POST OF THE DAY
QUALCOMM, Inc.
Multiplex and Conquer

Format for Printing

Format for printing

Request Reprints

Reuse/Reprint

By BRational
September 2, 2003

Posts selected for this feature rarely stand alone. They are usually a part of an ongoing thread, and are out of context when presented here. The material should be read in that light. How are these posts selected? Click here to find out and nominate a post yourself!

Multiplex and Conquer: Nokia's Ivvy Divvy Strategy

With much fanfare, Nokia recently announced "the world's first" cdma2000� 1xEV-DV high-speed packet data phone call completed with Nokia chipset:

SAN DIEGO - Nokia (NYSE:NOK) today announced that the world's first cdma2000� 1xEV-DV high-speed packet data phone call was completed at Nokia's CDMA product creation center in San Diego. The call, achieving a peak data rate of 3.09 Mbps, was made between a test set based on a commercially available Nokia 2285 handset upgraded with a Nokia 1xEV-DV chipset and a Racal Instruments, Wireless Solutions Group, 1xEV-DV base station emulator. This chipset is the world's first to support complete 1xEV-DV Release C functionality.

Hyperbole and embellishment are what PR writers get paid to do, and Nokia's writers can hold their own against the best in the trade. Note for instance the commercially available Nokia 2285 handset upgraded with a Nokia 1xEV-DV chipset intended to suggest that 1xEV-DV (affectionately referred to as Ivvy Divvy hereafter) is somehow close to commercial deployment. However, what I find somewhat ironic in this announcement is the suggestion that a data only call somehow demonstrates complete functionality of a data and voice standard. One could have just as easily used a commercially available and deployed Qualcomm EV-DO chip for the same feat... But we won't let this minor quibble stand in the way of this serious discussion.

Is this the latest manifestation of what psychologists in San Diego have diagnosed as a severe case of CCE (CDMA Chipset Envy)? Other symptoms of CCE include the obsessive compulsive need to add using its own CDMA chipset in every other sentence of a press release, as in the latest announcement of Nokia's entry-level CDMA phone for Unicom:

Shang Bin, Vice President of China Unicom. "We are pleased to see that Nokia has launched its first CDMA phone in China using its own CDMA chipset, indicating a strong commitment on Nokia's part. We warmly welcome Nokia's entry in the Chinese CDMA market."

"Nokia is the only mobile phone manufacturer in the world using its own CDMA chipset across its entire product line. This enables us to meet users' increasing needs for mobile features quickly and flexibly.

1. Back to the serious business of Ivvy-Divvy. Why, one might wonder, has Nokia so aggressively led the GPP-2 deliberations to adopt and standardize CDMA 1XEV-DV, or Releases C and D of the CDMA 2000 standard? (The difference between Releases C and D is primarily in the so-called backward link (from handset to base station); Release D offers about twice the data rate of Release C on the backward link, but the same rate on the forward link). Given Nokia's attempts to marginalize CDMA operators, and its limited presence in IS-95A/B CDMA (2G), one would question the motivation behind this great concern and enthusiasm for the evolutionary path of these operators. But the Ivvy Divvy strategy seems consistent with the way Nokia appears to behave every time it finds itself left behind by technological developments: stall for time by requiring lengthy "standardization" processes, in the name of "open standards", advocate technical features that are maximally distant from anything competitors may be offering�which provide desirable sounding features that everyone would agree with, yet require cumbersome technological trade-offs to accomplish in a stable manner, resulting in the kind of syndrome we have all too painfully witnessed with WCDMA. Furthermore, in the standard setting process, band with enough of your rivals to counter the more dominant rival, often resulting in eating the temporary allies' lunch (witness Ericsson first, and now Motorola).

