Research Of EEG Signal Based On Construction Of Network Of Recurrence Plot

Human brain is the network structure,which is made up by over 10 billion nerve cells.And the brain show high levels of complexity,non-stationary and nonlinear behavior in neural activity.Recurrence Plot(RP)and complex network can be used to study the interaction between different areas,the topological structure and dynamic information,and physiological functions in the process of neural activity.Electroencephalogram(EEG)is one of the most important physiological electrical signal for human,which contains a great deal of physiological and pathological information related to the brain.In this thesis,we analyse the difference between sleep EEG and waking EEG from the point of view of RP and complex network.And our work provides an important theoretical basis for the research of brain state under sleep state.Main work of this thesis is shown as below:(1)We use Recurrence Plot to analyse the difference between sleep EEG and waking EEG.Firstly,self-correlation function method and Cao method are used to determine the parameter of phase-space reconstruction.Then,we reconstruct phase space for 16 sets of EEG data,and draw its Recurrence Plot as well as compare the difference of two kind of EEG.To make our result more convincing,we use Recurrence Quantitative Analysis(RQA)to anslyse two kind of EEG.Result shows that the nonlinearity of waking EEG is much stronger than sleep EEG.(2)We adopt Order Recurrence Plot(ORP)to analyse sleep EEG and waking EEG.Firstly,we use ORP to analyse two kind of EEG,qualitatively.Then RQA is used to deal with two kind of ORP,quantitatively.Compared with RP,this mean has more obvious distinction for two kind of EEG.(3)In this part,we use the recursive matrix,coming from the RP,to construct complex network.For sleep EEG and waking EEG,the topological structure has obvious difference.Meanwhile,comparing the parameter of complex network produced by two kind of EEG,we can distinguish one kind of EEG from the other.The means adopted by this thesis can be used to analyse the difference between sleep EEG and waking EEG,and it reveal the dynamic behavior differences between two kind of neural network.