have a look to the functions wich will be included in this toolbox: reference manual

the toolbox contains MATLAB routines for computing recurrence plots and related problems as M-files

parameters in [ ] are optional

download: CRP.m, CRQA.m, tt.m;
sorry, software is only available for members of the nonlinear dynamics group

also visit to survey other useful toolboxes
still under construction,

routine description
recurrence plot, cross recurrence plot CRP(X [,Y [,M,T,E,A]])

creates a cross recurrence plot/ recurrence plot results could be stored into the workspace

allows to change the parameters interactively by using a GUI

the source-data X and test-data Y can be one- or a two-coloumn vectors (then, in the first column have to be the time); if the test-data Y is not specified, a simple recurrence plot is created

parameters: dimension M, delay T and the size of neighbourhood E; the flag A allows to switch between various methods of finding neighbours or CRPs, resp.:
A=1 -> Euclidean norm
A=2 -> Euclidean norm between normalized vectors
A=3 -> fixed amount of neighbours
A=4 -> distance coded matrix (global CRP)
parameters not needed to specify

current limitation: for higher speed in output the whole matrix of the recurrence plot is in the work space - this limits the application of long data series

complexity measures based on (cross) recurrence plots CRQA(X [,Y [,W,M,T,E,L,A]])

recurrence quantification analysis (WEBBER) vs. time (away from mean diagonal) of cross-recurrence with the first vector X and the second Y Lmin is the minimal length for a line if W=[] then the whole plot will calculated B is the maximum forwarded (backwarded) time J=time scale
Y(1)=RR of diagonalline

length of vertical lines in recurrence plots TT mean trapping time and its distribution.

A=TT(X) computes the mean of the length of the vertical line structures in a recurrence plot, so called trapping time TT.

[A B]=TT(X) computes the TT and the distribution of the length of the vertical line structures, stored in B.
estimation main diagonal path of a RP [X, Y]=TRACK(RP)

estimates the path of the main diagonal line in a RP
RP is a N x N-Matrix

heaviside function Y=H(X)

heaviside function y=H(x) with
y=1 for x>=0 and
y=0 for x<0

X can be an one-column or one-row vector

transformation to pretended distibutions Y=TRAFO(X [,A])

transformation of data X to a pretended distribution Y, where
A=0 normal distribution (defeault),
A=1 uniform distribution,
A=2 exponential distribution

X is an one-column or two-column vector (then, in the first column have to be the time)

mutual information S=MI(X1, X2, T, [,A])

mutual information between the data in the vectors X1 and X2
(A=0 -> without plot)

X1 and X2 are one-column vectors

auto mutual information M=autoMI(X, T)

computes the auto mutual information of the vector X

X is an one-column or two-column vector (then, in the first column have to be the time) vectors

estimation entropy E=ENTROPY(X)

estimates the entropy E of a vector X

phase space diameter [Y Z]=PSS(X,M,T)

computes the maximal (Y) and the averaged (Z) phase space diameter of data X, with the embedding parameters dimension M and lag T

walsh transformation Y=WALSH(X)

computes the walsh transformation s=walsh(x) where
Y and X are one-column vectors

walsh power spectrum Y=WALSHPOWER(X)

computes the walsh power spectrum s2=(walsh(x))2 where
Y and X are one-column vectors

etc. under construction