2. The saga that led to WCDMA is well known to this board; while all authorities agreed that CDMA was the way to 3G, and that TDMA and its GSM variant were destined to the 2G museum, Nokia and its European allies at the time concocted an ill-conceived prematurely cumbersome air interface in a 5MHz band, which they called WCDMA. Never mind that most operators did not have such bandwidth at their disposition (we shall have grand auctions and allocate the spectrum to the highest sucker); never mind that no research lab had anything close to a working prototype in sight�and never mind that the CDMA leader already had a working, low risk, low investment technology working in the field in the form of CDMA2000. Still, Nokia and its allies succeeded in splitting for good the future 3G universe, into CDMA2000 and WCDMA camps. And while more than 60 million CDMA 2000 handsets are now in use, WCDMA still has to crack its first million, and to successfully hand off between WCDMA and GSM... Think of the unrealized market potential that has been deferred by at least two years for Nokia and other handset makers. Yes, we know, Dr. J. told them so, but we won't rub it in...

Still, what was meant to be an end run on Qualcomm's IPR claims proved fruitless, as each and every significant player, including Nokia, has recognized these rights, however reluctantly.

3. Now back again to Ivvy Divvy. You'd think the GSM giants would leave the small fry CDMA markets of North America, Japan, and oddball Korea alone in their CDMA2000 un-roamed world. But Nokia had a simple evolutionary solution for them�while their GSM cousins could all afford 5MHz of additional spectrum to run WCDMA in addition to the GSM voice bandwidth, CDMA2000 operators certainly wanted to squeeze data and voice in their single 1.25 MHz channel. And they should wait a few more years for the privilege�why mess with that simple commercially deployed high data rate Ivvy Doo solution that is already supporting video streaming in Korea and Japan?

Nokia's enthusiasm for Ivvy Divvy is commendable. After recognizing its lagging role in IS 95A/B, and subsequently 1X (with all the talk of phones using its own chips, Nokia still does not field a commercial model that supports 1X data rates, what Sprint calls "data vision"�only voice), it has determined that it is losing a growing slice of the global telecom pie�in fact, the fastest growing slice, in relative terms, with the likes of China and India joining the CDMA evolution). Hence, time to standardize, and encourage the industry to adopt "open" standards, on its own terms. Qualcomm caught the world by surprise with EV-DO�derived from its research in HDR (High Data Rate); a clever, relatively simple and robust utilization of a separate narrowband channel for optimal data transfer. Surely Nokia couldn't control any part of the DO agenda, hence the Ivvy Divvy approach.

4. Conceptually, EV-DV could be reasonably attractive, in the sense that it dynamically allocates bandwidth to both voice and data transfer, exploiting the burst-y nature of the signals�really, much along the lines of why CDMA is efficient in the first place. However, data and voice have different characteristics, and dynamically blending them to achieve the desired high data rates requires carefully tuned algorithms. An enthusiastic presentation of the benefits of EV-DV can be found in this Nokia white paper, published this past June.

The document is more of a brochure, replete with general features and marketing slogans, rather than with technical facts. But it gives the picture. A better document on the technology and its specific features is this Motorola white paper, published more than a year ago.

5. So what is Qualcomm's position on Ivvy Divvy? Some of the answer lies in this white paper from 2001.

In its usual no-nonsense pragmatic approach to wireless telecom issues, Qualcomm states:

Furthermore, optimizing voice and data on different carriers is advantageous for both services: it simplifies system software development and avoids difficult load-balancing tasks.
1x/1xEV systems proven technologies are the lowest risk alternative.

It begins to sound like CDMA 2000 vs. WCDMA all over again�a practical, robust, here and now solution, against another conceptually elegant approach that requires a clumsy implementation�and a few more years until it sees the light of day commercially.

Qualcomm is not alone in its assessment. One of the conclusions of a recent 77-page report, CDMA2000 1xEV-DO: Opportunities, Challenges & Competitive Strategies, released by Datacomm Research Company is that:

The success of CDMA 'data only' (EV-DO) will reduce future demand for CDMA 'data-voice' (EV-DV). The latter will mainly be relegated to operators with insufficient spectrum.


