Why 802.11ac Will Kill The 5 GHz Wi-Fi Band

[kk22]

Reference:
http://www.smallnetbuilder.com/wire...4-why-80211ac-will-kill-the-5-ghz-wi-fi-band/

That was an interesting piece. Thanks Tim.

"The gist of all this is that 802.11ac, even in the first draft devices appearing this year, can reduce the effective number of available 5 GHz channels from nine to two!"

That's indeed a scary thought. I can already detect 30+ networks in the 2.4Ghz band in my townhouse complex neighborhood, with about 12 on Channel 1, 8 on Channel 6 and 7 on Channel 11, and a few using the in-between channels. The end result is my 2.4Ghz wireless N speeds aren't the best and there are times when the connection would drop.

I've since moved to the cleaner 5Ghz space and got more stability and speed for my laptops, but it looks like 802.11ac will make crowded neighborhoods a bigger nightmare for stable wireless connectivity.

Is there anything else in the works for wireless connectivity besides 802.11ac?

 

[darksamus]

Reference:
http://www.smallnetbuilder.com/wire...4-why-80211ac-will-kill-the-5-ghz-wi-fi-band/

That was an interesting piece. Thanks Tim.

"The gist of all this is that 802.11ac, even in the first draft devices appearing this year, can reduce the effective number of available 5 GHz channels from nine to two!"

That's indeed a scary thought. I can already detect 30+ networks in the 2.4Ghz band in my townhouse complex neighborhood, with about 12 on Channel 1, 8 on Channel 6 and 7 on Channel 11, and a few using the in-between channels. The end result is my 2.4Ghz wireless N speeds aren't the best and there are times when the connection would drop.

I've since moved to the cleaner 5Ghz space and got more stability and speed for my laptops, but it looks like 802.11ac will make crowded neighborhoods a bigger nightmare for stable wireless connectivity.

Is there anything else in the works for wireless connectivity besides 802.11ac?

Once all the vendors move to 802.11ac, the 2.4ghz will be back to less crowded!

One thing I don't like about this new standard or any wifi is that if you want higher bandwidth, you have to use 80mhz wide bandwidth and you know everyone is going to use it and it will start to overlap your neighbor's channels.

I forgot to mention that with new TVs that are going to be wifi enable from its base will be using 5ghz which means if your neighbors has that type of TV, you can bet that you will be able to pick up what they are watching.

 

[thiggins]

Once all the vendors move to 802.11ac, the 2.4ghz will be back to less crowded!

Not likely. 11ac products will all be dual band, implementing bgn in 2.4 GHz.

 

[thiggins]

Is there anything else in the works for wireless connectivity besides 802.11ac?

60 GHz technologies (WiGig and WirelessHD) are out there. But are short-range only.

 

[irev210]

I live in an urban environment and just fired up the latest version of inSSIDer.

In my apartment I pick up 133 access points.

87 at 2.4GHz (24 of them are a local university)
46 at 5GHz (35 of them are a local university)



Sad that 5GHz will become overcrowded.

 

[RogerSC]

Well, on the other hand, 5GHz. signals don't propagate nearly as well as 2.4GHz. signals. So I guess that the largest effects will be on people in apartment building. I see lots of 2.4GHz. networks, too, but I wouldn't expect to see more than one or two 5GHz. networks, from my immediate neighboring houses on either side of me. And not everyone jumps onto the new stuff, my immediate neighbor on one side of me, the strongest 2.4GHz. signal, is still using wireless-g with WEP *smile*. I'm guessing that at least they won't be jumping on the newest 5GHz. tech.

 

[DaveMcLain]

It appears that 11ac must take about an order of magnitude more processing power over 11n to implement so maybe it'll also have a lot more sophisticated "good neighbor" behavior when other networks are around competing for air time.

Great article Tim, thanks.

 

[stevech]

what we Don't need is more people with one-way over-powered WiFi routers/access points. That increases the radius of coverage, but only one-way... to the client.
The from-client radius of coverage is unchanged.

