Nifty New Cellular Phone Systems Race to Capture Japan's Consumers

Yet the slower system, starting from behind, is way ahead

5 min read

In February 1999, Japan's cellphone giant NTT DoCoMo Inc. (Tokyo) introduced the innovative i-Mode cellular phone system, with ingeniously designed handsets and feature menus that permitted customers to do things like download ring tones and exchange text messages much more easily than with earlier mobile phones. It was the i-Mode's meteoric takeoff, more than any other factor, that convinced communications companies around the world that ”third-generation” (3G) cellular technology--with features like i-Mode's, but even better and faster--was the wave of the future. European cellphone companies paid upwards of US $100 billion for 3G licenses, and the International Telecommunication Union (Geneva, Switzerland) allocated new spectrum generously to accommodate what was often called at the time the ”wireless Internet.”

Now all eyes once again are on Japan, where two 3G cellular systems have been introduced, one by DoCoMo, the other by its long-standing competitor, KDDI Corp. (Tokyo). The two systems are in a race for world domination, and the first lap is taking place in Japan. One system, wideband code-division multiple access (WCDMA), is preferred by the big European players; the other, CDMA2000, is backed by many of the U.S. heavyweights. Both are based on technology commercialized by Qualcomm Inc. (San Diego, Calif.) that allows signals to occupy the same bandwidth at the same time without becoming confused.

Commercialization of 3G systems in Europe has been much slower to start than expected, though the British affiliate of Hutchison Wampoa Ltd. (Hong Kong) has introduced limited service in the UK, Italy, Sweden, and Austria based on WCDMA. Even in Japan, the race was slow to get started, and it is far from over. Nevertheless, it has now entered an unexpectedly exciting phase, and that's because the late-starting horse on the outside track, KDDI, looks to be winning so far. Its CDMA2000 subscribership is rapidly approaching 10 million, whereas DoCoMo's WCDMA system has yet to win one million users .

While it's too soon to draw any hard and fast conclusions about what the status of the race implies for the two contending cellular technologies globally, this much can be safely said: so far the 3G contest is not playing out in Japan according to script.

KDDI's commanding start

Given DoCoMo's success with i-Mode, many assumed that when it introduced its 3G system, called FOMA, in October 2001, it would meet with a warm reception. (FOMA stands for Freedom of Mobile Multimedia Access--perhaps not the catchiest name.) But DoCoMo had had to pay a pretty penny to build a new cellular infrastructure for WCDMA, its previous (2G) system--the foundation of i-Mode--having been based on U.S. time-division multiple-access (TDMA) technology, in which each call signal is assigned a separate time slot.

Why did DoCoMo opt for the more expensive WCDMA, which was designed to be the natural upgrade from Europe's hugely successful 2G Global System for Mobile Communications (GSM), considering DoCoMo didn't use GSM? Evidently, like AT&T in the United States, it calculated that GSM and its proposed 3G successor were evolving into de facto world standards, given that GSM was universally used in Europe and very widely used in Asia. Accordingly, DoCoMo decided that it was better to join ’em than to fight ’em. WCDMA's higher data rate, compared with that of the CDMA2000 family, was an important consideration as well.

NEC and Fujitsu built the switching systems and radio network controllers for FOMA, and, together with Matsushita, supplied the base stations for it. Unfortunately for DoCoMo, however, the initial FOMA handsets were expensive and unexciting, and geographic coverage was incomplete. Instead of the 350 hours of standby time consumers were used to, they were being offered expensive initial sets with a miserly 55 hours of standby time, or even less, which wouldn't work outside Tokyo and a few other major cities. The upshot was that subscribers signed up by the thousands, not the hundreds of thousands or millions, as they had with i-Mode. After a year, fewer than 150 000 had bought the new service.

Meanwhile, ironically, it was KDDI that replicated the i-Mode experience. Using application software provided by Openwave Systems Inc. (Redwood City, Calif.), KDDI introduced its CDMA2000 1x in April 2002, offering such nifty features as improved photo and movie messaging, position locating, and games, all aimed at well-defined markets--like the teenage girls who had flocked to i-Mode. ”KDDI put all the essential components in place--a fast network, advanced handsets, and cool applications--to drive a dramatic increase in consumer demands,” says Richard Wong, senior vice president of Openwave.

