We use cookies to deliver our online services. Details of the cookies we use and instructions on how to disable them are set out in our Cookies Policy. By using this website you agree to our use of cookies. To close this message click close.

LTE-Unlicensed - The New WiFi?


30 April 2014

By Tom Peters

Over the past few months, a debate has been quietly raging in the international standards group 3GPP over the possibility of using the 4G wireless technology LTE in unlicensed spectrum.  LTE is the de facto standard for 4G in licensed spectrum – the latest iPhones and Android devices support it on every major US operator’s network - and WiFi is the wireless standard used over unlicensed spectrum in homes, local businesses and public “hot spots” – it is built into every computer, tablet and smartphone out there.  So the possibility of expanding LTE into unlicensed spectrum where the WiFi standard is ubiquitous and dominant brings several questions to mind:  “What’s wrong with the current arrangement?”, “Why does LTE want to infringe on WiFi’s territory?” and “How could this affect me in the future?”  

Let’s start with a brief description of 3GPP and its process:  3GPP, the Third Generation Partnership Project, is an agreement between six international standards bodies to produce consistent worldwide standards for 3GPP technologies.  The actual work is done by technical representatives of member companies, which include wireless operators, infrastructure vendors, antenna manufacturers, test equipment suppliers, industry organizations, regulators, and other stakeholders – basically anyone with an interest in the future of the wireless industry.  In quarterly plenary meetings, the group debates and then agrees on the work that is to be done and the schedule for completing it.  Then the actual technical work is done by contributing member companies and is presented, reviewed and often debated in quarterly Working Group meetings.  Completed work then goes back to the next plenary meeting for approval and subsequent incorporation into the standards produced by the six standards organizations.

Decisions in 3GPP are made by consensus.  That said, there are a great number of issues that are very contentious; thus, it is typical for companies to negotiate to a compromise, although sometimes this process can take years.  Of course, companies take positions in 3GPP that best represent their business interests – which are often non-public –and some of these positions can represent great future value to the company.  So sometimes an open compromise is not an option.  Given the numerous conflicting interests that arise when you put competing representatives from all parts of the wireless industry supply chain in the same room, disagreement is far from unusual.  And often, the reason driving the disagreement is not obvious.  As a result, companies will negotiate offline and strike private agreements that help them reach compromise and consensus in 3GPP.  Somewhat miraculously, this process has developed the wireless technology that powers your smartphone.

In the case of LTE- Unlicensed (or LTE-U), the public debate started when Qualcomm and Ericsson submitted a presentation at the December 2013 plenary meeting in Busan, South Korea.  This contribution closely matched information that Qualcomm had posted on their website just a couple weeks earlier that advocated the use of LTE in unlicensed spectrum to augment a primary service over licensed spectrum.  In other words, the use case for LTE-U that Qualcomm and Ericsson presented required both licensed and unlicensed spectrum:  the user’s primary LTE channel would use licensed spectrum, and through carrier aggregation, an LTE channel in unlicensed spectrum would be used to increase the throughput.  The presentation cited the exploding growth in wireless usage and the challenges operators face as they try to keep up with demand, and argued that operators will need to rely on unlicensed spectrum in the future.  The presentation also asserted that because Unlicensed LTE would use the same base station equipment and core network as licensed LTE, LTE-U would build on LTE’s scale and ecosystem.  Managing a single LTE network, it implied, was much easier and cheaper than managing two.

Although many wireless equipment manufacturers supported the Qualcomm/Ericsson proposal, only two wireless operators did:  Verizon and China Mobile.  One concern that many operators saw was that LTE-U could be a threat to the WiFi ecosystem, potentially reducing the value of the significant investments that many had already made in WiFi hot spots.  This concern did not prompt wholesale rejection of the idea, but rather led to a recommendation that 3GPP take time to think through all the implications, and generally tread slowly.  Another concern was potential distraction from ongoing work on Release 12 which is scheduled to wind down around September 2014.  As a result, these opposing operators wanted to hold off any work on LTE-U until then.

At the heart of the technical concerns is the fact that WiFi currently includes a “politeness protocol,” which basically means that WiFi will back off when it senses interference from other users.  The current version of LTE could be considered “rude” in comparison since it has no such protocol.  So a major concern is that if a rude technology operates in the same band as a polite technology, the rude technology will eventually take over the band.

