MESH speed/distance experience

  • 5
  • Question
  • Updated 3 years ago
I would like to get your experience on the speed/distance of your MESH networks. I'm specifically looking for any tangible data on MESH links you have set-up especially in non rural and open areas with little or no interference.

I need to make the best educated guess I can

Anyone that posts an answer please specify at least the following:

1. Dostance between APs
2. AP models used
3. Antennas used (internal or external) - If ext. specify gain and angle.
4. Speedtest result (UDP and/or TCP)
5. General description of the are where APs were placed (city/rural, open/closed, etc)
5. Any known interference present (WiFi or non WiFi)

Thanks
Photo of Primož Marinšek

Primož Marinšek, AlphaDog

  • 413 Posts
  • 48 Reply Likes

Posted 4 years ago

  • 5
Photo of Max O'Driscoll

Max O'Driscoll, AlphaDog

  • 343 Posts
  • 87 Reply Likes
Thought this was interesting but no one has volunteered any real world data.
Don't use MESH so cannot contribute.

One time gentle bump!
Photo of Primož Marinšek

Primož Marinšek, AlphaDog

  • 413 Posts
  • 48 Reply Likes
Thnks for the bump :)
Photo of Alan

Alan

  • 5 Posts
  • 1 Reply Like
Let me see if I can help a bit, I manage a ifi area using ruckus 7762 ap's. We have 11 of them, distance between them range from 245ft to 940ft. Ruckus specs on these ap's say they can reach a distance of 300 meters. This is not true in our area. we are covered with trees, trimmed at a 16ft height. But, the ap's seem to only reach a maximum of 720ft. Our antennas are the 5gz Omni directional. Ruckus also says these ap's can reach 150 meters and maintain 50mbps, again, not so in the real environment. I have found the max to be 300ft or 100 meters. The decibel reading at 450ft or 150 meters will put you into the mid to upper 70's not allowing your customers to obtain a consistent connection. Speed test result using ruckus speed flex at 1 to 100 ft from ap is an upload of 1/2 meg and download regardless of my distance never completes. I am in a large area, and the park is covered with trees. We use ruckus smart mesh wifi technology, but this seems not to be a good idea in this type of environment. We will be wiring them into the router this year. The mesh system cause's the ap's to make to many jumps to get to the gateway.
Photo of Primož Marinšek

Primož Marinšek, AlphaDog

  • 413 Posts
  • 48 Reply Likes
Those omni antennas are 5dB i think so they will not mesh that good. We have tested mesh in a congested city area and could r un about 30Mbps up&down at 300m, but without trees. Trees can be a bitch and there aren't that many buildings that could bounce the signal around, so maybe that's your problem.
Photo of Alan

Alan

  • 5 Posts
  • 1 Reply Like
Should I try a higher gain antenna's? Why doesn't the signal bounce of the rv's?
Photo of Primož Marinšek

Primož Marinšek, AlphaDog

  • 413 Posts
  • 48 Reply Likes
Hard to say what to do since I can't see the actual area, but trees tend to attenuate the signal hard. 900ft isn't a distance a MESH shouldn't cover. I would say in a "normal" environment you should be getting about 30Mbps.

RVs are usually made of wood so they don't bounce enough signal for MIMO to work properly. The majority MIMO effect you'll be getting there is polarization diversity MIMO, but if trees are in the way that will be limited also. I'm also not too sure if PD-MIMO works on APs on 5GHz. It probably should but I don't know how if antennas are oriented perpendicular to each other in the AP.

It's hard to say whats wrong in your environment but you can contact me with some more details and we can have a look.
Photo of Kristijan Fabina

Kristijan Fabina

  • 14 Posts
  • 0 Reply Likes
Primož,

do you need results between only 2 x AP's in a mesh on various distances or you want results for more than 2 AP's in mesh?
I could set one example in rural area. But this rural area has also some ADSL's with Wi-Fi around so does that count or you want urban area?

One example of mesh is inside building of 4 floors + ground level. In total 5 AP's. Height between floors is 4 meters. AP's (ZF7372) were placed on the hallway on each floor so many concrete walls/floors were in between. Lots of Wi-Fi networks around the building which were heard by those AP's inside.

Anyway we have tested 20 MHz vs 40 MHz channels and 20 MHz was giving a little bit less but was more stable. On the 4th floor the latest AP had 25-28 Mbps of available aggregate throughput. There are few appartments on that floor, very good coverage on 2.4 GHz and enough capacity for all the clients. Actually more than enough since the Internet connection was around 5 Mbps tops.
Photo of Com1 NL - Bas Sanders

Com1 NL - Bas Sanders

  • 32 Posts
  • 9 Reply Likes
Hi All,

Please keep in mind that it is of no use to talk about distance and speed/stability if you don't keep the noise floor in mind!

We have delivered WiFi with 7782's and 7762-S at a large DoD event last week and found that we had perfect and stable connections at 300m of an AP at 30-40mbit. In one area we only were able to get 5mbit at 100meters.

With the SpecAn in hand (a real one!) we noticed a 16dB difference in noise floor.

Running at -70dBm receive level still gives you 40dB C/N if your noise floor is at -110dBm.
That same -70dBm receive level only gives you 20dB C/N if the noise floor is at -90dBm.

Another thing to keep in mind is that trees attenuate quite a bit as Primož mentioned.. When it is raining, it could increase attenuation of a tree with 3-6dB easily.

It's all about link budgets. And do them BOTH ways of course! :-)
Photo of Alan

Alan

  • 5 Posts
  • 1 Reply Like
ok, been a while since I have been here but, still on the same issue.
when testing the ap's speed test through the zone director, I get a fair up load and down load, But using speed flex in the field I get an upload test and the down load can not connect to the server.
I now have 4 of the 11 ap.s hard wired (cat6) and still no down load test available in the field
Wiring them in has reduced the hops to 1 - 3 instead of 3 - 5
One of the ap's does not let you connect if you are further then 250ft from it, there is no trees in the area for this ap's signal to be blocked. It has a direct connection to the zone director, Still no down load test available, upload test has an average speed test of 160Mbs with an 45% packet loss
Todays test on this AP was 250Kbs and 59% packet loss and down load can not connect to server.
The graph in the zone director for all AP's show the down load flat lined a 0 and no upload line at all.