Confidence limits for noise measurement
 

On Sun, 20 Aug 2006 23:09:39 -0400, "J. Mc Laughlin"
wrote:

Dear Owen and Richard:
....
At Ohio State I did something similar with, probably for the first time,
actual digitizing that went to a computer (punched paper tape!). As I
recall, I did worry about sample rate, but it was much faster than any
changes being observed because a heavy LPF was used. (With the slow
computers in use, I needed not to overdesign the rate too much.) It seems
to me that if you have any reasonably fast filter fall-off, 11 kHz is plenty
fast enough. But then, I am not too sure that I understand your concern and
I am starting to ramble (though I am stone sober).

Mac, I collected my thinking on the subject in an article at
http://www.vk1od.net/fsm/FsmNoiseConfidenceLimits.htm .

Owen
--