Wireless Bridge 4~5 miles

I am going to try to bridge a network 4-5 miles using wireless equitment. I
am looking to advice as what to buy at a reasonable price. I was planning on
using wap11 ap and clients with a parabolic or panel antenna such as this
http://cgi.ebay.com/ws/eBayISAPI.dll...745305365&rd=1

Will this work? Anyone have better advice?

Thanks!!

"root" <(E-Mail Removed)> wrote:
>I am going to try to bridge a network 4-5 miles using wireless equitment. I
>am looking to advice as what to buy at a reasonable price. I was planning on
>using wap11 ap and clients with a parabolic or panel antenna such as this
>http://cgi.ebay.com/ws/eBayISAPI.dll...745305365&rd=1
>
>Will this work? Anyone have better advice?
>
>Thanks!!

It depends on just how clear your line of sight actually is, and
perhaps on what kind of weather you experience.

If you can stand where one antenna is to be located and, using
10x binoculars, see the point where the distant antenna is to be
located, that's what it takes. Any trees, buildings, hills, or
whatever that get in the way will reduce your chances of it
working directly in proportion to just how much in the way they
are.

The problem with a 4-5 mile path is that losses are fairly high
anyway, so you don't have much room for "margin".

"Margin" can be defined as "Shit Happens". You'll get path
fades for certain types of weather conditions (heavy snow,
*really* heavy rain), atmospherics (inversion layers can reflect
radio signals), trees that are bushier in the summer, and odd
things like reflections from large vehicles temporarily in the
area between the antennas.

Essentially, if you set up a link that is just 5 dB above the
threshold, there will be times when the path has 10 dB more
loss, and your link will fail.

Higher gain antennas, with narrower beam width, are good. A
dish antenna is particularly good because it has fewer sidelobes
than many other types.

Also, in the US you can use some fairly high power transmitters
for wireless and still be quite legal. (Consider that the first
microwave I ever worked on cost more than $100,000 per site;
while by the 1970's it could be done for merely $10,000! And
today you can put in something with even better bandwidth for
less that $500 per site.) I saw an ad for a 1 watt transmit
amplifier and an 18 dB gain receive pre-amp all in one unit for
$200. A wifi AP, a 1 watt amplifier, a dish antenna (and buy
the best semi-rigid 1/2 inch coax you can to connect to the
antenna, and keep it as short as possible), and it should work
like a charm.

Note that commercial microwave links that cover distances like
30 miles use 1 to 2 watts of power but use low loss waveguide
rather than coax to connect to the antenna. If you can keep the
feed line short enough, with a 1 watt transmitter you have the
equivalent.

If you do have line of sight, that is.

--
Floyd L. Davidson <http://web.newsguy.com/floyd_davidson>
Ukpeagvik (Barrow, Alaska) (E-Mail Removed)

"Floyd L. Davidson" <(E-Mail Removed)> wrote in message
news:(E-Mail Removed)...
> "root" <(E-Mail Removed)> wrote:

>
> Also, in the US you can use some fairly high power transmitters
> for wireless and still be quite legal. (Consider that the first
> microwave I ever worked on cost more than $100,000 per site;
> while by the 1970's it could be done for merely $10,000! And
> today you can put in something with even better bandwidth for
> less that $500 per site.) I saw an ad for a 1 watt transmit
> amplifier and an 18 dB gain receive pre-amp all in one unit for
> $200. A wifi AP, a 1 watt amplifier, a dish antenna (and buy
> the best semi-rigid 1/2 inch coax you can to connect to the
> antenna, and keep it as short as possible), and it should work
> like a charm.
>

> If you do have line of sight, that is.
>
> --


I believe I do have line of sight. As the ground is lower between the two
points. I will have to double check this evening. Where do you sugest the
best place to pick up some of this equitment is?

Thanks

On Mon, 24 Jan 2005 16:35:50 GMT, "root" <(E-Mail Removed)> wrote:

>I am going to try to bridge a network 4-5 miles using wireless equitment. I
>am looking to advice as what to buy at a reasonable price. I was planning on
>using wap11 ap and clients with a parabolic or panel antenna such as this
>http://cgi.ebay.com/ws/eBayISAPI.dll...745305365&rd=1
>
>Will this work? Anyone have better advice?

$54/dish is a good price, but do you really need 5 of them?

See:
http://www.fab-corp.com
for a better selection of antennas.

I would not use the WAP11, but instead use WAP54G radios. The WAP11
client mode has a nasty habit of only talking to other WAP11 access
points.

If you don't want to waterproof your access points or deal with lossy
coax cable, try an integrated bridge/antenna combination:
http://www.ydi.com/products/bridges/index.php
http://www.ydi.com/products/bridges/bridge-in-a-box.php
Yeah, they much more expensive but they save time and effort in
construction.

