If there's one telecom company that knows about the follies of supporting two distinct types of wireless technologies at once, it's Sprint.
After all, the carrier's decline as a top competitor in the wireless industry can largely be traced back to its 2005 merger with Nextel that left Sprint running both its own CDMA network and Nextel's iDEN network. The iDEN network proved to a major headache for the carrier, with Sprint posting net losses of just under 4 million iDEN subscribers over the last two years alone. Ditching the iDEN network hasn't proved an easy feat, as the company has just started this year to move its iDEN push-to-talk services off the old network and onto its CDMA network.
So if the reports are true that Sprint this month will announce plans to host LightSquared's LTE network on its own infrastructure, it means that Sprint will be hosting both LTE and the WiMax network it has built in conjunction with Clearwire. With Sprint potentially once again supporting two different types of wireless technologies for its 4G services, does this mean the company is repeating the mistake it made when it acquired Nextel?
DEBATE: LTE vs. WiMax
Not necessarily. The difference between WiMax and LTE is far smaller than the difference that existed between CDMA and iDEN, so the technical challenges of hosting both networks shouldn't be as difficult. But from a cost standpoint, says Gartner analyst Phil Redman, it will still pose significant challenges since Sprint's rivals will have a comparatively simple setup where they only need to support their own LTE networks rather than the LTE and WiMax networks of two different companies.
"There's no doubt that the cost of having two networks is different from the costs of having one," he says. "It's technically feasible, yes, but it still ... stretches out your engineers."
Sprint bet big on WiMax back in 2006, as it earmarked $5 billion to build a nationwide network with the assumption that having 4G services up and running before Verizon and AT&T got around to launching their own LTE networks would give Sprint a major competitive advantage in the wireless data marketplace. But with Verizon getting its LTE network fired up in 38 markets last year, Sprint's time-to-market advantage expired before the company had made significant progress in upping its customer base relative to Verizon and AT&T.
In fact, the drag of the iDEN network has resulted in Sprint actually losing wireless subscribers in the period since its WiMax network launched. In September 2008, when Sprint fired up its first WiMax network in Baltimore, the company had 50.5 million total wireless subscribers. By the end of last year, Sprint was left with just under 50 million wireless subscribers. And with LTE poised to become the dominant wireless technology in the U.S. for consumer handsets, Redman says that Sprint will be forced to gradually migrate from WiMax to LTE to remain competitive.
"There's no doubt they're going to start diminishing their investment in WiMax," he says. "It's a big challenge but the alternative is even worse so it's something they really have to do."
ABI Research analyst Phil Solis also sees Sprint and Clearwire plowing more money into LTE in the coming years, although he notes that Clearwire's recent financial troubles could limit the amount of cash it has on hand to spend on new technology. Clearwire's biggest problem stems from the fact that the increased revenues from subscriber additions haven't at all been able to keep up with operating expenses that have soared over the past three years, going from $514 million in 2008 to $1.5 billion in 2009 to $2.8 billion in 2010. Even so, he says that Sprint and Clearwire have more than enough spectrum to set up their own LTE networks if they can find the resources to do so.
"Sprint is already migrating its iDEN network off the 800MHz band, so they could use that for LTE," he says. "Clearwire has a lot of 2.5GHz spectrum so they could use that for LTE. Funding is the only roadblock for what Clearwire could do there."
But supporting two networks at the same time will only be part of the challenge for Sprint. The other part will be in getting device manufacturers to create dual-radio handsets capable of supporting both technologies for a reasonable price.
"The device side will be the greater challenge," says Forrester analyst Charles Golvin. "Instead of having a device like [HTC's] Evo 4G that has to carry support for the 3G network and the WiMax network, it will have to support the LTE network now as well."
Golvin notes that Sprint could get some help from Qualcomm in this regard, as the company has developed an LTE chipset that is backward compatible with the CDMA-based EV-DO Rev. A networks that Sprint uses. This means that Sprint devices could merely replace its traditional CDMA chips with the new Qualcomm LTE chips rather than having to add a third wireless data chipset into the mix. Even so, Golvin thinks that such a chip configuration will still leave Sprint at a disadvantage compared with Verizon and AT&T, whose handsets won't need as many inputs since they don't have to worry about supporting WiMax.
But as Redman has noted, Sprint really doesn't have much of a choice at this point. It can't simply abandon its WiMax services and boot all of its data users back to its EV-DO Rev. A network while it makes the switch to LTE. And it can't stick with WiMax alone indefinitely since LTE has become the de facto standard for 4G data services on consumer handsets. So hosting two different wireless standards for its 4G services is far and away the best option for Sprint right now.
"For a tier-one provider trying to compete against other tier-one providers, going your own way is expensive in an industry where standardization is important," says Redman.
Read more about anti-malware in Network World's Anti-malware section.