[FieldTrip] Sharp transition in multitaper coherence when moving from one taper to two

[Mircea Stoica]

Dear fieldtrippers,

I'm interested in coherence analysis with mtmconvol and multitapers. My
frequencies of interest are spaced logarithmically (steps of 1/16 octaves)
and I adjust the time window and number of tapers to yield a constant
frequency smoothing of 0.75 octaves (half that for cfg.tapsmofrq of
course). This results in an ugly step-like transition of coherence when
moving from one taper to two, as you can see in the following image.

https://photos.app.goo.gl/5uWy05RzZxaA9gci1

This sharp transition is not visible in the power output. I calculated the
actual frequency smoothing as fw = (K + 1) ./ cfg.t_ftimwin / 2 and in a
semilogy plot it gives this:

https://photos.app.goo.gl/xQsrY1jr8L0qLLDm2

And the number of tapers for each frequency:

https://photos.app.goo.gl/LqaeUVW6ymBWEKEx2

There are no discontinuities in the frequency smoothing. Any ideas what
could be the cause? Or is there perhaps nothing to be done about it?


Best,
Mircea
--

_____________________________________________________________________

Universitätsklinikum Hamburg-Eppendorf; Körperschaft des öffentlichen Rechts; Gerichtsstand: Hamburg | www.uke.de
Vorstandsmitglieder: Prof. Dr. Burkhard Göke (Vorsitzender), Prof. Dr. Dr. Uwe Koch-Gromus, Joachim Prölß, Martina Saurin (komm.)
_____________________________________________________________________

SAVE PAPER - THINK BEFORE PRINTING
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://mailman.science.ru.nl/pipermail/fieldtrip/attachments/20171011/9577d5e8/attachment-0001.html>