On Sat, 14 Jun 2008 06:55:28 GMT, "William R. Cousert"
<(E-Mail Removed)> wrote:
>http://www.apple.com/airportextreme/
>
>Would it be possible to blanket a city or metropolitan area with these
>things and build a city wide intranet?
Yes. The procedure is well established. This is the perfect time
because it's an election year. Simply purchase a suitable politician
searching for re-election. The politician proposes to supply *FREE*
wireless internet to those that need it least (low income, poverty,
etc) and bury the rest of the users in advertising. The politician
will use it as a campaign promise. There were previously several
companies willing to actually setup and install such a system. While
Earthlink has dropped out, there are still a few that seem to think
they can make money on free wireless.
Once you have your politician and free service company in place, you
need to generate interest in the project. Press releases, promises,
and claims should keep a PR person busy. Lies and science fiction
performance claims are acceptable as nobody really understands the
technology.
>It appears that the major ISP's are about to switch to metered service
>(charge for each gigabyte transferred).
Yep. The fat pipe just isn't fat enough for P2P, IPTV, and other high
bandwidth services. We're well past the point were user restraint and
common sense can be relied upon to prevent overload.
>I'd like to use 802.11n to create an alternative to the Internet.
802.11n is all about speed, not range. Typical range for the slowest
802.11n speeds are about 30-50 meters maximum (depending mostly on
antenna gain). To supply such speeds, you'll need a huge number of
Apple Airport Extreme routers. The current science fiction technology
for 802.11b/g is about 3-7 access points per square mile. That's what
the initial technical proposals usually contain. When installed, it
usually turns out to be more like 40-50 per square mile. If you
insist on 802.11n, it will be even more. How many square miles were
you thinking of covering?
>Apple's Airport Extreme can handle 50 connections... What if you were to
>create groups of 25
>Airport devices, each with a persistant connection with each of the
>remaining 24, plus a connection to a neighboring group of 25 Airport
>devices. This leaves each Airport with 25 open channels for wireless users.
Real outdoor access points are made to handle 2000 MAC addresses. Not
all of these are active, but the MAC address table has to be
sufficiently large to handle this number. In terms of traffic, the
rule of thumb is:
100 users doing light email and web browsing.
10 business users doing whatever business users do.
1 file sharing user hogging ALL the bandwidth
In short, the Airport router is not designed for a large number of
users. I just checked an AP in the middle of a large shopping mall.
It shows 15 active connections and about 80 inactive MAC addresses.
That's about typical.
>Could this work?
Nope. Never mind the political, financial, or logistical reasons. It
won't work on the technical reasons. Figure out what you want to do
first. Then figure out what equipment you need. Very last, you
figure out what vendors hardware you're going to buy.
>If not, are there any low cost (< $300 each person?) alternatives that could
>scale up to a large city?
Actually yes. See any of the WiMax solutions. They're made for doing
exactly that. The problem is that you need to buy or rent a very
expensive WiMax license. Think millions of dollars for just the
license. Despite this, Sprint-Nextel and Clearwire are throwing
millions into WiMax.
--
Jeff Liebermann
(E-Mail Removed)
150 Felker St #D
http://www.LearnByDestroying.com
Santa Cruz CA 95060
http://802.11junk.com
Skype: JeffLiebermann AE6KS 831-336-2558