Bounce off a condo tower

[jeremy]

Archived from groups: alt.internet.wireless (More info?)

I'm looking to connect my WAP to a client less than 1 km away. We
would have perfect line of sight except for the senior's complex
directly in between. I understand that this is a complete no-go.
However, approximately 1/2 way between and off to the side there is a
cluster of large condos (mostly concrete & glass on the outside) that
I think I can bounce a signal off of.
My questions are:
1. Point A (WAP) is on the 2nd floor, point B (client) is on the 4th,
but in order to clear some trees and other houses, we'd have to aim
for about the 10-15th floors of the condo. Principles of reflection
tell me my signal will end up several hundred feet above my target. Am
I going to end up with enough "splatter" to make it back down?
2. What sort of additional hardware will I have to work with for a
reasonable test? Can I use a cantenna-type deal on both ends and
expect a decent result?
3. If I put a directional antenna on the WAP, will I still be able to
access it locally (client within the same room) even though the
antenna is pointing in the other direction?
4. How much better will my result be if I befriend someone in the
condo and get him to put in an omni?

Sorry if these issues have been beaten to death already.

TIA
Jeremy.


--There are no stupid questions, but there are a lot of inquisitive
idiots.

 

[Guest]

Archived from groups: alt.internet.wireless (More info?)

On 2 Dec 2004 08:43:23 -0800, jeremy.haydraude@gmail.com (Jeremy)
wrote:

>I'm looking to connect my WAP to a client less than 1 km away. We
>would have perfect line of sight except for the senior's complex
>directly in between. I understand that this is a complete no-go.
>However, approximately 1/2 way between and off to the side there is a
>cluster of large condos (mostly concrete & glass on the outside) that
>I think I can bounce a signal off of.

Won't work. Do the math. Let's say your 24dBi dish antenna has a
-3dB beamwidth of 8 degrees. That's half your power going into 8
degrees. At 500 meters away, your reflector would need to be:
diameter = 2 * tan(4) * 500 = 70 meters
That's a circular area of:
Pi * (35 meters)^2 = 3900 sq meters
to capture only half the power from the antenna. If you prepare a
realistic midpoint reflector of perhaps 1 meter square (ignoring
reflection angle), your reflector would capture only 1/3900th of the
signal resulting in a loss of:
loss = 10*log(1/3900) = -36dB
Reflection efficiency is at best 50% (-3dB). So, our total loss
coming off the reflector is:
3 + 36 + 3 = -42dB loss
Now, that's the loss in *addition* to the path loss between endpoints.
It won't work simply because you can't build an antenna narrow enough
to concentrate the signal on a reflector, which you can't build large
enough to capture all the signal.

>My questions are:
>1. Point A (WAP) is on the 2nd floor, point B (client) is on the 4th,
>but in order to clear some trees and other houses, we'd have to aim
>for about the 10-15th floors of the condo. Principles of reflection
>tell me my signal will end up several hundred feet above my target. Am
>I going to end up with enough "splatter" to make it back down?

No. Search the net for calculations for "periscope antennas".
http://www.w1ghz.org/antbook/chap8.pdf

>2. What sort of additional hardware will I have to work with for a
>reasonable test? Can I use a cantenna-type deal on both ends and
>expect a decent result?

A cantenna has a best case gain of about 8dBi and a beamwidth of about
100 degrees. You need a narrower beamwidth.
http://members.cruzio.com/~jeffl/pics/antennas/coffee2400/index.html

>3. If I put a directional antenna on the WAP, will I still be able to
>access it locally (client within the same room) even though the
>antenna is pointing in the other direction?

In theory, no because most of the energy is going toward your midpoint
reflector. In reality, most barbeque grill dish antennas leak badly
(sidelobes) and you'll have enough signal to for nearby
communications. You won't get far, but near the dish, it will work.

>4. How much better will my result be if I befriend someone in the
>condo and get him to put in an omni?

Much better assuming you both have line of sight.

--
Jeff Liebermann jeffl@comix.santa-cruz.ca.us
150 Felker St #D http://www.LearnByDestroying.com
Santa Cruz CA 95060 AE6KS 831-336-2558

 

[Guest]

Archived from groups: alt.internet.wireless (More info?)

On Thu, 02 Dec 2004 09:15:26 -0800, Jeff Liebermann
<jeffl@comix.santa-cruz.ca.us> wrote:

>On 2 Dec 2004 08:43:23 -0800, jeremy.haydraude@gmail.com (Jeremy)
>wrote:
>
>>I'm looking to connect my WAP to a client less than 1 km away. We
>>would have perfect line of sight except for the senior's complex
>>directly in between. I understand that this is a complete no-go.
>>However, approximately 1/2 way between and off to the side there is a
>>cluster of large condos (mostly concrete & glass on the outside) that
>>I think I can bounce a signal off of.
>
>Won't work. Do the math. Let's say your 24dBi dish antenna has a
>-3dB beamwidth of 8 degrees. That's half your power going into 8
>degrees. At 500 meters away, your reflector would need to be:
> diameter = 2 * tan(4) * 500 = 70 meters
>That's a circular area of:
> Pi * (35 meters)^2 = 3900 sq meters
>to capture only half the power from the antenna. If you prepare a
>realistic midpoint reflector of perhaps 1 meter square (ignoring
>reflection angle), your reflector would capture only 1/3900th of the
>signal resulting in a loss of:
> loss = 10*log(1/3900) = -36dB
>Reflection efficiency is at best 50% (-3dB). So, our total loss
>coming off the reflector is:
> 3 + 36 + 3 = -42dB loss
>Now, that's the loss in *addition* to the path loss between endpoints.
>It won't work simply because you can't build an antenna narrow enough
>to concentrate the signal on a reflector, which you can't build large
>enough to capture all the signal.
>
>>My questions are:
>>1. Point A (WAP) is on the 2nd floor, point B (client) is on the 4th,
>>but in order to clear some trees and other houses, we'd have to aim
>>for about the 10-15th floors of the condo. Principles of reflection
>>tell me my signal will end up several hundred feet above my target. Am
>>I going to end up with enough "splatter" to make it back down?
>
>No. Search the net for calculations for "periscope antennas".
> http://www.w1ghz.org/antbook/chap8.pdf
>
>>2. What sort of additional hardware will I have to work with for a
>>reasonable test? Can I use a cantenna-type deal on both ends and
>>expect a decent result?
>
>A cantenna has a best case gain of about 8dBi and a beamwidth of about
>100 degrees. You need a narrower beamwidth.
> http://members.cruzio.com/~jeffl/pics/antennas/coffee2400/index.html
>
>>3. If I put a directional antenna on the WAP, will I still be able to
>>access it locally (client within the same room) even though the
>>antenna is pointing in the other direction?
>
>In theory, no because most of the energy is going toward your midpoint
>reflector. In reality, most barbeque grill dish antennas leak badly
>(sidelobes) and you'll have enough signal to for nearby
>communications. You won't get far, but near the dish, it will work.
>
>>4. How much better will my result be if I befriend someone in the
>>condo and get him to put in an omni?
>
>Much better assuming you both have line of sight.

--
Jeff Liebermann jeffl@comix.santa-cruz.ca.us
150 Felker St #D http://www.LearnByDestroying.com
Santa Cruz CA 95060 AE6KS 831-336-2558