LTE-U: A quick explainer
OK, so what’s LTE-U
LTE-U is a system of wireless communication designed to use unlicensed spectrum – which is open to everybody, within certain limits – to ease the burden on big mobile carriers’ networks. Regular LTE is the system they use to transmit and receive information across their licensed spectrum – to which only they have access. LTE-U (short for Long-Term Evolution in unlicensed spectrum) uses the same “language” to operate on the unlicensed spectrum, which the carriers don’t have to spend billions of dollars to acquire.
Billions
Oh, goodness, yes. The FCC has been busy auctioning off the rights to various parts of the spectrum lately – companies bid for the rights to such-and-such a frequency in specific geographic locations in the U.S. – and the last auction took in almost $45 billion.
+ ALSO ON NETWORK WORLD: Cisco acquires IoT analytics company | Apple's Wi-Fi Assist iOS 9 feature subject of $5M lawsuit+
Cripes, that’s real money, even for Verizon and AT&T.
Sure is. And the reason they’re willing to spend it is that their networks are creaking under the truly crazy demand for data that they’re facing – all that Netflix and YouTube and Twitch and even the stuff that isn’t video (although video is the biggest issue by a long shot) is creating serious capacity problems for the big carriers.
That’s why they’re doing everything they can to stay ahead of it – building new infrastructure, acquiring new spectrum and trying to impose data caps without looking like they’re imposing data caps. LTE-U is part of that, since it would let them offload some of the spiraling demand onto the unlicensed band.
Swell. So what’s the big deal
Well, there’s this thing called Wi-Fi that operates on the unlicensed band in the 2.4GHz and 5GHz ranges. Which is exactly the piece of spectrum that LTE-U wants to use. Two radio waves in the same physical location at the same frequency means interference, which means crappy service and “ugh-why-doesn’t-this-stupid-thing-work”
Oh, well, that’s going to annoy just about everybody, huh
Yep. Qualcomm – which invented LTE-U – swears up and down that they’re incorporating coexistence features that will prevent it from harming existing Wi-Fi installations, and to be fair, it seems highly unlikely that they’re just planning to throw LTE-U out there, your home Wi-Fi be damned. The problem is, though, that we don’t really have any way of knowing that for sure, nor any guarantees that the system will operate the way it ought to.
How come
Qualcomm didn’t present LTE-U to either of the big wireless industry standards bodies – 3GPP or IEEE – for formal testing and approval, even though they’ve been relatively up-front about what the technology is going to entail. The idea will be to use a system called CSAT (Carrier Sense Adaptive Transmission, before you ask) to make LTE-U stations pause their transmissions for tiny periods to allow Wi-Fi to make use of the same frequencies. The principle is called duty cycling.
That sounds fair enough – but people still have a problem
Yeah – the thing about duty cycling is that the carriers are the ones in charge of scheduling those pauses, and they’re under no real obligation to provide a decent window of time for Wi-Fi to coexist. Remember, it’s unlicensed spectrum! But since Wi-Fi is what’s known as a polite protocol, it will politely stop talking when the LTE-U is transmitting – and even if it didn’t, all that would happen is the signals crashing into each other and getting garbled. Basically, LTE-U’s coexistence mechanisms aren’t very convincing to some people, and there’s no standards group that has the authority to force it to play nice.
Well, that’s no good – what are the alternatives
There’s a technology called Licensed Assisted Access or LAA that does roughly the same thing as LTE-U, but folds in a standard called “listen before talk” (LBT), which does pretty much what it sounds like. (Wi-Fi does this.) It’s not a perfect solution to the main coexistence problem, but LTE-U critics say it’s considerably more even-handed than Qualcomm’s plan. LBT is actually a legal requirement in the EU and Japan, so LAA is the only game in those particular towns.
Huh. So why not just use that instead
Because LAA is a 3GPP standard, and as such is going through a lengthy process of testing and approval – which means that, in places that don’t legally mandate the use of LBT, including the U.S. and China, companies could rush LTE-U to market quicker and help take the pressure off their networks.
If the carriers have this huge demand problem, wouldn’t they want to make it easier for people to use Wi-Fi instead
Certainly, and LTE-U’s backers have been making this very point at great volume as evidence that LTE-U won’t pose a coexistence problem. Realistically, it doesn’t seem likely that any version of LTE-U that the carriers would release would cause Wi-Fi Armageddon, and the problem seems more likely to be a matter of degrees – if LTE-U helps ease a carrier’s network load, even if it has minor deleterious effects on Wi-Fi networks in an area, they can probably live with that, given that there aren’t any real consequences for them.
Uh-oh.
To be fair, there’s no need to freak out just yet – Verizon and T-Mobile, the strongest advocates for LTE-U, have said that they’re not planning to roll the technology out until next year, and a lot can happen between then and now. Discussions among industry players are continuing, and FCC chairman Tom Wheeler has hinted that that agency could get involved if the companies can’t come up with a more convincing solution to the coexistence problem.