Everything to play for
By Mike Hibberd
Head to head technology battles are familiar territory in the wireless industry. Even today, if you listen carefully, you can still hear voices expounding the technical benefits of CDMA over GSM. Never mind that GSM (and its ‘family’ of standards) enjoys around 80 per cent of the world market, rendering the rallying cry futile – competing technologies are lobbied for at great expense and with considerable passion.
The great GSM/CDMA battle may have died out now – not least because the GSM community’s evolutionary process took it onto a variant of CDMA anyway – but that doesn’t mean the industry has seen its last technology tussle.
Vodafone Group CEO Arun Sarin was trying to motivate both the operator and vendor community during his speech at the 3GSM World Congress a year ago, when he warned that LTE (The 3GPP’s Long Term Evolution plan for cellular access) risked losing a time to market battle with mobile WiMAX. “LTE is still at the standards stage, while WiMAX is a commercial reality,” said Sarin. Whatever his motivation, Sarin framed the situation in competitive terms.
WiMAX has emerged as a fixed wireless solution but the companies behind the technology – represented by the WiMAX Forum – are keen to push the 802.16e mobile version of the standard. Cellular players may have their data demand covered with HSPA or EVDO (depending on their location) for now. But the time will come when some will have to upgrade to an OFDM technology, which will involve a new network deployment. And then they will be faced with options.
For every industry voice that talks up the face off between technologies, though, there is usually a counterpoint. These solutions are complementary, not competitive, according to this alternative perspective. And this is the first problem in trying to ascertain how WiMAX and LTE might fare relative to one another: not everybody wants to see the situation defined as a choice between the two.
Kerl Haslam, chairman of the Mobile WiMAX Acceleration Group – a
Steve Pusey, global CTO of Vodafone, which plays in more markets than any other mobile operator, believes the two technologies will ultimately converge. “If you look at it, they’re basically the same,” he says. “What’s the foundation of LTE and WiMAX? OFDM as a modulation technique is the biggest step, and the use of MIMO antennas. More likely than competition, you’ll see convergence.”
The two technologies do have separate conceptual beginnings. LTE is designed to offer a cellular service with great enhanced data rates, while WiMAX is a broadband technology looking for an inroad to mobility. It had to be conceived differently to LTE, according to its proponents, because it needed its own USP.
The WiMAX Forum’s world view sees WiMAX as much more than just a way for people to access the internet. Intel’s vision is that there will be a WiMAX chip in all manner of consumer devices: cameras, camcorders, games consoles, mp3 players, cars, pretty much anything one might think of. From this perspective, and by comparison, LTE is anchored in the handset play that has been the mobile industry’s historical sphere of operation.
But the two technologies are destined for competition in one sense because each is being positioned to move into the other’s space. Just as it is hoped that WiMAX can deliver enhanced mobility, LTE is envisioned as a means of broadband access provision as well as a mobile technology.
Paul Senior, CTO of Airspan Networks, was a founder member of the WiMAX Forum, and still participates as a member of the organisation’s Marketing Working Group. While he says that Airspan will respond to market demand in terms of technology, the firm is a big backer of WiMAX technology. “This is definitely a competitive technology to LTE,” he says.
What WiMAX clearly is not, however, is a competitor to HSPA. One of its proponents’ favoured campaign cries is that it has a time to market advantage over LTE. But even the evangelists concede that there’s no turning the head of the GSM-family players just yet. “Everybody that jumped on the [WCDMA] 3G bandwagon has HSPA in front of them,” says Senior. “They’ve made too much investment to make the switch from CDMA in 2008.”
So, what is the timeframe? The WiMAX community suffered a blow last year when its flagship deployment – planned as a joint effort by
This year will therefore be crucial for the WiMAX camp. Both Sprint and Clearwire have indicated they are keen to continue developing services based on the technology and the success of Sprint in particular will be scrutinised as bellwether. Sprint recently claimed it was on track to launch commercial services in April this year [2008].
Elsewhere, the Forum estimates that some 300 operators in 65 countries have run – or are running – pilots and trials of mobile WiMAX. As 2007 drew to a close, a consortium led by carrier KDDI won one of two Japanese 2.5GHz licences and plans to operate mobile WiMAX, launching in 2009. The other licence went to PHS service provider Willcom, which is planning a next generation PHS service.
