An Improved Automatic EEG Signal Segmentation Method based on GeneralizedLikelihood Ratio

It is often needed to label electroencephalogram(EEG) signals by segments of similar characteristicsthat are particularly meaningful to clinicians and for assessment by neurophysiologists. Within eachsegment, the signals are considered statistically stationary, usually with similar characteristics such asamplitude and/or frequency. In order to detect the segment boundaries of a signal, we propose animproved method using time-varying autoregressive (TVAR) model, integral, basic generalizedlikelihood ratio (GLR) and new particle swarm optimization (NPSO) which is a powerful intelligentoptimizer. Since autoregressive (AR) model for the GLR method is valid for only stationary signals,the TVAR as a valuable and powerful tool for non-stationary signals is suggested. Moreover, toimprove the performance of the basic GLR and increase the speed of that, we propose to use movingsteps formore than one sample for successive windows in the basic GLR method. The purpose of usingNPSO is finding two parameters used in this approach. By using synthetic and real EEG data, theproposed method is compared with the conventional ones, i.e. the GLR and wavelet GLR (WGLR).The simulation results indicate the absolute advantages of the proposed method