Slashdot Mirror
Ultrawideband May Stall Before It Starts
judgecorp writes "The IEEE's group for faster Wi-Fi, 802.11n has reached the end-point, with the Intel-backed TGn Sync proposal taking the lead. This is a contrast to the ultrawideband world 802.15.3a, where the competing proposals are slugging it out. Indeed, the vendors could be in for more trouble than they expect getting UWB past regulators in Europe." From the article: "Within the next two years, we should start to see fast wireless links based on ultrawideband (UWB), taking the place of short-range connections such as USB and Firewire, and providing fast data links between consumer goods. Chipmakers are now on the verge of creating the silicon, and vendor groups are completing the standards.But the technology may have trouble getting a world market, as regulators wrestle with the objections of the cellphone industry. UWB standards are in deadlock at the IEEE; but what the regulators say matters far more to the future of the technology."
Bluetooth is much slower, typically around 700kbps. Bluetooth consumes much less power, so don't expect Bluetooth to be pushed out on your headset or wireless mouse, for example.
-- Andyvan
"The problem with UWB is that it works great for one single device, but not so great once you have 100 million of the buggers running around.There's only so much bandwidth in the whole spectrum, so the "low noise due to wide-band modulation" argument would not hold once millions of these devices got made."
I don't think you really understand the concern here. UWB's main caveat is that it would raise the noise floor, making traditional wireless signals *possibly* harder to decode. UWB has extremely short range, so there would be very few devices within interference range with each other; also since UWB sends data using impulses, traditional TDMA technology (which is used on cell phones - you don't see cell phone carriers supporting only one cellphone per tower, do you?) can be used to have many signal streams in the same area.
"In the software world we're used to super-duper-ultra-wideband spaces: MD5 hashes are a good example."
This is totally irrelevant. MD5 has no bearing to UWB.
"It's the same problem as those RF-excited plasma light bulbs that were all the rage a while ago: the first 10,000 or so work great -- but by the time you deployed 10 of 'em to every household in America, nobody's radio would work any more."
The power spectral density of UWB is extremely low; crappy cd-players and consumer electronics devices can cause more interference than a properly-design UWB transmitter.