[Tinyos-devel] CTP and LPL on tmote

[Geoffrey Werner-Allen]

FWIW, my experience was that at least the original CC2420 LPL+CTP seemed to
work OK as long as the node density wasn't too high. The ETX numbers were
way off the charts but it didn't seem to impact routing performance too
much.

Best,

-gwa-

geoffrey werner-allen :: 617.694.7261 :: www.eecs.harvard.edu/~werner


On Wed, Nov 26, 2008 at 12:41 PM, Philip Levis <pal at cs.stanford.edu> wrote:

>
> On Nov 26, 2008, at 8:00 AM, Andreas Köpke wrote:
>
> > Philip Levis wrote:
> >> On Nov 25, 2008, at 12:41 PM, Andreas Köpke wrote:
> >>> Hi, has anyone ever got CTP to run with LPL on  a tmote sky at a
> >>> duty cycle
> >>> smaller than 10%? In our tests with 5% the ETX values are way above
> >>> any
> >>> reasonable value.
> >>
> >> net2 is actively working on this now.
> >>
> >> Phil
> >
> > Are there any theories why there is a problem? Anything we could
> > test/do to
> > help in the process?
>
> Very good question. Unfortunately, it's the day before Thanksgiving,
> so responsiveness from Americans is kinda low until Monday. Let's
> talk about it then?
>
> To give a bit of background, in developing CTP we sketched our
> performance goal timeline like this:
>
> 1) Accurate ETX estimation and minimizing data path (data packets/
> delivery) cost
> 2) Minimizing control traffic
> 3) Integrating and testing with LPL
>
> Right now we are on 3. Part of the trick with 3 is that it depends on
> the LPL implementation. It works very well with the CC1100 and CC2500
> implementations (the ones David Moss supports most actively). We're
> just now collecting experimental traces with the CC2420 one.
>
> Phil
> _______________________________________________
> Tinyos-devel mailing list
> Tinyos-devel at millennium.berkeley.edu
> https://www.millennium.berkeley.edu/cgi-bin/mailman/listinfo/tinyos-devel
>
>
-------------- next part --------------
An HTML attachment was scrubbed...
URL: https://www.millennium.berkeley.edu/pipermail/tinyos-devel/attachments/20081202/24f348e6/attachment.htm