ceemd(measurements, xvec; num_imfs=6)

Compute the Complete Empirical Mode Decomposition of the time series with values measurements and time steps xvec. numimfs is the Number of Intrinsic Mode Functions. Returns a list of numimfs + 1 Vectors of the same size as measurements.

eemd(measurements, xvec, numtrails=100)

Return the Intrinsic Mode Functions and the residual of the ensemble Empirical Mode Decomposition of the measurements given on time steps xvec.

emd(measurements, xvec)

Return the Intrinsic Mode Functions and the residual of the Empirical Mode Decomposition of the measurements given on time steps given in xvec.

maketestdata(seed)

Return a simple example time series with the composing parts

Iterator for the Empirical Mode Decomposition. The time series values are an AbstractVector of type T, and the time positions are an AbstractVector of type U.

SiftIterable{T, U}

Iterator for the sifting algorithm. The time series values are an AbstractVector of type T, and the time positions are an AbstractVector of type U. Fields:

colominas2014_s()

First testdata from Colominas 2014 et al. http://dx.doi.org/10.1016/j.bspc.2014.06.009

colominas2014_x()

Second testdata from Colominas 2014 et al. http://dx.doi.org/10.1016/j.bspc.2014.06.009 It is not exactly the same, because ϕ is not feasable because the arccos is only defined for values between -1 and 1.

fosso2014()

Make testdata from Fosso 2014 et al.

get_edgepoint(y, xvec, extremas, pos, comp)

Compute the edgepoint which should be used as the extrema on the edge for the spline computation.

ismonotonic(x::AbstractVector)

Check wether x is monotonic. This means, every value is either larger or smaller than the preceding value.

localmaxmin!(x, maxes, mins)

Detect the local extrema of x. Push the maxima into maxes and the minima into mins.

sift(y, xvec)

Sift the vector y whose points have x coordinates given by xvec.

zerocrossing!(y, crosses)

Compute the indices of zerocrossings of a vector It searches for elements which are either zero or near a signflip and pushes the indices into crosses.

Iterator for the Complete Empirical Mode Decomposition. The time series values are an AbstractVector of type T, and the time positions are an AbstractVector of type U.

Intermediate results of the CEEMD Iteration

SiftState

Handle the intermediate results of the sifting.