Is the Industry Ready for LTE? - Connected Planet

Is the Industry Ready for LTE?
Connected Planet
Kevin Fitchard

Jan. 21, 2010

This next-generation wireless technology has arrived, but how long will it take for all the pieces of the LTE ecosystem to fall into place and for 4G to live up to its promise?

Long-term evolution officially ceased being long term a little more than a month ago.

It was then, just around the turn of the new year, that TeliaSonera launched the first two LTE networks in the capitals of Sweden and Norway, beating even the industry’s most optimistic estimates of when LTE would go commercial. Here in the U.S., Verizon Wireless and MetroPCS are set to follow in the back half of the year, as is NTT DoCoMo in Japan. It won’t be long before long-term evolution begins to recede in the rearview mirror.

LTE has finally arrived, but what does that mean exactly?  Will a revolution in mobile data services occur overnight, or are we only at the beginning of protracted developmental process that will make long-term evolution live up to its name?

When 3G first emerged at the beginning of the millennium, the industry preceded it with an enormous build up of hype upon which it utterly failed to deliver. Almost a decade later, the industry can now claim to have lived up to 3G’s promise, but only after delivering more powerful 3G networks and years of trial and error with applications and devices. For the first several years of their existence, UMTS and 1X networks were basically glorified voice networks. 4G, too, has suffered from similar hype, but this time around the industry has been far more restrained in its claims. While 3G’s hype centered largely on technology, boasting about 4G centers on concepts like cost-per-bit efficiencies, greater capacity per user and better customer experience. 4G may eventually turn the industry on its head, but right now it seems mainly to be about a better way of delivering data services.

Verizon Wireless will have one of the world’s first large-scale LTE networks in the world with plans to networks cover 25 to 30 markets and 100 million pops by the end of 2010. The networks will be ready - that’s almost a certainty - but whether all of the other pieces that go into making an LTE service hum will fall in line is a bit harder to gauge. It depends on what question you ask, said Brian Higgins, VZW executive director for LTE ecosystem development.

If the question is: will Verizon Wireless be able to offer relatively straightforward commercial mobile broadband access service via laptop data cards, then the answer is definitely yes, Higgins said. To meet that goal, Verizon Wireless and its vendor partners have had to cover a lot of new ground - not only deploying new radio access and core networks, implementing a service delivery architecture and upgrading capacity in its backhaul and transport, but also introducing a new subscriber identity module (SIM)-based billing and subscriber management system as well as work with manufacturers to ensure it has devices available. All of those elements have fallen into their proper place, Higgins said, allowing Verizon to offer simple broadband access from day one.

But if the question is whether a new mobile broadband revolution, complete with a myriad of new devices, applications and use cases, will emerge on day 1, then the answer, Higgins said is no. The first set of devices are going to be dongles, which will be used with the types of data plans customers are already used to, Higgins said. But when it comes to the new services and devices 4G is expected to shepherd into the wireless industry, that will take some time, Higgins said.

Verizon and its ecosystem partners are just beginning to explore the possibility for new devices and applications over 4G, and he’s not just talking about phones with better data connections. While some smartphones will invariably make their way to LTE network, the new generation of devices could come in any number of form factors, which could range from video devices that take advantage of LTE’s wide open pipe to machine-to-machine modules that don’t need massive bandwidth but can take advantage of LTE’s lower cost of data delivery. It will be more of an iterative process, Higgins said. The trickiest problems will be with embedded solutions that don’t necessarily need to take advantage of higher throughput but could benefit from a more efficient network.

MetroPCS is targeting its LTE launch for the same time period as Verizon, but its plans for that network differs greatly from Verizon’s. Metro plans to use the 4G network much the way other operators use their 3G networks, offering smartphones and data plans. If it is looking more toward the short term benefits of 4G, Metro has a good reason. It has no 3G mobile broadband data network over which to offer those services today. Metro has built CDMA 1X throughout its network, but never turned any of its 1X channels into EV-DO data channels, allowing it to offer narrowband data as well as voice, but not any higher-bandwidth services.

