Instantaneous VEP signal frequency analysis.

Various methods have been used in the analysis of visually evoked potential signals. Due to the high noise content of the ssVEP signal, the signal is usually filtered using averaging or an order statistic filter, and then analyzed for frequency content. A frequency analysis technique often employed is the short-time Fourier transform (STFT) which yields information about both the frequency content and the spatial localization simultaneously. In 1946, D. Gabor modified the STFT by using a Gaussian window instead of a Hanning, Hamming, etc. for optimal simultaneous location in time and frequency. The common problem with the windowed DFT/FFT is the ambiguity of the estimate of frequency at a point (frequency localization). This paper will introduce a frequency analysis technique utilizing the WMMR (weighted majority with minimum range) filter, which, while in the time domain, will determine the frequency and spatial properties of a ssVEP signal within one period. This technique has been shown to be robust to impulsive noise of up to 40% and also robust to independent and identically distributed (i.i.d.) noise and DC shifts, for frequency content and spatial localization analysis for sinusoidal signals. The WMMR technique will be compared to the Gabor STFT by application to steady state visual evoked potential data.