I had posted the following on another site and thought it may be of interest:
A group of us in the Wickenburg/Congress, AZ area have tested with both the Linksys and Ubiquiti devices. We have used IP phones and IP camera for our testing.
1. The low power of the Linksys devices make them suitable for the last (mile??) local hook-up. The Ubiquiti devices have performed out in the desert with good quality phone audio and video at 5 miles. The setting was foothills on one end and valley on the other end. It appears that you do have to pay attention to the Fresnel zone. Height on the units is your friend.
2. We have had some surprising results on the back side of the sector antennas - nanostation M2 to rocket 120 deg sector at about a mile.
3. The video and audio drop out shortly after 5 miles with the Ubiquiti Rocket 120 deg sector to NSM2.
4. The Linksys have worked well to multiplex phone and camera behind a sector antenna - better to direct connect if you can.
5. Our testing at 8 miles did not yield good results, but then we were aiming over the top of 30+ other 2.4Ghz routers.
6. We have not had good success with amps on the Linksys.
7. I have encountered several problems between the chair and the keyboard - but I am pleased with our successes.
Thank you for the result feedback.
I would say most of what you have touched on strikes as what I would expect (and that they are good items to see happen first hand in the real world to understand how they affect the network design)
Next step comes tweaking and improving which I'm sure your group is likely under way in trying (changing bands, using more directional dishes for one side, etc) to improve the links.
Keep up the deployment testing and the knowledge will grow in the group. We all will learn new items as we deploy these nodes, its a new territory for many of us, a spot the Microwave groups have experimented in for years and gain great knowledge that rest of us now are starting to pick up.
PS: I'm sure I saw this on the other site as well, no one here is going to be upset on the mention of the site, or reference to discussions that been had (or is being had.) When it comes to growing these networks knowledge and skill are important and limiting either of those will degrade the quality of the networks we can deliver so share away on the knowledge you gain as you continue to experiment and learn what works and what doesn't.
I wonder, did you model your network with the Airlink tool? If so, do you have any comparison data on theoretical vs. actual to share?
No, we did not model. We read and reread as much as we could, choose a sterile RF environment and worked for distance. Then we experimented with up one side of the hill, repeater on top and down the other side. Hope to get more into models and theory over the summer.
"6. We have not had good success with amps on the Linksys."
I have found that a ch1 bandpass filter is the primary factor that improved the long distance performance of the linksys. We have an 8 mile link operational on our local mesh with 24dB grid on the linksys side.
It was difficult to determine if also adding an amp helped (as it also amplifies the noise).
BTW, I have 3 filters and 3 amps I'd be willing to sell at half cost. I don't use linksys anymore (the cost of the filter is the cost of a UBNT device...).
Joe AE6XE
Joe,
I would be interested in getting one of the filters, for testing. Still available?
tx
Ron K7OPA
Yes, still available. Send me email at callsign @ soara.org to connect up.
http://www.rflinx.com/products/filters/2400/bpf/
Joe AE6XE