Re: [PLUG] Wireless network - Swarthmore

Chris "Magnus" Hedemark <chrish@trilug.org> · 2003年6月11日 12:55:03 -0400

-----BEGIN PGP SIGNED MESSAGE-----
Hash: SHA1

On Wednesday, June 11, 2003, at 11:27 AM, William H. Magill wrote:
I doubt that you will get any commercial roof space for free, 
especially in the city.

I've already gotten four commercial roof tops, though none yet in the 
city (keeping in mind I haven't *tried* in the city yet)
It is a well known fact that roof tops are prime real estate for Cell 
sites, and agents charge accordingly. And any "reasonably situated" 
building has been contacted many times since the cell build-out began 
several years ago.

Yep. One of the four roof tops I've gotten has already got a cell 
phone site on it, but there is no exclusive arrangement, and there is 
plenty of room for additional antennas.
You allude to the back-haul problem at the central hub... each 
suburban site requires a dedicated hub site.

Yes & no.
Setting up a backbone network will greatly reduce the number of hops 
from town to town within then network, thus greatly improving latency 
when traversing the network. Also, early on, the node operators are 
rather spread out and we won't have a chain of omnidirectional hops 
from town to town for quite some time (if ever). One or more dedicated 
hub sites per town would probably be the best way to get all of the 
towns linked but not necessarily the only way.
So with a 2 antenna limit per AP, that means the central hub is going 
to need a lot of real estate for not only multiple antennas, but for 
multiple electronics as well.

Negligible. The electronics all fit into a small box that can be pole 
mounted. Having gotten a good close look at a cell site yesterday, and 
the associated electronics, I can tell you with great confidence that 
our requirements will be far less. Indeed we should be able to fit all 
electronics and antennas on a single 15'mast if we were forced to. 
Ideally the antennas should be at least a little more spread out but it 
is still a relatively small amount of real estate.
Also, long-haul directional antennas need to be substantial devices. 
They have to be capable of being aimed precisely and then anchored in 
such a way as to preclude ANY movement of the antenna.

Anchoring a mast to a brick wall strikes me as a pretty solid mounting 
surface.
Since the slightest movement of the antenna will cause a tight-beam to 
shift dramatically at the receiving end over 15 miles, and a wide-beam 
requires much more transmitter power for the same distance.

Understood.
Don't forget, that the proximity of the AP to the Antenna is an 
important factor. Transmission losses attributable to the line from 
the AP to the Antenna can be just as devastating as foliage and fog.

Yep. There is no reason for the AP to be more than 2 meters from the 
antenna. Likely even closer. The idea is to put the AP in a NEMA box, 
mount it on the pole right next to the antenna.
- --Magnus
-----BEGIN PGP SIGNATURE-----
Version: GnuPG v1.3.2 (Darwin)
iEYEARECAAYFAj7nXusACgkQYPuF4Zq9lvZF5gCgxmA/I4dDq0er7VxJVxsPc/SZ
YKEAoK20kuGw0Ip3rEDHDTwtsPTl1Faz
=D5Wu
-----END PGP SIGNATURE-----
_________________________________________________________________________
Philadelphia Linux Users Group -- http://www.phillylinux.org
Announcements - http://lists.netisland.net/mailman/listinfo/plug-announce
General Discussion -- http://lists.netisland.net/mailman/listinfo/plug 