On Tue, Apr 15, 2008 at 1:48 AM, Michiel Konstapel
<[EMAIL PROTECTED]> wrote:
>
>
> We have found that CTP is sensitive to short bursts of packet loss on
> otherwise good radio links (ETX 1.0 to 1.2) and found something that may be
> of interest. LinkEstimatorP uses an ALPHA parameter of 2 (actually 0.2) for
> decaying its exponential average of eetx:
>
> void updateEETX(neighbor_table_entry_t *ne, uint16_t newEst) {
>
> ne->eetx = (ALPHA * ne->eetx + (10 - ALPHA) * newEst)/10;
>
> }
>
>
> Written as it is, this means it takes 80% of the new estimate and 20% of the
> old – is it possible that this was supposed to be the other way around? We
> find that a short burst of lost packets will quickly increase ETX to levels
> where CTP will select a new parent, which causes excessive beaconing as CTP
> resets its beacon interval and other nodes start switching parents as well.
>
This was fixed in December to use an alpha of 9. Depending on what
version you are using, you might be missing this and other important
updates to CTP. In a day or two, there will be a new commit that
dramatically reduces the number of beacons.
- om_p
_______________________________________________
Tinyos-help mailing list
Tinyos-help@millennium.berkeley.edu
https://www.millennium.berkeley.edu/cgi-bin/mailman/listinfo/tinyos-help
