wireless relays

> I don't have access to cable or dsl. My friend down the
> road (we are in a rural hilly area north of Knoxville TN)
> does have cable. It stopped about 1/2 mile from my house. I
> am exploring some way to network with his house to get
> higher speed. I discussed it with the cable company and was
> informed that due to cost they would not be extending it my
> way any soon and the person I talked to didn't see any
> problem with the sharing, since they would/could not
> provide it.

Extending CATV about 1/2 mile is not a big deal. At worst, you'll
need a line amplifier. Instead of connecting to your friend, is there
someone near the end of the cable that can be convinced to attach a
connection?

> I am still looking into some type of wire connection, it is
> about 1.2 miles.

Lots of ways to do 1.2 miles. The easiest is 2 twisted pair and a
pair of SDSL modems. Not all SDSL modems will work. I have a list
somewhere. The catch is that you may be limited to about 3Mbits/sec
thruput.

> I am going to check with BellSouth about
> getting a dry pair that is looped out here somewhere. If it
> goes to the CO, it will be too far to be of any use. I am
> about 10 mitles from there. But from what I have read,
> this is probably a no-go.

Good luck. We can't get dry pairs in SBC country at all. Totally
impossible including alarm companies. I've tried. All drops go via
the CO. Sorry.

> I am also going to check with the utility company about
> stringing my own cable. Don't know how that will go, or
> what type of expense I am looking at, but asking is free.

I've done it for short distances without permission. They don't like
it but it's too big a bother to tear them all down. As long as it's
done professionally, they don't seem to care. I've also strung fiber
legally on the former Pacific Bell poles. The cost was nominal (about
$2/pole) to the utilities, but the insurance costs was really awful.
Something like $20,000/year for $10million in liability. That's how
they keep independents off the poles legally.

> That leaves me wireless. I can get line of sight from the
> top of the ridge behind me. I own the property but there is
> no power anywhere around.

No problem. Most wireless repeaters only burn about 5 watts maximum.
A gel cell and a solar charger can handle that.

> The idea of the passive repeaters
> sound really good.

No. They don't work. Please re-read whatever I wrote. The beamwidth
necessary at both ends is very small. The reflector will by necessity
be VERY big.

> And I see posts all over the place
> recommending them, but your math and reasoning looked
> pretty sound at ruling it out in the basic case.
> Are these people recommending something that just sounds good but they
> have not actually tried to get it to work?

I've gotten it to work under limited conditions. One used a 4ft
diamter dish at each end. The reflector was a solid piece of 3x6ft
aluminium sheet metal. Losses were horrible, but it was good enough.
Big problem in the morning when water condensed on the reflector and
screwed up the signal.

> In my case I would estimate 1 mi from the relay to my friend
> (A) and . 5 mi from the relay to me (B). I used the page you
> referred to to do some calculations, but am not sure if I
> did things correctly. If I wanted a passive repeater to work,
> how much power would I need if I had 24dbi antennas all
> around? Is it feasible with amplifiers?

Maybe. Let's see your calculations and I'll check them. However, I
would prefer you re-post your question in alt.internet.wireless so I
can deal with it at my leasure.

> How about if I put a bi-directional amplifier between the
> relay antennas and powered that with a battery/solar
> charger. That should require such little power I don't
> think it would be a problem. A 250mw doesn't seem too expensive.

A 250mw (24dBm) amplifier at each end will improve the signal over the
typical +15dBm radio by about 9dB. That's quite a bit and will help.
However, the beamwidth will kill you no mattery how much power you
have. A 6degree beamwidth 24dBi antenna at 0.6 miles (3200ft) will be
over 300ft wide. Got a 300ft wide reflector? Cut the area in half
and you lose 3dB. Half again, another 3dB. Do it often enough and
you have no signal left.

My question about the amplifier is in putting it in the field with the
relay. Instead of having a totally passive relay, hook up a
bidirectional amplifier between the 2 antennas. It seems that shooting
the 1 mile with 2 good antennas. But it has been shown there is not
enought signal to feed into the next antenna as a passive relay. Would
amplifying the signal then feeding that into the second antenna provide
enough power to be usable.



radio - antenna -------------------- antenna - amp - antenna
-------------------- antenna - radio

But even if this works, I ran across another fact that might be my
downfall. I read a post discussing the ACK time for 802.11 and they
showed how 1 mi was about the limit due to timing. I couldn't tell if
they meant that just for the 11MB or also applied to the lower speeds.

As for me if I can get 2MB or better I would be happy.

On 6 Feb 2006 19:15:23 -0800, "TKMitchell"
<(E-Mail Removed)> wrote:

>My question about the amplifier is in putting it in the field with the
>relay. Instead of having a totally passive relay, hook up a
>bidirectional amplifier between the 2 antennas. It seems that shooting
>the 1 mile with 2 good antennas. But it has been shown there is not
>enought signal to feed into the next antenna as a passive relay. Would
>amplifying the signal then feeding that into the second antenna provide
>enough power to be usable.
>radio - antenna -------------------- antenna - amp - antenna
>-------------------- antenna - radio

Ok, let's do the math.

