[time-nuts] 1/f^2 noise

Jim Lux jimlux at earthlink.net
Mon Nov 24 10:06:43 EST 2014


Started plotting some sampled data from my experimental system, and 
interestingly, it seems to have a 1/f^2 slope, rather than a 1/f slope. 
  I had expected the amplifier noise to dominate, but perhaps one of the 
myriad other noise sources is contributing as well (e.g. sample clock 
jitter, ADC quantization).  ADC quantization should be white, though.

only 6000 samples at 200 Hz here (30 seconds) so the data below 1 Hz is 
getting pretty sketchy..

The input to the system wasn't necessarily AWGN or DC (it might be an 
oscillator beating against itself in a mixer with a very short 
difference in time (nanoseconds).  I'll run another test with the input 
shorted.



-------------- next part --------------
A non-text attachment was scrubbed...
Name: phasenoise1.png
Type: image/png
Size: 42138 bytes
Desc: not available
URL: <http://www.febo.com/pipermail/time-nuts/attachments/20141124/c600b70f/attachment-0001.png>


More information about the time-nuts mailing list