[SI-LIST] Re: Quantization and Statistical Averaging
- From: "john lipsius" <johnlipsius@xxxxxxxxx>
- To: <slund347@xxxxxxxx>, <si-list@xxxxxxxxxxxxx>
- Date: Tue, 27 Aug 2002 02:01:04 -0700
Steve,
I haven't worked with anything like you describe but, if I understand
your description, I suspect that the time periods where you have jitter
in the measurement are the periods when the "crystal ctld pulse generator"
and the measurement circuit clock references are strongly in phase.
That is, they are two different references so their frequencies are constantly
drifting in relative phase. Given the right conditions, they line up and
stabilize
and you get a deterministic phase offset "detected" by the measurement
circuit.
Or, maybe I'm just dreaming.
-John
----- Original Message -----
From: "Steve Lund" <slund347@xxxxxxxx>
To: <si-list@xxxxxxxxxxxxx>
Sent: Tuesday, August 20, 2002 3:46 PM
Subject: [SI-LIST] Quantization and Statistical Averaging
>
> SI Brain Trust,
>
> My appologies as this question deals more with data integrity than signal
> integrity. It is an issue that has myself and several application engineers
> completely stumped. I am hoping that one of you has seen this before and
> could point me to some resources to help me understand what is going on.
>
> We have instrumentation that outputs data as timed pulses to represent a
> measured quantity. We are measuring this time via an interrupt on a
> processor. The processor clock provides a raw resolution of about 18nS. We
> are attempting to increase this base resolution by statistically averaging
> 128 samples.
>
> In order to qualify this new design we are feeding the output of a crystal
> controlled pulse generator into the circuit and collecting the averaged data
> over a fairly long period of time (hours to days). When we analyze the
> collected data we are seeing large blocks of time where the P-P "noise" is
> around 1.35nS. This is what we expect to happen from the averaging process.
> Unfortunately, we are also data intervals from a minute to 30 minutes in
> length where the P-P "noise" is considerably worse than the background
> 1.35nS. At worst case it looks like the averaging does absolutely no good
> and the P-P noise equals the 18nS base resolution.
>
> Based on this it would appear that data averaging to increase resolution
> runs into problems when the data in question is quantized and not Gaussian
> distributed. Does anyone know if there is a name for this phenomenon or
> where I could learn more about it or how to prevent it?
>
> Thanks for listening.
>
> Steve
>
> P.S. I won't be able to respond quickly to answer any questions as I am
> posting this from home. Apparently our domain at work got black listed for
> having an open mail relay (potential SPAM source). Freelists must subscribe
> to one of the blacklist services which intercepts and bounces the e-mails
> from blacklisted domains. It took our IT guys several hours to figure out
> what was going on.
>
> Steve Lund
> Emco Electronics
> (An OPW / Petro Vend Company)
> e-mail: slund@xxxxxxxxxxxxx
> (919) 460-6000 x2133
> (919) 460-7595 FAX
>
>
>
> ------------------------------------------------------------------
> To unsubscribe from si-list:
> si-list-request@xxxxxxxxxxxxx with 'unsubscribe' in the Subject field
>
> or to administer your membership from a web page, go to:
> //www.freelists.org/webpage/si-list
>
> For help:
> si-list-request@xxxxxxxxxxxxx with 'help' in the Subject field
>
> List archives are viewable at:
> //www.freelists.org/archives/si-list
> or at our remote archives:
> http://groups.yahoo.com/group/si-list/messages
> Old (prior to June 6, 2001) list archives are viewable at:
> http://www.qsl.net/wb6tpu
>
------------------------------------------------------------------
To unsubscribe from si-list:
si-list-request@xxxxxxxxxxxxx with 'unsubscribe' in the Subject field
or to administer your membership from a web page, go to:
//www.freelists.org/webpage/si-list
For help:
si-list-request@xxxxxxxxxxxxx with 'help' in the Subject field
List archives are viewable at:
//www.freelists.org/archives/si-list
or at our remote archives:
http://groups.yahoo.com/group/si-list/messages
Old (prior to June 6, 2001) list archives are viewable at:
http://www.qsl.net/wb6tpu
Other related posts:
