It is almost two years since Intel’s chief technology officer, Pat Gelsinger, first coined the slogan ‘Radio Free Intel’ for his vision of a world where wireless connectivity is in every device down to the wristwatch and where these devices run on adaptive radios that move intelligently between available networks according to requirements. It was a big idea, but one that has been taken up all around the industry. It was an even bigger idea that Intel – a company with virtually no RF experience - should be the provider of those ubiquitous radios.

While the general concept may have become common wisdom, the jury is still out on whether Intel can implement the vision in silicon ahead of established experts like Motorola and Texas Instruments. In the past few months, the chip giant has suffered from a string of uncharacteristic delays that have led some to speculate that it has bitten off more than it can chew in its wireless strategy. And last week, Sean Maloney, head of the communications division and generally seen as Intel’s third in command, seemed keen to damp down expectations around the adaptive radio, which only a year ago was seen as the company’s most important project as well as Gelsinger’s personal crusade.

When Gelsinger first mooted his big idea of a silicon-based digital adaptive radio - small enough to be incorporated into a tiny cellphone and, uniquely, implemented in cost efficient CMOS - one faction in the company, apparently including CEO Craig Barrett, was concerned that he was promising something beyond Intel’s capabilities to deliver. Last June, when preliminary results of the project were presented at an international semiconductor symposium in Japan , it seemed that Intel was far down the road towards its goal and Gelsinger’s star was in the ascendant.

The flexible radio

Now Maloney is more commonly heard expounding wireless strategy than the CTO, and his more cautious approach appears to be gaining ground. In a briefing on wireless at Intel’s California headquarters last week, he said the concept of a reconfigurable radio ‘superchip’ was now considered unlikely. "I'm trying to not sound like a hippie, but I think that it is the journey, not the destination," he said and that the company would now seek to achieve similar goals, with the less ambitious project of creating a ‘flexible radio’.

Although the idea of the full software defined radio is not dead, according to Maloney, it may well have proved a far more difficult project than Intel had first hoped, and the company cannot run the risk of raising expectations too high, given its recent experiences with Centrino and other wireless delays.

The flexible radio is likely to cover a more limited range of spectrum than the adaptive radio, and Mike Chartier, Intel’s director of spectrum policy, said the company was working with the International Telecommunications Union to develop rules that would govern the flexible radio technology by next year.

Wi-Fi/3G chips

Intel may have pulled in its multi-network horns, but it certainly has not backed off completely. At the briefing, it promised its first Wi-Fi/3G combination chips for dual-mode handsets for mid-2005. Although Intel is behind TI and Motorola in offering this facility, it has made significant advances in terms of silicon design, notably with a 10GHz frequency synthesizer implemented in CMOS – a step towards a hybrid controller for both Wi-Fi and cellular networks. With a 10GHz synthesizer, Intel's processor will handle both the 5GHz bands required by 802.11a Wi-Fi and unlicensed WiMAX, as well as 2.4GHz for 802.11g and 3G protocols.

"We're taking the approach that 10GHz covers all the bands of interest,” said Krishnamurthy Soumyanath, director of the communications and circuits lab at Intel.

Also focused on flexibility is the media access control layer for the architecture, which will hook up to a 3G physical layer and to radios tuned to various wireless frequencies; and the Adaptive Radio Architecture, the software for managing these hybrid devices. This combines two layers - one designed to manage the radios themselves, and an applications layer to facilitate functions like handing off a voice-over-IP call to a cellular network. The Adaptive Radio Architecture is being built in to Intel's Personal Communicator, a proof of concept device that supports multiple wireless protocols.

Intel said, at the same briefing, that Bluetooth would be incorporated into Centrino this fall, a move flagged when the company acquired Bluetooth/Wi-Fi coexistence specialist Mobilian last November.

Unsurprisingly, one protocol that is not in the Intel multi-radio plan is 802.20, the IEEE Mobile-Fi standard, which increasingly overlaps with the better supported 802.16e initiative. Maloney said: “The WiMax community is full of people that compete with each other but it's a functional family. It seems to me that the arguments around 802.20 are much more complicated.”

However, he took the chance to stress, once more, Intel’s commitment to WiMAX. "It's early and [there's a lot of] skepticism around it. We're trying to be cautious but we're six months more confident than we were six months ago,” he commented.

© Copyright 2004 Wireless Watch

Wireless Watch is published by Rethink Research, a London-based IT publishing and consulting firm. This weekly newsletter delivers in-depth analysis and market research of mobile and wireless for business. Subscription details are here.

Related stories

Intel preps chip to link 3G, Wi-Fi networks
Intel announces death of copper
Radio Intel - future or fantasy?