January 28, 2002

9:50 AM

Ultra Wide Band (UWB) communications: The end of communication technology as we know it, and the dawn of an entirely new paradigm, or just the empty promise of yet another 100 mile-per-gallon carburetor? Robert X. Cringely holds forth...
posted by verdezza (16 comments total)

The downside of UWB for users is range, which is generally limited to around one kilometer with high gain antennas, and for the fastest data rates, can be measured in tens of feet. UWB trades bandwidth for distance, so longer links are slower

Isn't that quite a steep downside? In all my ignorance, it sounds like they're just better walkie-talkies with very long lasting batteries.
posted by MiguelCardoso at 9:57 AM on January 28, 2002

I know that he has a fairly large contingent of readers, but I've always thought that Cringely's name was perfect for his column. This article is much like his homebrew DSL article last year that made the rounds. It's all fine and dandy if you just want to contact the neighbors, but requires a giant infrastructure change if you want to make it useful. I agree with Miguel, at best his experiment will get him a walkie talkie for the PC.
posted by eyeballkid at 10:20 AM on January 28, 2002

This reads like a sci-fi novel. I have no doubts as to the article's accuracy, but it seems that new things are constantly being touted that never hit the market. UWB? I'll believe it when I see it.
posted by ashbury at 10:29 AM on January 28, 2002

Cringely's name isn't Cringely, of course; it's Mark Stephens.
posted by kindall at 11:08 AM on January 28, 2002

Eh? kindall, please kindly provide a link. I'm interested to know what you mean.

I think I was the one to start the Cringely bashing on MeFi a while back. I read over the article (skimmed actually -- too much bullshit for my mind to handle), but I think the real question is, "Why the hell are you talking about UWB when broadband hasn't even succeeded yet?"

Cringely definitely has some interesting ideas (read: conspiracy theories) into the tech industry, but just once I'd wish he'd at least provide some links to other more credible sources to give himself more credibility rather than just relying on hearsay. "I'm told by a good friend in the military..." or "A source deep within Cisco..."

Really. Until then, he's just another whacked out fanatic standing up on a soapbox trying to deliver a whacked out message to the world.
posted by PWA_BadBoy at 12:50 PM on January 28, 2002

OK, I did some legwork myself. A quick google search netted me this gem. Damn. What a f***in' loser. People like this really need to get out of the damn business.
posted by PWA_BadBoy at 12:55 PM on January 28, 2002

Hmmmmm. I think Accidental Empires is an excellent book, as I did when I read it the first time in '92.

And I believe the point of the homebrew wireless connectivity story was the fact that the infrastructure wasn't available where he was, so he cobbled together his own way to get to the infrastructure. Nice little piece of hackishness, from my way of looking at things.

And why bother kicking Cringely, when the insufferable Larry Magid is around?
posted by dglynn at 2:00 PM on January 28, 2002

PWA_BadBoy, why the rabid beef? Sheesh, the guy does have some credentials. You sound as though you lost money betting on some column of his or something.

The technology discussed in the article sounds to me like simply one of so many different solutions to data transmission. They all have drawbacks. UWB has range limitations, but low power requirements. For more bandwidth, add more power.

He's trying to say that UWB will challenge the current data transmission offerings somehow, which I doubt. It doesn't sound economical to deploy for wide-scale use. Unless people set it up themselves to share a last-mile connection, I don't see the competition. And while Bob may be the type of guy who would, most people won't. Too hard.

This is like the pitting of Bluetooth and 802.11 against each other. They don't really compete, exactly. Close, but not quite.

Bob does get a little nuts with the "technology for the little guy to stick it to the man" angle. He's that kinda guy, though.
posted by dammitjim at 2:41 PM on January 28, 2002

Drawbacks:

1. Transceiving on so many frequencies at such high frequencies requires having a tuner that is capable of oscillating over 5 million times a second. Sadly, such a thing doesn't exist (yet?). I'm a ham radio op (extra class) and the newest TNC's barely hit 28.8 kbps under low wattage and in lower frequencies.

2. Sending and receiving the data is enormously expensive. Not only is the demand on the processor immense, the cost of buying such an embedded chip is as well.

3. Extreme latency problems. 1 KM is (at sea level), suprisingly, a long distance for radio waves (even high freq ones) to travel in order to maintain a 'broadband' connection.

4. The noise floor is raised, impacting other signals and even the UWB component's impedence.

5. Negotiating the packet timing is damn near impossible across so many frequencies. A protocol that allows the transceivers to take turns and shake hands like this would be so convoluted that it would add more overhead to its already limited capability.

6. Even with single-side-band negotiations, too much bandwidth would be used, rendering the network instable and squished. We'll have to see how the FAA handles it, the band allocation is precious.

7. Without reserved bands, other transmissions would most certainly interrupt UWB. Transmitting in tandem with other non-UWB broadcasting stations is bullshit.