If you wanna do it thyself, see:
http://www.sveasoft.com/articles/armored/
(about 7 pages. See tiny "next" in lower right).




--
Jeff Liebermann (E-Mail Removed)
150 Felker St #D http://www.LearnByDestroying.com
Santa Cruz CA 95060 AE6KS 831-336-2558

On Mon, 24 Jan 2005 18:01:46 GMT, "root" <(E-Mail Removed)> wrote:

>I believe I do have line of sight. As the ground is lower between the two
>points. I will have to double check this evening. Where do you sugest the
>best place to pick up some of this equitment is?

I don't believe it. At 5 miles, you'll have to be over about 20ft (my
guess, not calculated) high at each end to avoid hitting the earths
curvature and maintaining the Fresnel zone clearance. You'll need to
have more than just optical line of sight. You'll need to have
Fresnel zone clearance, which is about 2.1 meters radius at midpoint.
http://www.ydi.com/calculation/fresnel-zone.php

Your major problem might be interference. 24dBi dish antennas are a
good choice as they have a very narrow beamwidth. However, if you
have any source of interference between the endpoints or inline with
the path, you're going to have big problems. If you're shooting over
a metropolitan area, I don't think it will work well. If rural,
probably no problem. Do a site survey (sniff for other users) first.

Describe the area, path, bandwidth requirements, etc and I can run the
numbers. If you don't have optical line of sight, forget it. If you
have an obstructed path (trees, buildings), forget it.


--
Jeff Liebermann (E-Mail Removed)
150 Felker St #D http://www.LearnByDestroying.com
Santa Cruz CA 95060 AE6KS 831-336-2558

"Jeff Liebermann" <(E-Mail Removed)> wrote in message
news(E-Mail Removed)...
> On Mon, 24 Jan 2005 16:35:50 GMT, "root" <(E-Mail Removed)> wrote:
>
>>I am going to try to bridge a network 4-5 miles using wireless equitment.
>>I
>>am looking to advice as what to buy at a reasonable price. I was planning
>>on
>>using wap11 ap and clients with a parabolic or panel antenna such as this
>>http://cgi.ebay.com/ws/eBayISAPI.dll...745305365&rd=1
>>
>>Will this work? Anyone have better advice?
>
> $54/dish is a good price, but do you really need 5 of them?
>
> See:
> http://www.fab-corp.com
> for a better selection of antennas.
>
> I would not use the WAP11, but instead use WAP54G radios. The WAP11
> client mode has a nasty habit of only talking to other WAP11 access
> points.
>
> If you don't want to waterproof your access points or deal with lossy
> coax cable, try an integrated bridge/antenna combination:
> http://www.ydi.com/products/bridges/index.php
> http://www.ydi.com/products/bridges/bridge-in-a-box.php
> Yeah, they much more expensive but they save time and effort in
> construction.
>
> If you wanna do it thyself, see:
> http://www.sveasoft.com/articles/armored/
> (about 7 pages. See tiny "next" in lower right).
>

Very nice... exactly the kind of stuff I was looking for. Are there any
websites with expermentation of antenna's and their capablities?

"root" <(E-Mail Removed)> wrote:
>
>I believe I do have line of sight. As the ground is lower between the two
>points. I will have to double check this evening. Where do you sugest the
>best place to pick up some of this equitment is?

I have no idea what the best sources for you might be. Are you
located in Borneo, Tokyo, or closer to Moscow? :-)

The best advice I can give you is to use google and spend a few
hours reading up everything you can find on the subject.

--
Floyd L. Davidson <http://web.newsguy.com/floyd_davidson>
Ukpeagvik (Barrow, Alaska) (E-Mail Removed)

"Floyd L. Davidson" <(E-Mail Removed)> wrote
>
> It depends on just how clear your line of sight actually is, and
> perhaps on what kind of weather you experience.
>
> If you can stand where one antenna is to be located and, using
> 10x binoculars, see the point where the distant antenna is to be
> located, that's what it takes. Any trees, buildings, hills, or
> whatever that get in the way will reduce your chances of it
> working directly in proportion to just how much in the way they
> are.
>
> The problem with a 4-5 mile path is that losses are fairly high
> anyway, so you don't have much room for "margin".
>
> "Margin" can be defined as "Shit Happens". You'll get path
> fades for certain types of weather conditions (heavy snow,
> *really* heavy rain), atmospherics (inversion layers can reflect
> radio signals), trees that are bushier in the summer, and odd
> things like reflections from large vehicles temporarily in the
> area between the antennas.
>
> Essentially, if you set up a link that is just 5 dB above the
> threshold, there will be times when the path has 10 dB more
> loss, and your link will fail.
>
> Higher gain antennas, with narrower beam width, are good. A
> dish antenna is particularly good because it has fewer sidelobes
> than many other types.
>
> Also, in the US you can use some fairly high power transmitters
> for wireless and still be quite legal. (Consider that the first
> microwave I ever worked on cost more than $100,000 per site;
> while by the 1970's it could be done for merely $10,000! And
> today you can put in something with even better bandwidth for
> less that $500 per site.) I saw an ad for a 1 watt transmit
> amplifier and an 18 dB gain receive pre-amp all in one unit for
> $200. A wifi AP, a 1 watt amplifier, a dish antenna (and buy
> the best semi-rigid 1/2 inch coax you can to connect to the
> antenna, and keep it as short as possible), and it should work
> like a charm.
>
> Note that commercial microwave links that cover distances like
> 30 miles use 1 to 2 watts of power but use low loss waveguide
> rather than coax to connect to the antenna. If you can keep the
> feed line short enough, with a 1 watt transmitter you have the
> equivalent.
>
> If you do have line of sight, that is.

