Qualcomm Inc (QCOM)

[khemara_qc]

How Qualcomm and Flarion could trump WiMAX

Nov 8, 2005
By: John C. Tanner
Telecom Asia


http://www.telecomasia.net/telecomasia/article/articleDetail.jsp?id=196740

The race for next-generation wireless broadband arguably got a little more interesting a couple of months ago when Qualcomm announced that it was acquiring FLASH-OFDM vendor Flarion Technologies. Qualcomm's move was a surprise to some, not least because Qualcomm's top brass, including founder Irwin Jacobs and CEO Paul Jacobs, had spent no small effort dismissing wireless broadband - particularly WiMAX - as overhyped and redundant. The basic position for Qualcomm was: "We already have wide-area wireless broadband today. It's called 1x EV-DO, and it works just fine."

On the other hand, those who have paid attention to Qualcomm's activities weren't surprised at all. For all of Qualcomm's ire toward WiMAX, it has never questioned the role of OFDM (the technology underlying WiMAX and indeed most next-gen wireless broadband standards) in future wireless standards. Indeed, Qualcomm already uses a form of OFDM for its MediaFLO technology and uses an OFDM waveform in combination with CDMA in its cdma2000 1xEV-DO Platinum Multicast solution.

The big question now is what Qualcomm will do with its new acquisition (assuming the deal is completed), and what it will mean in the wireless broadband scheme of things. At the very least, it could mean more formidable competition against the WiMAX movement - and not just in terms of fighting for contracts with service providers. First off, there's Qualcomm's interest in hybrid OFDM/CDMA solutions that could give CDMA operators a competitive option to WiMAX. Second, Qualcomm could be in a better position to sell its long-held position that WiMAX - which is to say mobile WiMAX, a.k.a. 802.16e - is tomorrow and unproven, while existing technologies like EV-DO (and now FLASH-OFDM) are today.

Thirdly, and possibly most important, Qualcomm's ownership of Flarion also gives it something else - control over Flarion's OFDM patents, to include technology for making OFDM mobile. The upshot of that could be that Qualcomm is poised to directly take on Intel for the future direction of wireless broadband.


Mobility issues

It's worth clarifying at this stage that the chief battleground here is mobile wireless broadband - that is, wireless broadband that allows terminals to move around within the coverage area. WiMAX 802.16d, which is standardized today, isn't designed for mobility. WiMAX 802.16e is designed for mobility but is not yet a standard.

The distinction is crucial, not least because 16e is considered to be "real WiMAX", says Paul Sergeant, marketing director for alternative wireless access networks at Motorola, which supports WiMAX. In fact, Motorola isn't even bothering with 16d and is going straight to 16e, because that's where the larger market lies.

"16e is also a superior technology in terms of spectrum efficiency," Sergeant adds. "It will be a more cost-effective solution for both the carrier and the customer compared to 16d."

Either way, it's mobile WiMAX that Qualcomm has been vocally attacking over the past year or so. In fact, one reason why its purchase of Flarion seemed a bit jarring at first was that Qualcomm's WiMAX criticisms could easily be extended to any wireless broadband technology.

For example, in September - a month after the Qualcomm/Flarion deal was announced - Qualcomm's senior marketing VP Jeffrey Belk authored a white paper entitled "Why MAX?” that reiterated Qualcomm's skepticism of the WiMAX hype machine. With the caveat that none of this reflects how Flarion will eventually figure into all this, Belk spends a lot of time arguing that, because radio is messy to integrate into devices and just as messy to implement, WiMAX will ultimately prove to be far more difficult to implement than WiMAX hypesters are letting on. It will all happen given enough time, money and smart people, but the bottom line, he argues, is that whenever the 16e standard comes about, it will be years after that before WiMAX will have its act together. And even then, there's still no proven business case for it.

"We go to Starbucks and rarely see anyone on a PC. We travel through airports and really don't see folks in the hotspot zones," Belk says.

While WiMAX is sorting itself out and looking for a business case, Belk argues, operators will already have the option of existing technologies.

"Vast areas - not airports, not streets, not urban cores - are being covered by networks that can efficiently provide hundreds of kilobits of reliable wireless access. Not in 2007, 2008 or beyond, but today," he says. "In our view, mobile WiMAX is potentially caught in the middle between the rapidly developing WWAN standards, such as cdma2000/1xEV-DO and W-CDMA/HSDPA and the WLAN standards that are evolving from 802.11 a/b/g to 802.11n, and other evolving standards such as Flarion's FLASH OFDM."


