Biometrics Based On Specific Patterns Of Brain Activity

With the rapid development of the intelligent age,how to safely and reliably identify individual identity information has become an increasingly important issue.In recent years,Electroencephalograph which reflects brain electrical activity has recently been found to have great potential in the field of biometric identification due to its unique advantages compared with traditional biometric(fingerprint,face)identification methods,such as strong concealment,unable to steal,and difficult to forge.The key to EEG biometrics is that the features extracted from EEG signals need to have significant differences between individuals and can be repeated in multiple experiments.This article focuses on the study of individual specific brain activity patterns,studies the individual specific brain activity patterns under spontaneous EEG signals and evoked EEG signals,and proposes a variety of feasible biometric methods.The main research work of this article is as follows:Design experimental paradigms and create EEG data sets with multiple subjects across time.This article first conducted continuous EEG data monitoring on multiple subjects in different periods of six months,and created a biometric data set under resting potentials and steady-state visual evoked potentials.Explore the biometric technology based on the resting state function to connect the brain network.The individual specificity and persistence of EEG connection patterns in the resting state are determined.Records were made 6 times at different intervals over a period of 6 months to check the difference and persistence of the EEG function in the resting state over a long period of time.The results show that the EEG-FC resting state network has significant individual specificity,and the identification accuracy on a single experimental data set is as high as 90% or more.At the same time,this individual specificity remains stable,and the Granger causality index has a smaller downward trend after 6 months.The biometric technology based on the multi-spectral curve characteristics of steady-state visual evoked potential is explored.By discovering the difference in individual response to different frequencies,using a variety of frequency combinations to construct individual differences of multi-spectral curve features for identification,verifying the feasibility of identification based on steady-state visual evoked EEG.Explored the biometric characteristics of the brain network based on the task state function.For the first time,we use the functional connectivity indicators between different brain regions of the SSVEP signal for identification.The spectral coherence connectivity and phase lock value phase synchronization index based on SSVEP signal have significant specificity among individuals.The individual identification accuracy of the SSVEP functional network is as high as 98%.Our work has enriched the biometric technology based on EEG and verified the specific brain activity patterns of individuals.