Part 1 of a 4 part series on design considerations for today’s hospital Wi-Fi environment.
Part 1: 802.11 ac
802.11ac is the fifth generation of Wi-Fi standards, building on IEEE 802.11n in the 5GHz spectrum. Its purpose is to improve data rates into the multi Gbps speeds, increase RF bandwidth utilization efficiency, and support denser AP deployments. To learn more on the technical side of 802.11ac I would suggest reading my colleague Ali Kafel’s recent article: 802.11ac Wi-Fi – Planning for better channel bonding and more
Lately our industry has been a buzz discussing 802.11ac and how it will bring wired speed performance to the WLAN and therefore naturally must bring benefits to healthcare organizations. When asked that question I respond that yes it will bring benefits but as with most things in life its not a black and white answer. Absolutely, we are dealing with an explosive demand for more bandwidth, and the challenge of ultra dense client environments needing airtime fairness. The number of wireless clients in a hospital seems to multiple faster than rabbits and applications such as imaging and video are requiring more from the WLAN. Much like 802.11n gave our WLAN infrastructure a much needed boost in performance 802.11ac promises to do the same.
However, the adoption of technology on its own very rarely resolves business challenges. The key to applying new technology is with first leveraging the necessary expertise to successfully plan for the technology’s adoption into the environment. When it comes to healthcare, we have a very interesting environment in which technology adoption when successful has very tangible meaningful results. The flip side is that when technology is applied incorrectly or not in alignment with workflow we can cause more harm than good. Realizing the benefits of 802.11ac will not be as simple as just replacing access points. Instead it will require coordination in the following areas:
Design and Planning
Therefor when healthcare customers ask me about 802.11ac rather than focus on the end benefits I instead like to first start with a conversation about their current WLAN; its design, usage, and most importantly any current issues.
The current WLAN environment.
Before jumping into 802.11ac I highly recommend first making sure your infrastructure and RF environment is ready. For 802.11ac that means proper design in the 5GHz spectrum and wired network architecture. Two significant design issues come to mind based on our past experience with 5Ghz 802.11n adoption in hospitals. First, is your network optimized for leveraging 5GHz? I bring this up for a variety of reasons. Some hospitals WLANs were deployed with only the older 2.4GHz 802.11b/g radios. Other times I have witnessed the results of deployments that skipped RF surveys due to budget constraints, lack of tools, or because the IT staff were forced to wear many hats and WLAN support was an “extra” task assigned to someone. Often the WLAN works “sorta” but is far from optimized for supporting mission critical systems. Co-channel interference, poor AP placement, roaming issues, etc. create a perception that wireless is inherently unreliable. Upgrading to a faster access point will not fix these fundamental design issues. With 802.11n bonding channels in 5GHz was also not without its own challenges in terms of channel planning and legacy 802.11a client support. 802.11n brought bonding two 20 MHz channel to a single 40 MHz channel for higher throughput into the equation. With that, came challenges in channel planning and supporting legacy devices. The 802.11ac standard will support 80 and 160 MHz channels for higher throughput. This will lead to an even more complicated channel plan and challenge with supporting legacy devices. The second issue that became apparent with the adoption of 802.11n was how to handle the client loads back to the controllers. If all the traffic is tunneled from the APs back to a controller the larger WLANs can easily create traffic bottlenecks a wired network. Best practice for solving this challenge has been to evaluate application usage and when possible bridge traffic locally at the AP. For many hospitals that first required a redesign of their edge switching architecture and VLAN topologies. With 802.11ac bandwidth capabilities vendors are planning to use two-gigabit Ethernet ports to provide enough capacity back to the wired switch network. This will require additional project planning, cable costs, and wired switch ports to accommodate. The reality for IT is that this duel run design could also add complexity in support troubleshooting.
To make things interesting for those looking forward to the expected bandwidth benefits of 802.11ac, the standard is being released in two waves.
Wave 1 – Single User – Multiple Input Multiple Output (SU-MIMO): An AP has the ability to use multiple antennas to send data to a single client. The benefit is more bandwidth than what was available in 802.11n the downside is that Wave 1 hardware is not software upgradeable to Wave 2. Wave 1 is what we have available today in consumer products and some enterprise gear.
Wave 2 – Multi User – Multiple Input Multiple Output (MU-MIMO): An AP will now have the ability to use multiple antennas to send data to a multiple clients. This will be the closest Wi-Fi has ever gotten to replicating a wired Ethernets full duplex capabilities. In order to maximize the investment my opinion is that healthcare organizations should really hold off on 802.11ac until the Wave 2 infrastructure is available.
This brings us to client devices, their support of 802.11ac and performance. When 802.11n first came it took a number of years before supporting clients began to overtake legacy devices. It has taken even longer for tablets, and now smartphones to support 5GHz 802.11a/n. So the theoretical maximum performance gains of 802.11n took longer to realize than many had at first expected. What’s more, consumers soon learned that the bandwidth capabilities of 802.11n were limited to how many antennas a client radio contained multiple input multiple output or MIMO. The smaller the device the fewer the antennas is a good rule of thumb. Additionally the first generation 802.11ac will most likely carry on the tradition in needing multiple firmware updates to address bugs. Consumer grade USB adapters tend to be designed for connectivity to a stand-alone access point. Early 802.11n USB adapters performed horribly when in a multiple access point environment. Reports from the field are that early 802.11ac USB adapters have followed suit.
As 802.11ac client devices begin to permeate the market, healthcare organizations that have an 802.11 a/b/g/n network can begin to fully leverage the 5GHz spectrum. This trend alone should result in improved performance in the congested 2.4 GHz space while giving the IT organization time to design for the pending 802.11ac Wave 2 infrastructure and more importantly address how best to maximize the impact of this new technology within the healthcare environment.
Want to learn more about optimal channel utilization and simplifying the migration to 802.11ac, join us for a free July 31 webinar as we talk about this and other 801.11 factors.