The potential for anxiety over the “politeness protocol” was especially evident in comment included in the Qualcomm/Ericsson proposal:

LTE Unlicensed The New WiFi

This statement tied Qualcomm/Ericsson’s idea to another proposal submitted by Verizon in the same meeting entitled, “New Band for LTE deployment as Supplemental Downlink in unlicensed 5.8 GHz in USA.”  Specifically the Verizon proposal targeted the band 5725-5850 MHz and stated, “In USA, there are no requirements for unlicensed deployment that require changes to LTE air interface.”  Taken together, these statements seemed to cause quite a bit of concern and reaction about the potential for interference to WiFi, not the least of which came from a US operator with tens of thousands of WiFi hotspots -- AT&T.

Several operators also expressed other concerns about the LTE-U proposal, such as (i) the potential effect on the value of licensed spectrum around the world, (ii) the need for a global solution upfront including international harmonization of the unlicensed bands used for LTE-U, and (iii) whether the technology should be limited to supplemental downlink of a licensed band (as proposed by Qualcomm and Ericsson) or if uplink and/or standalone TDD operation should be included.  These topics were hotly debated for several hours, and although no company disagreed with the general concept of LTE-U, there was much disagreement over the details and the timing.  The end result was a fundamental agreement to disagree.  The 3GPP members left Busan with no plan to move forward.

With no formal 3GPP way forward for LTE-U, the group scheduled a nearly unprecedented, “unofficial” two-day meeting in Paris in January 2014.  More than 60 weary 3GPP members from all over the world traveled to the City of Lights to see if they could enlighten the “Je ne sais pas” that came out of Busan.  Twenty companies presented their views, focusing on such subjects as the motivation for LTE-U, the potential benefits, possible use-cases, the worldwide regulatory landscape, potential requirements, possible bands, performance evaluation, potential features, and the timeline.  This meeting seemed to produce more points of agreement than the Busan meeting, including the need to develop features in LTE to support inter-technology co-existence (e.g., LTE-U and WiFi in the same band) as well as inter-operator coexistence (e.g., two LTE-U operators in the same band).  But as in Busan, the members continued to disagree on the timing.

The next 3GPP plenary meeting was in Fukuoka, Japan in March 2014, and yet again, the crux of the debate centered on the timing.  This time, there was broader support from operators for a more accelerated timeline, as Verizon and China Mobile were joined by five others in support of Qualcomm/Ericsson’s idea.  The operators included NTT DoCoMo, T-Mobile USA, Deutsche Telekom, TeliaSonera, and China Unicom.  In addition, most infrastructure and device vendors supported Qualcomm and Ericsson, including Nokia, NSN, Alcatel-Lucent, LG, Huawei, ZTE, Hitachi, Panasonic, and others.  But the opposition was also strong, led by Orange, with Telefónica, Vodafone, AT&T, Sprint, SouthernLINC, US Cellular, DISH and a handful of vendors backing up the proposal to slow the process down.  Again, the end result was a stalemate.

However, the group did decide on one thing at the Fukuoka meeting:  there will be a half-day workshop devoted to more LTE-U discussions on Friday, June 13 after the next 3GPP plenary meeting that will take place in the hills of Sofia Antipolis, just outside of Nice, France.  Unlike the Paris meeting, this will be an official 3GPP meeting that many hope will advance the topic.  However, as September 2014 draws ever closer, one could say that the group wanting to put the brakes on LTE-U has been successful so far.

Whether and how 3GPP and the wireless industry resolve the many challenges and conundrums presented by LTE-U will have a major effect on both licensed and unlicensed spectrum.  The champion of the WiFi standard, the IEEE, is carefully monitoring the debate its progress, but wireless carriers, manufacturers, WiFi operators, and wireless broadband consumers around the world have a stake in the outcome of the debate taking place later this year.  


Cambridge White Spaces Trial

Gerry Oberst of the Hogan Lovells Technology, Media and Telecoms Practice moderated the afternoon sessions of the Cambridge White Spaces Trial Summit on 25 April.

16 May 2012

White Spaces in the U.S.

Despite some uncertainty regarding the potential impact of the FCC's proposal for incentive auctions of TV spectrum on the availability of this spectrum for unlicensed use, the Commission...

28 October 2011
Loading data