All Things 802.11ac - Question 10: What are the Preferred (and perhaps even optimal) Deployment Strategies for 802.11ac?

  • 2
  • Question
  • Updated 5 years ago
  • Answered
While greenfield buildouts of 802.11ac are certainly going to happen, it’s far more likely that IT staff will want to deploy .11ac into areas already covered by 802.11n (or, worst case, 802.11g/a). What are the issues here, and what strategies do you suggest for deployment? Augment current coverage with an overlay? Rip and replace, providing 2.4 coverage via a second radio in the .11ac AP? What coexistence and evolution approaches will work best?
Photo of Craig Mathias

Craig Mathias

  • 63 Posts
  • 0 Reply Likes

Posted 5 years ago

  • 2
Photo of Matthew Gast

Matthew Gast

  • 284 Posts
  • 63 Reply Likes
The first question to answer -- and this may be viewed as kind of heretical, given how much time I devote to developing Wi-Fi technology -- is to figure out whether need the boost from 11ac. With 802.11n, there was a clear business case to move to Fast Ethernet-equivalent connectivity.

With 802.11ac, it's not a dead-simple justification. In many deployments, what we're seeing is that there's a desire for increased speed, but not everywhere. (And with first-wave 802.11ac, benefits tend to be localized to within a room given the need for high SNR for 256-QAM.)

One of the major costs in building a network is the cabling, so it's pretty common to look at an existing 802.11n network and figure out which APs would benefit from becoming 802.11ac APs.
Photo of Matthew Gast

Matthew Gast

  • 284 Posts
  • 63 Reply Likes
Overlay coverage can be used to add additional capacity into a network. One really important tip for building overlay 11ac coverage is to pull two wires to each new drop location, since you might want that second wire as 802.11ac becomes more advanced.
Photo of Matthew Gast

Matthew Gast

  • 284 Posts
  • 63 Reply Likes
In general, 2.4 GHz coverage will be fine in a network designed for 5 GHz coverage. That goes double for 802.11ac because the effective range of "good" 802.11ac data rates is shorter than it is with 802.11n -- after all, if what you wanted was 802.11n data rates, you could just buy cheaper 802.11n APs.
Photo of Craig Mathias

Craig Mathias

  • 63 Posts
  • 0 Reply Likes
Small cells are ultimately good because they're the best way to provision more capacity.
Photo of Craig Mathias

Craig Mathias

  • 63 Posts
  • 0 Reply Likes
That's a good point - 802.11n has a long and happy life ahead of it. Our forecast is that we won't see much in the way of wholesale replacement of operational .11n networks before 2018 (thus overlay is the predominant deployment model). We can even see adding some additional .11n APs in some cases. But an upgrade to 802.11ac is, ultimately, inevitable in every case; the industry in moving in that direction; prices will fall, etc. And I still believe that .11ac operating in .11n mode will yield a "better n than n" effect in many cases, so the future-proofing of buying a .11ac AP, coupled with improved .11n performance, means the market uptake for .11ac will be much greater than many have forecast - again, I think you'll see .11ac as half the market by the end of 2015.
Photo of Matthew Gast

Matthew Gast

  • 284 Posts
  • 63 Reply Likes
Good point! In past transitions, a new PHY comes out at a 20-30% price premium and is adopted only by those with lots of extra money or the need for significant extra performance. Once those early adopters get volumes going up, the cost curve comes down incredibly fast and the market flashes over. I don't see that happening quite as fast with 11ac, but I can't wait to ride the cost curve down!