Bet hedging

That last sentence is key, perhaps since without it, Qualcomm's anti-WiMAX/pro-EV-DO stance could easily apply to any wireless broadband technology similar to WiMAX. Certainly Qualcomm has been at loggerheads with Flarion in the past, particularly in Europe where Qualcomm has lobbied for 450-MHz spectrum to be allocated for CDMA450, only to see it go to Flarion customers.

Now, however, Qualcomm's Belk says that Flarion has a role to play in Qualcomm's strategy - whatever it might be. Belk says the company is still evaluating its options, but does allow that the company is, in a sense, hedging its bets by broadening its technology portfolio.

"We believe CDMA will provide the most advanced, spectrally efficient wide area wireless mobile networks for the foreseeable future, but with Flarion we can now more effectively support operators that prefer OFDMA or a hybrid OFDM/CDMA track for differentiating their services," Belk told Telecom Asia. He adds that OFDM has figured into its strategy for some time.

However, Belk makes no bones about where Qualcomm sees the most value in Flarion. "Flarion has a world-class team and an expansive portfolio of OFDMA intellectual property," he says. "The combination of Flarion and Qualcomm's engineering resources strengthens Qualcomm's position as a leader in advanced wireless technology development and enhances Qualcomm's ability to design and license OFDMA systems, components and products for operators interested in OFDMA for certain present and future networks."

In short, at a time when the wireless industry is gearing up for OFDM-based network deployments, Qualcomm just acquired a potential royalty bonanza. And if Qualcomm understands anything, it's how to run a business on royalty revenues.


CDMA cellcos sewed up

If its competitors are worried, they aren't showing it. Many vendors have described the deal as a positive move for the industry as a whole in terms of OFDM's future. The more companies take OFDM seriously, the better.

Some analysts agree. "Qualcomm's acquisition of Flarion validates the wireless broadband market in particular, and creates more competition in the wireless industry overall," says ABI Research senior analyst Philip Solis. "In the long term, Qualcomm can position itself to compete in the market for 4G technologies. With Qualcomm's R&D and marketing strengths, Qualcomm will help Flarion's Flash-OFDM technology achieve more acceptance than it would have alone."

Another way of interpreting that, of course, is that Qualcomm is now a bigger threat to WiMAX than it was before. For example, it allows Qualcomm to gain customers it might have lost to Flarion. More to the point, however, it also gives Qualcomm's CDMA customers an alternative to WiMAX: a hybrid CDMA/OFDM solution.

The danger as Caroline Gabriel, research director of Rethink Research Associates observes is that "all this could shut the CDMA world off from adopting WiMAX as a parallel system or future migration path to 4G."


Gunning for Intel

That might be acceptable for the WiMAX camp, given that CDMA operators are a small market compared to GSM carriers, and there will be plenty of telcos looking for a wireless arm, and greenfield plays as well. However, the area where Qualcomm could make a deeper impact is the IPR.

Julien Grivolas, an analyst at Ovum, says that the IPR is at the true heart of Qualcomm's strategy. "OFDM will be an important component of next-generation radio technologies, and the value of the deal, up to $800 million for a company with virtually no sales, is a proof of Qualcomm's confidence in OFDM."

According to a research note from The Shosteck Group, one of the key patents that Flarion holds is for its technology to give OFDM mobile capabilities. And mobility is the name of the WiMAX game. Shosteck reckons that Qualcomm's beefed-up OFDM patents "may prove to be useful in delaying or controlling development of the rival mobile WiMAX (802.16e) technology."

While Qualcomm is bound to put Flarion's OFDM technology to good use in its own Platinum Multicast and FLO technologies, it will also undoubtedly use its IPR set its sights directly on Intel, the driving force behind WiMAX and particularly 802.16e, Shosteck says.

"It is no secret that Intel is the major competitive target for Qualcomm," observes Shosteck. "Given the use of OFDM in WiMAX, and that Flarion probably owns IPR in mobilizing OFDM, it may be a trump card for Qualcomm. Designers of mobile WiMAX 802.16e chips may have a hard time avoiding infringing Qualcomm patents."

