How to cover our campus

I do some small time IT work for a school in Honolulu, HI and they
decided to provide wireless on the campus. The school in on 3.5
acres, and has tons of trees, ect. we want to cover the whole campus,
but don't want 1) to setup more than 2 access points 2)melt the paint
off our the houses in the neighborhood by over driving our signal.
Where can I find recommendations on how to best setup our campus? A
local installer recommened an antenna with 12bd gain and a 250 mw amp,
but i want to stay legal. Thanks for your help, Cal-

In article <(E-Mail Removed) >,
Calvin Paxson <(E-Mail Removed)> wrote:
:I do some small time IT work for a school in Honolulu, HI and they
:decided to provide wireless on the campus. The school in on 3.5
:acres, and has tons of trees, ect. we want to cover the whole campus,
:but don't want 1) to setup more than 2 access points 2)melt the paint
ff our the houses in the neighborhood by over driving our signal.
:Where can I find recommendations on how to best setup our campus? A
:local installer recommened an antenna with 12bd gain and a 250 mw amp,
:but i want to stay legal. Thanks for your help, Cal-

Remember that the signals from the mobile systems have to be
recievable at the antenna. Most kinds of trees are quite effective
absorbers of wireless signals. I gather that even with a
fairly good omnidirectional receiving antenna, that your mobile signals
are probably only going to survive one or two trees in the path.

What I would conclude is that if you want to have a chance of doing
what you want, then you are going to have to put the APs quite
high -- high enough that everywhere on campus can see the
receiver by looking -up- instead of trying to go across through
tree after tree.

You'd be better off if you could require a parabolic antenna on
each mobile unit, that had to be adjusted to point to the AP.
You probably can't do that, though. And if there are tons of
trees, then unless they happen to be particularily spindly (few leaves)
I would tend to doubt that 2 AP is going to be sufficient for you.
--
millihamlet: the average coherency of prose created by a single monkey
typing randomly on a keyboard. Usenet postings may be rated in mHl.
-- Walter Roberson

(E-Mail Removed) (Calvin Paxson) wrote in news:8f962c50.0311051829.28b94307
@posting.google.com:

> I do some small time IT work for a school in Honolulu, HI and they
> decided to provide wireless on the campus. The school in on 3.5
> acres, and has tons of trees, ect. we want to cover the whole campus,
> but don't want 1) to setup more than 2 access points

make it 20

> 2)melt the paint
> off our the houses in the neighborhood by over driving our signal.

forget the amps, good antenna is far better than any amp

Pozdrawiam.
--
RusH // [502-20-14-27 tylko SMS]
http://kiti.pulse.pdi.net/qv30/ <-- to prawdziwy ja
Pent-up passive-aggressive dork alert! Whoop! Whoop!
Whoop! Whoop! Boy, you're really lighting up this alarm here!

> Remember that the signals from the mobile systems have to be
> recievable at the antenna. Most kinds of trees are quite effective
> absorbers of wireless signals. I gather that even with a
> fairly good omnidirectional receiving antenna, that your mobile signals
> are probably only going to survive one or two trees in the path.

What about a return amp? so that i am only amping in singnal that is
incoming at the ap? Does such a thing exist? I see that many of the
amplifiers out there do 2 way amplification, switching between tx and
rx, is their one that is a receive amp only?

Thank again, cal

On 7 Nov 2003 01:58:25 -0800, (E-Mail Removed) (Calvin Paxson) wrote:

>> Remember that the signals from the mobile systems have to be
>> recievable at the antenna. Most kinds of trees are quite effective
>> absorbers of wireless signals. I gather that even with a
>> fairly good omnidirectional receiving antenna, that your mobile signals
>> are probably only going to survive one or two trees in the path.
>
>What about a return amp? so that i am only amping in singnal that is
>incoming at the ap? Does such a thing exist? I see that many of the
>amplifiers out there do 2 way amplification, switching between tx and
>rx, is their one that is a receive amp only?
>
>Thank again, cal

Doesn't help the clients at all. Most of the time they don't help the AP
because they also amplify noise just as well as signal, and if not well
made will add a major amount of noise. If you have a good antenna, the
problem is not often signal strength, but signal to noise ration. If you
only have a 10db SNR, amplifying it still leaves you with a 10db SNR. A
better antenna improves that by picking up less noise. All the noise that
is not in its pattern is not picked up.

Reading this thread, there is no way you are going to reliably cover campus
with only 2 APs. One point that hasn't been addressed is how many
connections do you think you will have at any one time? Even with very good
signals, at about 10 connections you are probably back down to modem
speeds. Add in a couple of weak signals that connect at 2mbs, and your
performance goes down real fast. So how many people do you expect to be
connected at one time Peak? Average? That is as important, if not more so,
to figuring out what you need. What level of performance is going to be
acceptable? No sense being worse than a 56k modem.

JT

In article <(E-Mail Removed)> ,
Calvin Paxson <(E-Mail Removed)> wrote:
[quoting me]
:> Remember that the signals from the mobile systems have to be
:> recievable at the antenna. Most kinds of trees are quite effective
:> absorbers of wireless signals. I gather that even with a
:> fairly good omnidirectional receiving antenna, that your mobile signals
:> are probably only going to survive one or two trees in the path.

:What about a return amp? so that i am only amping in singnal that is
:incoming at the ap? Does such a thing exist? I see that many of the
:amplifiers out there do 2 way amplification, switching between tx and
:rx, is their one that is a receive amp only?

Yes, I believe receive-only amps do exist, but recall that they
are amplifying the noise along with the signals. The trees are not
just attenuating the signal (leaving a weaker but distinct signal),
they are scattering it badly (leaving a weaker and noisy signal).

We can't give you any figures about how many trees your signal
could usefully survive: it would depend upon the density of the leaf
covering, on the thickness and water content of the leaves, on the
time of year (different water contents at different times of year),
on the exact angle the signal would have to go through the leaves,
on the distance from the source to the leaves (closer in blocks
a greater angular radius and is blocking a higher power signal).
Then there's the effect of bounces off of existing buildings... and
cars and park benches...

You mentioned having lots of trees, but you didn't say anything about
what kind of trees they are. If they are all towering trees whose
branches start 100 feet up, then you would have clear line of sight
below the branches, but if they are more like thick bushes then
they are going to block the signal quite thoroughly.
--
Cannot open .signature: Permission denied

Super. thanks for all the info guys. we only at this time expect to
have 3 to 4 clients connecting at the same time. as it grows we will
add more ap's. we will put up a good antenna and see what we can do
with it.

Mahalo, cal