[aprssig] Weather Stations and Net Neutrality

Robert Bruninga bruninga at usna.edu
Fri Aug 15 16:07:20 EDT 2008


> So as temp slowly dropped, it is common to 
> flip back and forth between two values. 
> Each makes APRSdos send a packet?

We anticipated that and required a change of 2 degrees before
temperature would reset the decay algorithm and then we also set
the minimum refresh rate for weather at some minimum.  It was
generally agreed I think to be 5 minutes. 

> How is a receiving station supposed to know whether 
> conditions are stable or the weather station is off 
> the air?

Because there is also the APRS standard maximum decay value of
30 minutes.  If nothing at the weather station changes or
exceeds any of the pre-set thresholds (rare), then the
unchanging weather will decay down to a once every 30 minute
rate.  But as soon as anything changes, APRSdos goes back to an
immediate report and then resets the decay algorithm to the
weather minimum (usually 5 minutes I think, though it could be
set less if unusual conditions were anticipated.

> Regardless, this is different than the way the 
> meteorologist usually report, which is at regular 
> intervals. Maybe you want all of them to  
> change their paradigm too?

No, we tried to meet their requirements without loading down the
network with repititious unchanging duplicative report.  APRS
from the beginning defined two maximum net-cycle times for
decayed unchanging data (posits, messages, status, WX,
whatever).  10 minutes for local direct or one hop activities
and information, and 30 minutes for regional activities.  This
was so that all applications could assume consistent performance
based on these fundamental timings, and users could have
consistent expectations.  In the case of weather I think we all
felt that 5 minutes was a reasonable refresh rate for changing
weather data, though it was not codified in the spec the way the
10 and 30 minute rates were.

Also, APRSdos automatically chose the 10 minute or 30 minute
final decay rate based on the users selected path length.  This
again made all APRS functions more or less independent of user
settings whether he was doing a local real-time critical event,
or was just keeping the lights on at home.

Bob, WB4APR





More information about the aprssig mailing list