03-14-2017 10:22 PM
Hey,
Has anyone toyed with data decimation and statistics in LV? I.e I'm thinking something like an adaptable moving average filter for decimation, that stops decimating once something was outside of 4.5 sigma or so of the other measurements, and once it returns to that bound starts decimating again. Basically something that decimates a straight line but not anything that is a glitch (voltage spike, etc). This would be for scope waveforms - similar to LeCroys top function vs max.
Thanks!
David
03-15-2017 02:31 AM
03-15-2017 03:29 AM
I am not seeing anything that couldn't be done fairly easily. What do you want done in your waveform when outside the 4.5 sigma? Delete the time sample? Make it NaN? Use previous average?
03-15-2017 09:24 PM
Time to see if I can get it working 🙂
Just wanted to make sure it wasn't done before - I am a fan of code reuse.