Research of EEG identification computing based on AR model
Zhendong Mu, Jianfeng Hu · 2009
This paper proposes the use of EEG as an identification tool, With the help of AR model, the Time-domain EEG signals are converted to frequency-domain. The Characteristic are extracted according to the Fisher distance, and then analyses the Characteristic of the distribution in the frequency bands. The results show that Characteristic are mainly distributed in between 14 ∼ 30Hz.Computed with BP network technology. The results show that the average rate of the identification up to 86.7%.