We saw that for the applications our customers were demanding, except for downloads and video streaming, our network was enough, said Roger Linquist, CEO of MetroPCS. LTE will give Metro the opportunity to offer this missing piece of the data puzzle as well as support future high-throughput apps, while allowing it to skip a generation jumping straight from 2G to 4G, Linquist said. I don’t want to invest in a technology that we’ll have to phase out in a few years.  We see this approach as advantageous for us going forward from both a cost standpoint and an efficiency standpoint.

By moving directly from 1X to LTE, though, Metro is much more dependent on developing an immediate ecosystem of devices. Verizon can experiment with new devices and business models (it’s already begun with its LTE Innovation Lab, the first result of which was a connected home solution developed by 4Home and unveiled at CES). But Metro needs 3G devices for its 4G network. That means smartphones.

LTE-ready smart phones, however, will require not only dual-mode chipsets so they can fall back on Metro’s 1X network, but they’ll need to have innards that are much more sophisticated than the silicon that goes into a USB card. Furthermore, handset vendors will also need the incentive - meaning the promise of volume sales – to make them, incentives that took years to emerge for 3G networks. Linquist, however, is confident that development in the LTE world will be accelerated. While it won’t be offering handsets at launch, MetroPCS has already selected Samsung as its primary handset partner and is expecting the Korean vendor to fully deliver on its promise of a dual-mode 2G-4G smartphones in the next 18 months.

Are operators being too optimistic? Depending on where one puts the date of widespread 3G deployment and the date of widespread mobile data adoption, 3G took anywhere from three to seven years to mature into a true ecosystem. In fact, every generation of technology, whether it was analog, GSM-CDMA or HSPA-EV-DO, followed a pretty typical curve of development and ramp up before reaching maturity typically eight to ten years after the initial standard was finalized, said Eran Eshed, vice president of marketing and business development of Altair Semiconductor, a silicon vendor specializing in 4G chipsets.

“I don’t think there is any cellular technology that can go from standardization to mass-market commercial services in less than four or five years,” Eshed said. “LTE isn’t any different. It will go through a pretty typical development curve.”

That might sound odd coming from a vendor that prides itself in beating its larger competitors to market with commercial grade 4G silicon, but Eshed said Altair is merely being realistic. Even with silicon shipping in volumes, there is still a long period of device development and testing that needs to be coupled with large-scale network deployment and intensive application development before any technology ecosystem can fully emerge. If any further evidence is necessary, one need turn only to the Clearwire WiMax network. Though much earlier to the market, WiMax faced the same demand for mobile broadband as LTE and its vendors carried the same lessons over from the deployment of 3G. The first large-scale WiMax networks weren’t deployed until three years after the standard was approved and there hasn’t been a proliferation of new applications and devices in the WiMax ecosystem. “They offer pure Internet access services,” Eshed said.

That’s why it’s so impressive that the radio access vendors have managed to produce commercial equipment so soon after the LTE standard was passed, an accomplishment all the more impressive considering that capital expenditures in telecom all but dried up during the recent economic downturn, said Keith Higgins, vice president of global marketing for telecom consulting and software development outsourcing firm Aricent. But Higgins also fears that the radio aspects of the ecosystem may have outpaced all of the other elements necessary to develop a full 4G service. All of the work necessary to create an operational network – from billing and support systems to device silicon – is still under development. Basic questions like how voice and SMS will be handled on LTE remain unanswered. The network radios are definitely ready, but it will take time for other elements to catch up, Higgins said.

“There will be a lot of LTE deployments in 2010, but LTE will be more of a marketing initiative this year,” Higgins said. “The fact is that a lot of work done on LTE so far has been on the radio.”

Analysts are also skeptical that LTE will have an immediate impact on the industry. “What I believe is in 2010 we’re going to see LTE networks deployed in a meaningful way that will put us on a trajectory that will see 4G services delivered to us in the next few years,” said Phil Asmundson, vice chairman of Deloitte and leader of its US technology, media and telecommunications practice.

