1KM 802.11b @ 2MB
OffTheRack writes "Check out this web site to see how a guy in Egypt built his own line-of-site (H:Get? It's Punny.) 1KM broadband connection. Plenty of nice pictures." Pretty cool set-up.
← Back to Stories (view on slashdot.org)
Although I live in a small-ish town with presumably little electromagnetic interference, it is worth mentioning that this high-gain, 3 meter tall GSM network monstrosity is located about 7 meters away from my AP antenna, on the same roof. I really can't tell whether or not it is affecting my antenna.
I think the more important question would be, "Is your unlicensed amplified antenna affecting their transmissions"
http://trevormarshall.com/biquad.htm
I read direct TV dishes are a lot easier to build and go alot further.
Linksys outputs 50mw. If you were to go to the FCC regulation maximum of 100mw (1 watt), you would do better.
If you use Cisco's Wireless Calculator Excel spreadsheet, you will see that a 21 dbi parabolic dish (as seen here could grant you 76.1 kilometers.
Now you have problems with this distance. Line of sight, for one.
The common calculation calls for a cone shaped space from each antenna meeting at the middle. This is called the Fresnel Zone The calculator says that this zone, for 76km is 28.2 meters (92 feet). So, you have to have 92 feet of clearance. No trees. No buildings. Nothing.
At this distance (44.5 miles) The earth curvature is 90 feet, somehow your signal must peak over that!
So, you are going to need antenna heights of over 150 meters to set up a 802.11b wireless like of this long.
Most engineers say this is simply not feasable. Other issues, such as antenna alighment, water (nasty multipath bounce), WiFi hotspot interferance, and the actual time it takes for a signal to travel that far are issues as well.
So just remember line of sight. If you have clean line of sight to the other endpoint, and clarity through the Fresenel Zone, you might just have a chance for point-point 802.11b wireless.
Cheers!
I may be way off base here...but I was under the impression that 128 bit encrypted products couldn't be exported without a permit of some kind. The author mentions that a friend in California brought the products to Egypt. Wouldn't that be illegal to export such a product?
And yes, I do find it cool what he did... just curious about the legality by US laws.
Mike
I'm using a HyperLink omni and amp on a 20 foot mast on
top of a two story farmhouse. The amp is fed by an Airport
base station. My rev. A iBook gets 2mb from 2 miles away
with it's internal antenna if it can LOS the antenna.
I'm in rural flatland Indiana so it's pretty easy to spot the
house and antenna from far away.
I'm running below legal power limit but I'm feeding the
antenna with a 75 ft. feed from the amp so I'm very
sub optimal. Gotta get things spiffier in the spring
but for now, it's cold out there. No way I'm going up
on the roof for a couple months.
Don't believe me? Check out this. Look at the section called "Microscopic View of Copper Wire".
The electric FIELD in the wire moves at nearly the speed of light. The electrons THEMSELVES are barely moving at all!
It turns out that what matters is the dielectric constant of the medium in which the electric wave travels.
In the case of wireless, that medium is air, which has a dielectric constant very nearly the same as free space (vacuum).
But electrical signals travelling in twisted pair wire (like category 5 cabling) travel a bit slower because the dielectric material they are travelling in is really the insulation in the wire. That's right, the signal is really travelling in the dielectric, not in the wires, per se.
So the speed of propagation doesn't depend on the properties of the wire conductor, but on the dielectric in between the two conductors.
You can calculate the speed if you know the relative permitivity of the dielectric. This is very closely related to the index of refraction, by the way.
Anyway, in most practical situations (i.e., in non-magnetic materials) the velocity equals the free-space speed of light divided by the square root of the relative permitivity of the dielectric. So in typical fiberglass circuit board, called FR4, where the '4' represents the relative permitivity, the speed of light is C/sqrt(4), or about half the free-space speed. Since the free-space speed is 300 Million meters per second, half of it is 150 million meters per second. You could also say that it is 150 meters per microsecond, or 0.150 meters per nanosecond, or 150 mm/ns.
But this is really only true for signals which are not on one of the surface layers. Surface layer signals experience a medium partially of FR4, and partially of air, so they travel a little faster.
--
MM
By including this sig, the copyright holders of this work or collection unreservedly place it in the public domain.
cheaper + easier.. dunno where i found the link.. 10mbps at 1 km and gpl:ed design.. works perfect even with heavy rainstorms. http://ronja.jikos.cz/