Informa Telecom & Media expects there to be fewer than ten million WiMAX subscribers (fixed wireless and mobile) globally by the end of 2008, growing to 65.6 million in 2012. The WiMAX Forum estimates that there will be 410,000 mobile WiMAX subscribers by the end of this year [2008].
Othmar Kyas, director of strategic marketing, network diagnostic, at test and measurement house Tektronix, provides more information on the lead time that WiMAX enjoys. “With LTE we are just in the phase where proprietary demo systems are being phased out and the first pre-commercial LTE gear is being made ready for 3GSM this February.” Kerl Haslam’s assessment is that mobile WiMAX has a two-year lead on LTE.
A time to market advantage, though, is only as good as what is done with it. One popular assessment is that mobile operators that evolved along the CDMA2000 route – and are therefore unlikely to want to adopt a technology from the GSM family – will represent a prime market for WiMAX.
“These are the people who have a choice,” says Airspan’s Paul Senior. “Almost all of those players say that what they need is something that does mobile broadband very well, and that mobile WiMAX is the best for that,” he adds. Clearly Sprint is one of those carriers. KDDI and SK Telecom are others.
There is one glaring exception to Senior’s trend, however. Verizon Wireless, the
Othmar Kyas expects LTE and WiMAX to split the market along similar lines to GSM and CDMA, with an 80/20 divide in favour of the 3GPP’s solution, indicating that he shares Senior’s outlook for the CDMA camp.
Another means by which WiMAX might become established is as a data add-on to existing voice services. “Even in developed countries there are relatively small numbers of operators with UMTS licences,” says Kyas. “There are a number of scenarios where two or three operators are planning to acquire inexpensive GSM licences for voice and build a WiMAX network for data. UMTS licences were sold for billions of dollars. The relatively inexpensive WiMAX infrastructure in combination with inexpensive GSM voice costs might create a business case for a number of operators out there.”
These are business models and concepts that generate debate. There is more consensus around the chances that WiMAX has for success in emerging markets. In these countries, fixed infrastructure is often found wanting and broadband internet access is sorely needed. “Where you’ve got a situation with very little DSL penetration and you want to provide a much more basic level of data connectivity,” says Aircom’s Margaret Rice-Jones, “then I think WiMAX is a technology that can do very well. That is the early market for WiMAX; nomadic rather than mobile.”
Non-GSM cellular players, a modular data offering for mobile operators and nomadic wireless broadband access in emerging markets: the business models differ but the overall purpose is the same. Each of these options allows WiMAX to be seeded in the market so that it can evolve technologically and strategically to the point when GSM family cellular carriers may begin to look at their options for the fourth generation.
“That,” says Paul Senior, “is when we’re going to see defection. Because WCDMA is pretty much finished in a couple of years’ time in terms of its roadmap. It will be out there forever, of course, but in terms of what operators then install, the choice is much more interesting.”
Not everybody would agree that the end of the WCDMA line (including the HSPA upgrades) is quite so imminent, however. “We’ve got a few years’ of legs ahead of us,” says Vodafone’s Steve Pusey. “I can sense peak speeds of 28.8Mbps in the not too distant future. And that’s quite a head of steam to offer our customers. There’s an awful lot you can do with that for consumer and business services before you need to worry about upgrade. We’ve got a nice path for the next few years.” If this is the case, and LTE matures concurrently, the WiMAX time to market advantage could be negated.
Pusey’s view on the headroom afforded by HSPA is shared by many WCDMA operators, but not by Paul Senior. “That’s just not true,” he says, continuing: “If you look at how much spectrum they’ve got and the spectral efficiencies of those technologies, and the lack of smarts, it’s simply not going to be enough. Go to
If LTE requires a couple of years to mature before it is ready for deployment, then so does mobile WiMAX – certainly if the companies behind it want to see it evolve into a technology that can take on cellular standards. And this is something that the WiMAX community does not seek to hide. “WiMAX today is more of a broadband technology than a mobile technology,” says Senior. “We call it mobile WiMAX and, yes, you can drive round and do handovers. But it’s not really architected for mobility first. It will always be about broadband capacity – whether you could do that while travelling on an TGV through
This will change, he says, as the technology evolves to the next iteration, 802.16m, where mobility will be the trust of development work. Even so, it’s a fact that could lend a feeling of security to the LTE camp.
