IT managers should focus on capacity, not wireless throughput

Wireless throughput is getting faster, but capacity is the measure you should really be studying.

It's difficult to believe, but the first IEEE 802.11 standard, issued back in 1997, specified throughput of only 1 and 2 Mbps. Today's gigabit-class 802.11ac-based systems are seeing explosive demand as IT managers seek not just higher wireless throughput, but also the vast array of improved features and services embodied in current products. And while 802.11ac can theoretically go almost to 7 Gbps, work now underway in other 802.11 task groups promises a future where throughput will no longer be a concern.

Today's 802.11ac products commonly offer 1.3 Gbps in 80 MHz channels using three multiple input multiple output (MIMO) streams. Appearing soon, the so-called "Wave 2" products add a fourth stream, the potential for 160 MHz channels (for up to about 3.5 Gbps) and a new feature known as multi-user MIMO. MU-MIMO enables multiple client devices to receive unique data streams during a single transmission cycle from an access point (AP); however, this technique requires new APs and new client devices.

Capacity, not wireless throughput, is the real issue for today's wireless LANs

The advent of MU-MIMO illustrates an interesting phenomenon in wireless LANs today: While the talk is usually all about throughput, the real issue is capacity. This means the emphasis going forward will not be on provisioning multi-gigabit throughput to a single client device. Instead, the goal will be to share all that wireless bandwidth to meet the needs of a diverse and rapidly increasing number of users, devices and applications.

Additionally, having more wireless throughput to divide up in such a fashion requires improvement in the efficiency of radio transmissions. And while 802.11ac specifies throughputs of up to almost 7 Gbps, products capable of supporting that speed won't be available anytime soon. This is because that level of wireless throughput within .11ac requires eight MIMO streams and that isn't feasible in mobile devices. Instead, the next big development here will be 802.11ax, known as high-efficiency WLAN, which isn't due until 2018. While technical details are understandably sketchy at this time, the goal for .11ax is 10 Gbps, using the same 5-GHz spectrum as .11ac.

In the meantime, concerns are growing in some quarters that even the large (around 1 GHz) amount of bandwidth available in the 5 GHz spectrum will become oversubscribed as the adoption of .11ac continues and as competing and incompatible services like LAA-LTE (unlicensed operation of LTE cellular services) and 5G cellular deploy in these frequencies. But never fear -- there's some great spectrum, and a truly vast amount of it, available at 60 GHz.

100 Gbps wireless? Yes, it's on its way

In fact, the 802.11ad standard for use of 60 GHz predates 802.11ac. But misconceptions surrounding the nature of 60 GHz and a few engineering challenges for those designing components operating at those frequencies have muted demand to this point. However, new developments underway will result in a good number of .11ad products in 2015. Yes, it's true that open office environments are an optimal venue for making the best use of 60 GHz; though propagation through walls will be problematic in many cases. But the throughput and capacity available here are truly amazing: four channels with up to approximately 7 Gbps each. And given the very high frequencies involved, it's actually easier to build the multi-antenna designs that are essential to this level of performance today.

Not enough for you? Well, the brand-new 802.11ay effort hopes to reach 100 Gbps within the next few years

Not enough for you? Well, the brand-new 802.11ay effort hopes to reach 100 Gbps within the next few years. I expect that wireless throughput range will be -- at least slightly -- compromised in this case, but we're talking fiber speeds. Again, amazing capacity here.

And there are other efforts underway within 802.11 to enhance functionality in a number of end-user-visible domains. 802.11ah promises new life for the sub-1 GHz frequency bands (like 900 MHz, where wireless LANs initially gained a footing); this time, with Internet of Things applications in mind. 802.11az will push the envelope for location, tracking and positioning; 802.11aq is working on pre-association service discovery and 802.11ai is developing a standard for fast link setup designed to minimize the time required to get connected to a WLAN.

And while all of this plays out, here are few tips for the rest of this year:

  • If you already have a capacity crunch, there's no real equity in waiting for .11ac Wave 2. The alternatives are limited to either living with the problem (including reduced end-user productivity) while waiting to upgrade both infrastructure and client base, or purchasing more 802.11n products, which are now clearly on a terminal trajectory.
  • Don't optimize for raw wireless throughput. Consider using 40 MHz channels for your 802.11ac deployments that make it easier to divide up bandwidth by class or quality of service and/or class of user and thus to optimize for capacity. It's also less difficult to roll out .11ac when the channel plan is the same as for .11n.
  • Don't automatically rule out 60 GHz. Both throughput and range can be excellent in open-office environments and in many other in-room applications. Give it a try as soon as you can.
  • Talk with your vendors about their WLAN development plans. A standard is one thing; but products implementing these standards can vary dramatically in feature sets, performance and, of course, availability.

Next Steps

80211.ac: How did we get here?

What Gigabit Wi-Fi means to you

What to look for in .11ac access points

This was last published in May 2015

Dig Deeper on Wireless LAN (WLAN)