LONDON — With personal navigation devices (PNDs) growing in popularity and location-based services moving onto cell phone platforms, knowing where you are has never been easier. And with developments in the market proceeding at a whirlwind pace (witness Nokia's $8.1 billion acquisition late last year of digital map provider Navtaq), it can only get simpler to find your way--or for a canny marketer to find you--via your cell phone.

But for designers of phones that will have GPS capability, the path isn't so clear.

Conventional wisdom would hold that the natural migration of global positioning technology into handsets at minimal cost means the integration of RF and digital functions into the same device, with a "no chip" GPS solution as the end game. The most notable proponent of that architectural approach is Qualcomm Inc. (San Diego), which for years has been integrating GPS processors into digital baseband chips for CDMA handsets.

But there is an alternate approach: providing standalone GPS silicon and intellectual property (IP) for integration, as has been done for Bluetooth and Wi-Fi functionality.

By far the biggest supplier of standalone GPS silicon and associated software is SiRF Technology (San Jose, Calif.), which has enjoyed significant market share for standalone PNDs and is seeing its chips increasingly designed into mobile phones. SiRF has been busy consolidating its lead through licensing and joint product development deals with other chip makers, including NXP Semiconductors, Freescale Semiconductor and Intel.

Given the emerging market's potential volumes, it's not surprising that several startups are snapping at SiRF's heels. They include Swiss group Nemerix; GloNav, a spinout of the GPS activities of Ceva Inc.; u-Blox (Thalwil, Switzerland); u-Nav Microelectronics (Irvine, Calif.); and Bluetooth chip pioneer CSR (Cambridge, England), which about a year ago bought NordNav and Cambridge Positioning Systems. Two of these companies are poised to change hands: NXP is acquiring GloNav for at least $85 million, and Atheros is picking up u-Nav for about $54 million.

Even less surprisingly, given Qualcomm's participation in the sector, Broadcom Inc. moved to extend its own presence late last year through the acquisition of Global Locate, a long-term supplier of GPS and assisted-GPS devices and associated software, for about $150 million.

So which approach is the likely winning formula for GPS on handsets?

"There are many trade-offs, both in terms of technology and business cases," said Kanwar Chadha, SiRF's founder and vice president of marketing. "But a standalone approach clearly gives designers the most flexibility and makes the most sense when it comes to attach rates, which vary significantly in different countries and with different cellular technologies."

Attach rates for GPS are notoriously difficult to ascertain, but Chadha believes they could approach 100 percent in the United States for CDMA handsets and that they are also high in Japan and South Korea. By contrast, they are "low" in Europe and are even worse in China and India, he said.