Window and overlap size question with coherence and mscohere

4 visualizaciones (últimos 30 días)
Alan Armstrong
Alan Armstrong el 4 de Jul. de 2013
Hi,
I've look at a few other posts regarding this question but i wanted to double check that my settings are correct.
From what I read on this forum I gather that the window size will adjust the frequency resolution of my coherence. Low resolution = small windows and accuracy of results are better with small windows as well.
Also the window size should be related to the sample rate in order to target a specific frequency range. The relationship may look something like this, but I'm not sure e.g. sample_rate/window_size = expected frequency range of data.
So here are my settings:
Data is from an intentional coordination task so I would expect relatively good level of coherence (around 0.7) at frequencies ranging from 0.5 - 1 Hz, which is the range of frequencies that the stimuli in my experiment were presented at for the synchronisation task.
data_length = 3000;
sample_rate = 100;
window = 200; (Expecting frequencies between 0.5 - 1 Hz, 100/200 = 0.5)
overlap = 100; (overlap of 50%)
L = length(data)
nfft = 2^nextpow2(L);
[CSC, f] = mscohere(x,y,window,overlap,nfft, sample_rate);
If anyone could let me know if this looks correct I would greatly appreciate it.
thanks,
Alan

Respuestas (0)

Categorías

Más información sobre Specifying Target for Graphics Output en Help Center y File Exchange.

Productos

Community Treasure Hunt

Find the treasures in MATLAB Central and discover how the community can help you!

Start Hunting!

Translated by