Wide-scale adoption of 4G services is unrealistic in the next few years, but that doesn’t necessarily mean that the industry will have to wait four or five years for the ecosystem to develop, Asmundson said. LTE will probably take just as long as 3G to mature, but its development along that curve may be front-loaded. In 3G, the industry had to create a business around mobile data. It took years of trial-and-error before the right networks, the right devices and the right services converged to create the massive market we have today for smartphones and mobile broadband.

“I think the device makers will be quicker on the uptake with LTE because they have more experience in the business models surrounding data-enabled phones,” Asmundson said. “They don’t have to do an awful lot to stimulate demand for these things.” On the software development side, there’s also a precedent. Application developers have now been building robust applications for 3G networks for years and have a much clearer understanding of what services and software the market wants. Unlike in the 3G world, they aren’t inventing a market from scratch.

So far device makers seem much more aggressive in their support for LTE. Samsung had USB dongles ready for TeliaSonera’s early launch of LTE in Oslo and Stockholm and has committed to producing dual-mode handsets next year. Almost all of the silicon vendors have begun sampling or shipping their first-generation LTE chipsets, many of which are still single-mode 4G platforms designed for data cards, but all of them have dual-mode and handset silicon in the pipeline. When LTE silicon is mentioned, though, most eyes turn to Qualcomm, which is not only the world’s largest 3G device chipmaker, but will play a crucial role in early LTE launches, since many of the first LTE carriers have CDMA networks. Qualcomm, however, has been taking it on the chin from critics about its development timelines, which have lagged many of its competitors; at least as far as single-mode silicon is concerned. Their concern is valid—without Qualcomm a major piece of the LTE ecosystem is missing—but their conclusions are off, said Peter Carson, senior director of product management for Qualcomm CDMA technologies.

Qualcomm has sampled its first chipset, the MDM 9000 aimed at data cards, which is on target to support the 2010 launches, Carson said. And it’s more advanced handset silicon, with embedded software and application processor, is on track to support smartphone releases in 2011. Qualcomm may be a bit later to market, but Carson said Qualcomm in both cases has created a far more integrated dual-mode solution while its competitors are using single-mode solutions to get to market more quickly. The utility of a single-mode solution is questionable in a network where 4G coverage is a long way from ubiquitous, Carson said.

“We’ll have the first integrated solution in the market,” Carson said. “We took the approach that we’re not going to get caught up in the hype. We set a realistic date for LTE and we’ve stuck to it.  … We don’t need to state fictitious dates to validate the ecosystem. Our customers know we’ll be on time.”

The industry has learned a lot from its mistakes since the deployment of 3G, said Ken Wirth, president of LTE/4G networks for Alcatel-Lucent. Frankly, Wirth said, 3G deployments were driven “by some weird economics.”

“Carriers, as they looked at 3G, weren’t quite sure what to do with it,” Wirth said. “They just knew they had a lot of bandwidth. They didn’t investigate the services and business models that would go along with it.”

This time around, though, both operators and vendors are better prepared, Wirth said. Operators are planning far ahead of their deployments the services they plan to offer and the business models they plan to implement, Wirth said. For instance, long before Verizon Wireless turned its first trial networks on in Boston and Seattle, it had device and application developer programs and its LTE Innovation Labs initiative in place. It and other operators have been putting together the LTE business case for the last 12-to-15 months. Vendors like Alcatel-Lucent have also branched out from merely supplying equipment to building an application ecosystem. Wirth pointed to ALU’s Ng Connect program, which looks to link operators, device makers and network providers with application providers outside of the industry, exploring projects ranging from digital signage to the connected car, Wirth said.

Vendors have also focused much more on device interoperability. Shortly after the LTE standard passed in late 2008, infrastructure vendors began pairing off with their handset and chipset counterparts in attempt to head off compatibility issues, said Erik Ekudden, vice president of technology and industry for Ericsson. Vendors have also simplified network planning processes by implementing self-organizing network technologies, which greatly reduce the site planning and engineering delays and costs associated with any new deployment, Ekudden said.