I don't have any need for 802.11 long-hauls, but these recents threads on
the subject has sparked an interest on the subject.

I wonder what type of 802.11 activity is going on within the ham radio
community. I would imagine that if you downconvert 802.11 to HF for
transmission, impressive distances could be obtained (?).

I haven't worked with any commercial microwave stuff, but have worked
tactical and fixed microwave (SHF) terminals when I was in the military.
To extend beyond the visual and radio horizons, these particular terminals
could crank out 1 to 2 KW in order to take advantage of the troposcatter
phenomenon. Links ranging between 100 to 150 miles were easily obtainable,
and (for the most part) pretty reliable and stable -- although they were
also subject to same limiting conditions as you noted (weather, temperature
inversions, ducting, multi-paths, terrain, etc). For the most part, they
were pretty rock solid. It wouldn't be uncommon for a terminal, over in
the desert, to be in operation for literally years. Once a link was
established, the most common "tweaking" I would have to do would be perhaps
to slightly raise the antennas (parabolics) in the early AM in order to peak
up on the signal. Even then, the receive signal levels were still strong
enough to keep the link. Peaking up just ensured that the bit error rate
didn't drop off. The connection between the antennas and the terminal were,
of course, waveguides. (Also worked with satellite long-haul terminals,
which required more coordination but didn't rely as much on RF "voodoo" as
troposcatter.)

One of these days I need to get involved with ham. HF's characteristics
aren't too familiar, so a lot of it comes across to me as "black magic". I
remember our HF guys regarding HF as much of an art as a science. Cool
stuff though!

"Floyd L. Davidson" <(E-Mail Removed)> wrote in message
news:(E-Mail Removed)...
> "root" <(E-Mail Removed)> wrote:
>>
>>I believe I do have line of sight. As the ground is lower between the two
>>points. I will have to double check this evening. Where do you sugest the
>>best place to pick up some of this equitment is?
>
> I have no idea what the best sources for you might be. Are you
> located in Borneo, Tokyo, or closer to Moscow? :-)
>


Im in a mountain valley called heber valley near SLC, Utah. The two
locations are one in Midway Utah, and the other in Daniels, Utah. Because
they are both kind of on the benches I believe line of sight is possible.

In article <(E-Mail Removed)>,
Jeff Liebermann <(E-Mail Removed)> wrote:
>On Mon, 24 Jan 2005 18:01:46 GMT, "root" <(E-Mail Removed)> wrote:
>
>>I believe I do have line of sight. As the ground is lower between the two
>>points. I will have to double check this evening. Where do you sugest the
>>best place to pick up some of this equitment is?
>
>I don't believe it. At 5 miles, you'll have to be over about 20ft (my
>guess, not calculated) high at each end to avoid hitting the earths
>curvature and maintaining the Fresnel zone clearance. You'll need to
>have more than just optical line of sight. You'll need to have
>Fresnel zone clearance, which is about 2.1 meters radius at midpoint.
> http://www.ydi.com/calculation/fresnel-zone.php
>
>Your major problem might be interference. 24dBi dish antennas are a
>good choice as they have a very narrow beamwidth. However, if you
>have any source of interference between the endpoints or inline with
>the path, you're going to have big problems. If you're shooting over
>a metropolitan area, I don't think it will work well. If rural,
>probably no problem. Do a site survey (sniff for other users) first.
>
>Describe the area, path, bandwidth requirements, etc and I can run the
>numbers. If you don't have optical line of sight, forget it. If you
>have an obstructed path (trees, buildings), forget it.
>
>
>--
>Jeff Liebermann (E-Mail Removed)
>150 Felker St #D http://www.LearnByDestroying.com
>Santa Cruz CA 95060 AE6KS 831-336-2558


FWIW Here's an "open source" design and parts kit for a 10mb optical
datacomm link system. The range is a little short but there may be a
speed/distance tradeoff or someone may have hackked a long range
version.

http://ronja.advel.cz/main.php

--

a d y k e s @ p a n i x . c o m

Don't blame me. I voted for Gore.