多尺度散布熵matlab代码.zip

function [Out_DisEn, npdf]=DisEn_NCDF_ms(x,m,nc,mu,sigma,tau) % % This function calculates dispersion entropy (DisEn) using normal cumulative distribution function (NCDF) with defined mean (mu) and standard deviation (sigma) values. % % Inputs: % % x: univariate signal - a vector of size 1 x N (the number of sample points) % m: embedding dimension % nc: number of classes (it is usually equal to a number between 3 and 9 - we used c=6 in our studies) % tau: time lag (it is usually equal to 1) % % Outputs: % % Out_DisEn: scalar quantity - the DisEn of x % npdf: a vector of length nc^m, showing the normalized number of disersion patterns of x % % Ref: % % [1] H. Azami, M. Rostaghi, D. Abasolo, and J. Escudero, "Refined Composite Multiscale Dispersion Entropy and its Application to Biomedical % Signals", IEEE Transactions on Biomedical Engineering, 2017. % [2] M. Rostaghi and H. Azami, "Dispersion Entropy: A Measure for Time-Series Analysis", IEEE Signal Processing Letters. vol. 23, n. 5, pp. 610-614, 2016. % % If you use the code, please make sure that you cite references [1] and [2]. % % Hamed Azami, Mostafa Rostaghi, and Javier Escudero Rodriguez % hamed.azami@ed.ac.uk, rostaghi@yahoo.com, and javier.escudero@ed.ac.uk % % 20-January-2017 %% N=length(x); y=normcdf(x,mu,sigma); for i_N=1:N if y(i_N)==1 y(i_N)=1-(1e-10); end if y(i_N)==0 y(i_N)=(1e-10); end end z=round(y*nc+0.5); all_patterns=[1:nc]'; for f=2:m temp=all_patterns; all_patterns=[]; j=1; for w=1:nc [a,b]=size(temp); all_patterns(j:j+a-1,:)=[temp,w*ones(a,1)]; j=j+a; end end for i=1:nc^m key(i)=0; for ii=1:m key(i)=key(i)*10+all_patterns(i,ii); end end embd2=zeros(N-(m-1)*tau,1); for i = 1:m, embd2=[z(1+(i-1)*tau:N-(m-i)*tau)]'*10^(m-i)+embd2; end pdf=zeros(1,nc^m); for id=1:nc^m [R,C]=find(embd2==key(id)); pdf(id)=length(R); end npdf=pdf/(N-(m-1)*tau); p=npdf(npdf~=0); Out_DisEn = -sum(p .* log(p));