wireless networking questions

[twist]

I need to connect to a wireless network. I bought a Linksys WAP (whatever model number it is that costs around $140). I am in the process of getting a 13.x dbi yagi antenna. I plan on mounting the antenna outside my site, pointing it towards a yagi antenna at the second site, which should be easy to get a near direct on line of signal (with room to spare due to the 30 degree signal arc). The distance between the two antennas will be approximately half a mile...a little less I think. Only concern I have is that there are a number of trees in the way. To my thinking, some trees shouldn't matter to the 2.4 gHz signal, right?
My plan is to run the yagi antenna into one of the connectors on the WAP, then run CAT5 cable to a linux box to use as a firewall/router for several other computers. The small dipole antenna on the other connector of the WAP will be used for 1 close range wireless pcmcia nic card one room over and one floor down, with it's own IP address since it can't be inside of the firewall/router loop.
Should this configuration work with no problems?

Twist

 

[Soybomb]

From my experience trees provide a significant problem to wifi signals. Let me know how it goes, I'm curious.

 

[EricHagen]

Line of sight means line of sight.

If you can shine a flashlight from one antenna and see it at the other, it should work.

If you can't, it probably won't.


I can block an 802.11b signal with my bare hand by covering the transmitter. Wood is very dense- moreso than most walls.


It may work, but if you can't actually SEE the other antenna, you're pushing your luck.

Sorry,
Eric

 

[blstriker]

2.4 can go through most wood houses, for only a short distance though, that's why the ratings for inside versus outside are so vastly different. Are you thinking of building the pringles can yagi? I'm planning on doing this when I have time!

 

[ktwebb]

At a half mile with the weak Linksys power output, if it very very doubtful you will get a signal at all from point to point.

 

[IaPuP]

2.4 can go through most wood houses, for only a short distance though


Well, since most houses have only wood FRAMES, but porus drywall WALLS, yeah- that's true.


The wood itself is only 1/12th of the wall and only 1 1/2 inches wide, meaning the majority of the time you're just transmitting through porus drywall, which resembels compressed foam in density.

If you made solid 3 1/2 thick wooden walls, you would not be able to transmit more than 2 rooms over- i think it just wouldn't work. You cannot really transmit through two floors/cielings very well because they ARE solid wood (only 1 1/2 - 3 1/2 inches thick usually, but solid).

Eric

 

[JackMDS]

Wireless Networking Reference - Antennas / Range Boosting

 

[IaPuP]

Cool link!

Eric

 

[IaPuP]

... stop dead at trees, which look like hanging globs of stationary water to radio signals. So if you can't go through the trees, you'll have to go around them,


hahaha

I thought that was a great analogy

Eric

 

[twist]

Yeah, I had a friend of mine explain the whole trees/concrete/etc absorbancy issue...guess I'll just have to get a chainsaw and cut down a ton of trees
In all likelihood, looks like I'll end up moving the mounting position of my antenna and running more cable between it and the Linksys WAP. Not exactly what I want to do, but it's really the only choice left.
Do leaves pose the same absorbancy problem with the rf signal as wood does?

Twist

 

[ktwebb]

You could make it through a tree or two under the right circumstances. Not likely with a Linksys product. The problem is, even if you were able to get the signal through the trees, any substantial movement in the trees, wind etc... would cause an RF disruption. Rain, not so much the actual rain, but water on the leaves also causes serious issues with reception.

 

[IaPuP]

I saw an interesting article yesterday about an ISP in northern Maine that distributes bandwidth via a series of 802.11 links.

They say that for them a 20-30 mile link is routine and they have several that are well over 40 miles and still maintaining a steady 3Mbps connection. Not too bad!

Good luck with that hookup.

Eric

 

[ktwebb]

I wouldnt believe that until I saw it. We have had some 20 miles links but right at the breach of FCC power regulations. At 40 miles I just could not see it, but who knows. THat is certainly point to point with very good transmission hardware if it is legitimate. With any Joe Network having the ability to buy 802.11 a, b, whatever equipment, a 802.11 WAN is a recipe for a very shaky network. It may work today, but there is no guarantee about tomorrow. Part of the beauty and the failure of the unlicensed bands.

 

[IaPuP]

Jones says the company thinks nothing of a 15- or 20-mile link, and that longer connections have worked with some tweaking. From Westport to Rockland, for instance, MIS runs a 40-mile point-to-point connection with a 300-foot tower on one end and a 100-foot tower on a 450-foot hill on the other.

It's on the second page of this fascinating article from Orielly. They have some open-ocean 15-mile links to islands in the atlantic ocean. THAT is innovative, since running wires there costs A LOT.

Wireless Backbones for a remote ISP?

The customer service via wireless isn't unique, but the fact that they carry one of their T1 backbones over a long-haul wireless link with a series of repeaters is pretty impressive.

Eric

 

[ScottMac]

Yeah, well, maybe so, but it ain't 802.11, and it ain't 100milliwatt.

AND it's gotta be operating in licensed spectrum, probably at least one DS3 (45Mbps).

You can build a rocket in your backyard too, but I bet you don't get to the moon in it.

How much transmission line are you using to get from the AP to the antenna? Do you know what kind of loss you get a 2.4 GHz?

Give it up, you're wasting your money and time.

FWIW

Scott