UDC 004.724
APPLICATION OF VECTOR AUTOREGRESSIVE MODEL FOR SPECTRAL ANALYSIS OF ENCEPHALOGRAMS
V. G. Andrejev, Dr. in technical sciences, full professor, RSREU, Ryazan, Russia;
orcid.org/0000-0003-3059-3532, e-mail: This email address is being protected from spambots. You need JavaScript enabled to view it.
V. A. Belokurov, Dr. in technical sciences, full professor, RSREU, Ryazan, Russia;
orcid.org/0000-0002-8893-550X, e-mail: This email address is being protected from spambots. You need JavaScript enabled to view it.
N. V. Belokurova, Ph.D. (medical sciences), assistant, RSMU, Ryazan, Russia;
e-mail: This email address is being protected from spambots. You need JavaScript enabled to view it.
The problem of estimating spectral power density of biological signals is considered. A vector auto regressive model for estimating spectral power density is proposed. The criterion of minimum first autocor relation coefficient at the output of whitening filter is proposed for estimating model order. The aim of the work is to investigate the feasibility of calculating spectral power density of encephalogram signals using a vector autoregressive model. The authors show that model order depends not only on the number of enceph alogram leads used, but also on the location of the leads on the head. A relationship between the order of selected vector autoregression model and the magnitude of autocorrelation coefficients of whitening process is found. The effect of an increase in the number of signals from leads on the order of selected model is con sidered. The model order is shown to lie in the range from 5 to 14 for the number of leads used from 4 to 12. The proposed algorithm is tested using real EEG data.
Key words: electroencephalogram, alpha rhythm, non-stationarity, power spectral density, vector auto regressive model, estimation of power spectral density of a random process.