Qualcomm has single-mindedly focused on its patented Code Division Multiple Access cellular technology, which is only now starting to bring useable wireless data links to mobile enterprise users.
So it may come as a surprise that Qualcomm has just as single-mindedly been on an acquisition binge for the last 18 months for technologies that often have little to do directly with the CDMA, Wideband CDMA (WCDMA) and cellular radio chipsets that have been its extremely profitable bread and butter for 15 years.
Qualcomm has been adding a range of wireless, content and media technologies designed to make the mobile phone both more personal and networked than the personal computer.
The acquisitions coincide with Paul Jacobs taking over in mid-2005 as Qualcomm CEO from his father, and Qualcomm founder, Irwin Jacobs. In a statement at the time, Jacobs said that new products and technologies will "enhance the mobile phone's role as the most personal electronic device in a world where wireless, computing, entertainment and consumer electronics are converging."
"Paul's looking ahead and saying 'what are the other technologies and applications that can make our original franchise even more valuable?'" says Howard Anderson, who co-founded Battery Ventures and is the William Porter Distinguished Lecturer at the Massachusetts Institute of Technology.
Shortly after Jacobs took over, Qualcomm forked over US$600 million for Flarion Technologies, with its load of patents around Orthogonal Frequency Division Multiplex Access technology, a foundation for both future cellular and WiMAX broadband wireless. Flarion developed Flash- Orthogonal Frequency Division Multiplex, a proprietary cellular broadband technology for all-IP, mobile connectivity.
Other acquisitions have included:
-- U.K.-based Elata, for its software to organize, control and deliver a wide range of content and applications over various types of cellular nets.
-- Qualphone, a maker of an application, based on IP Multimedia Subsystem, that can integrate voice, text, image and video on 3G handsets on WCDMA and CDMA2000 networks.
-- nPhase, a Chicago vendor of machine-to-machine telemetry software that will let enterprise wirelessly monitor over cell networks everything from truck fleets to assembly-line robots.
Qualcomm closed 2006 with two other purchases. One was Airgo Networks, creator of the first WLAN chipset based on multiple-input multiple-output technology, which is the heart of the emerging 802.11n wireless LAN (WLAN) standard. That standard will make possible WLAN connections with throughput of 100M to 300Mbps. The Airgo group is shipping what it says is the first chipset that complies with the second draft of the emerging 11n standard. The second purchase was the Bluetooth division of RF Micro Devices, focused on embedded Bluetooth radios for headsets and handsets.
But Qualcomm has continued its own aggressive R&D. Verizon in a few weeks will launch a nationwide mobile TV broadcasting net based on Qualcomm's MediaFLO technology, unveiled in 2004, and the nationwide UHF channel 55 spectrum that Qualcomm owns and operates to run the network. MediaFLO uses a far more spectrally efficient point-to-multipoint model instead of the point-to-point technique used in most of the early mobile TV services. Verizon's mobile subscribers will be able to tune into 24-hour broadcast TV content from participating media companies such as CBS, Comedy Central and FOX.
Toward a better mobile phone
These acquisitions and Qualcomm's own innovations are paving the way for a mobile phone that's a powerful computer with extensive voice and data capabilities, which can tie directly to enterprise data and applications via wireless broadband networks.
"In a converged world, the [3G] WAN needs other pieces," says Michael Concannon, vice president of strategic products with Qualcomm's CDMA Technologies division, which is responsible for 3G chipsets, systems software and development tools. "One key push is to beef up connectivity outside the cellular modem." That focus led directly to the Airgo and RFMD acquisitions, he says.
The second key push is to make the phone itself a powerful computer. Qualcomm's efforts include improved power efficiency, vastly expanded memory, new always-on displays and more intuitive user interfaces.
There also is Qualcomm's next-generation processor platform: its ARM-based SnapDragon battery-powered chipset, which handles a wide range of processing chores for the phone. It consists of the 1GHz Scorpion CPU coupled with the company's VeNum vector processor, and with a new digital signal processor running at 600MHz.
The goal, Concannon says, was to maximize processing power in million instructions per seconds (MIPS) while minimizing electrical power in milliwatts. Qualcomm rates Snapdragon at 2100 Dhrystone MIPS, using just 240 milliWatts. By comparison, Qualcomm says Intel's XScale PXA270 chipset delivers 780 DMIPS using 480 mWatts. The first samples of the SnapDragon chipsets are scheduled for the third quarter, and the first devices equipped with the chipsets are expected in 2008.
The company prides itself on a systems approach: designing the full complement of radio, processor, antenna and power management components, then working hand-in-glove with handset makers, base station suppliers and the carriers that deploy both, to ensure the chipsets work flawlessly. "When we deliver a chipset to LG [Electronics] and the others, it's been banged around and proven," says Behrooz Abdi, senior vice president of the CDMA Technologies division. "A lot of the tier 1 network problems [and issues] have already been fixed. That improves time to market for the manufacturers."
That's partly the reason why Abdi and Qualcomm in general professes to be unconcerned about possible rival technologies such as IEEE 802.16 fixed and mobile WiMAX broadband wireless.
"WiMAX is a good, big pipe for fixed [wireless] transmission, and on the mobility side, [those companies] are working hard to address those issues," Abdi says. "But a lot of the cost with any protocol is associated with the mobility piece of it: handing off from tower to tower for example, boosting receiver power, the backend networking and switching."
WiMAX will address those issues, he says. "But by the time they do, the cost is going to be pretty much comparable with 3G [cellular] and they'll still have to build the towers and all the rest," Abdi says. "The business model for WiMAX doesn't make sense to us at the moment."