform (HHT) and common space pattern (CSP), which are commonly used to process abnormal electroencephalogram (EEG) signals in brain-computer interface technology. This paper also introduces their respective principles, characteristics and application fields, analyzes and compares the advantages and disadvantages of these algorithms, and obtains the development direction of feature extraction algorithms in the future. Finally, it also briefly discusses the ethical issues brought about by brain-computer interface technology.