passive reflectors

[Tony]

Robert
I think you need to get out and try this yourself :) Coming from a
background of doing licensed link this can me done. While this will not
work in all cases, if does work for shorter link. Using ISM equipment we
have made this work at 3 mile.

Sincerely
Tony 

-----Original Message-----
From: wireless-admin at lists.samba.org
[mailto:wireless-admin at lists.samba.org] On Behalf Of robert scholten
Sent: Sunday, June 09, 2002 6:02 PM
To: wireless at lists.samba.org
Cc: eqeeu at yahoo.com
Subject: passive reflectors


This passive reflector stuff is baloney.  It's basic inverse-square law.
Consider a setup with 2 dishes, e.g. 24dB Conifers.  Beam widths are 7.5
and 10 degrees (-3dB widths).  Area is about 0.5m^2.
Imagine you have a really good link with two 24dB dishes over 5 miles =
8km.  Let's say you get 1 nanowatt (i.e. -60dBm).  I've managed about
-80dBm from a good site with two 24dB dishes, one site on the side of a
small mountain, the other on the plains, length about 20km.  That
translates to -73dBm for 8km, but let's be generous to Cringely.
Now let's put a passive reflector at that point, instead of a single
dish.  I.e. a reflector consisting not of Pringles tins, but good 24dB
dishes with zero loss cable between them.  Ignore the fact that both
dishes in the reflector will radiate (i.e. instant 3dB loss).  So the
second dish re-radiates 1nW of power into a "cone" of 7.5 x 10 degrees.
That power is spread over an area of approx. d*tan(a1)*tan(a2) where a1
and a2 are the angles, 7.5 and 10 degrees.  At 1000m (again, let's be
generous to Cringely and assume 1.5 miles is 1000m!).  The power is
spread into 23000m^2 so at our final 24dB dish we have 0.5/23000 * 1nW =
2e-14 watts, or about -107dBm.  More realistically you should use -73dBm
at the repeater, include at least 3dB loss in the repeater, add another
3dB because he's 1.5 miles away from the mountain (2km), and come up
with a very optimistic -119dBm.  The best cards (Cisco 350's) are good
down to about -94dB so it's pretty much irrelevant anyway.
How big would Cringely's pringles tins need to be to make it work?
Optimistically he'd need an extra 11dB, i.e. 5dB on each side, so 30dB
(about 1.5m^2).  More realistically he'd be needing a good 25dB, 12.5dB
per side, 9m^2.  Dishes of 3.4 m diameter (about 11 feet).
Forget it!



To: Eqeeu <eqeeu at yahoo.com>
Cc: wireless at lists.samba.org
Subject: Piercing a mountain? Help!


Eqeeu writes:
 > Hi
 > 
 > I want to setup a 802.11b link between my house and a
 > friends, the only problem is that there is a large
 > mountain (I think its Ainslie mountain) smack in the
 > middle of our two houses and its impossible to get
 > line of site.
 > 
 > Is this a problem?
 > 
 > Would we have to go around or over the mountain with
 > an extra repeater WAP somewhere, or is it possible to
 > use two high gain 24dBi+ directional antenna and
 > "pierce" straight through the centre of the mountain?
 > 


An article on doing passive repeaters is at :


 http://www.pbs.org/cringely/pulpit/pulpit20020207.html.


I think you can get more doco. from there.


wes




Do You Yahoo!?
Sign-up for Video Highlights of 2002 FIFA World Cup