Local maximum and local minimum by spline about pyhht

[jaidevd]

Yes, for signal that has been sampled at a very low frequency, the local extrema can be misleading. I'd be happy if you'd send a PR. But I don't fully understand. If we are working on extrema of the splines, how do we construct the splines in the first place, since we need local extrema on the raw data to build the upper and lower envelopes?

Hi, I mean that you can interpolate raw data by spline and interpolate the spline's extrema points for upper and lower envelopes.

here shows the difference between spline's local extrema and local extrema by index of array
http://docs.scipy.org/doc/scipy/reference/tutorial/interpolate.html#d-interpolation-interp1d

[jaidevd]

All right. So we just add more samples in the raw data by interpolating it, and then perform sifting on this data. Is that right?

No, adding more samples is needless, we can find exact extrema points on spline at x axis by first order derivatives' root and determine whether the points is maxima or minium by second order derivatives' sign(negative : maximum, positive : minimum)

http://docs.scipy.org/doc/scipy/reference/generated/scipy.interpolate.splder.html#scipy.interpolate.splder
http://docs.scipy.org/doc/scipy/reference/generated/scipy.interpolate.sproot.html#scipy.interpolate.sproot

but why there's several files for the same purpose? which one should I modify?

I sent pull request, please check

[jaidevd]

Closing due to inactivity.