[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]
Re: Linux & Wireless networking
"James L. Burke" wrote:
> Great addition to this thread, Mike. Funny, I remmember you posting
> something about the price of these access points a while back, but it really
> didn't mean anything to me at the time ... too bad for me ...
As I said, I'm willing to make another order as long as the price holds.
We've also been able to find the RG-1000 on pricewatch for around $154.
But for $140 you get the AP *and* an extra silver card. If you crack the
shell on the RG-1000, you effectively get 2 silver cards for $70 each.
Let me know if you're interested ASAP - Randy? James?
BTW, these cards are stock PCMCIA cards for your laptops.
> I'm interested in what the max range would be for a Orinoco card ... In
> addition to pulling the plug on my laptop and being able to roam the house
> and work, I'd like to give my business partner access to our office through
> the high speed access at my house ... his house is about 50 yards away ...
> will the silver Orinoco cards have a 50 yard range under normal conditions?
We've had firsthand reports of 600+ yards at 10mbps from Node#1 here in town.
The OReilly book on (Building) Wireless Community Networking is really
good. And *all* of his examples are using Linux - software, firewall rules,
AP configuration, etc.
www.wavelan.com has 'cliproxy' which is Linux software for configuring the
RG-1000. It also works on AirPorts as well. The RG-1000 does NAT and DHCP.
I've heard that the WAP-11's can also be used as an air-to-air repeater,
which if it works, is pretty cool. These are radios, so think
walkie-talkie - they work in half-duplex mode and are switching back and
forth between transmitting and receiving all the time.
As for reception, well, it depends. YMMV. Get yours working and then walk
over to your neighbor and see if works. Major factors affecting signal -
1) Antenna orientation. Laptops have a crappy orientation and location
for good signal characteristics. Try turning the laptop sideways.
Or even better get an external antenna for your laptop.
This make war driving especially fun.
2) Water degrades the signal big time. You (people) are mostly water.
That means standing between your laptop and the antenna is not good.
Trees with leaves and foliage suck signal, too. So it might work in winter,
but be a no-go in spring/summer.
3) Line of sight is king. If you don't have it, reception will suck. You can
do tricks with bouncing signal off buildings etc.
4) External antennas are good. Even hooked up to your laptop.
Again, orientation matters.
5) Connectors, long cable runs, and skinny cables are great sinks for
signal strength. Your card is only radiating <100mW anyway.
He gives an example on range calculations: (I've altered it slightly)
Setting up a five mile link using channel 6 (2.437GHz):
Path loss at 2.4GHz is: L=20log(d)+20log(f)+36.6
L=20log(5)+20log(2437)+36.6 = 118dB signal loss betwen the two points.
Add up all your gains (radios+antennas+amps) and subtract your losses
(cable, connectors, arrestors, and misc). Assume you are using Orinoco
Silver cards (just like the RG-1000 uses) which are 15dBm, no amps,
a 12dBi gain sector antenna on one side and a 6dBi omni on the other.
Assume 1m of LMR-400 and an arrestor on each side, .25dB loss at
each connector, and 1dB for each pigtail. Using simple math:
radio - pigtail - arrestor - connector - cable - connector + antenna
Site A: 15 - 1 - 1.25 - 0.25 - 0.22 - 0.25 + 12 = 24.03
Site B: 15 - 1 - 1.25 - 0.25 - 0.22 - 0.25 + 6 = 18.03
A+B = 42.06dB total gain. Subtracting path loss = 42.06 - 118 = -75.94dB
Receiver Sensitivity for Orinoco Silver cards:
11 Mbps => -82dBm
5.5 Mbps => -87dBm
2 Mbps => -91dBm
1 Mbps => -94dBm
So, because 82 - 76 = 6dB, we have a slim fudge factor (usually 20dB) for our
11Mbps link - i.e. maybe if conditions are optimal.
But this is for a *five mile* link, remember. It's also for two antennas
oriented properly for optimal pickup. Hence a 20dB fudge is good to have.
I hope that gives you a taste of what you're in for trying to predict
whether or not it will work. Math makes my head hurt, so I opt for
just trying it out and measuring the results. You should also plan on
doing a signal survey of your AP, and think about how the pickup patterns
of an omni antenna vs. a sector (directional) antenna fit your desired
perl -le "$_='7284254074:0930970:H4012816';tr[0->][ BOPEN!SMUT];print"
To unsubscribe, send email to firstname.lastname@example.org with
"unsubscribe silug-discuss" in the body.