Even with all of that advanced preparation, there is another key factor accelerating LTE, according to Ekudden: the LTE ecosystem is a much simpler ecosystem.  The radios and the new end-to-end IP architecture are new, but 4G—at least in its early stages—is borrowing the rest of its elements from 3G, whether it be its billing systems, applications or device form factors.  Most importantly the fundamental mobile broadband business model is already in place—it must only be implemented on a new type of network, Ekudden said. “3G has really paved the way for 4G,” he said.

Does that mean that LTE is just a more efficient and faster way of delivering 3G data services? Initially yes. In fact, there seems to be a definite parallel there between the way 3G and 4G services were initially deployed. Just as voice and SMS were the proven applications in 2G, making them the first applications widely used on 3G, 3G data services are the proven applications that will initially take hold in 4G. The consensus is that the industry has existing applications it can get started quickly on LTE while exploring the true 4G applications of the future. It might be years before the industry sees its first true 4G app or 4G device. Even if those apps and gadgets were ready today, there is little likelihood they’d be deployed or adopted immediately anyway, according to Juha Lappalainen, head of LTE for the North American market for Nokia Siemens Networks. Offering 3G services over 4G networks is an easy leap for operators since they have robust 3G footprints to fall back upon. But while 4G network coverage is still limited they’ll be reluctant to launch a high-bandwidth application that sputter and die when a customer travels outside of 4G coverage, Lappalainen said.

“If you start introducing data-hungry services immediately, you will have problems,” Lappalainen said. “Some of these networks will be coverage limited and will have to fall back on 3G networks. If you start running applications designed for LTE on HSPA, you will pretty much overload those networks.”

What these eventual 4G apps and devices will look like is anybody’s guess, just as nobody could have predicted the iPhone in the early days of 3G. But the industry isn’t standing around waiting for the 4G-equivalent of the iPhone to emerge. Carriers, as well as infrastructure, device and silicon vendors, all have a clear idea of the 4G route they want to pursue. Intel is hoping to use its expertise in processing to shepherd in a new category of mobile computing and Internet devices, dependent on the high-throughput of a 4G connection. Alcatel-Lucent’s Wirth said he is confident that the rich media services that have proliferated in the home with IPTV can be transplanted to the 4G network and that cloud computing can find its ultimate use with a mobile end point. Verizon Wireless makes no secret of the fact that it wants to embed an LTE chipset into just about everything from home appliances to industrial machinery, taking advantage of LTE’s greater cost-per-bit efficiencies. On the other side of the equation, VZW wants to use LTE’s 100 Mb/s of capacity as an alternate residential access technology and as means of offering connected home services.

Everyone in the industry is still a bit cagey about when exactly these new apps and devices will emerge—an element of caution resulting from the enormous hype surrounding 3G. But we may not have to wait too long before we see application developers testing the limits of the LTE network. At least one developer already has a true 4G app developed and is just itching to bring it to market.

Panasonic has developed for its Toughbook computer line a video application called the Arbitrator 360, which allows a member of law enforcement to simultaneously view six different video streams from vehicle mounted cameras, giving an officer sitting in a car what amounts to 360 degree view of his or her surroundings. Panasonic has built remote viewing capabilities into the application, which would allow a dispatcher or supervisor to view the video streams live. But in order to fully realize those capabilities, Panasonic needs 4G connectivity, said Victoria Obenshain, director of wireless strategy for Panasonic Computer Solution Company.

Obenshain said Panasonic had originally planned to include LTE modules in its designs in 2011, targeting 2012 for commercial release, but with the acceleration of LTE deployments, it’s re-evaluating its plans. Panasonic might wind up being the first laptop maker with an embedded LTE solution and 4G applications suite. Several factors will go into determining whether it pushes LTE into notebooks in early 2011, including the availability of the appropriate LTE module and the size of the footprint (ideally 65% of the major markets in the US would be covered, Obenshain said). If the industry can meet Panasonic’s basic criteria, Obenshain sees no reason not be very aggressive in 4G considering the enormous and immediate benefits it could deliver to one of its core customer bases, emergency services.

“We’ve built our whole business model around embedded communications,” Obenshain said. Every generation of wireless has greatly expanded the capabilities of its emergency and field-services applications, Obenshain said, “but I don’t think anything we’ve seen previously will come close to touching what we’ll see with 4G.”