Unless, of course, China's Ministry of Industry and Information Technology (MIIT) continues to ram the standard down the throat of a reluctant China Mobile, while at the same time further delaying the issuance of licenses for the third-generation cellular technologies the rest of the world is using: wideband CDMA and cdma2000. Such a delay would give TD-SCDMA the breathing room it needs to overcome certain technical hurdles and extend its coverage.

It's not that the standard itself is inherently flawed. On the contrary, time-division multiplexing offers some advantages over the frequency-division duplex (FDD) wideband-CDMA mode. (W-CDMA supports both TDD and FDD, but the latter is being emphasized in current deployments.) With TDD, it's easier to assign time slots dynamically to adjust for asymmetric Internet traffic. TDD is also more spectrally efficient, and it simplifies frequency planning. Further, the use of a single band obviates the need for FDD's expensive and bulky filters, and allows more-accurate channel estimation techniques.

Like all technological innovations, however, TD-SCDMA requires testing and improvement. That's where the problem arises. Mobile wireless tests take time and money. The latter seems to know no bounds, and the former is running out as China falls further behind the rest of the world in its 3G rollout.

Early TD-SCDMA trials revealed coverage and capacity problems, and the data rates--2.8 Mbits/second in the downlink and 384 kbits/s for the uplink--need to be improved by going to a multicarrier format. (Putting multiple carriers together yields higher data rates.) But that creates its own set of problems, which have yet to be resolved.

So, why did China pick TD-SCDMA in the first place? Therein lies the irony.

In 2006, Zhou Huan, chairman and president of Datang Technology, a major backer of TD-SCDMA, pointed to the slow progress and birthing pains of W-CDMA--performance issues and the late arrival of handsets--as a chance for China to get ahead of the rest of the world with a more spectrally efficient standard that, theoretically, would cost less to deploy. It wouldn't be the first time a major nation went down its own road. In an earlier wireless generation, the United States chose CDMA over the more widely used GSM, and fell woefully behind Europe and the rest of the world in cellular proliferation and usage. Likewise, Japan opted for PDC over GSM in 1991, and its citizens remained isolated with poor-quality voice and low-quality handsets.

But rather than an opportunity, W-CDMA's problems should have been a red flag for the Chinese. Wireless, especially 3G, is just plain hard to do. Ask any of the operators or equipment providers that have been steeped in W-CDMA for the past 10 years.

Even though W-CDMA and cdma2000 are now proven and deployable technologies that are rapidly moving down the cost curve, China Mobile, under pressure from the MIIT, continues commercial trials of TD-SCDMA. "The Chinese government will ensure China Mobile sticks to TD-SCDMA no matter what happens; it has to make TD-SCDMA work," said Hwai Lin Khor, wireless analyst for ABI Research (Singapore).

Indeed, "The MIIT needs to have some achievement to show the country. If they give up, they lose face," said Kevin Wang, senior manager of China research at iSuppli Corp.

So motivated, China Mobile announced in May that it had invested up to $2.04 billion to install TD-SCDMA networks in eight Chinese cities and that coverage had reached 95 percent that of 2G networks (see story, page 46). These, however, are still commercial trials. Forecasters see TD-SCDMA in 30 cities by year's end, translating to 500,000 users by the end of 2008 and 3 million by the end of 2009. The MIIT originally wanted 10 million users by the close of next year.