WiFi Goes Gigabit-802.11ac

user-pic
The Interop trade show wrapped up recently in New York City and featured an excellent lineup of sessions on wireless and mobility. The topic of effective strategies for dealing with BYOD clearly topped the list (and popped up in virtually every session I attended), but there was also a lot of talk about developments on the Wi-Fi front.

The one session on Wi-Fi that stood out was "Wi-Fi: Gigabit Performance (and a lot more!)." It provided an overview of the emerging 8023.11ac and ad radio links, along with other interesting developments. The CEO of the Wi-Fi Alliance, Edgar Figueroa moderated the panel.  It featured the Director for Standards and Business Development at Realtek, Sean Coffey talking about 802.11ac. The Marketing VP at Wilocity, Mark Grodzinksy discussed developments with 802.11ad. We'll look at 802.11ac in this TechNote and 802.11ad in the next.

What Network Managers Need to Know

Lisa Phifer of Core Competence wrote about 802.11ac and ad back in March, but Sean Coffey of Realtek got into the details at Interop. The new standard will be the follow-on to 802.11n and will operate exclusively in the 5 GHz band delivering data rates into the gigabit range. At the raw bit level, 802.11ac will support data rates from 6.5 Mbps to 866.7 Mbps on a single stream using a combination of wider bandwidth channels and more efficient coding mechanisms.

Where 802.11n could run on 20 MHz or 40 MHz channels, 802.11ac can be deployed on 20-, 40-, 80-, and 160-MHz channels. Bigger channels translate into more potential bits per second, but it also means there will be fewer channels available. That's something network managers will have to take into account in their planning.

To pack more bits into the available radio spectrum, 802.11ac also uses a more efficient coding system. Like 802.11a, g, and n, 802.11ac uses OFDM signal encoding. However, where 802.11a, g, and n maxed out at 64-QAM (six-bits per symbol), the 802.11ac defines modulation up to 256-QAM (eight-bits per symbol), a 33% increase in efficiency. As we try to send more bits per cycle of radio bandwidth, the probability of error increases. So some of those bits are used to provide forward error correction to improve the reliability of the transmission.

More On MIMO

802.11ac also adds important enhancements to the multiple input-multiple output (MIMO) capability. MIMO allows a transmitter to send multiple simultaneous data streams over the same radio channel and it was first included in the 802.11n standard. However, where 80.11n can send up to four-streams (most existing products use a max of two or three), 802.11ac can go to eight-streams. Eight streams at 866.7 Mbps adds up to a total maximum data rate of 6.934 Gbps versus a maximum of 600 Mbps for 802.11n. In other words, we're looking at a 10x increase in capacity.

The high capacity per stream will be very important to devices like smartphones. The downside of running multiple data streams is that it calls for multiple radios. That means more power draw. To conserve battery, devices like smartphones generally use a single stream, but now that stream will be able to carry a data rate of 866.7 Mbps versus 150 Mbps in 802.11n.

The other key feature in 802.11ac is multi-user MIMO capability. In the 802.11n MIMO implementation, all the streams are used to communicate between the access point and a single device. With multi-user MIMO, different MIMO streams could be directed to different users. This could provide a real boost in an environment where the access point has multiple radios, but the devices are single stream. The access point could now be communicating with multiple users simultaneously, greatly increasing the total network throughput.

While the standard has not yet been finalized, chipsets for 802.11ac are already going into production and we can expect to see enterprise products rolling out early next year. While users are typically reluctant to purchase "pre-standard" implementations, the Wi-Fi Alliance's track record in ensuring compliance with the eventual standard is without blemish, so we fully expect that the lack of an official standard will not deter sales.  That's the scenario we saw with 802.11n.

