Single motor imagery electroencephalogram signal recognition method based on multi-linear principal component analysis

Abstract

The invention provides a single motor imagery electroencephalogram signal recognition method based on multi-linear principal component analysis. According to the method, a projection matrix can be found from each of a time domain, a frequency domain and a space domain to project the three-dimensional EEG tensor data, so that dimensionality reduction of the original EEG tensor data is realized, andthen classification is carried out by combining a linear classification method. Compared with a traditional principal component analysis method, according to the multi-linear principal component analysis method provided by the invention, dimension reduction is directly carried out from multiple dimensions in the multi-dimensional tensor; spatial structure information of the signals is reserved, the signals are expanded into a one-dimensional vector form for classification after dimension reduction, and therefore compared with a traditional method based on principal component analysis, the method provided by the invention reserves the spatial characteristics of the EEG signals; compared with EEG time-domain analysis, frequency-domain analysis, time-frequency analysis or space-time analysis, the EEG signals are subjected to multi-modal analysis from the time domain, the frequency domain and the space domain, more comprehensive features can be extracted, and the electroencephalogram recognition effect is still high under the condition of small samples.

Description

technical field [0001] The invention relates to the field of biological signal processing, in particular to a single motor imagery EEG signal recognition method based on multi-linear principal component analysis. Background technique [0002] Electroencephalography (EEG) is a signal generated by the activity of brain neurons, which contains a wealth of brain state information. In order to realize the brain-computer interface (brain-computer interface, BCI), it is necessary to effectively decode the EEG signal. The decoding process includes feature extraction and pattern classification of the EEG signal. In recent years, many research groups in the world have devoted a lot of energy to the research on the feature extraction method of single motor imagery electroencephalography (MI-EEG). Extracting features directly from the time domain is the earliest developed method because it is intuitive and has a clear physical meaning. However, because the waveform of the EEG signal i...

AI Technical Summary

Problems solved by technology

However, the effect of wavelet transform depends heavily on the selection of center frequency and bandwidth during continuous wavelet transform. At present, the selection of these parameters often depends on experience or experiments. For EEG signals with large individual differences, the effect is not stable enough; The spatial distribution of time and space, the time-space analysis method of fusion analysis of time and space information is conducive to revealing and enhancing the hidden features in multi-conductor EEG signals

The spatio-temporal analysis method can provide people with more information, which is an important research direction in EEG signal analysis, but it ignores the frequency domain information contained in the EEG signal.

Images