[aprssig] bad PHG, or lots of big towers?

[Robert Bruninga]

>>> daron at wilson.org 12/14/04 1:11:17 PM >>>
>We have several mountain top digis, that... hear [too much]
>...having one on a high mountaintop does not (by default) 
>make it the best choice.

Since this is HAM radio the fun part is designing systems.
One suggestion that might work is to put the really high
ones with a UHF output pointed back to an IGate.

This really high one hears all surrounding digis for 200 
miles and puts all that traffic down on a nearly 100% busy
UHF link to the IGate.  Thus users everywhere are 
only 2 hops from the IGate.  But the BIG cross band digi 
does not QRM any of the other ones.  for a drastic improvement.

If the BIG digi still hears too many collisions from all the
surrounding digis, then the next step is to put  up
four beams each with their own receive TNCs lower on
the tower so that each TNC (even though they are all
listening on 144.39 to the other digis) hears its preferreed
direction at least 10 dB better than any other direction.
This way all 4 surrounding digis can all TX at the same
time and yet the multiple TNC's and beams at the central
BIG digi will hear them all without collisoin and can combine 
them into the single UHF downlink to the IGate.

Lots of fun RF stuff to play with up there in the cold and
ice!  But it would allow a 4 fold improvement in reliability
for the IGate to capture those 2 hop packets from the
entire state...

ALSO, remember with the status quo, a station trying to
get a packet from area A to area C through the big digi B
which can hear A, C, D, E and F digis also will have a 
FAR worse chance of success getting throgh B on one
hop then he would by using the digi path of A-C.  This
is because A,B,and C can all hear each other and so
CSMA will work well. 

THus when the station trys to get into B direct, he is
competing with all the traffic on A,C,D,and F for a
quiet slot.  But when he tries to get into his local digi
A, he is only competing with B and C, not the
others.  And once he gets into A, then A only
has to complete with B and C to finish the 2 hop path.

In many cases, this might be more reliable...  Thus, the
loss of digipeating through B (if the UHF link is implemented)
is no big loss to the RF users in the area...

de WB4APR