Conclusion

Interoperability has been a challenge in many parts of our industry, but the Wi-Fi Alliance has managed to get all of the key vendors working out of the same playbook, which is a big reason Wi-Fi continues to flourish. Mr. Figueroa's statistics are pretty impressive. 1.2 billion Wi-Fi devices were shipped in 2011 and the technology has seen 10% year-over-year growth every year since 2001,

Suffice it to say that Wi-Fi is not going away any time soon, and the work of the Wi-Fi Alliance in ensuring multi-vendor interoperability has been one of the keys to its enormous success.

Email and Social Media Links: Share securely via email |  |

10 Comments

Suppose a company is only half-way through an upgrade to 802.11n. Is support for 802.11ac going to be a hardware (a.k.a. forklift) or a software upgrade?

Unfortunately, it's hardware. Equally unfortunate is the fact that with the advantage 802.11ac brings to Wi-Fi on smartphones, you can expect it to show up in a big hurry- the pundits are saying sales may equal 802.11n by 2015 and eliminate it by 2018. That assumption is based on the normal two-year replacement cycle for smartphones. I'd get with my vendor to talk about a strategy, but that may involve slowing down your 802.11n rollout and waiting for the 802.11ac. Vendors don't like to hear about deferring a purchase, so you might do well to seek out advice from someone who doesn't have a vested interest in a particular outcome.

Although upgrade involves hardware, it's not necessarily a forklift upgrade. From a technology standpoint, with the exception of MU-MIMO, 11ac is more of an incremental upgrade than replacement for 11n.

One sensible strategy to move an 11n deployment towards 11ac might be to position 11n APs in locations where they meet design specs for 5GHz coverage and capacity. If you haven't already, start pushing clients towards 5GHz to largely deprecate use of 2.4GHz where possible. These two moves will leave you well-positioned to deploy 11ac APs as a drop-in-replacement for 11n APs when your WLAN utilization reaches the point where 11ac is needed.

Another (complementary) strategy is to look carefully at your WLAN infrastructure vendor's migration path to 11ac. Some enterprise AP vendors are already offering pre-purchase discounts, trade-in coupons towards 11ac APs, and modular platforms that allow 11n radios to swapped out for 11n radios (or added to an array). Taking advantage of these offers can protect investment in 11n APs that you plan to use for awhile.

Also, think about where/how you can tap the advantages of 11ac. For example, you may only need those high data rates for video in selected areas, but do fine with 11n for a long time elsewhere. You may only really need higher-density coverage in public spaces, but be fine with lln in many offices. And so forth. So think about 11ac deployment in terms of gradual enhancement of your existing WLAN and look at your existing infrastructure to determine impacts of adding 11ac APs where and as-needed.

user-pic

Hi, Michael -

Quick question on your math with regard to single stream smart phone being able "to carry a data rate of 866.7 Mbps versus 150 Mbps in 802.11n." Isn't the greater bandwidth, in large part, a function of the additional number of streams supported by 802.11ac? And if a smart phone just has one stream, will it really get the higher speeds? Or will they see little or no difference from 802.11n? Thanks for any insight you can provide. - Joanie

11ac achieves higher data rates in many ways - more efficient encoding, wider channels, more simultaneous streams. Just using an 80 MHz channel instead of a 40 MHz channel doubles max data rate. Using 256 QAM boosts max data rate about 33%. And so forth. So all these 11ac improvements add up, even for single stream devices. Caveat: As always, it's a matter of rate-vs-range, with the highest rates achieved only by clients very close to the 11ac AP.

Lisa, a couple of questions: Are you assuming that the client is 802.11ac-capable (not just the AP infrastructure)? How likely is that? Also, I understand your comments about rate improvements made possible beyond just the channel width. But without the multiple channels, would a single-stream client really be likely to experience nearly 1Gbps speeds? Or would it likely see a smaller performance improvement, but not the max, even in ideal environmental conditions?

Even older 11n clients will benefit somewhat from newer 11ac clients using less airtime to transmit.

Yes, a client would have to be 11ac capable to access 80 or 160 MHz wide channels or use 256 QAM. But no, a new 11ac client doesn't have to use multiple streams to obtain noticeably better throughput.

