Signal processing for brain-computer interface: enhance feature extraction and classification

Abstract-In this paper we present a new scheme for brain imaginary movement invovles sophisticated spatial-temporalsignal processing and classification for electroencephalogram spectral dynamics in the brain. For example, a synchronization based brain-computer interfaces, by emphasizing the extraction of higher frequency components embeded in a desynchroof space-time-frequency feature as well as the combination of ofclassifiers. In particular, we use wavelet packets as a time- npzatio olowe frequency come canbf on a frequency analysis tool and employ sparse component analysis to specific electrode location at the same moment of time [7]. recover source components in the brain signals. We subsequently The understanding of such complex patterns, as well as an apply multi-class common spatial pattern filters to the signals appropriate signal processing method to capture the important and thus obtain important space-time-frequency features for structures of the patterns, should be crucial ...

• Type: Conference
• Year: 2006
• Where: ISCAS
• Authors: Haihong Zhang, Cuntai Guan, Yuanqing Li
• Keywords: Electroencephalogram Spectral Dynamics, ISCAS 2006, Motor Imagery Signals, Space-time-frequency Features