Hi sw,
Just an FYI -
if you're doing something
like calculating integrated-phase-noise from 10Hz to 1MHz from a CW, there's a way to
obtain a good approximation - and it executes %50 faster than the
Phase-noise utility on the mass-memory module. We make 6
spot-frequency measurements at, say, 10Hz, 100Hz, 1KHz, 10KHz, 100KHz,
and 1MHz (from CW) then calculate/sum the phase-noise for each
interval. An accurate approximation may be possible by starting with
the assumption that dBc vs log(F) is linear over each
interval. The straight-line approximation (dB/log(F)) is
common-knowledge if you search this subject - but I couldn't find the integral calculation in the public domain. If you're a math-wiz
(or know one) start with:
m = (dB
F1 - dB
F2)/(F
1dB-F
2dB) => dBc as a function of F, or: dBc = dB
F1 - mF
1dB + mF
dB. This eventually leads to:
double-sideband [integrated] Phase-Noise = 2*K/(m+1)[F
2^(m+1) - F
1^(m+1)] <rad^2>
The integral equation was derived/supplied by an EE where I work, and
simply handing-it-out seems wrong... but it (and K) can be derived from
what's here.
then again...
Before giving-up on automating the phase-noise Utility, I concluded it was
probably possible to write/edit functions stored on the mass-memory module - if so, then
you have
power over the Phase-noise Utility!
Luck/Cheers. tbd (formerly Dynamik)
Message Edited by tbd on 12-13-2006 03:17 AM
"Inside every large program is a small program struggling to get out." (attributed to
Tony Hoare)