Instantaneous Mean Frequency

[gaiter]

I have an EMG Burst that I would like to calculate the Instantaneous Mean Frequency on.  I have used the Signal Analysis Toolkit in LV7.1. 

Attached is a simple program in which I use the TFA Mean Instantaneous Frequency.vi but do not get the desired results.  I would expect a line around 80 Hz when the muscle fires. 

Any help would be appreciated

[eek]

Hi

I'd like to help you with your issue but I cannot view your vi's block diagram. You've locked editing in the properties of your vi. I'll need to see your code in order to offer suggestions. Any chance you can post the unlocked files?

Cheers,
Emilie

[gaiter]

I think this will work....................Thanks

Run - IMF_Test2.vi

[johnsold]

I cannot run your vi because some dlls are missing, but I see one thing which may be part of the problem: The time axes on the graphs are different. The RAW EMG has a maximum of 3500 (ms?) while the spectrograms have 8.5 and 9.8 seconds and the IMF is 500. The Time-Freq Sampling Info cluster has the Time steps set to -1 which results in the Distribution function VIs calculating a default time step. It looks like your sampling rate is 1000 samples/second based on the number of points in RAW EMG. What happens if you tell the distribution functions what the actual sampling rate is?

Lynn

[gaiter]

Sorry, my bad.  Your right.  I took this out of a bigger program and didn't copy the sampling rate (2000Hz).  Putting that back in doesn't change much.  Looking at the spectogram, it looks as what I would expect.  The TFA Instantaneous Mean Frequency.vi in the signal processing toolkit is not giving me what I would expect - even playing with the time-frequency sampling info.  The time-frequency axis are really scale and translation. 

Do you have the signal proessing toolkit?  Is that why there are missing .dll's are missing or am I doing something incorrect?

Thanks

[johnsold]

I do not have the toolkit (well, actually, I have a copy of it, but I cannot use it because it does not run on my platform!). So I cannot dig into it any further.

Lynn

[gaiter]

Thanks anyway.......