Wireless Propagator: A Significant IEEE Standards 'Variation'

In what can be mildly termed a significant 'variation' in the IEEE standards process,27 companies including chipset vendors Atheros, Broadcom, Intel and Marvellannounced Monday the official formation of the Enhanced Wireless Consortium(EWC) in an effort to speed up the IEEE 802.11n standards process. Thisannouncement comes only three months after these firms publicly committed todeveloping a joint proposal to the IEEE task group.

The IEEE 802.11n task group has been mandated to create a wireless standard witha throughput of at least 100 Mbps, which is at least three to four times the speedof the current standards-based 802.11a and g. The foundation of the standard isMIMO (multiple in, multiple out), technology created and developed by AirgoNetworks. Although the market leader in MIMO-based technologies--the company isalready in its third generation of hardware--Atheros and Ruckus Wireless (formerlyVideo54), which have developed similar technologies, have challenged Airgo'ssuccess.

Three significant proposals emerged from IEEE 802.11n meetings: MitMot(Mitsubishi-Motorola), TGn Sync (Task Group 'n' synchronization) and WWiSE (World-Wide Spectrum Efficiency). Influential industry players such as Atheros, Broadcom,Intel, Motorola, Nokia, Nortel and Texas Instruments were split among theseproposals. Some argued for the use of beamforming, different types of preamblesand more or less spectrum, among myriad highly technical details. TGn Sync gainedmajority support in a March 2005 meeting, but in May failed to gain the 75 percentneeded for the second confirmation vote. As part of the regular standardsprocess, the three proposal groups were then reinstated.

In July, the IEEE sanctioned a joint proposal group made up of the trio, whose goalwas to present a proposal at a November meeting in Vancouver. According topublished reports, it was later discovered that Broadcom and Intel, with Marvel andAtheros joining sometime later, had secretly met outside the IEEE process to drafttheir own proposal. This raised the ire of Airgo Networks, which desired full andopen participation in the standards process. Pundits and analysts commented thatthis diversion would cost significant months in the selection and ratification of an802.11n proposal.

The Enhanced Wireless Consortium (EWC) came out of the closet this week, with alarge group of promoter and adopter companies working together to "acceleratethe IEEE 802.11n development process and promote a technology specification forinteroperability of next-generation wireless local area networking (WLAN)products." By bringing in additional companies, some that are not even involved inthe IEEE 802.11n standards process, this group is likely to deflate attacks over itsmembers' earlier extra-curricular activities and gain public support for theproposal. The consortium specifically plans to submit a draft specification at theNovember IEEE meeting, but the core specification is already complete.

Noticeably absent from the EWC membership list is Airgo Networks. During an EWCinterview where member firms Conexant, Marvell, Atheros and Broadcom werepresent, it was emphasized that the proposal includes elements of TGn Sync andWWiSE and that there has been a lot of OEM support in both the consumer and theenterprise (Cisco, for example) space. Membership is open, but it was suggestedduring this interview that Airgo was developing its own non-interoperable andunknown specification. Ironically, the EWC has not made its proposal publiclyavailable at this time.

The obvious question is why? Why create a consortium outside the standardsprocess when a joint proposal group had already been achieved, with by all accountsa significant level of consensus between the WWiSE and TGn Sync proposals? Twoanswers are possible. First, the EWC claims that a consortium outside the standardcan generate more momentum behind a specification than can be garnered withinthe slower standards process. Although the dynamics between private and publicgroups can be debated, there is nothing to suggest that the joint-proposal group,which is made up of many of the same members as the EWC, cannot work just asquickly. And yet the EWC was quick to point out that, at other times, groupsoutside the standards process (such as the Gigabit Ethernet Alliance and ZigBee)have formed to accelerate standards development and remove logjams. But there isno standing precedent that secret coalitions made up of leading chipset vendorsform into public groups. Speculation abounds as to whether this violates anti-trustlaw.

The second, and more cynical answer to this extra-standard work is that thechipset vendors have aggressive time-to-market plans that would be delayed if thestandard's work followed the regular IEEE standards process. Atheros stated asmuch in its press release, saying the company has "plans to deliver chipsets basedon the new WLAN specification introduced today by the EWC."

In an interview I had with Greg Raleigh, president and CEO of Airgo Networks, he re-iterated his company's commitment to the open standards process and gave noindication that he would join the EWC. Raleigh is convinced that the EWC draftspecification does not include features that serve consumers, such as support forWi-Fi handsets, multimedia devices and real-time video streaming. He said thatconsumers will only buy products that comply with open standards and that anyproduct not in compliance will not be accepted by the marketplace. In what hecharacterized as a 'sideshow,' the EWC "has only served to delay and confuse theprocess," he said.

EWC members claim that they are not trying to subvert the standard; rather, theyhope to build on the momentum their group has achieved so they can quickly reachstandards process milestones. In a briefing I had with Atheros, Todd Antes, vicepresident of marketing, preferred to characterize the EWC as a catalyst forgetting things moving and helping to make the draft proposal more solid than itmight have developed otherwise.

Although how EWC membership could translate into IEEE 802.11n votes has notbeen tallied, the EWC is confident that with its existing 27 members--and more tojoin--it will present a proposal that can be confirmed with a 75 percent majorityvote. Even if that does happen, Airgo's Raleigh shared in my interview with him thatthe proposal will only describe minimum interoperability and that Airgo's technologywill still easily serve this "subset" of all possible consumer features. He believes thefinal standard will be something other than what the EWC has in mind, because hehas faith that the open standards process will ultimately do what's best.

Product based on any version of 802.11n is at least one year away, and it'sprobably farther out. Enterprises should calmly watch this standards fiasco fromthe sidelines, as vendors thrash things out on and off the court. Let's just hopethat vendors don't use this off-season "All-Star game" to change the rules of theregular league.

Frank Bulk is a contributing writer to Network Computing Magazine covering wireless and mobile technologies and works for a telecommunications company based in the Midwest.