Blogs

Is 2.4GHz Wi-Fi really dead for enterprises?

By mverbloot posted Apr 04, 2016 11:46 AM

  

The use of the 2.4GHz band for Wi-Fi dates back 17 years to the invention of 802.11b. If you were indeed partying like it was 1999, you might not remember that people were excited about Windows 98 and Star Wars: The Phantom Menace. And that 11Mbps of Wi-Fi was more than you’d ever need.

Flash forward to 2016: Apple and Cisco have declared that 2.4GHz is not suitable for any business.

 

We all know that it is better to use 5GHz – more capacity, more throughput and less interference but is 2.4GHz really dead?

 

The argument is that these older devices are relics, and the 2.4GHz band is slow, crowded and noisy. Older clients can choke off performance for the rest of us. Newer, dual-band clients quite often still connect at 2.4GHz simply because the 2.4Ghz signal almost always appears stronger than the 5Ghz signal from the AP. Once they connect they tend to stick on 2.4Ghz when they could be enjoying the performance of 5GHz.

 

Using a 2.4GHz client is a bit like driving a Ford Model T on the autobahn, but the reality is there are still a lot of them out there. Free public Wi-Fi often uses the legacy technology because it’s less expensive. Manufacturers depend on 2.4GHz barcode scanners to keep production lines flowing. Early adopters of IP PBXs may still have legacy IP phones and schools can’t begin to close the digital divide if they declare that students can’t use older tablets and laptops—or cheaper devices—in the classroom.

 

Create a slow lane and a fast lane

Organizations can take less drastic action and do better than cutting off the air supply for their users with 2.4GHz devices by ensuring that the 2.4Ghz band is only used by single band legacy devices. It’s a simple matter of putting the network in control with some smart RF management and tuning techniques like Client Match, Band steering and helping encourage dual band clients to always prefer 5Ghz by setting the transmit power 9dB lower on the 2.4GHz radio than the 5GHz radio.

 

Band steering encourages dual-band-capable clients to associate on the 5GHz band by essentially ignoring probe requests made by the client on 2.4GHz (if it is dual band capable), which of course helps free up the 2.4GHz band for older devices. Despite it’s best intentions Band steering does not always move every client that is dual-band capable. Some clients are just plain stubborn! This is where the 9dB radio power differential is very helpful. To the client, when the transmit power is set this way on the Access points, 5GHz always appears to have a higher signal level and

almost every Wi-Fi client out there will try and associate to the strongest signal it can find. In practice, this method is very effective in ensuring the 5GHz band is fully utilized by all 5Ghz capable clients.

 

Beyond the technology, it’s important to keep users informed. Be clear that if they have 2.4GHz clients, they are on the “best effort” network, but if they want zippy performance for critical applications, it’s time to look at those shiny new devices.

 

Using some smart RF tuning can deliver a better experience for everyone. Legacy clients stay in the slow lane and newer mobile devices can zip by, and everyone gets the air they need.

2 comments
1 view

Comments

Apr 06, 2016 06:01 AM

Having 2.4GHz only clients would still be fully supported with the power differtial discussed in the Blog. If they can't connect at 5GHz they will connect at 2.4G. I am not by any means suggesting that 2.4G should not be supported, merely that we need to do all we can to discourage dual band clients from using 2.4G. Lowering the power on 2.4G is generally best practice for most deployments as you want to minimimse CCI and maximise channel reuse. Irrespective of whether you have a majority of 5G or 2.4G clients you should implement best practice RF design and configuration on both bands.

Apr 05, 2016 05:17 PM

Quite frankly we have to do it the other way around: 2.4 gets most of the priority and 5 gets to squeeze by when it can. Of my 1500+ clients currently connected about 30% can do 5GHz, and at that we're talking G or A - less than half of them can go N and maybe 2% can do 1 or 2 stream AC.

The money (bottom-line supporting activity, not actual monetary transactions) in our wireless network moves in telnet sessions on ABG devices.

We are installing gear which can do all of the new-fangled stuff, but the forklift required to move our clients info the future is overwhelmingly expensive.

All we can do is be ready for them when the do upgrade.