Reasons for KDDI's success

Having already introduced a precursor CDMA system in 1998, cdmaOne--built in its entirety by Motorola Inc. (Schaumburg, Ill.)--and having already written off much of the costs, KDDI was able to offer its 3G system on attractive terms. Because the new 3G system was backward-compatible with cdmaOne, geographic coverage in Japan was very nearly complete. And though only half as fast as FOMA in terms of downstream data transmission (crucial for downloading things like film clips from the Internet), the CDMA2000 system was still fast enough to carry features with enough appeal to lure its cdmaOne subscribers to upgrade to the new system.

It may be that with 3G there's no one killer application, as Kirk Boodry, an analyst with Dresdner Kleinwort Wasserstein Ltd. (Tokyo) suggested to IEEE Spectrum. What seems to count to a great extent, he says, is ”providing the same services you got in the past,” but in even more satisfying ways.

Lately the gap between KDDI and DoCoMo has been narrowing somewhat, with DoCoMo now picking up new subscribers at an accelerated rate. DoCoMo has continued to build its WCDMA network, so that the FOMA signal is now available to 96 percent of the population. At the same time, and at least as important, it made known that it would share up to half the R&D costs for new handsets, which average about $80 million.

To date, DoCoMo has divided on a 50-50 basis the costs of developing six new handsets--which came to roughly $480 million, a company spokesman told Spectrum--with Fujitsu, Matsushita, and NEC. NEC and Matsushita had supplied the original handsets, and Mitsubishi, Sharp, and Toshiba are now developing phones as well.

With FOMA's high speed, DoCoMo is marketing it as the service that can deliver the more sophisticated applications--complex games, richer ring tones, and screen savers, not to mention video communications and e-mail with high-resolution images embedded--as well as much better voice than its 2G system, in itself a significant attraction.

The company is banking on the eagerness of Japanese subscribers in general, and younger users in particular, to frequently upgrade handsets to ones with newer features. Sooner or later, it hopes, FOMA will prove itself to be the next must-have upgrade for its current 45 million non-FOMA subscribers, of whom 39 million use i-Mode. But, of course, KDDI is upgrading its systems, too.

The pace accelerates

This fall, KDDI plans to introduce an upgraded CDMA2000 1x system called Evolution-Data Optimized (CDMA2000 1x EV-DO), initially in the Tokyo, Osaka, and Nagoya regions. The technology will be capable of a theoretical peak transmission speed of 2.4 Mb/s, though average download rates are expected to be around 600 kb/s.

DoCoMo's upgrade plans are further out, as Seizo Onoe, managing director of DoCoMo's Radio Network Development Department, told Spectrum. In the first half of 2005, it plans to start deploying 3.5G high-speed downlink packet access (HSDPA) technology, though Onoe says the company has yet to decide when it will actually introduce a 3.5G service based on it. HSDPA will be able to boost packet transmission speeds as high as 14 Mb/s in principle. ”But 14 megabits per second is the peak speed,” Onoe cautions. ”Depending on usage conditions and according to simulation, the average speed is likely to be [about] 3 to 4 megabits per second.”

By the time such services are available in Japan, the rest of the world may be catching up to where Japan is now. Perhaps then we will know whether GSM and WCDMA will dominate, with CDMA2000 merely a niche player, or whether it will be a more even split.

This article is for IEEE members only. Join IEEE to access our full archive.

Join the world’s largest professional organization devoted to engineering and applied sciences and get access to all of Spectrum’s articles, podcasts, and special reports. Learn more →

If you're already an IEEE member, please sign in to continue reading.

Membership includes:

  • Get unlimited access to IEEE Spectrum content
  • Follow your favorite topics to create a personalized feed of IEEE Spectrum content
  • Save Spectrum articles to read later
  • Network with other technology professionals
  • Establish a professional profile
  • Create a group to share and collaborate on projects
  • Discover IEEE events and activities
  • Join and participate in discussions