Showcasing wireless strategies and innovations has been the centrepiece of the biannual Intel Developer Forum (IDF) for a few years, and this time the focus has been on an accelerated WiMAX roadmap and the emerging WiMedia platform. Both of these, of course, are immature wireless technologies that Intel has largely driven and which are critical to its objectives in 2006-2010 - to provide the market with platforms that will ensure ubiquitous wireless access and new applications, and so spark increased acquisition of Intel-powered devices and servers. At this early stage, success is far from assured. Both technologies have powerful challengers and unproven real world markets, and both are dependent for truly global support on regulator spectrum decisions.The IDF showed Intel seeking to increase the chances for both platforms.

In WiMAX, Intel's main pressure is to maintain industry confidence in the roadmap towards mobile 802.16e - the version of the WiMAX standard that has most interest for large scale vendors and operators, but which is in a race against time to achieve market presence before its functionality is leapfrogged by the next generation 3G-based platforms, HSxPA and LTE.

WiMAX laptop roadmap speeds up:

Intel said it had brought forward its roll-out schedule for WiMAX client cards for incorporation in laptops and other devices, with a PCMCIA and mini-PCMCIA product now scheduled for late 2006 rather than 2007. "Our aim is for a global standard for WiMAX," Sean Maloney, head of Intel's mobile products group, told IDF. "We're getting close now, though, and expect data in the 2.3GHz, 2.5GHz, 3GHz and 5GHz bands to cover the entire world." Maloney gave the first public demonstration of an integrated Wi-Fi/WiMAX chip that allows for a smooth transformation between the wireless technologies, critical to operator models and consumer uptake. He showed a 2Mbps link to a PC attached to a scooter, receiving live traffic updates, directions and streaming video over a range covering most of San Francisco, the conference's venue.

Maloney also unveiled Intel's Wi-Fi/WiMAX integrated radio, which will give notebooks full access to both networks. This will be important to encourage deployment of WiMAX base stations even before 802.16 clients are common; to enable WiMAX to leverage the success of Wi-Fi rather than threatening it; and to support the creation of Wi-Fi/WiMAX metrozones - a deployment type that is becoming increasingly important in boosting consumer uptake of wireless devices, by offering low cost, easily available access.

Beceem:

Intel has also invested an undisclosed but "substantial" sum in WiMAX chipset maker Beceem, which has been a prime mover in the WiMAX Forum and already boasts capital from Samsung, the main force behind the main precursor to 802.16e, Wi-Bro.

Shahin Hedayat, CEO and co-founder of Beceem, said: "This collaboration is aimed to accelerate the trial and deployment of mobile WiMAX networks and establish Mobile WiMAX as the highest performance, most spectrally efficient wireless broadband technology available."  The company shipped its first WiMAX digital baseband and integrated radio chipset in January. The MS120 chipset is built to the 802.16e profile, to which Beceem contributed, though there will be no certification of this until mid-2006. It is targeted at handsets and PDAs and Beceem is also shipping a reference design kit for the WiMAX modem, including host driver software.

The Beceem alliance shows how Intel needs to build up its ecosystem rapidly, in order to keep mobile WiMAX on track and to add to its functionality at a greater rate than the 3G platforms can. In Wi-Fi, Intel was keen to move quickly to a state where it made all elements of its chipset, including the radio, an ambition that caused it several hiccups in Centrino's progress, such as being slow to market with 802.11g support. In WiMAX, it is being far more pragmatic, assembling technology from many quarters to get a solution to market quickly and with sufficient features to impress sceptical OEMs and carriers. It already has important chip level alliances with PicoChip and Arraycomm, and development partnerships with Motorola, Sprint Nextel, Clearwire/NextNet and others.

Wi-Bro in Brazil:

There is also an element of competition with Samsung, which stands to be the main beneficiary from Wi-Bro's early leap into the personal broadband world. The fact that Wi-Bro, though far more limited than full blown 802.16e, will be the basis of mobile WiMAX, gives Samsung and, to a lesser extent, LG a headstart in the market and the chance to build a global business. Any notion that Wi-Bro would remain within its original remit of being a specifically Korean network has been destroyed by a series of trials, most recently last week's announcement of a major pilot in Brazil, and most notably the Sprint Nextel testing programme with Samsung.

Samsung is working with Brazil's largest media group, Abril, to pilot Wi-Bro, initially with Abril's cable TV subsidiary TVA. The tests will start in May with commercial services scheduled for November. Samsung has a similar deal with Venezuela's Omnivision, which will go live in the third quarter. Wi-Bro is also in trials in Korea, the US, Japan, Italy and Croatia.

While Wi-Bro is good for bringing at least a semi-WiMAX solution to reality, it has also raised fears among US suppliers that the WiMAX agenda, once so firmly controlled by Intel and other American players, will slip to the Asian challengers.

Centrino futures:

Intel also showed off a new iteration of Centrino, codenamed Santa Rosa, with improved security and graphics and an 802.11n adapter for the forthcoming 100Mbps-plus evolution of Wi-Fi. The platform will arrive, together with a new graphics chipset called Crestline, in the second half of 2007. Santa Rose is the follow-on to the current platform, Napa, and will feature a mobile microprocessor, codenamed Merom.

The platform will also include Intel's NAND flash-based platform accelerator, codenamed Robson, which enables more rapid boot-up time and power savings.

WiMedia:

Meanwhile, WiMedia - the UltraWideBand-based personal area networking system backed by Intel, Texas Instruments and a large sector of the consumer electronics industry for digital home and wireless USB applications - has now been standardized, but by the European ECMA body, not the IEEE, which abandoned its own UWB efforts following a protracted stalemate between WiMedia and the Freescale-led UWB Forum. Intel sought to spark developer and vendor interest with demonstrations of high definition video over WiMedia, but aware that the challenge from Freescale has not gone away. While the USB Alliance, which governs the main technology for PC-peripheral interconnect, backs WiMedia for its wireless incarnation, Freescale partners will come to market earlier with their alternative, CableFree USB. The two networks will not speak to each other, and this will be the first test in the real market, as opposed to the standards committees, of whether Freescale's Direct Sequence USB will find a significant role despite the stronger bluechip backing for WiMedia. (Both systems have about 200 supporters, but the only real household names in the UWB Forum are Samsung and Lucent, and Samsung also supports WiMedia.)

The wireless USB debate also shows up the differences in objective between Intel and Freescale. Intel's primary objective is to drive PC upgrades as it did by incorporating Wi-Fi into the laptop and as it hopes to do again with WiMAX, reducing the average time between notebook refreshes. Therefore, WiMedia will require new drivers and hardware, and so will only work with the newest PCs, encouraging new purchase. By contrast, CableFree USB works with existing USB hardware and drivers, since Freescale's aim is to generate a mass market for itself in the actual UWB chips, which Intel has dismissed as a low margin sideline compared to boosting processor sales (it is possible it will not manufacture WiMedia chips itself at all). The WiMedia platform will really fly when the driver is incorporated into Windows Vista, itself a major driver for PC replacement, and while CableFree USB will be a useful short term option, the Microsoft support will ultimately prove decisive, at least in any sectors where Windows is important, which clearly includes PC interconnect.

This article originally ran in Wireless Watch, a publication of Rethink Research. Reproduced with permission. For information on the weekly Wireless Watch Newsletter and other Rethink Research products and services, click here.