Carrier WiFi

Jury Still Out on LTE-Unlicensed

In less than 20 years of widespread availability, broadband radio technologies in the unlicensed spectrum, such as Bluetooth, WiFi, and ZigBee, have achieved a number of milestones. (See Playing by the Rules: The Success of Unlicensed Spectrum.)

The current successes of unlicensed technologies have been achieved primarily in the ISM band at 2.4 GHz -- a band spanning only 80 MHz. However, over the coming years, we will increasingly see these technologies expand their operation to new spectrum bands, whose characteristics will enable entirely new types of application.

  • The TV White Spaces are the unused sub-1 GHz bands in the broadcast TV bands providing anywhere from 30 to 200 MHz on average per person across the US. Their use permits long-distance non-line-of-sight communication and is already being tested in rural and urban broadband projects around the world, as well as new generations of long-range sensor networks.
  • The 60 GHz band provides an extremely capacious 7 GHz of unlicensed spectrum. Due to its extremely high frequency, it will be used for short-range, ultra-fast communication to mobile devices and somewhat longer-range fixed uses such as line-of-sight backhaul from small cells and point-to-point uses.
  • The 5 GHz band provides propagation possibilities most similar to the 2.4 GHz band. However, it also contains an order of magnitude more spectrum -- 775 MHz in the US. The 5 GHz technologies will be used to build superfast multi-gigabit LANs and longer-range, point-to-point gigabit links, which will be particularly important for rural WISPs in the developed world and for building the high-capacity distribution systems that will extend the Internet across developing nations.

These developments will likely increase the already impressive economic value delivered by unlicensed technologies. However, WiFi will not be the sole technology seeking widespread deployment in the 5 GHz band.

LTE Advanced in unlicensed spectrum (LTE-U) is an innovative proposal from Qualcomm Inc. (Nasdaq: QCOM) to deliver LTE with small cells using unlicensed 5 GHz spectrum. Previously, Qualcomm has argued that only licensed spectrum can support the quality of communications that consumers expect. "Quality of service predictability is linked to the exclusivity and the binary access to a given spectrum resource, at a given location and a given time," the company said in November.

LTE-U is an abrupt change of direction. However, it does not mean that LTE networks can be deployed completely in unlicensed spectrum. The specification will require that a control channel is implemented in a licensed band. Whether or not this is necessary to ensure that, as Qualcomm says, "the crucial signaling information is always communicated properly," a side effect is that the only deployers of LTE-U will be licensed mobile operators.

The eventual success of LTE-U will depend on mobile operators and their willingness to take on large-scale deployments of the technology alongside or instead of carrier WiFi. This remains an open question.

In its favor, LTE-U could allow mobile operators to offer very high speeds by tapping into the immense spectrum resources of the 5 GHz band -- especially attractive to weaker operators and new entrants with limited spectrum resources. Due to its deep integration with operators' current networks, fast handover between LTE-U cells and traditional macrocells should be possible.

However, LTE-U adoption will also face substantial obstacles. It is unlikely that LTE-U can be as cost effective as WiFi. Base stations are likely to cost substantially more, and the operators will always need to bear the backhaul costs. The premium features provided by LTE-U (seamless voice and data roaming) may not prove sufficiently more valuable than those offered by WiFi technologies such as Hotspot 2.0.

Perhaps the greatest drawback of LTE-U is that it will, of course, work only with LTE-capable devices. The future will contain many more devices that feature WiFi connectivity than LTE, including the majority of tablets, laptop computers, cameras, and nearly every other connected device. Will operators choose a standard (LTE-U) that locks them out of addressing the greatest possible market of connected devices?

Should LTE-U prove a success, then the technology will be a prominent addition to the rule-based spectrum bands. As the history of Bluetooth and WiFi shows, conflicts between technologies can arise in unlicensed spectrum. Nonetheless, these conflicts have been resolved, largely through the creation of mutually non-exclusionary rules of operation embedded in each technology -- the protocol standards themselves.

If LTE-U takes this conciliatory approach, then it can become an important part of the immense patchwork of unlicensed technology and usage. However, should the operation of LTE-U seriously degrade the possibility for other users to take advantage of the 5 GHz band, then this could precipitate the first true crisis of governance of the spectrum commons. This would pitch Qualcomm and some licensed stakeholders against an immensely broad coalition of unlicensed wireless users, including private citizens, retailers, governments at every level, healthcare providers, universities, and nearly every single business that uses networking technology.

