On 3/29/06, Olin Lathrop <olin_piclist@embedinc.com> wrote:
> David VanHorn wrote:
> > I have some analog channels that I'm reading, and I was worried about
> > getting "unlucky" and aliasing the sample rate against unknown
> > external variables,


At high sampling rate this could be corrected with antialising filters.
I'm always use this LPFs for MHz sampling rate.
But on PICs sampling rate is often small so a simple RC filter may help.

Vasile


so what I did was to use a PN generator to spin a
> > random number which is used to determine the time between samples. As
> > set, it goes from 64mS to 128mS, which seems to suit the app.
>
> If you want to formalize this, sample according to a poisson interval
> distribution.  Any aliases then become white noise added to the sampled
> signal instead of appearing having a fixed frequency relationship with the
> undersampled signal.  Note that you can never make the energy in
> undersampled frequencies go away.  This technique essentially spreads it out
> over the output spectrum.
>
>
> ******************************************************************
> Embed Inc, Littleton Massachusetts, (978) 742-9014.  #1 PIC
> consultant in 2004 program year.  http://www.embedinc.com/products
> --
> http://www.piclist.com PIC/SX FAQ & list archive
> View/change your membership options at
> http://mailman.mit.edu/mailman/listinfo/piclist
>

-- 
http://www.piclist.com PIC/SX FAQ & list archive
View/change your membership options at
http://mailman.mit.edu/mailman/listinfo/piclist