Now that the 2.4GHz is Dead gauntlet has been thrown down, I thought it prudent to help those who are trying to move to a Pure-5GHz Wi-Fi infrastructure. While there could potentially be 10 different strategies on how to move from a mixed (2.4GHz + 5GHz) environment to a Pure-5GHz, it’s my guess that many of them would be deployment-specific and would only reveal themselves to network managers who are facing uncommon challenges.
The 5GHz UNII bands are a huge swath of underutilized spectrum, ripe for the using, and represent the next horizon of Wi-Fi performance. In the US, we currently have 25 available channels, and hopefully 12 more on the way. The client and infrastructure equipment supporting all of these channels are also forthcoming.
If you operate a sizable Wi-Fi infrastructure, I would like to request that you post comments outlining:
- Size of your network (number of APs and users)
- Your vertical market (e.g. EDU, Healthcare, Legal, Manufacturing, etc.)
- Percentages of 2.4GHz users and 5GHz users on your network today
- Whether you are using band-steering or other manipulation techniques (and to what extent you feel they are working)
- The average utilization of 2.4GHz channels versus the average utilization of 5GHz channels that are in use (if known)
Your comments would be helpful to all readers who are considering a move away from 2.4GHz for Wi-Fi infrastructure.
I have recently called on Wi-Fi infrastructure manufacturers to give us dual-radio APs capable of operating on two 5GHz channels simultaneously. Such software-defined radios are also much more flexible and cost-effective than today’s fixed-configuration 2.4GHz + 5GHz APs.
The first major migration strategy that I propose is what I call the “Single 2.4GHz SSID” strategy. We need to have a common lexicon for such things, and after my years at CWNP, generating names and acronyms is second nature. It extends even into my home, where my wife has nicknamed my library of sometimes-odd terminology the “Devinator Dictionary.”
The Single 2.4GHz SSID strategy starts by reducing the number of 2.4GHz SSIDs to just one, that is secured by WPA2-PSK, where this unique SSID is not available on 5GHz channels. Of course, this could also mean using iPSK (individual PSK) mechanisms such as those currently offered by Ruckus and Aerohive. It’s noteworthy that I have also recently called on Wi-Fi infrastructure manufacturers to offer such iPSK mechanisms if they don’t already.
The next step in this strategy is to secure this single SSID appropriately. PSK by itself isn’t known for its overwhelming security or scalability, so additional infrastructure features can help. Such security and scalability features could include a per-SSID or per-USER firewall, iPSK, WIPS, and more.
The simplest methodology that I’ve found to implementing the separation of 2.4GHz-only and dual-band capable clients is to leave the dual-band clients where they are, on whatever SSID(s) are already implemented, and then manually move the 2.4GHz-only clients (typically in the minority) over to the new SSID. Since 2.4GHz-only clients are often uncommon and vertical-specific devices, the migration may entail manual reconfiguration. Yes, that’s a bummer. Look on the bright side, your network performance is about to go up an octave or two while your troubleshooting goes down several notches.
There won’t be a need for band-steering functionality now, so there’s no need for the infrastructure to have it enabled. Turn it off.
Only enable 2.4GHz radios where they are absolutely needed. Of course, additional radios can be used for WIPS or Spectrum scanning if desired.
If you find that disabling 2.4GHz for guest and BYOD access causes a logistical problem, you might consider suggesting:
- BYOD users purchase a dual-band USB adapter
- Your organization buys several dual-band USB adapters for loaning out
My personal preference of dual-band USB adapters is the Edimax EW-7822UAC which can be purchased on Amazon.com for roughly $30 USD.
I look forward to hearing from the audience regarding this migration strategy or others they may have implemented (or be considering implementing).
As a bonus, I thought I would add that if you really want to go with or stick with “1 AP per Classroom”, then Pure-5GHz is the only way I can think of that it makes sense to do so. Let the arguments begin! 🙂