King Solomon was inspired when he resolved an argument over who was the true mother of an infant by threatening to divide the child in two. Only the true mother would offer to relinquish her claim in order to save the child. Problem solved. Few would dispute the logic of individual sacrifice for the sake of the greater good. But fast-forward a few thousand years and you'll find a community to which the logic seemingly doesn't apply: the ultrawideband-focused IEEE 802.15.3a task group.
Still deadlocked after two years of bitter, fractious and often asinine debate, the group still can't decide whether multiband orthogonal frequency division multiplexing (MB-OFDM) or direct-sequence ultrawideband (DS-UWB) is the best path to 480-Mbit/second short-range wireless communication. The MB-OFDM proposal is backed by the Multiband-OFDM Alliance Special Interest Group (MBOA-SIG), led by Intel, Texas Instruments, Philips and Staccato, with a who's who of semiconductor and consumer electronics companies on the roster and an "ecosystem" centered on WiMedia. At the other end of the spectrum, the DS-UWB proposal is being championed by Freescale Semiconductor, which is a member of the UWB Forum, with its ragtag assembly of lesser-known companies and startups.
Now, marketing machines in tow, both groups are ready to take it to the street and let the market decide while still paying lip service to the IEEE process.
Problem is, they're going to be joined this week by Pulse-Link, which has decided to jump into the fray by setting up its own C-Wave Alliance to promote its own, completely different flavor of UWB (see story, page 14). So now we have three coalitions and still no standard.
Nonetheless, the two .15.3a camps will make their bimonthly pilgrimage to the task group meeting (the next one is March 13-18), argue the same points, accept their penance and return home, all with the full a priori knowledge that it's a complete waste of time and money. (I wonder if the shareholders are aware of this?)
Still, neither camp can afford not to go. There's simply too much at stake to let the other side win. Freescale has years of investment in its impulse-based technology and wants to protect what it perceives as a two-year lead in the market. For its part, the MBOA has what it believes is a more CMOS-friendly design for lower cost and power consumption, and of course it wants to obliterate any time-to-market advantage Freescale may (or may not) have by resetting the clock with an MB-OFDM-based standard. One must get 75 percent of the vote to win, and there's not a snowball's chance in Hades of that happening.
So now what? Maybe it's time to put away Robert's Rules and revisit Solomon's code: Consider the greater good, and either find a way to get the standard done or disband the task group and let UWB develop organically. Unfortunately, the only way to get the standard done now is to split the baby adopt a dual-physical-layer approach. That's historically the way such debates have been settled in the IEEE. Then the market will decide, and the .15.3a task group members will get to stay home and focus on engineering a novel thought.
As is typical of this debacle, one side thinks compromise is a good idea and the other does not. The MBOA believes there should be only one UWB approach, since more than that would create confusion and slow adoption. But as Pulse-Link will help make clear this week, there will not just be one UWB approach and if a standard isn't decided quickly, confusion will surely reign.
Of course, it could be argued that since this is a nascent technology anyway, maybe it should be let loose for a while so designers can experiment with various implementations before the formal standard is set. In that case, this editorial was a waste of time.
Patrick Mannion covers wireless technology and DSPs for EE Times and is editor of the Design Currents section.