[aprsspec] rain fall in hour

[Scott Miller]

Yeah, I'm familiar with the problem.  Like for tracking wind gusts.  If you 
don't need it to be 100% accurate there are some tricks you can do.  For the 
gust problem, I track two gust values and their age so that when the older 
one expires (ages past 5 minutes) you can switch to the next highest.  The 
problem there is that the next highest value might be only a second away 
from expiration itself, so then you've lost all the other gusts that came in 
the last 5 minutes.  Deciding when to dump the next-highest value for a 
newer one basically turns into an optimal stopping problem:

http://plus.maths.org/issue3/marriage/

The rain fall problem is similar to the 1-minute average calculation for my 
Geiger counter.  Rather than keep 60 readings it just assumes the count rate 
was steady across the 60 second period and decays accordingly.  I think this 
is technically a single-pole IIR filter (an integrator).  I haven't added 
rain gauge support to the OpenTracker yet, but when I do I'll probably use 
the same algorithm.

You could improve the accuracy a bit by breaking it up into, say, four 
blocks of 15 minutes.  I'm sure there are more elegant ways to do it, too. 
Was never too big on calculus myself.

Scott
N1VG

----- Original Message ----- 
From: "James Jefferson Jarvis" <jj at aprsworld.net>
To: "Curt Mills, WE7U" <archer at eskimo.com>; "TAPR APRS Specification 
Discussion List" <aprsspec at lists.tapr.org>
Sent: Monday, July 11, 2005 2:08 PM
Subject: Re: [aprsspec] rain fall in hour


> That's kinda how I read it as well. I was just hoping to avoid 60 bytes of 
> memory for rain fall counters, plus 23 more bytes of memory to do the rain 
> in the last 24 hours. Yep, doing some embedded work.
>
> Thanks,
>
> -Jim KB0THN
>
> Curt, WE7U wrote:
>
>>On Sun, 10 Jul 2005, James Jefferson Jarvis wrote:
>>
>>
>>>How does the rain fall in hour part of a weather packet work? Is it in 
>>>the
>>>last wall clock hour (ie between 8pm and 9pm) or in the last 60 minutes?
>>>
>>
>>The spec says "rainfall (in hundredths of an inch) in the last
>>hour".
>>
>>I read that as:  Rainfall during the current minute plus the 59
>>previous minutes.
>>
>>--
>>Curt, WE7U.   APRS Client Comparisons: http://www.eskimo.com/~archer
>>"Lotto:    A tax on people who are bad at math." -- unknown
>>"Windows:  Microsoft's tax on computer illiterates." -- WE7U
>>"The world DOES revolve around me:  I picked the coordinate system!"
>>
>>_______________________________________________
>>aprsspec mailing list
>>aprsspec at lists.tapr.org
>>https://lists.tapr.org/cgi-bin/mailman/listinfo/aprsspec
>>
>
>
> _______________________________________________
> aprsspec mailing list
> aprsspec at lists.tapr.org
> https://lists.tapr.org/cgi-bin/mailman/listinfo/aprsspec
>