6. Here again, unable to master a technology, Nokia seeks to divide and conquer; now it's the turn of the CDMA2000 camp, into those who DO, and those who Divvy. Will the strategy succeed? Not likely. There is already a value chain around Qualcomm that is highly unlikely from rallying behind a Nokia-led initiative. Furthermore, commercial success at the systems level in CDMA goes beyond a chipset in a handset. It involves mastering an entire system that Qualcomm and its infrastructure partners have fine-tuned over the years. Just as it has emerged as a formidable contender on the WCDMA chipset scene, Qualcomm's QCT will rise to the challenge of Ivvy Divvy, if need be. After ignoring it for some time, QCT recently announced a timeline for Release D chipsets (no messing with Release C�why bother?)

As specified by 3GPP2, the CDMA2000 Revision D standard is backward compatible with IS-95 and the CDMA2000 Release 0, A, B and C standards, providing today's CDMA2000 1X operators a seamless network evolution. CDMA2000 Revision D is a significant advance over CDMA2000 Revision C supporting not only similar peak forward rates to 1xEV-DO but also high data rate reverse links to better support high-resolution cameras, video streaming, video telephony and other multimedia services. CDMA2000 Revision D and 1xEV-DO Revision A, also being developed in 3GPP2, will allow the wireless operator two different complementary choices for deployment of high rate forward and reverse link packet services.

Note the language�CDMA2000 Revision D; not EV-DV. And of course, DV is just an alternative to DO. In any case, whatever the operators end up choosing, QCT will have the chips.

"CDMA2000 1X operators that are evaluating emerging wireless standards such as 1xEV-DV can continue to look to QUALCOMM as the CDMA industry leader for complete, easy-to-migrate, end-to-end system solutions," said Sanjay Jha, president of QUALCOMM CDMA Technologies. "We are uniquely positioned to drive the global expansion of CDMA2000 1X and to develop highly-integrated products that enable the easiest possible technology transitions while providing leading performance for both modems and applications processing."

The chipset in question is the MSM 6700, described further.


7. For the carrier with limited but underutilized spectrum, e.g. in rural areas, the Ivvy Divvy route may be a natural extension of the current 1X narrow band service. As such, these carriers will want to maximize compatibility with existing services, including handset functionality. Hence Qualcomm will again be the natural supplier of chipsets for those markets, as it has mastered all the features to which users have become accustomed in these markets, and will continue to lead the way for the CDMA carriers.

The parallels for Nokia with the WCDMA strategy are too evident to miss. Except that in this case it is dealing with operators who do have other choices. And Nokia knows that there is no getting around paying Qualcomm royalties (though it believes that with the patents it holds on EV-DV it can have a stronger bargaining position). Still, the EV-DV market, if it materializes, is likely to be rather small, hence affecting Nokia's ability to achieve much by way of economies of scale, as it has done so effectively with GSM. Will it be as "successful" at Ivvy Divvy as it has been at Dubya CDMA? As a long-term shareholder in the company, I shudder at the thought.


I personally continue to believe that any serious commitment that Nokia makes to CDMA will help grow the overall CDMA share of the wireless market. Yet such growth will come at the expense of its core constituency of GSM operators. Herein lies the fundamental conflict that will preclude Nokia from having the trust of the CDMA carriers. Feature for feature, Nokia's handsets for GSM carriers will continue to be ahead of what it makes available for CDMA, where it is likely to remain an entry-level supplier. On the other hand, Nokia serves several GSM carriers in many markets, so why not add CDMA operators to the client list?


Then again, if Nokia really believes all the great things its brochure says about Ivvy Divvy, it stands to reason that operators should prefer it to DubyaCDMA. Perhaps Qualcomm could help them bring the Ivvy Divvy vision and benefits to all these operators who have been misled down the UMTS path-- Anyone for GSM 1X-EV-DV?

BR


Become a Complete Fool
Join the best community on the web! Becoming a full member of the Fool Community is easy, takes just a minute, and is very inexpensive.