Let's pretend we have a symmetrical bi-directional amplfier with
perhaps 15dB of gain. I don't know if such a thing exists, but you
can use Google to find something. Remember, it has to be symmetrical,
not a bi-directional amplifier that will belch lots of power in one
direction, and just run the receiver in the other. It also has to
have it's own transmit/receive switching mechanism to switch
direction. Let me know if you find something similar.

Running it through the link calculations at:
| http://en.wikibooks.org/wiki/FAQ_for...k_Calculations
I would guess we have:

TX power = +17dBm
TX coax loss = 4dB (3ft LMR-240 plus a mess of connectors)
TX ant gain = 24dBi parabolic dish
Distance = 0.6 miles
RX ant gain = 24dBi parabolic dish
RX coax loss = 4dB (same at other end)
RX level = unknown

Plugging into:
http://www.terabeam.com/support/calculations/som.php
I get an RX level of -44.8dBm.
With a 15dB amplifier, the tx level is now:
-44.8 +15 = -29.8dBm or -30dBm (close enough)

Now, we do the other half of the link:
TX power = -30dBm
TX coax loss = 4dB (3ft LMR-240 plus a mess of connectors)
TX ant gain = 24dBi parabolic dish
Distance = 0.6 miles
RX ant gain = 24dBi parabolic dish
RX coax loss = 4dB (same at other end)
RX level = unknown

Plugging into:
http://www.terabeam.com/support/calculations/som.php
again we get a receive level of about -90dBm.

In order for such a link to function, we need a minimum fade margin of
20dB. Less will work, but not very reliably. The basic receiver
sensitivity is about -84dBm (at 11Mbits/sec) which yields a fade
margin of -6dB. That's 26dB short of being useable. If you've got a
way of gaining 26dB more gain in there, it might work. A 1 watt
(maximum legal) power amplifier on both ends will gain about 13dB but
that's still not enough.

Conclusion: Unless I screwed up the math (it happens), a
bi-directional amplifier won't work.

>But even if this works, I ran across another fact that might be my
>downfall. I read a post discussing the ACK time for 802.11 and they
>showed how 1 mi was about the limit due to timing. I couldn't tell if
>they meant that just for the 11MB or also applied to the lower speeds.

It is my understanding that the timing limit is about 10 miles. There
are numerous links running between 1 mile and further, that work just
fine.

>As for me if I can get 2MB or better I would be happy.

Find a Wi-Fi repeater, range extender, or whatever. Install on the
hilltop. Solar power with a gel cell, charge controller, and possibly
a DC to DC regulator. Most such repeaters suck about 5 watts. Try
bashing the numbers into my solar calculation speadsheet at:
http://802.11junk.com/jeffl/rf-calc/...peater-206.xls
The yellow numbers are the one's you can tweak. I'll help after I
throw all the visitors out of my palatial office.

--
# Jeff Liebermann 150 Felker St #D Santa Cruz CA 95060
# 831-336-2558 (E-Mail Removed)
# http://802.11junk.com (E-Mail Removed)
# http://www.LearnByDestroying.com AE6KS

My question about the amplifier is in putting it in the field with the
relay. Instead of having a totally passive relay, hook up a
bidirectional amplifier between the 2 antennas. It seems that shooting
the 1 mile with 2 good antennas. But it has been shown there is not
enought signal to feed into the next antenna as a passive relay. Would
amplifying the signal then feeding that into the second antenna provide
enough power to be usable.



radio - antenna -------------------- antenna - amp - antenna
-------------------- antenna - radio

But even if this works, I ran across another fact that might be my
downfall. I read a post discussing the ACK time for 802.11 and they
showed how 1 mi was about the limit due to timing. I couldn't tell if
they meant that just for the 11MB or also applied to the lower speeds.

As for me if I can get 2MB or better on the link I would be happy.

My question about the amplifier is in putting it in the field with the
relay. Instead of having a totally passive relay, hook up a
bidirectional amplifier between the 2 antennas. It seems that shooting
the 1 mile with 2 good antennas. But it has been shown there is not
enought signal to feed into the next antenna as a passive relay. Would
amplifying the signal then feeding that into the second antenna provide
enough power to be usable.



radio - antenna -------------------- antenna - amp - antenna
-------------------- antenna - radio

But even if this works, I ran across another fact that might be my
downfall. I read a post discussing the ACK time for 802.11 and they
showed how 1 mi was about the limit due to timing. I couldn't tell if
they meant that just for the 11MB or also applied to the lower speeds.

As for me if I can get 2MB or better on the link I would be happy.