Prev: split one line computation without duplicating data
Next: multi peak surface 2D Gaussian fitting
From: Debora on 9 Aug 2010 16:01
Hi, Jan
Yes, I do understand what you are asking. But this is interesting, though.
I am using emg as an timing parameter for optimization and I will need to tell when the muscle is active and when is not. I am used to do emg analysis on a specific software that operates on dos. But I wanna optimize my work doing this on matlab, that is what I am asking here, obviously. Well, the thing is that I've did a few tests on that software and on previous works of my group that were looking for on and offset, and we have always used this processing, or something alike, and it have always worked fine...
Curious, no?
Debi

"Jan Simon" <matlab.THIS_YEAR(a)nMINUSsimon.de> wrote in message <i3pkq8$s6c$1(a)fred.mathworks.com>...
> Dear Debora,
>
> > In fact, when I used a bortec electromyographer I did not have this low frequencies noise, but this new equipment saturates in really low frequencies.
>
> Ok, I assume, you are sure that this is noise. And I think, you've understood my question, because for the kind of investigations I perform, the 20 to 500Hz are just the modulation, while the signal I'm searching for is the on and off switching of the limb muscles, which have a frequency of 1 Hz (equivalent to 1 stride per second). So If I remove the 1 Hz part of the signal, the muscles seem to be active ever and with almost constant force.
> Anyhow, there can good reasons also to remove this part.
>
> Kind regards, Jan
 | 
Pages: 1
Prev: split one line computation without duplicating data
Next: multi peak surface 2D Gaussian fitting