By allowing rule-based access, the unlicensed spectrum bands have enabled innovation and business deployment by entities ranging from the world's largest companies to the smallest of startups. The rules of this commons have been sufficient to generate the most varied, the most efficient, and potentially the most economically valuable bands in wireless communication.

If LTE-U operation substantially disrupts the ability of other technologies to operate, the regulatory authorities must leave no doubt that they will intervene to help strengthen the rules governing the great spectrum commons.

— Richard Thanki, telecommunications economist and PhD candidate, University of Southampton

mendyk 3/12/2014 | 11:06:53 AM
We interrupt this broadcast ... At what point will over-the-air video broadcast spectrum be repurposed for mobile service? I'm guessing it will start in 10 to 15 years.
DHagar 3/12/2014 | 4:25:24 PM
Jury Still Out on LTE-Unlicensed @mendyk - good question - I think you have a good estimate - especially if the market is driven by Qualcomm and others.

I have another question for Richard, as it appears then that the best market players are those who use LTE-U along with other services and packages, what is the best model of ROI for them?  (Since you are an economist)  And what is the best economic model for the customer?

Great blog.

Sarah Thomas 3/12/2014 | 7:32:45 PM
LTE-U: friend or foe? Interesting post, Richard. Thank you. I get the impression that LTE-U is still very much in the R&D phase, but that operators don't yet know what to think of it. If the app is making the decision which network to use, it takes that control away from them. But, wouldn't it only be used in worst-case scenarios when the network can't handle the load? In that case, it's not necessarily an alternative to WiFi, but just a way to augment it where it's not enough. Or, am I misreading it?
WilliamofOccam 3/13/2014 | 12:42:59 AM
LTE-U vs. WiFI Operators with substantial investment in WiFi are unlikley to support LTE-U while exactly the reverse is true for those operators who have not depoloyed ot tied up with a WLAN network. It will be interesting to see who emerges as the winner in the standards war that seems bound to happe in 3GPP.
Sarah Thomas 3/13/2014 | 12:43:45 PM
Re: LTE-U vs. WiFI I think I'm confused -- wouldn't they have to have a WiFi fotprint to take advantage of LTE-U? 
Richard Thanki 3/17/2014 | 8:01:53 AM
Re: LTE-U vs. WiFI It'd certainly be much easier for an operator that had an extensive Wi-Fi network to roll out LTE-U as they would already have acces to the small cell locations that the latter will need. However, an operator without extensive Wi-Fi could choose to deploy LTE-U + backhaul in a small cell formation. However, it is exceedingly unlikely that the equipment chosen by such an operator would be combined LTE-U and Wi-Fi - Wi-Fi is low cost and allows an operator to address devices that aren't LTE-U compatible.
Richard Thanki 3/17/2014 | 8:13:26 AM
Re: Jury Still Out on LTE-Unlicensed The return for the deployers of LTE-U will come from deferring the need for building new macro, micro and pico cells and reducing the need to spend vast sums in spectrum auctions.

However, that's the crux. If operators see that Wi-Fi + Passpoint (which allows seamless connection and roaming) provides comparable benefits to LTE-U for a fraction of the cost then LTE-U's future doesn't look too rosy!
Richard Thanki 3/17/2014 | 8:16:31 AM
Re: We interrupt this broadcast ... Very likely correct, however, if TV White Space unlicensed technology prooves a real success over the same timeframe then there will be a strong argument to make ~100MHz or so of the band dedicated unlicensed specrtum.
zyj_1410 3/17/2014 | 9:38:16 AM
Re: LTE-U vs. WiFI Compared with WiFi, I think LTE-U is more suitable for operator. 1) easy to manage and control. 2) easy to operate 3) approximate cost if to reduce LTE-U small cell complexity, ect.
DHagar 3/17/2014 | 12:32:50 PM
Jury Still Out on LTE-Unlicensed @Richard, it appears that the key will be the packaging and pricing that will best determine the market positioning.  If so, establishing a continuum of services that offers choices along the way may be the best position.

Sign In