mrsussex DD-WRT Novice
Joined: 10 Feb 2011 Posts: 1
|
Posted: Thu Feb 10, 2011 3:54 Post subject: Very long range, marginal wifi experiment, advice needed :) |
|
Hi,
I work in an office in a Spanish town and live high-up in the mountains overlooking the town. As the crow flies, it is about 6 kilometers and I have clear line of site - my home is about 1000ft higher-up than the office.
There is no water, sewerage, gas or mail where I live, in fact we have only had electricity for the last six years! A standard internet connection is not an option!
I have done my link budget many times, and it seems that it may be just about possible to get a connection at my home from the office. This is all at 2.4ghz G.
This is my plan:
I would use a high power Ubiquiti Bullet adapter connected to a 24dbi yagi with an unobstructed view of the town. At the other end is a standard, boring linksys router that I am not allowed to touch, although I may be able to make configuration changes without being seen.
Running the connection at 6mbps gives me a receive sensitivity of -94dbm on the Ubiquiti Bullet and once I have done the maths I have a theoretical -4 dbm to spare. Obviously, this is not enough.
Getting the signal from my home to the router is no issue at all, I have -30dbm to spare in the budget. Clearly, I will knock the TX power back a little so that this figure is less outrageous.
My problem is receiving the routers signal at my home. I have seen some wifi amplifiers that claim to give -8 to -20db receive gain which would solve all my problems, if they work.
So my questions are:
Might an amplifier work? I live in the middle of nowhere, my nearest nieghbour is 2kms away and he is a goatherd. I can't imagine there will be much interference and noise at my end of the signal chain. The far-end, in the office, is just a quiet, small town, no real industry and again I reckon pretty quiet electro-magnetically.
Why doesn't a higher gain antenna improve reception from the direction it is pointing? I thought gain was beneficial to Tx and Rx, but when I put the figures into the online calculator, the gain of my antenna makes no difference to Rx sensitivity. Is there a limit to the amount of Rx gain an antenna can provide?
Is a Yagi a good choice? I was considering a parabolic reflector fed by a helical antenna, would this be a better choice? What about unwanted signal rejection differences between the different antenna designs?
Roughly, what are the pros and cons of Yagi, Parabolic and waveguide antennas? Also, which is best to feed a parabolic reflector, waveguide, bi-quad etc?
This might be a bit mad, but would a twin antenna, MIMO setup work more effectively using a couple of yagis?
I cannot find any routers or network adapters that claim great Rx sensitivity, about 94dbm @ 6mbps for 2.4ghz G is about the best I have found, does anyone know of anything better?
That's it for now, this is very experimental, but I used to be into Cb radio and I find it interesting. I'd be interested to hear anybodys thoughts about any aspect of this project and I will, of course, post my progress and results here for future reference.
Sorry it's so long, hope you can help |
|
TinCanWaveguide DD-WRT User
Joined: 05 May 2008 Posts: 62 Location: Near Area 51, Nevada
|
Posted: Sun Feb 20, 2011 8:45 Post subject: One of my favorite subjects, I could go on and on...... |
|
First, I HIGHLY recomend the Ubiquity Wifistation, at 1000mw, and driven by a 9xxx atheros chipset, it is without exception the best I have ever used.
http://www.streakwave.com/Itemdesc.asp?ic=WifiStationEXT&eq=&Tp=
Next, try a waveguide, if possible, see my sig. _________________ Perfect Cantenna: Use the big spagetti sauce cans, 3-1/4" diameter. Make the element 31mm long, and place the element exactly 2-1/2" from the bottom of the can. The dimentions are critical, and these are perfect. |
|