Channel 1,6 and 11 Myth or Truth?

[Tech Focus]

Up until recently, I have been relying on auto channel selection by my router/access points. However, I know there are many experienced network user talks about 1, 6 and 11 for 2.4 GHz. The reasoning here makes total sense, which is because these 3 are only channels that do not overlap in 2.4 GHz 20 MHz channel width setting.

The part I am still not convinced/understand is the whole point of above is to avoid adjacent channel interference. So I believe the true goal should be finding a channel that has minimum interference. Let me illustrates this by couple example.

Case 1

Let's say, this is is what I see from my neighbors. Clearly Channel 6 is the most saturated/utilized. Based on 1, 6 and 11 rule, if I had a two APs, I would have easy time choosing as only 2 channels left after 6 are 1 and 11.

But technically speaking (may be a bit hard see on this image), channel 8, 9 and even 10 are less congested/utilized than channel 11. All those channels won't overlap with channel 1 that I am about to use. So if I go by congestion/utilization those channels look to be a better option.

Non-overlap rule: 1 & 11
Least utilized rule: 1 & 8/9/10

Case 2

Here Channel 6 is wide open and if I have one AP, it make sense to use it. But if I have two APs, I feel 3 & 8 (two are non-overlapping channel) is better because if I use channel 1 and 6, 1 is more congested than either 3 or 8.

Non-overlap rule: 1 and 6.
Least utilized rule: 3 and 8

So I feel like in an ideal world, every neighbor are elite students and follow 1,6 and 11 rule, then everyone should be using one of those channel perhaps but as soon as someone starts to violate the rule, utilization/congestion are harder to predict.

One question I have is which one would be considered a better practice, for situation like if I had 3 APs, say left, middle and right in home. Based on less congestion, choosing 1, 6 and 1 where two of own APs are using the same channels or forcefully separating all 3 own AP and use 1, 6 and 11 even if I know 11 is congested/more utilized.

I'd appreciate if there is a good reference/test someone has performed for congestion/utilization based vs. non-overlapping based channel setting practice.

Thanks

 

[degrub]

just move to 5 GHz or wired if that is what you see.

 

[thiggins]

What are you using to show channel congestion? Do you know how it is being measured?

You're right, in a world where everyone followed the 1,6,11 rule of thumb, adjacent channel interference would be minimized.

If you do have a tool that shows actual channel congestion, then choosing a least congested channel is always the way to go. But channel congestion is dynamic. So you may have to periodically change the channel.

Moving to 5 GHz isn't necessarily a solution. With 80 MHz wide channels and without access to DFS channels, there are only two effective channels available (in the U.S.). All you are doing when you change a 5 GHz channel is changing the primary channel where management frames are sent (beacons, CTS/RTS, beamforming sounding, etc.)

11ac/ax are supposed to dynamically modify bandwidth to minimize effects of co-channel interference, but not all products do this effectively.

 

[Tech Focus]

What are you using to show channel congestion? Do you know how it is being measured?

You're right, in a world where everyone followed the 1,6,11 rule of thumb, adjacent channel interference would be minimized.

If you do have a tool that shows actual channel congestion, then choosing a least congested channel is always the way to go. But channel congestion is dynamic. So you may have to periodically change the channel.

Moving to 5 GHz isn't necessarily a solution. With 80 MHz wide channels and without access to DFS channels, there are only two effective channels available (in the U.S.). All you are doing when you change a 5 GHz channel is changing the primary channel where management frames are sent (beacons, CTS/RTS, beamforming sounding, etc.)

11ac/ax are supposed to dynamically modify bandwidth to minimize effects of co-channel interference, but not all products do this effectively.

The tool I captured images from is built in RF Scanner in UniFi Access Points. I couldn't finding a specific detail of how they compute but it says,

"the best channel found will be selected and automatically set, based on how busy the channel is, the RSSI of the channel, and whether or not the same SSID is already transmitting on that channel."

So I assume it does decent job?

I found this one old article last night by Cisco. Their conclusion was 3 channel setup is better than 4 channel setup, which partly infers overlapping channel vs. non-overlapping Channel but I do not think it answers my question of CONGESTED overlap channel vs. non-Congested channel. At this point it is more of my pure curiosity than anything else as I know we do not do any high demanding tasks on 2.4 GHz band and even channel look congested/highly utilized, I think my IoTs still seem to be working or at least I don't notice any impact from it.

just move to 5 GHz or wired if that is what you see.

