Warakorn - Fotolia
Wireless networking is a complicated beast, and the nuance that goes along with each installation is what keeps thousands of wireless LAN (WLAN) professionals employed. At the same time, it's natural for all of us to want to adopt rules of thumb to guide our daily work. On my end, we design all of our WLAN spaces to adhere to Cisco's VoIP wireless construct in 5 GHz as faithfully as we can (-67 RSSI cell edge, robust signal-to-noise ratio, proper cell overlap, etc.) while taking expected device counts and the desire to deploy low-power cells under consideration.
This gives us a technically sound, unifying approach that provides a reasonably consistent baseline throughout our large WLAN. Others adopt similar approaches and the best WLANs tend to have design roots in RF consideration. This is why the popular, "One access point (AP) per classroom" thing has got me a bit miffed at those who push it for K-12 environments.
Good WLAN design: Civic officials don't always understand technical issues
Outside of my position as a wireless network architect, engineer, trainer and supporter, I also participate in local government as the deputy mayor of a small village. In the past, I've sat on our local school district's technology committee. Both positions have given me insight into the marketing that often comes at public servants and K-12 administrators who have little understanding of technical issues.
Oftentimes, those holding the purse strings are completely out of their depth when it comes to judging the merit of proposed technical solutions, so hearing things like, "Everyone has a wireless device now. You really need one access point per classroom,” boils down what otherwise might be an extremely complicated issue to a one-liner that anyone can digest.
And this is where the problem lies.
Wireless networks need more than marketing slogans
Wireless networks need proper design, not simple marketing slogans. With "One AP per classroom," you get a slogan. You also get the strong potential for too many expensive APs getting purchased -- or perhaps not enough. It's just far too simple of a message and it is seldom delivered with the caveat, one AP per classroom might help you arrive at rough budget numbers, but you'd have to have a professional design to figure out what your school really needs. Even commissioning a predictive survey done by skilled professionals is better than letting a marketing mantra guide conversations regarding design.
In any other environment, Good WLAN Design 101 dictates a number of criteria be addressed when designing wireless networks. Among them:
- What are the expected client device types?
- What is expected client density (how many in a given area)?
- What sorts of applications will be in use?
- What is the projected bandwidth needed?
- What are the building's construction materials? Will you need Wi-Fi everywhere?
- What shape is your existing switching in?
- Do you have an environment suitable for hosting controllers or are you better off with a cloud-managed approach?
Schools get a free pass from this scrutiny
I'm not sure why K-12 environments get a free pass from this sort of scrutiny and somehow fall into a simple paradigm, regardless of what the answers to those unasked questions might be.
If I follow this logic, I can see doing away with survey tools and expertise altogether and instead devising a simple matrix that lists space types versus AP counts per room, or every couple of rooms.
Heck, anyone can do that, right? Wrong.
There is a reason why the same marketers that puff, "One AP per classroom," to the K-12 space don't also try to get away with it in higher-ed WLANs: Colleges and universities usually aren't that gullible. These institutions tend to have staff with varying levels of Wi-Fi systems competency, and they know a vibrant WLAN training and tool realm exists for a good reason. Wireless is just too complex to let slogans guide our decisions. Spend the time that's required to properly investigate your WLAN infrastructure. The result will be better wireless networks that are right-sized as well as evolvable.
Friends don't let friends fall into the "One AP per classroom" marketing trap.
About the author:
Lee Badman is a network engineer and wireless technical lead for a large private university. He also teaches classes on networking, wireless network administration and wireless security.