WiFi and Ubiquiti questions

Forums › LAN (ethernet/Wifi/routers/Bluetooth) › WiFi and Ubiquiti questions

Mark

1653 posts

Uber Geek

#243113 27-Nov-2018 14:49

WiFi is not something I know a huge deal about, anyone here able answer a question or three ?

I'm looking at getting some wifi into my second dwelling, which is about 500M away and behind a stand of trees/bush .. how feasible would it be to get something like a Ubiquiti Nanostation M2 at the main house, point it at the second dwelling and just cranking the power up to max ? Would a standard phone or laptop then be able to connect to the M2 access point ? Or would they need something equally powerful to reply back ? (or is the M2 sensitive enough to hear the them anyway?)

Basically I'm trying to be cheap and see if I can get away with just a powerful enough accesspoint :-) I know the M2s put out quite a lot of power as the iMac in my house can pick up my friends one who is about 1 or 2 km away with a clear line of sight, not sure if my iMac counld connect though, might ask for the password and try.

ResponseMediaNZ

518 posts

Ultimate Geek

ID Verified

Trusted

#2135282 27-Nov-2018 14:57

Standard RF principles apply. The thing that always lets us down is return path, with that. you could have all the power in the world transmitting but if the device doesn't have enough to come back to it then it won't work.

Best to create a wireless bridge between and put an AP in the dwelling

engedib

254 posts

Ultimate Geek

#2135317 27-Nov-2018 15:21

I would get a pair of Nanobeam 5AC Gen2 units for the P2P connection and a Unifi AC-Lite or AC-LR (if it is a large dwelling) for the devices.

Recently worked with a similar config, ~450Mbit of throughput capacity was achieved at a similar distance.

MCSE+M/S, MCITP

wratterus

1687 posts

Uber Geek

#2135400 27-Nov-2018 17:08

engedib:

I would get a pair of Nanobeam 5AC Gen2 units for the P2P connection and a Unifi AC-Lite or AC-LR (if it is a large dwelling) for the devices.

Recently worked with a similar config, ~450Mbit of throughput capacity was achieved at a similar distance.

Yep, do it once and do it right. The other solution is unlikely to give you a stable connection, and interference on 2.4GHz is only getting worse and worse.

The 5GHz devices work best with perfect line of sight - are the trees yours?

sbiddle

30853 posts

Uber Geek

Retired Mod

Trusted

Biddle Corp

Lifetime subscriber

#2135440 27-Nov-2018 18:23

Simple answer. No.

You also can't "turn the power up" because there are rules to have to abide by. Even if you didn't have trees it simply can't work because your antenna can't compensate for the free space loss.

The solution in your situation would be as mentioned above - a point to point link with an internal AP. With trees in the way that's unlikely to work either.

Mark

1653 posts

Uber Geek

#2135453 27-Nov-2018 19:34

Bums .. I was hoping to be a cheap skate :-)

So looks like I'll get two LocoM2s, bridge them and then have something like an AirGateway inside.

Luckily I live rural and there is barely any other wifi activity nearby (just my friend blasting across the valley) so the 2.4Gha range is clear, and hopefully it'll go through the trees.

sbiddle

30853 posts

Uber Geek

Retired Mod

Trusted

Biddle Corp

Lifetime subscriber

#2135457 27-Nov-2018 19:41

Mark:

Bums .. I was hoping to be a cheap skate :-)

So looks like I'll get two LocoM2s, bridge them and then have something like an AirGateway inside.

Luckily I live rural and there is barely any other wifi activity nearby (just my friend blasting across the valley) so the 2.4Gha range is clear, and hopefully it'll go through the trees.

WiFi doesn't go through trees. Period.

toejam316

1466 posts

Uber Geek

Trusted

Lifetime subscriber

#2135493 27-Nov-2018 20:49

Most of the solutions that are used for point to point links that are affordable are 5GHz devices - no hope of getting through the trees.

Best option would be to trim a path or find a clear line of site, and setup a point to point with a pair of Nanostation Loco5AC's - the cheapest and probably easiest option to boot.

Anything I say is the ramblings of an ill informed, opinionated so-and-so, and not representative of any of my past, present or future employers, and is also probably best disregarded.

Lenovo

Shop now for Lenovo laptops and other devices (affiliate link).

Crowdie

228 posts

Master Geek

#2138230 2-Dec-2018 20:01

There are a number of issues at play here:

Free Space Loss

As a wave passes through any medium (including air) there is a amplitude (signal strength) loss. This loss (attenuation) is called free space loss. The rule of thumb is that the higher the frequency the greater the free space loss all things being equal.

Doing some basic calculations the 5 GHz signal loss over 1 km is approximately 107 dB. This means that a signal propagated at 23 dBm will arrive 1 km away (assuming no obstructions) at approximately -84 dBm. To maintain a stable mesh connection you need the signal to arrive with an amplitude of -72 dBm or greater. Therefore, you will need to transmit at a minimum of 35 dBm to maintain a stable mesh connection over 1 km or 41 dBm over 2 km (assuming no obstructions).

Antenna Selection

One of the most important components is the antenna. It still amazes me when I get called onsite to resolve a wireless issue and the "qualified" engineer has completed screwed the antenna selection.

For a point to point (PtP) connection you want a directional antenna. A directional antenna propagates signal in a limited horizontal and vertical area - both outward and inward. This means that all the radio power is directed within that area resulting in a much higher amplitude (signal strength) in that area and a much lower (sometimes almost zero) amplitude outside that area (in comparison to an omni-directional antenna). Choose a directional antenna with the smallest horizontal and vertical beamwidth (the angle 50% of the signal is propagated in) possible while still allowing for antenna movement - if you make the horizontal and vertical beamwidth too small you will be up correcting the antenna mounting angle after every storm. A horizontal beamwidth between 7 and 30 degrees would probably be suitable for your requirements.

Antenna Mounting

As a few posters have raised you will struggle to propagate the 5 GHz signal through trees and hold a stable mesh connection. The situation will get worse in winter when the water on the tree's leaves will increase the attenuation (signal loss). Therefore, you need to mount the antennas on poles high enough that the signal is propagated over the top of the trees. If you deploy access points with external antennas the access point can be mounted inside the buildings with only the antennas mounted outside.

Be very careful to ensure that the vertical beamwidth is small enough to clear the trees. The further the trees are from the antenna the smaller the vertical beamwidth will need to be to clear them.

The Last Mile Problem

Once you have a stable mesh connection between the buildings you need to allow clients to connect.

The signal propagation requirements for the mesh (all signal going straight from one building to another) is different to the client serving access point (coverage through the internals of the external building). As a number of people have stated a unique access point for the external building would be the best option. If the building is relatively square or the coverage area can be serviced by a relatively circular pattern I would recommend deploying an access point with a 5 or 6 dBi omni-directional antenna in the center of the coverage area.