A ROBUST PARAMETRIC ESTIMATOR FOR SINGLE-TRIAL MOVEMENT-RELATED BRAINPOTENTIALS

Current estimators for single-trial evoked potentials (EP's) require a signal-to-noise ratio (SNR) of 0 dB or better to obtain high quality estimations, yet many types of EP's suffer from substantially lower SN R's, This paper presents a robust-evoked-potential-estimator (REPE) fa cilitating high quality estimations of single movement related EP's wi th a relatively low SNR, The estimator is based on a standard ARX mode l, enhanced to support estimation under poor SNR conditions, The REPE was tested successfully on a computer simulated data set giving reliab le single-trial estimations for the low SNR range of around -20 dB, Th e REPE was also applied to experimental data, producing clear single-t rial estimations of movement related brain signals recorded in a class ic scenario of self-paced finger tapping experiment.