REAL-TIME MULTI-CHANNEL EEG SIGNAL PROCESSOR BASED ON ON-LINE RECURSIVE INDEPENDENT COMPONENT ANALYSIS

摘要

A real-time multi-channel EEG signal processor based on an on-line recursive independent component analysis is provided. A whitening unit generates covariance matrix by computing covariance according to a received sampling signal. A covariance matrix generates a whitening matrix by a computation of an inverse square root matrix calculation unit. An ORICA calculation unit computes the sampling signal and the whitening matrix to obtain a post-whitening sampling signal. The post-whitening sampling signal and an unmixing matrix implement an independent component analysis computation to obtain an independent component data. An ORICA training unit implements training of the unmixing matrix according to the independent component data to generate a new unmixing matrix. The ORICA calculation unit may use the new unmixing matrix to implement an independent component analysis computation. Hardware complexity and power consumption can be reduced by sharing registers and arithmetic calculation units.

主权项

1. A real-time multi-channel EEG signal processor based on an on-line recursive independent component analysis (ORICA), the real-time multi-channel EEG signal processor comprising:
an inverse square root matrix calculation unit for providing a computation of eigen, eigen vector and inverse square root matrix; a whitening unit coupled to the inverse square root matrix calculation unit for covariance computation of a sampling signal to generate a covariance matrix, wherein the covariance matrix generates a whitening matrix based on the computation of the inverse square root matrix calculation unit; an ORICA calculation unit coupled to the inverse square root matrix calculation unit and the whitening unit for computing the sampling signal and the whitening matrix to obtain a post-whitening sampling signal, wherein an independent component analysis computation of the post-whitening sampling signal and a predetermined unmixing matrix is performed to obtain independent component data; and an ORICA training unit coupled to the inverse square root matrix calculation unit and the ORICA calculation unit for training the unmixing matrix according to the independent component data to obtain an inverse matrix of the unmixing matrix by the computation of the inverse square root matrix calculation unit, and computing the unmixing matrix and the inverse matrix of the unmixing matrix to generate a new unmixing matrix, wherein the new unmixing matrix is used for the ORICA calculation unit to perform a next independent component analysis computation of a next post-whitening sampling signal and the new unmixing matrix, so as to obtain next independent component data.