Your browser is not supported. Please update it.

16 January 2006

WiMAX chipmakers increasingly focus on 802.16e for innovation

One of the key dilemmas facing WiMAX vendors and operators is whether to support both flavors of the standard ‘ the current, fixed/portable 802.16-2004 and the recently ratified 802.16e, with its mobile potential ‘ even though these are based on different OFDM technologies. Many larger players are opting to go straight to 802.16e, even though it is a year or so behind its stablemate, because they see the fixed platform as one with a short life expectancy, which will be overtaken by ‘e’ in the most lucrative markets because of its ability to support fixed/mobile convergence, the root of many large operator business plans for 2006 onwards. 802.16-2004, they argue, will remain confined to traditional broadband wireless markets, solid but with limited growth, such as WISPs in 5GHz spectrum and rural access. Now one of the fiercest defenders of the future-proof nature of fixed WiMAX seems to be joining the race to ‘e’, as chipmaker Wavesat signs up with South Korea Telecom (SKT) to create silicon to support Wi-Bro, the Korean technology that is the basis of 802.16e.

Wavesat will work with SKT, which with KT will deploy Wi-Bro services this year, to codevelop technology for mobile devices based on the chipmaker’s forthcoming UMobile chips.

Vijay Dube, vice president of marketing and business development at Wavesat, explained the multitiered product strategy that is emerging. “Wavesat’s strategy is to provide a distinct solution for each WiMAX market segment: from backhaul to DSL extension with 802.16-2004 fixed OFDM solutions; to high speed connectivity for laptop basic mobility with OFDM Evolutive products; to full mobility with the upcoming 802.16e OFDMA UMobile product portfolio.’ Agreements like that with SKT are important to enable a start-up company to develop so many lines. Wavesat will not make chips for Wi-Bro base stations directly but will cooperate with other vendors to make sure they are compatible.

Although Wavesat is by no means leaving its fixed WiMAX work behind, it has jumped towards ‘e’ rather earlier than most had expected, a move that will reinforce views that the growth potential in the older market is limited. Chipmakers such as Wavesat and TeleCIS have argued powerfully that there is plenty of shelf life in 802.16-2004, with its OFDM256 physical layer, provided enhancements such as sub-channelization are supported. These enable it to work in a portable environment, and support most of the applications foreseen for 802.16e in its early years. 802.16e will not, in its first iteration, be highly mobile, and can be expected to thrive mainly in laptops and PDAs, so these companies, as they seek to protect their early investment in the original WiMAX system, are also making a valid point.

The flaw in the argument is that the roadmap to make 802.16e fully mobile (at vehicle speeds) is being accelerated by the interest of Samsung, Nokia and others, and pressure from major carriers, and so it could be highly differentiated from 802.16-2004 by the end of 2007. At this point, any service provider wanting to offer mobile services would need to make a forklift network change, or risk falling behind its rivals, in a timescale that will not have justified the initial investment in fixed WiMAX.

Wavesat, which has important alliances with commodity Taiwanese equipment makers, and TeleCIS have both been influential in speeding the way to get an 802.16-2004 laptop card on to the market, and this should appear in midyear. But given the provisos outlined above, this is likely to appeal mainly to localized providers or those in unlicensed spectrum, where mobility will not be supported ‘ it will be attractive as an enhanced option in classic Wi-Fi applications but will not appeal to the larger carriers. Hence, we will see the WiMAX pioneers increasingly forced to adopt Wi-Bro and 802.16e, more quickly than their own business and ROI plans might have dictated.

It was a busy week for WiMAX chipmakers. STMicroelectronics announced its first foray into the open 802.16 market, planning to bundle its STW51000 base station system-on-a-chip with a full set of inhouse software libraries, including 802.16e software for frequency division duplex and TDD modules. This approach of placing all the configuration options on the system in software reduces time and cost to market for systems vendors, says ST.

And UK-based picoChip continued its bid to be at the cutting edge of the emerging 802.16e market by adopting ArrayComm’s smart antenna Network MIMO technology. MIMO (Multiple In Multiple Out) is one of the performance enhancing antenna techniques approved by the WiMAX Forum in its 802.16e profiles, and ArrayComm is one of the most advanced exponents, with a major development deal with Intel ‘ also a customer of picoChip’s base station architecture for WiMAX.

The Network MIMO software will be incorporated into the physical layer of picoChip’s software configurable wireless platform and will be offered as an option for customers wanting to add MIMO to base station and CPE designs. Network MIMO implements all the antenna processing aspects of the 802.16e profiles including support for MIMO, adaptive antenna systems (AAS) and combined MIMO/AAS modes on both subscriber terminals and base stations. MIMO and AAS, used in combination, increase subscriber data rates, improve cell edge link budgets, manage interference, and maximize overall network capacity. ArrayComm is building a strong array of partnerships, including Texas Instruments and Intel.