Poor Man's Cellular Spins Wireless Gold

Manhasset, N.Y. — What's old is new again. After years of wandering in the backwaters of wireless communications, the Personal Handyphone System has resurfaced in developing markets from China to South America as a vibrant, low-cost wireless option for the hundreds of millions who cannot afford 3G or even GSM.

At one time handicapped by poor mobility, limited range and data capabilities, and a lack of roaming agreements, PHS has systematically overcome those limitations. At the same time, proponents have leveraged a lack of spectrum-licensing overhead and regulations — as well as steadily improving semiconductor integration and radio performance — to drive down handset, basestation and deployment costs. The upshot is affordable service for those on tight budgets.

"There's not much else in the world you can run for $4 or $5 per month," said Howie Frisch, director of wireless-product management for UTStarcom Inc., which holds 60 percent of the global market for equipment used by PHS service carriers (followed by ZTE Corp., with 30 percent). Frisch was referring to the average return per user per month (ARPU), the almighty bottom line for wireless carriers. "If you applied the ARPU model of $30 to $50 [for other cellular services] in developing countries such as China and South America, you're underwater," he said.

Though many silicon suppliers have sidestepped PHS for greener-looking third-generation cellular fields, savvy startups and others without a good entry point in cellular are starting to see its lure. And some have moved to capitalize on those potential volumes.

Developed by NTT and introduced in Japan in 1995, PHS moved to China in 1998 under UTStarcom's leadership as the Personal Access System (PAS). That turned PHS from a distinct, separate, widely deployed network into a wireline extension — essentially a glorified cordless phone. In this scheme, a picostation can be placed atop any copper line — anywhere — to encompass a small radius of typically 4 km per station. The picocell structure extends battery life beyond that of wider-area cellular networks.

"That's why the first cell phone wristwatch was PHS-based," said Alan Varghese, principal analyst for semiconductors at ABI Research Inc. (Oyster Bay, N.Y.). PHS' caller-party-pays billing has also contributed to its success, to the point where Frisch estimates a potential global subscriber base for PHS/PAS in the 2 billion to 3 billion range.

Far-flung rollouts
Aside from deployments in China, Japan and Taiwan, Frisch points to far-flung trials and deployments in Vietnam, Thailand and Bangladesh in Asia; Nigeria, Mali and Tanzania in Africa; and Honduras and Chile in Central and South America. "There are 40,000 subscribers in Haiti, where the cells are powered over the copper line directly from the central office," he said.

"There are 500,000 PHS subscribers in greater Bangkok," said David Gamba, senior marketing manager at Xilinx Inc., which worked with Willcom Inc. (the former DDI-Pocket) to expand its PHS basestation capacity in Japan using FPGAs. "Imagine the numbers if it took off in India, which is an ideal market for it."

The combination of low cost and potentially massive volumes caught the eye of Atheros Communications Inc. (Sunnyvale, Calif.), which has led in the wireless-LAN arena via low-cost RF CMOS integration. "At one-fifth the cost per call of GSM, it's the wireless technology for the 'other billions' who don't have a lot to spend," said Craig Barratt, chief executive officer at Atheros. "And it's perfect for a whole class of [wireline] carriers without a clear path to wireless, who would otherwise be limited by regulations and cost."

Barratt also saw that PHS was a market grossly underserved by mainstream silicon providers. Analog Devices Inc., for example, scuttled its foray into PHS. "We have our hands full fighting off competition in GSM, GPRS, Edge and W-CDMA," said Doug Grant, director of business development at ADI. He sees PHS as "a good way for a startup to get points on the board."

Spotting an opportunity, Atheros in February 2004 set to work on leveraging its radio-on-chip RF integration expertise to develop the AR1900, which it will formally announce this week (see story, page 76). The company believes the device is the market's first fully integrated, single-chip PHS phone. "By making a cheap technology 30 percent cheaper through higher integration, we've widened the cost gap between PHS and [mainstream] cellular handsets," said Barratt. He claimed the AR1900 can reduce the component count for a PHS phone from 400 to 130.

Although caught off guard, analysts quickly warmed to Atheros' move. "I almost fell out of my seat," said Will Strauss, president of Forward Concepts (Tempe, Ariz.). And, yet, the numbers, he said, make it "a no-brainer." According to an upcoming Forward Concepts report, there were 67 million subscribers in China at the end of 2004 — 90 percent more than the year before. Barratt cited EMC World Cellular Database figures that put the number of subscribers in Japan, Taiwan and Thailand at 6 million. In Japan, PHS is as ubiquitous as a 128-kbit/second data connection for laptops.

BDA China Ltd., a Beijing-based telecom consultancy, said PHS saw its biggest growth year in 2004, with about 27 million users signing up. Although the rate is expected to taper off this year, that hardly means the shine is off PHS. In China, slower growth can still mean big numbers. Forward Concepts expects PHS to add about 50 million users in China in 2005, while BDA expects about 20 million. "And even as growth slows, the replacement market is growing, so it is still a very viable market in China," said Ted Dean, managing director at BDA.

In addition, Barratt pointed to recent initiatives by the Ministry for Information Industry, China's IT regulator, that allow selective intercity roaming — essential for business users — and mandate Short Message Service interoperability between PHS and GSM/CDMA. Intercity roaming helps close the gap in usability between PHS and GSM/CDMA, Barratt said, while SMS interoperability represents a vote of confidence in PHS, which has long been in somewhat of a gray area with regard to regulatory compliance within China.

China Telecom and China Netcom, state-owned wireline service providers, lack licenses for wireless, so they loosely describe PHS as a fixed-line, limited wireless solution. But there is no guarantee they will stick with PHS once they get wireless licenses. That will happen later this year, with the advent of 3G.

Some observers have speculated that the two China telcos will upgrade the service to a niche "3G-lite." But there is a very real risk they will starve the service after they get wireless licenses, or if the GSM and CDMA units of wireless provider China Unicom are divided between Telecom and Netcom in a much-talked-about government restructuring.

That move would give the fixed-line providers the GSM-through-3G access they want and little incentive to continue investment in PHS. In addition, a recent BDA study indicated that if GSM or CDMA carriers offered CPP, 62 percent of PHS subscribers would defect.

Despite the risk, Atheros' Barratt expects PHS penetration will grow to at least 80 million users by 2008.

"We think PHS will be very competitive; it has no disadvantage," said Tim Richardson, CEO at Micro Linear Corp. (San Jose, Calif.), which started sampling PHS transceivers last October. "It's all about low cost, low power and time-to-market."

Mike Feibus, principal analyst at TechKnowledge Strategies Inc., said that Atheros is "definitely bringing something new to the party at a time when others are investing elsewhere," he said. "They have fewer barriers to entry and so will be more successful here than with GSM." — Additional reporting by Mike Clendenin.