At this point, I don't think that's an option for most of us as even new IoT gadgets still coming out use 2.4 GHz connection like Sonos, B&W WiFi Speakers. Even Wi-Fi 6E still keeps 2.4 GHz as part of Tri-band design. So it will be awhile to get rid of 2.4 GHz band entirely but may be in 10 years!

But I continued to search online and what I have found so far is there isn't clear cut science. One guy commented enterprise setting 1,6, and 11 for design is way to go, but not for necessary for home. His logic made sense as in enterprise setting area of coverage is big enough, you won't really have interference from other network. But for home setting especially in apartment, I think it's not the case. So his wording made total sense.. just wanted to see if anyone here knew good testing especially I admire scientific tests of SNB.

Thanks

 

[thiggins]

If your WiFi Scanner is really providing good channel congestion information, try the "uncongested" channels, but stay with 20 MHz bandwidth. Just realize that if you are in an area with lots of neighboring networks, you could be the cause of other people's Wi-Fi pain.

However, if your 2.4 GHz use is for low-bandwidth devices (audio streamers, home automation, you probably won't be a big bandwidth hog.

 

[leerees]

If you join channel 1,6 or 11 you will be subjected to the "exposed node" issue. This means your overall bandwidth will be significantly slower and latency higher due to having CTS, RTS and then ACKs delayed by other nodes (AP's). Had you been on a free overlapping channel, RTS/CTS/ACK will have all happened without delays, thus higher bandwidth/ lower latency.

Radio signals are not rigid bell curves on a chart and wireless AP's use direct sequence spread spectrum and OFDM to ensure the signal get through regardless of any interference. It's been this way since the original 802.11 standard was established in the 90's.

Always pick a free wireless channel not a shared one, cross channel interference doesn't matter in modern wireless networks.

 

[Razor512]

For most modern WiFi routers the auto channel selection for the 2.4GHz band will do well. If your WiFi router or AP offers a debug option, you will be able to collect detailed log files that will show you how often it is scanning for a clear channel.

As for channel reuse, depending on the environment, doing something like using channel 1, 6, 1, could have a slightly higher noise floor since in many cases, it is uncommon to have 2 of the APs so far that they can't see each other, often the signal will just be really bad, e.g., by the time you go from the 1st to the 3rd AP, the first AP may be at an RSSI of -80, and while still usable, it will be insanely slow, thus it is still adding to the noise floor for the 3rd AP if reusing the channel.

For me at least, the router does a better job than a human, especially in my environment where there is a massive number of APs (well over 250, due to it being an urban area, and being very close to a number of apartment buildings.
Many people forget to use their Ethernet ports, thus random channels can become congested because a bunch of people decide to all stream video at the same time. In the evening, the 2.4GHz band struggles significantly, but the 5GHz band handles it well, especially since the signal strength of the nearby networks are much weaker.

Only issue, is the FCC really wanted to give WiFi users a hard time, and did everything in their power to hinder the use of the 5GHz band, thus depending on your location, you could end up using a DFS channel where it would have to frequently check to make sure no radar is being used in the area, only issue 160MHz channel width for 802.11ax effectively ensures that you will have to suffer with DFS channels.


Channel 36
Channel 40
Channel 44
Channel 48
Channel 52 , RADAR Sensitive
Channel 56 , RADAR Sensitive
Channel 60 , RADAR Sensitive
Channel 64 , RADAR Sensitive
Channel 100 , RADAR Sensitive, Passive
Channel 104 , RADAR Sensitive, Passive
Channel 108 , RADAR Sensitive, Passive
Channel 112 , RADAR Sensitive, Passive
Channel 116 , RADAR Sensitive, Passive
Channel 120 , RADAR Sensitive, Passive
Channel 124 , RADAR Sensitive, Passive
Channel 128 , RADAR Sensitive, Passive
Channel 132 , RADAR Sensitive, Passive
Channel 136 , RADAR Sensitive, Passive
Channel 140 , RADAR Sensitive, Passive
Channel 144 , RADAR Sensitive, Passive
Channel 149
Channel 153
Channel 157
Channel 161
Channel 165