For those old enough to remember the hay-days of Citizen's Band Radio... we had too many people that used hundreds of watts. This and user behavior became so out of control that the US FCC abandoned licensing and managing use of the band. It more or less self destructed for all but the truckers who use it for quasi-valid reasons in this cell phone era.

My point is... if 2.4 then 5.8GHz are grossly misused, versus the original intent of low power, short range, SHARED-USE, then we all are impacted. The 40MHz mode of 11n is misuse, as it burdens 2/3 of the entire band. Not a problem if you are a light-use or rural user.

 

[RogerSC]

Yes Steve, I remember being bemused by the sales of massive linear amplifiers to CBer's. Really created an unusable zoo, but I guess it made the really loud guys happy. So it goes. Hopefully that "if a little is good, more is better" philosophy won't be carried into this arena.

 

[thiggins]

Well, on the other hand, 5GHz. signals don't propagate nearly as well as 2.4GHz. signals.

Good point, which I forgot to include. I'll add it.

On the other hand, for enterprise use, with lots of APs, 80 MHz wide channels are going to make it more difficult for channel assignment.

 

[stevech]

5.8 GHz has about 6dB disadvantage vs. 2.4GHz, in free space attenuation (laws of physics). Adding-back that 6dB via more antenna gain and/or more transmitter power on either end can offset. More transmitter power without increased distortion increases the cost significantly (competitive factor). The distortion of the transmitted signal at high bit rates is the key factor-- reduce the bit rate (modulation order it's called), and the transmitter power can generally increase. This is the nature of OFDM in 11g/n.

The attenuation from drywall etc. is a bit higher too in 5.8GHz - but not markedly so (per report from the NIST http://fire.nist.gov/bfrlpubs/build97/PDF/b97123.pdf). This increased attenuation + path length simply sum up in the path loss calculation -- one for each direction.

 

[sfx2000]

Good point, which I forgot to include. I'll add it.

On the other hand, for enterprise use, with lots of APs, 80 MHz wide channels are going to make it more difficult for channel assignment.

Actually, at the enterprise level, it's quite easy - set the AP's to 802.11n, and run 20MHz channels, alternately, you can run 40 MHz channels, but most enterprises I've seen run 5GHz 802.11n overlays at 20Mhz for the built-in 802.11 a/b/g clients... 802.11ac will run just fine in 40Mhz channels.

My big fear is not the enterprise where networks are centrally managed, it's the home market where AP's and Device's default to wide channels, and many AP's don't honor the intolerant bits (part of 802.11n)

 

[kk22]

I live in an urban environment and just fired up the latest version of inSSIDer.

In my apartment I pick up 133 access points.

87 at 2.4GHz (24 of them are a local university)
46 at 5GHz (35 of them are a local university)

Sad that 5GHz will become overcrowded.

I thought seeing 30-40 was bad. 87!!!

Well, on the other hand, 5GHz. signals don't propagate nearly as well as 2.4GHz. signals.

That's a good point. I'll have to wait & see what it's like when people start getting these devices in our townhouse complex...

 

[stevech]

maybe band crowding will fix its self - as did cordless phone problematic coexistence in (US) 900MHz, then 2.4GHz, then 5.8GHz and finally an effectively cordless phone-only band for DECT phones. This is near 1.9GHz, but not overlapping 2G/3G cellular is near that same freq. too.

Or the kinda screwy concept of "white space spectrum" will spill over to wireless LANs.

 

[Biker]

Here's some more read on this thread, different site, http://wifinetnews.com/archives/2007/01/5_ghz_or_bust.html And this, http://blogs.broughturner.com/2010/02/broadband-wireless-the-future-is-5-ghz.html

 

[stevech]

I wonder if North America regulations will change as they have in some EU/middle-eastern countries: 2.4GHz may not be used outdoors. Prohibited. Practically speaking, this means no outdoor access points. To me, this makes good sense, given the intended use of the band. We don't want metro-WiFi (APs on light poles) cluttering the spectrum.

