I would like to detect the peaks of a modulated signal using only the threshold level rather than the threshold level and the pulse width.

[devrv]

I would like to measure the peaks of the modulated signal using the threshold level rather than the threshold level + the pulse width as in the labview peak detector Vi program.  Since my waveform is continuously changing, it is hard to predict correctly the pulse width, Examples of the waveform are shown in figure 1. Y axis is amplitude and the X-axis is time in milliseconds.   Looks like pulse width is crucial thing in the peak detector Vi.  Based on the pulse width, the peak position changes. I was wondering if anyone has solved these issues based on the +Ve  and -ve slope crossing the threshold level rather than the using the pulse width. We are using Infinium 500 digital Oscilloscope to detect these signals and analyzing it

[Kenny_K]

I am using the peak detection vi to measure all of the peaks and valleys of a signal.  I have one instance of the VI detecting the peaks, with a threshold and another detecting the valleys with a threshold.  It has worked just fine that way.  I do not have the width terminal connected.  As with your case, my pulse width varies so I could not use the width terminal

Hope it works for you.

Let me know if you run into any problems

Kenny

[devrv]

If you do not specify the pulse width, don't you have problem with the peak detection? It doesn't look like the peaks are accurate.  Let me know what you think?

[johnsold]

If your data has a very high signal to noise ratio, you could look at the slope of the signal. When the slope passes through zero, you have located a peak. With noisy signals you will need to do some smoothing or averaging to find the peaks this way, because differentiating to get the slope accentuates the noise.

Lynn

[Kenny_K]

Here is a screenshot from my program after I ahve taken some data.  I also Included the block diagram of the peak detect portion.

[devrv]

Thanks for getting back to me. Lookslike you are using Labview 8.   I am using 7.1 which doesn't have the standard deviation for the peak..Are you using Labview Vi or you have written your own?  Thanks..

[Kenny_K]

I have a Std Deviation and Mean vi (same as I use in 😎 under LV 7.1 located in Analyze>Mathematics>  Probability and Statistics.

Kenny