Overlap for FFT averaging

[Chris_12345]

When doing averaged FFTs for a long timeseries, I hear that using a 50% overlap is needed to avoid spectral artifacts.  For example, doing an FFT from 0 sec to 1 sec, then 0.5 to 1.5 sec, then 1 sec to 2 sec, then 1.5 sec to 2.5 sec, etc.

Do the FFT averaging functions in LabVIEW do this overlap, or do I need to write this code myself?  Is a 50% overlap even necessary?

Thanks for your help.

[JLS]

Hello,

I'd like to help you here, but I'm not a signal processing expert.  Let me state your questions with some remarks, and I think we'll be able to make some progress here:

Q: Do the FFT averaging functions in LabVIEW do this overlap, or do I need to write this code myself? 

 - Which function are you referring to in LabVIEW - we can read about it in the LabVIEW help, and if it's not specifically declared there, I can do more research and hopefully find a more specific answer.

Q: Is a 50% overlap even necessary?

 - I don't know the answer to that question, but you should be able to google for some related articles to get some expert opinions and analysis.

Sorry I couldn't be of more direct help now - perhaps we can answer the first of your questions on the LabVIEW side after you indicate which functions you are referring to!

Best Regards,

JLS