 

[sfx2000]

Yes Steve, I remember being bemused by the sales of massive linear amplifiers to CBer's. Really created an unusable zoo, but I guess it made the really loud guys happy. So it goes. Hopefully that "if a little is good, more is better" philosophy won't be carried into this arena.

Hehe, I remember days of Citizens Band 100W linears and reverbs to make the voice even bigger...

It was fun listening to my toaster while my neighbour 3 doors down was chatting with his buds...

 

[sfx2000]

I wonder if North America regulations will change as they have in some EU/middle-eastern countries: 2.4GHz may not be used outdoors. Prohibited. Practically speaking, this means no outdoor access points. To me, this makes good sense, given the intended use of the band. We don't want metro-WiFi (APs on light poles) cluttering the spectrum.

I think the bigger thing to fear for Metro wireless is 802.16m in the unlicensed band at 5GHz... to an 802.11 LAN, it can be a problem... but for what was Metro Wifi applications, 802.16 is a good answer, and scales much better from a network load perspective - Wimax isn't dead just yet, just looking for a better business case...

 

[Moetop]

I know I am resurrecting this from the dead, but something that is little talked about is the difference in power the channel bands can transmit.

- 36,40,44,48 (50 mW)
- 52,56,60,64,100,104,108,112,116,120,124,128,132,136,140 (250 mW)
- 149,153,157,161,165 (1000 mW)

So if you want distance you are even more limited by the channels you can use. WiFi has been a mess almost from the beginning with the original 11 Channels (US) and the ability to only use 3 without overlap (1,6,11). The speed certainly has increased, but so has the confusion surrounding it.

From The Wiki http://en.wikipedia.org/wiki/U-NII

U-NII Low (U-NII-1[3]): 5.15-5.25 GHz. Regulations require use of an integrated antenna. Power limited to 50mW[4]
U-NII Mid (U-NII-2[3]): 5.25-5.35 GHz. Regulations allow for a user-installable antenna, subject to Dynamic Frequency Selection (DFS, or radar avoidance)[5]. Power limited to 250mW[4]
U-NII Worldwide: 5.47-5.725 GHz. Both outdoor and indoor use, subject to Dynamic Frequency Selection (DFS, or radar avoidance)[5]. Power limited to 250mW[4]. This spectrum was added by the FCC in 2003 to "align the frequency bands used by U-NII devices in the United States with bands in other parts of the world"[5]. The FCC currently has an interim limitation on operations on channels which overlap the 5600 - 5650 MHz band[6].
U-NII Upper (U-NII-3[3]): 5.725 to 5.825 GHz. Sometimes referred to as U-NII / ISM due to overlap with the ISM band. Regulations allow for a user-installable antenna. Power limited to 1W[4]

Reference:
http://www.smallnetbuilder.com/wire...4-why-80211ac-will-kill-the-5-ghz-wi-fi-band/

That was an interesting piece. Thanks Tim.

"The gist of all this is that 802.11ac, even in the first draft devices appearing this year, can reduce the effective number of available 5 GHz channels from nine to two!"

That's indeed a scary thought. I can already detect 30+ networks in the 2.4Ghz band in my townhouse complex neighborhood, with about 12 on Channel 1, 8 on Channel 6 and 7 on Channel 11, and a few using the in-between channels. The end result is my 2.4Ghz wireless N speeds aren't the best and there are times when the connection would drop.

I've since moved to the cleaner 5Ghz space and got more stability and speed for my laptops, but it looks like 802.11ac will make crowded neighborhoods a bigger nightmare for stable wireless connectivity.

Is there anything else in the works for wireless connectivity besides 802.11ac?

 

[richto]

FYI, there are 4 non overlapping WiFi channels in most of the world: 1, 5, 9, 13