A Faster Path to 802.11n?
A new vendor consortium hopes to ram through its own high-speed Wi-Fi standard, maybe in less than a year... and perhaps at the cost of the open standards process.
The stories of infighting in IEEE (define) Task Groups are becoming legendary. The most familiar of late is in the 802.11 Working Group's Task Group N (TGn). Members, entrusted with creating a specification that will move 802.11-based wireless networking up to speeds of 100 Megabits per second (Mbps) or faster, split into two camps with neither able to get the 75 percent super-majority needed to become the draft standard.
However, a new group has entered the fray, with plans to push through a hybrid proposal sooner rather than later. Called the Enhanced Wireless Consortium (EWC), it is made up of not just the four originally rumored Wi-Fi chip companies (Intel, Atheros, Broadcom and Marvell) but 27 vendors in total, including other chip makers (Conexant), equipment vendors (Cisco, Linksys, D-Link, Buffalo, Netgear, 3Com, Symbol, US Robotics), computer manufacturers (Apple, Lenovo, Sony, Toshiba) and others.
The EWC proposal would offer Wi-Fi speeds as high as 600 Mbps.
Notably absent from the EWC: Airgo Networks, the chip maker currently selling products through companies like Linksys and Netgear, using the MIMO (multiple in, multiple out) technology that will be the backbone technology of 802.11n.
Many see the formation of the EWC as a shot across the Airgo bow, though EWC denies it. Bill Bunch, director of product line management for wireless at Broadcom, says on behalf of the EWC that "one of the founding principals is that all companies are welcome to join... this alignment is companies that want to get things done."
And that's the mantra of the EWC: acceleration of the IEEE 802.11n development process. The faster the IEEE Task Group N pushes through a specification, the better for all the companies, they claim. However, Atheros vice president of marketing Todd Antes says, without naming Airgo specifically, that "there's some who benefit from the slowing down of the standard."
"Holding the Process Hostage"
Greg Raleigh, CEO of Airgo Networks, doesn't see it that way.
Instead, Raleigh says, "We are disappointed at the approach being taken to try and create an interoperable standard that goes around the open process."
A primary claim of the EWC is that, because the Joint Proposal (JP) group formed from WWiSE and TGn Sync doesn't have a joint proposal yet, the EWC came up with one fast to, again, move things along faster.
Raleigh believes that the EWC purposefully scuttled the JP ship when it started rumors in back door meetings. "We'd have had a preliminary draft by now" in the JP, he says. "The silicon competitor companies made it clear in subsequent statements that they'd push this specification regardless of what happens in 802.11n.... That doesn't sound like open standards with transparent compromise to us."
Raleigh says that the customer companies listed in the EWC group such as Linksys, Netgear and D-Link, those who buy the chips they need for retail products are only allied with the group inasmuch as it leads to an actual 802.11n standard in the end.
Despite the potential strength the EWC might have in pushing through its agenda, Raleigh refused to even hypothesize on a future where the group would be successful.
"We are focused on helping create an open 802.11n standard," he says. "The JP group will form the right compromise, then there will be more changes. There's never been any such thing as a Draft 1 that didn't have final changes.... It would be an outrageous prediction to say EWC will be the final 11n."
"That's how they want the process to work," Raleigh says.
Airgo is still seen by many as the leader on the WWiSE side, and Raleigh notes that there are several member companies in the two groups that aren't represented in the EWC, including big names like Texas Instruments and Philips.
Can the EWC Win?
From a technology viewpoint, EWC's proposal is nothing new, says Bob Wheeler, senior analyst at The Linley Group. "It's a subset of what would have been in either one of the two proposals" from WWiSE and TGn Sync, he says.
What's not obvious yet on the surface is intellectual property (IP) issues. "Who has the patents on some of these technologies?" Wheeler asks. "There's likely to be some Airgo intellectual property involved in any MIMO implementations."
EWC, even though it came up with this proposal outside of IEEE meetings, is not working completely outside the process. It will meet with the joint WWiSE/TGn Sync group this month for discussion. The goal is for JP concession before the next IEEE 802 Working Group meeting takes place in November.
The meetings probably won't be bloodless, as Airgo's Raleigh intimated. Wheeler concurs, saying, "[EWC's] goal is to force this down the throats of the combined group and hope they can get the 75 percent in November."
The IEEE 802.11 Task Group N has to vote in November, and if they don't get a super-majority, the whole selection process would start over again at the next meeting in January, causing further delay that the EWC hopes to avoid.
The EWC's planned timetable, if successful, would certainly speed things up for everyone, according to ABI Research. While the firm said a month ago that 802.11n would be delayed until mid-2007 at best because the JP group did not have a draft ready, senior analyst Philip Solis said in a statement today that success for the EWC would mean "companies can start building EWC-compliant Wi-Fi chipsets and products immediately, with EWC-compliant products probably reaching the market by the fourth quarter of 2006."
Wheeler agrees, but says even with a draft standard there will be issues with interoperability unless the EWC or someone else such as the Wi-Fi Alliance steps in to perform tests even before 11n is final.
The Alliance's stated stance is that it will wait for 11n to be ratified before testing; it has a policy to strip away Wi-Fi certification from so-called "pre-N" products that cause interference with current 802.11a/b/g products on the market.
That might, of course, be a moot point if the EWC can't get past the majority needed. Nothing is a given in the IEEE voting process, after all.
"You come down to not companies, but members," Wheeler says of the voting process. Anything can happen, despite the best-laid plans of EWC, Airgo or any other group.
Wheeler doesn't believe the EWC wants to torpedo the whole process. "They think they can get 75 percent, or they wouldn't have gone this route," he says.