Given the relatively quick uptick in LAA, the industry has to ask itself: Was it worth the time and trouble to develop LTE-U?
LTE-U was supposed to have a great time-to-market advantage over LAA but ended up upsetting the Wi-Fi community more than LAA. CableLabs pointed out that LTE-U was being developed outside of the 3GPP and was taking a different approach to coexistence, "using a carrier-controlled on/off switch known as 'duty cycling' instead of reliably fair listen-before-talk.” The FCC ended up getting involved and ultimately authorized the first LTE-U devices after the Wi-Fi Alliance almost exactly one year ago announced it was making its Wi-Fi/LTE-U coexistence test plan available.
Now that most operators seem to be embracing LAA, one has to wonder if LTE-U is getting relegated to the wayside. To hear Qualcomm—one of the biggest proponents of LTE-U—tell it, LTE-U is alive and well. In fact, the recent Gigabit LTE rollouts wouldn’t be possible without it.
“We are extremely pleased with the rapid rollout all over the world of Gigabit LTE and the increasing number of devices, all using Snapdragon chips and technology that Qualcomm developed to bring faster, better mobile broadband to consumers,” Dean Brenner, senior vice president of government affairs at Qualcomm, said in a statement provided to FierceWirelessTech. “Today, 38 operators around the world, supported by 14 devices, are offering Gigabit LTE, and these numbers grow every day.”
The rapid rollout of Qualcomm’s technology was jump-started in February 2017, when the first LTE-U equipment was approved by the FCC, he added. “That very day, T-Mobile US announced their intention to roll out of LTE-U, and operators in the US and around the world have never looked back.”
Indeed, on Feb. 22, T-Mobile announced it was deploying LTE-U technology after getting FCC certification of equipment from strategic partners Ericsson and Nokia. Starting last spring, LTE-U was going to allow T-Mobile customers to tap into the first 20 MHz of underused unlicensed spectrum on the 5 GHz band and use it for additional LTE capacity.
But by June, T-Mobile was bragging about having completed the nation’s first mobile broadband data session live in the field using LAA on its commercial network. The field testing, which began in Los Angeles on June 25, showed 741 Mbps download speeds using 80 MHz of aggregated spectrum.
At the time, T-Mobile said LTE-U was live in select locations in Bellevue, Washington; Brooklyn, New York; Dearborn, Michigan; Las Vegas, Nevada; Richardson, Texas; and Simi Valley, California, with more rolling out later this year. Asked, among other things, if T-Mobile is going to deploy LAA from here on, the company declined to comment on the matter this past week.
Let’s not forget Verizon, which publicly dug its heels into LTE-U and seemed to acknowledge its plans well before T-Mobile. In 2014, Verizon created the LTE-U Forum along with Alcatel-Lucent, Ericsson, LG Electronics, Qualcomm and Samsung.
Asked a similar set of questions, Verizon responded that LTE-U was always designed as a steppingstone to LAA, as LAA has the prioritization function to minimize interference between cellular networks and other traffic in shared spectrum space. “Verizon was out in front of LTE-U as an early technology and continues to be out in front of LAA as seen by our recent first-to-market activities,” said Verizon spokesperson Karen Schulz. “Also it’s important to note most products are software upgradable to LAA. Thus, being early to market with LTE-U has and will only enable quicker deployment of LAA.”
Point taken. As an equipment vendor, SpiderCloud Wireless was early to support LTE-U. Marketing Director Art King told me on the sidelines of Mobile World Congress Americas that engineers learned a lot from LTE-U, including much more about how Wi-Fi works.
“The reality is that we knew that LAA was going to replace LTE-U long-term,” he said, and that’s looking like sometime soon. “We wouldn’t have built it except the Qualcomm chip will be upgradable with just a new software code release from LTE-U to LAA,” so for SpiderCloud to build an LTE-U product at the request of a customer that wanted to test it near the end of 2016 wasn’t a huge deal. But he noted that if people deploy LTE-U right now, very few mobile devices actually support it, so there’s not a lot of utility there, and soon after deployment, LTE-U radios will switch to LAA.
So it wasn’t it a waste of time? “Not at all,” he said. “We learned a lot of lessons, and where we’ve got LTE-U deployed right now, we are learning so much about Wi-Fi.”
One of the things SpiderCloud does with LTE-U that carries forward to LAA is automatic band planning. Each of the devices it uses has a Wi-Fi chip on board as a listener and instead of listening on Wi-Fi for just energy like the specification says, “we actually listen to actual Wi-Fi and decode it,” he said. “That allows us from a band planning perspective to do a more intelligent job of grabbing channels and allocating channels to LTE-U. We do our best that we can to not conflict and not occupy a nearby channel” that Wi-Fi already has. “We’re improving our product” based on what is learned about Wi-Fi in the wild.
So, LTE-U is not exactly a moot point, although much of the world is heading down the LAA lane, one that AT&T opted for pretty much from the get-go. Lessons were learned and learnings applied—many of which will carry forward as the traditional licensed players continue to look for more ways to use unlicensed spectrum. — Monica | email | @fiercewrlsstech