LTE-U is a technology standard more correctly referred to by its 3GPP name of License-Assisted Access (LAA). The standard provides a framework for implementing carrier grade LTE services in unlicensed RF spectrum frequently used by Wi-Fi products (among others).  Specifically, the goal is to operate in the currently available 5GHz spectrum now relied on solely by 802.11ac. 

At this time, the standard is promoted largely by Qualcomm, a chip manufacturer who has invested heavily in related product development, and Verizon Wireless, the dominant North American cellular services provider.  Interestingly, AT&T has indicated hesitancy about deploying services in unlicensed spectrum.

Numerous issues are yet to be resolved – primarily how to ensure that carrier services will not consume all available bandwidth to the detriment of the incumbent Wi-Fi users.

One thing is for certain: Claims such as those made by Qualcomm CTO Matt Grob that LAA/LTE-U will actually increase available bandwidth should be viewed with skepticism.  Carriers will offload from congested licensed bands to unlicensed and free up their bandwidth, but this will most likely come at a cost to the other unlicensed users.  While it’s true that LTE is more efficient than 802.11 in terms of bandwidth required for like transmissions, LTE-U does not magically increase the usable bandwidth in the U-NII bands.  Moreover, the overall increase in utilization in these bands due to the huge number of radios (base stations and end points) controlled by carriers will almost certainly have a negative impact relative to how things sit today for 802.11 users.

However, the fact is that the same FCC rules regulating unlicensed spectrum – which enabled the development and widespread implementation of Wi-Fi – also allow for LAA. As disappointing as some may find it (myself included), 5GHz is about to get a lot more crowded. Between LAA and 80-160MHz 802.11ac channels, the notion that moving usage to 5GHz is a quick fix to 802.11 medium contention problems is about to become very outdated.

The good news is that chip makers and carriers do have both regulatory and commercial motivations to seek mechanisms which support LTE and 802.11/Wi-Fi coexistence. The question ends up being: How much trouble will they cause for the rest of us while determining the best implementation?

The two approaches most discussed at this time are listen-before-talk (LBT), which usually shares strong similarities to DCF/EDCA as currently implemented in 802.11; or the Qualcomm suggested approach (for FCC and select other domains) of Carrier Sense Adaptive Transmission (CSAT).

Qualcomm prefers CSAT because it doesn’t require any modification to the current LTE air interface; this brings their products to market faster. The problem is that it leaves all decision making up to the LTE radio and makes no attempt to cooperate with the incumbent 802.11 contention management algorithms. While some may argue that the notion of incumbents in unlicensed spectrum is a fallacy, to stake a claim on neighborliness based on a fundamentally one-sided protocol seems at least as questionable.

This isn’t to suggest that LBT based protocols like DCF/EDCA are perfect either. There’s no better illustration of this than the CableLabs report on LBT approaches. When read in conjunction with their study on CSAT, it’s clear that the challenge of how to optimize Wi-Fi and LTE coexistence is far from being solved.

Congestion in U-NII bands is coming; don’t let it stop you from moving your WLAN off of 2.4GHz. It’s unlikely 5GHz will ever be as difficult to operate in as that band. But if you’re building mission critical wireless infrastructure, this is a development you want to keep an eye on and, by the end of this year, have a chat with your WLAN vendor about.

Learn more about improving your Wi-Fi network performance and capacity in CDW’s white paper, Achieving a Best-in-Class Wireless Infrastructure.

3 thoughts on “LTE-U is Coming to a Neighborhood Near You

Leave a Reply

Your email address will not be published. Required fields are marked *

You may use these HTML tags and attributes: <a href="" title=""> <abbr title=""> <acronym title=""> <b> <blockquote cite=""> <cite> <code> <del datetime=""> <em> <i> <q cite=""> <s> <strike> <strong>

This site uses Akismet to reduce spam. Learn how your comment data is processed.