Order Tray | Contact Us | Home | SIG Lists

[aprssig] Fundamental APRS capacity

Robert Bruninga bruninga at usna.edu
Mon Oct 4 15:23:04 UTC 2004


>> >Great explanation, and the [ALOHA] algorithm seems reasonable...
>> >How do you use this...to come up with a reasonable path....?
>>
>> Ah, that's easy.  Just have your software display the PHG
>> circles for the digis and then you can see how many hops it
takes...
>
>Easy for you to say.  I don't run DOS any longer...  You wanna
>contribute the relevant code to the xastir project?

I assume that Xastir can already plot PHG circles to assist the 
user in conjuring up real-time paths during normal operations.
If you are speaking of the ALOHA algorithm, I tried to discribe
it in pseudo language so that it would be easy to implement in
any language:

* 1200 baud can handle about 1800 packets per 30mins
* Initialize the number of copies "C" to one (1).
* Sort all stations heard on RF by range
* Starting with closest, add the number of packets per 30 mins
   from each type station:
  - Mobile in motion = 15 per 30 mins
  - Other Mobile      =  7 per 30 mins (use APRS1.1 list of symbols)
  - WX station         = 6 per 30 mins
  - DIGIs                =  3 per 30 mins
  - Home stations  =   2 per 30 mins

* Multiply each of these by the number of copies (C).
* C begins as 1, but as you encounter each digi, further
  and further away, increment C by 1 because of the added copy.
* When the cumulative total gets to 1800, STOP
* The range of that last station is the limit of the local
  network.  And an ALOHA circle should be plotted always on 
  every map view as a reminder to the user of his "real" maximum 
  possible communications range.  No, it is never a perfect
  circle, but this circle is only "representative".  The user 
  is supposed to have enough local sense to figure out how to
  apply the represetation to his area.

Remember also, that this is a LIBERAL maximum.  And
my guestimate is that it is about the 50% reliability circle.

>Besides, this is all theoretical at this point.

Well, the number of packets transmitted by those stations
is real.  And the limit of the 1200 baud chanel is real.
Now before someone says, "but I can ALWAYS communicate
far beyond that distance"...  remember that  this is TRUE,
but always AT SOMEONE ELSE'S EXPENSE.  A strong
station close to a digi will ALWAYS win and can get through,
but he is stepping on others to do so..

>My only tracker is a l[ow-power] pockettracker... It 
>wouldn't matter if I ran an obnoxious RELAY, WIDE9-9 
>when I can't even hit the RELAY, which is 10 miles away 
>on an 85 foot tower.  Those packets that do
>manage to hit the RELAY (when I drive right past the tower
> on the X-way) seem to make it to the internet without going 
>thru a WIDE at all...

Lots of ideas here:
1) It shows that high power next to a digi will always
    work (but only because it jams someone else)
2) This is a clear example of how putting up a 144.99
    alternate input (if available in your area) will eliminate
    95% of the competition that your tracker is having
    to contend with
3) You just proved the case that RELAY,WIDE7-7 will 
    be no more reliable than just RELAY

Thanks
Bob





More information about the aprssig mailing list