The voice legacy of this camp is important to LTE. Voice revenues are still the major breadwinner for cellular players and WiMAX is a data technology. As the Forum itself says, “WiMAX technology is designed to supply data bandwidth only, increasing the bandwidth without compromising voice service quality, since voice services are not operating on WiMAX bandwidth.”
Othmar Kyas sees this as a potential drawback for WiMAX. “Nobody really knows how well voice services are going to work over WiMAX. This is the biggest unknown and the biggest limitation. At the moment, the large operators deploying mobile WiMAX plan to only launch data services, which implies that, in any case, for mobile voice you’ll have to have another technology.”
Spectrum is another point of contention, one that Vodafone’s Steve Pusey sees as a potential obstacle for WiMAX. “WiMAX uses TDD spectrum primarily, while LTE is FDD. Most of the operators that are looking at these technologies have an FDD footprint, like us,” he says. “There is a natural choice for us and it’s difficult to see, because of spectrum, how WiMAX could play for us in our Western European footprint. The spectrum we have naturally lends itself to an evolution towards LTE.”
Vodafone is in both camps, though, and committed to pushing both standards; a point Pusey is keen to stress, although he won’t give a view on which is the better technology. He also points out that LTE could be delivered in a TDD flavour, and that an FDD WiMAX could also be developed.”
Exactly when that will happen is a matter for some debate. “The WiMAX Forum will have an FDD profile for mobile WiMAX inside six months,” Paul Senior says. “We’ve been working on it for the last 12 months. We’ve been a bit quiet about it because we wanted to get the IMT 2000 decision. And if we had gone to IMT with an FDD profile, we probably couldn’t have got it through. We decide to go for something that was a little less threatening, which was a TDD profile.
“There will be an FDD profile, it will sit at 2.5GHz FDD allocations just as well as any other technology. And in terms of implementation, we’re only 12 months away from products. If our competitors are betting on [the FDD/TDD issue] then they’re going to be quite surprised,” he says. This could be big news and, arguably, gives the strongest indication yet that WiMAX is indeed going to be positioned as a serious competitor to LTE.
But when these comments were run in MCI’s weekly email newsletter, A Week in Wireless, the resulting attention drew the following from WiMAX Forum chairman, Ron Resnick: “Contrary to any of the unofficial statements made recently, the WiMAX Forum has not made any Board approved policy or determination of when FDD mobile WiMAX system or certification profiles will be created. All profiles will be determined by the WiMAX Forum, driven by market demand, and the WiMAX Forum is exploring mobile profiles for FDD certification and is defining a network architecture to support FDD. However, no decision has yet been made when to propose an FDD evolution of the ‘WiMAX’ IMT-2000 air interface to the ITU, nor has it been decided what specific profile might be proposed by WiMAX Forum in the future.”
Spectrum is a crucial issue, as is technical performance. Time to market advantage cannot be underestimated. But in all probability, each of these will be trumped by cost. In this industry lower cost comes with scale and the perceived enthusiasm among the operator community for each of these technologies will play a part in setting the vendors’ pricing strategies.
Othmar Kyas reveals that the engineering budgets of the major vendors reveal “a clear preference” for LTE. The leading vendor in the mobile space, Ericsson, is a very visible absentee at the WiMAX Forum, while its competitors are all members.
Cost is a serious issue for the vendors too, of course. And that’s why Steve Pusey reckons the two technologies really will converge in the end. “The vendors themselves will look at the two portfolios and say that, basically, the ingredients are very similar. The only difference will be addressing different spectrum needs so they’ll just aim for one solution with different spectrum,” he says.
Obviously vendors will do what their customers ask of them. But by the time LTE is being deployed, and 802.16m (which Paul Senior promises will do everything that LTE will do) is ready, the respective roles of operator and vendor may have changed significantly. Perhaps by then, the operators will be increasingly prepared to outsource decisions as to which technologies are best to the vendors that are running the networks day by day.