Thus, Shosteck concludes, "Qualcomm probably wins either way - whether mobile WiMAX becomes established or not."


GSM vs CDMA redux

Not everyone shares such a dire view over WiMAX's prospects in terms of IPR. Paul Dittner, an analyst at Gartner Research, for instance, points out that many companies hold patents on OFDM, and it remains to be seen how the royalty structure will evolve.

"Qualcomm will not have the dominant royalty position with OFDM that it has with cdma2000, but it now has a better negotiating stance," he says.

However, Shosteck argues that there's another key factor in all this: Qualcomm's own reputation within the wireless industry. Its relationship with GSM vendors has ranged from turbulent to antagonistic, and the GSM community has never been easy with the idea that Qualcomm is getting a percentage off their migration to W-CDMA even as its executives publicly proclaim 1x EV-DO to be the far superior technology. Even Qualcomm's own CDMA customers in South Korea have complained about Qualcomm's royalty licensing scheme.

"Cellular operators may now be much more wary of Flarion and FLASH-OFDM," Shosteck says. "These operators will be looking for a more favorable IPR licensing regime to allow multiple suppliers to license the technology, with 'reasonable' royalty fees on infrastructure and devices."

The same may well be true of wireless vendors, which could go out of their way to back mobile WiMAX just to avoid paying more royalties to Qualcomm.


Not too late

Whether this is a good or bad thing remains to be seen. That said, some healthy competition from Qualcomm / Flarion could eventually make WiMAX a better and more competitive technology than it might otherwise become.

However the royalty issue works out, the more immediate challenge for WiMAX is one of Belk's strongest points: the fact that by the time mobile WiMAX sees serious commercial deployments in 2008, it will be up against both established commercial networks supplied by Flarion and IPWireless (the latter of which uses IMT-2000-approved TD-CDMA and has 14 live commercial networks to its credit today) and EV-DO and even HSDPA networks that have the extra advantage of being cross-subsidized by voice revenues.

Sergeant of Motorola argues that, aside from the fact that Motorola will have its first pre-standard 802.16e products ready early next year (all software upgradeable over the air to the final standard, incidentally) and that WiMAX's standards-based ecosystem approach will give it the advantage of economies of scale in the long run, it's early days for everyone in the wireless broadband market.

"Wireless broadband in general is at the early stage of the market, so I don't think that having a solution ready in 2006 is too early or too late," he says. "It's a good time to get in the game."

Fiona Chau contributed to this article




Everyone can play

WiMAX critics may have a point that mobile WiMAX will be at a severe disadvantage when it finally goes commercial in the next couple of years. By that time, it will be up against more firmly established players running FLASH-OFDM, TD-CDMA and of course HSDPA and EV-DO.

WiMAX proponents, on the other hand, dismiss such criticism by pointing out that wireless broadband isn't a zero sum game where only one technology wins. There will be room for everyone.

Even Motorola, which plays in the cellular infrastructure space as well as WiMAX (and still offers its proprietary Canopy technology as a fixed wireless product for unlicensed spectrum), maintains that these different technologies are complementary, and that different customers have different needs, depending on the market, the available spectrum and the service being offered.

To that end, Motorola counsels customers on what technology is the best fit for whatever service they have in mind, says Paul Sergeant, marketing director for alternative wireless access networks at Motorola.

"For example, if you already have 3G spectrum and you're already going down the GSM path toward UMTS and HSDPA and you want to offer high mobility, then UMTS/HSDPA is your best bet," he explains. "If you want to offer a portable broadband service with cheaper voice via VoIP and you have spectrum that's unsuited for mobility - for instance, higher spectrum bands aren't really suited for mobility because the propagation is lower - then you could consider WiMAX."

Peter Humphreys, Asia-Pacific VP of IPWireless, adds that not all technologies are created equal when it comes to cell capacity, which in turn determines what kind of service you can realistically support.

"HSDPA is the most limited of the three. You can only fit about ten customers per segment, or 30 per base station. That's okay for short-burst services where the customer is only using the network for a few minutes - short video calls or downloads," Humphreys says. "EV-DO Rev. A, by contrast, can support four times as many users per cell. TD-CDMA can support 750 to 1,000 users per cell, which means it can support users with tablet-PC like devices, always-on VPNs, that sort of thing."

- John C. Tanner