The max MCS rate with single-stream (256 QAM/160 GHz) is just under a Gbps (866 Mbps). Over a more realistic 80 MHz channel, the max is 650 Mbps. Yes, that's nowhere near a Gbps. BUT "better" is relative to today's data rates. Even 650 Mbps is a tenfold improvement over 11n's single-stream over 20 MHz and 4x better than single-stream over 40 MHz.

Rates decline with distance etc, but it stands to reason that a single-stream 11n client and a single-stream 11ac client used in the same location, for the same applications, could experience this DELTA in performance.

Re: timing. 11ac isn't going to wait for IT upgrades - it's is already starting to show up in the workplace on BYODs and other consumer electronics. Because chipset manufacturers are moving quickly on 11ac and consumer electronic devices incorporate current generation silicon, 11ac is likely to become common in smartphones, tabs and laptops even sooner than it will get broadly deployed in enterprise WLAN infrastructure.

user-pic

Great thread. 802.11ac with MU-MIMO represents a significant benefit for capacity stressed networks, however, a few thoughts here; 1) Most of Wi-Fi clients today are single stream with 20 MHz channels so they are not really taking full advantage of 11n today. Do you expect to see this change with 11ac enabled smartphones and tablets running 40/80 MHz channel widths when power consumption is a consideration? If so, wouldn't 11ac clients with 40 MHz channel widths benefit on an 11n/40 Mhz network? 2) the first chipset implementation of 802.11ac will not support MU-MIMO and initially most clients will be legacy 11n/20 MHz clients (seems strange to be saying this already) so it would seem that network engineers can expect their 11n networks to still have quite a bit of life and capacity into the future (3+ years). 3) As pointed out, the potential capacity of a 11ac with MU-MIMO at 80 MHz is quite significant, which could exceed the Gigabit Ethernet links used today for APs. Network engineers will have to consider the impact to their wired infrastructure to take the full advantage of 11ac.

Maybe you also need to upgrade your PoE infrastructure from 802.3af to 802.3at.

1) With smartphones and tablets, most client are single stream, but laptops are typically are two stream. With 802.11ac I expect smartphones and tablets to stay one stream (for the power consideration you raised), and use the wider channels to get the increased capacity. 2) It is weird that 11n is already "old news", but the predictions I'm hearing is that 11ac sales will surpass 11n sales in 2015. Of course there will still be a lot of legacy gear around, and enterprises are not going to be doing an 11n to 11ac upgrade overnight. 3) Absolutely! We learned that lesson with 11n!







Join the Webtorials Community
Subscription Maintenance


Featured Sponsors























Recent Comments

Webtorials TechNotes

Featured Analysts

Gary Audin, Delphi, Inc.

Michael Finneran, dBrn Associates

William A. Flanagan, Flanagan Consulting

Douglas Jarrett, Keller and Heckman LLP

Jim Metzler, Ashton, Metzler & Associates

Lisa Phifer, Core Competence

Dave Powell, Independent Technical Writer

David Rohde, TechCaliber Consulting LLC

Steven Taylor, Distributed Networking Associates, Inc.

Joanie Wexler, Technology Analyst/Editor


Publisher

Steven Taylor

TechNotes is a special program of Webtorials and Distributed Networking Associates, Inc.

Notices

Please note: By downloading this information, you acknowledge that the sponsor(s) of this information may contact you, providing that they give you the option of opting out of further communications from them concerning this information.  Also, by your downloading this information, you agree that the information is for your personal use only and that this information may not be retransmitted to others or reposted on another web site.  Please encourage colleagues to download their own copy after registering at http://www.webtorials.com/reg/.  Continuing past this point indicates your acceptance of our terms of use as specified at Terms of Use.

Webtorial® is a registered servicemark of Distributed Networking Associates. The Webtorial logo is a servicemark of Distributed Networking Associates. Copyright 1999-2013, Distributed Networking Associates, Inc.