(Sorry for the long post, feel free to agree or disagree. I'm just stating what I believe to be true.)
posted by bloggboy at 3:00 PM on January 28, 2002

There's a lot of misconceptions about UWB. Here's my attempt at clearing them up:

UWB does not modulate the frequency but rather sends *extremely* short pulses and then varies when exactly it sends them. A short squarewave pulse occupies a very wide frequency range. That's where the name Ultra Wide Band comes from.

The receiver only listens for pulses when it's expecting one. It calculates when it's expecting one according to a code. Different transmitter/receivers have different codes so bandwidth can be shared (rather like Direct Sequence Spread Spectrum).

It's an old idea, the hard part was the precise timing. This has been solved by new chips that can operate at these extreme frequences ( 0 to greater than 5 billion hertz). These chips can now be made cheaply.

It does affect existing radios because these small pulse seem like an increase in ambient noise. Whether it's too much interference is what's being tested.

Another difficulty is antenna's that can operate over such a wide bandwidth, some kind of fractal design's seem to work well.

All in all UWB looks very interesting and some novel applications will emerge pretty soon.
posted by Amrik at 4:15 PM on January 28, 2002

PWA: You found the same Web page I did (I wanted to make sure I had his name right so searched on "Cringely real name" or something like that.)
posted by kindall at 5:36 PM on January 28, 2002

Here's an article on UWB from the Los Angeles Times from earlier this month, which says two of the technology's most prominent backers are Intel and Sony; and here's a very readable m-commerce times (?) article from last April, in which an industry exec claims that "in five or ten years' time, everybody will be using this technology in the same way as they use the transistor today."

I also found the home page of an organization called the "ultra-wideband working group," whose about page states that UWB devices are "... used for precise measurement of distances or locations and for obtaining the images of objects buried under ground or behind surfaces... [as well as for] wireless communications, particularly... short-range high-speed data transmissions suitable for broadband access to the Internet."

They then go on to offer slightly more detailed information on possible UWB applications for radar, tracking, and communications; you can find their FAQs here.
posted by verdezza at 9:02 PM on January 28, 2002

PWA_BadBoy, why the rabid beef? Sheesh, the guy does have some credentials

Credentials? Did you read the link he posted? The man is practically a fraud. He lives his life as a person who doesn't exist. He claims to be a PhD graduate, but has been proven to only hold a Master's degree. In communications.

He was hired at InfoWorld to write a gossip column. His work to the computer industry is what the National Inquirer is to the New York Times. Although the National Inquirer might have a bit more credibility than this man does.
posted by PWA_BadBoy at 9:17 PM on January 28, 2002

...when broadband hasn't even succeeded yet?

In America, and other places, perhaps. But here in Korea, it's almost to the point of a utility already. Pervasive, faultless (in my experience), and cheap DSL and cable. 4+ Mb pipe into my house, about $US17 a month. Uncapped.

They anticipate next-gen rollout to be complete by 2006, with fiber everywhere, and somewhere on the order of 5 Gb into the home. And they'll do it, too. They don't mess around with that stuff here - they just get it done.

*quiver

Meanwhile, poor Oz, still getting screwed by Telstra's monopoly power. Like a textbook lesson in how to guarantee that an otherwise vibrant country becomes a technological backwater...
posted by stavrosthewonderchicken at 10:44 PM on January 28, 2002

The man is practically a fraud... He was hired at InfoWorld to write a gossip column. His work to the computer industry is what the National Inquirer is to the New York Times...

PWA_BadBoy, I'm with dglynn and dammitjim -- I don't really understand the vitriol you're leveling against Cringely/Stephens. He seems like an incredibly resourceful opportunist -- and I mean that in a good way -- who isn't hurting anybody, and who has an obvious and unusual talent for spinning fun, folksy (and, to my ear, credible) tales from tech and techbiz arcana.

And as far as the weird backstory on his ID -- which I'd never heard before kindall mentioned it -- I find it kind of endearing. I mean, he built up the brand equity of the Cringely name almost singlehandedly, by writing eight years of InfoWorld columns under that moniker, and then successfully stood up to "the Man" to share in its value after they parted ways. (But then, I'm writing this from Manhattan, where personal reinvention is almost as praiseworthy -- and essential to success -- as it is in Hollywood, so my view is admittedly skewed.)

And stavrosthewonderchicken, what you tell us is already up and running and in the pipeline for where you live is absolutely awesome.
posted by verdezza at 4:08 AM on January 29, 2002

verdezza : Forgot about the 3G cell network already in place, and wireless broadband net access being rolled out on all subway lines in Seoul so you can check your email etc. while commuting to work...it is pretty amazing, actually. One of the things that makes living here more enjoyable, particularly for a geek like me.
posted by stavrosthewonderchicken at 4:56 AM on January 29, 2002