Visit Your Local PBS Station PBS Home PBS Home Programs A-Z TV Schedules Watch Video Donate Shop PBS Search PBS
I, Cringely - The Survival of the Nerdiest with Robert X. Cringely
Search I,Cringely:

The Pulpit
The Pulpit

<< [ Trust me, I'm From Microsoft ]   |  The 100 Mile-Per-Gallon Carburetor  |   [ Soon, It's Gonna Rain ] >>

Weekly Column

The 100 Mile-Per-Gallon Carburetor: How Ultra Wide Band May (or May Not) Change the World

Status: [CLOSED]
By Robert X. Cringely
bob@cringely.com

A few weeks ago, I wrote about my desire for Wendy, my personal supercomputer (since completed and works fine, thanks), and mentioned that one of the projects I planned for her was an experiment in Ultra Wide Band (UWB) communications. This single reference brought from the web world many reader responses, most of them from engineers alarmed at the whole idea of UWB. They are convinced that UWB will be bad for us all, making useless the GPS satellite system, and maybe even your mobile phone. I am not at all convinced of that, but I am sure that UWB will shortly create a thorny business problem. What it reminds me of most is the proverbial 100 mile-per-gallon carburetor.

The 100 mile-per-gallon carburetor is supposed to be a gizmo that does exactly as its name suggests, only of course, we can't buy one. That's because the oil companies and the car companies and who knows what other companies are supposed to have bought-up the patents on the 100 mile-per-gallon carb specifically to make sure it never hits the market. Entire industries would be destroyed, we're told, if this beast ever hit store shelves. Certainly, Exxon-Mobil would have to sell a few corporate jets to get profits back in line once gasoline consumption dropped in half.

There is actually some precedent to this particular conspiracy theory. America had a well-developed light rail system at the end of the Second World War that was systematically bought up and destroyed primarily by General Motors, which promptly replaced rail cars with buses. If GM would buy one industry just to destroy it, why not buy another?

But what is it that makes UWB so scary?

Ultra Wide Band will either be the beginning of a new age of communication or the end of an old one, and probably both. UWB is true digital radio communication, a series of very short electrical pulses (billionths of a second) that exist not on any particular frequency, but on ALL frequencies simultaneously. It is just a blast of electrical noise. The key to turning that noise into communication lies in the timing of the pulses that beat out a code. But in order to hear the information in that code, a UWB receiver has to know the exact pulse sequence used by the transmitter. In other words, with UWB you have to know exactly WHEN to listen in order to hear. This makes UWB very secure, which is why the military likes it so much (UWB technology was originally developed for the military and various U.S. and Soviet spy agencies as early as 1960). UWB is pretty much immune to eavesdropping, is equally immune to interference or jamming, and because its broad frequency range includes the ultra-low frequencies used to communicate with submerged submarines, UWB can be used easily in buildings and even underground. And because of the random nature of UWB timing schemes, there appears to be virtually no limit to the number of UWB signals that can share the same airwaves.

In many ways, UWB is the successor to spread spectrum radio, a World War II technology for splitting a broadcast among many radio frequencies to avoid jamming. Spread spectrum, which was patented in 1942 by actress Hedy Lamar and composer George Antheil (I am not making this up), operates today in every mobile phone. But where spread spectrum used just a few dozen frequencies and used them one at a time, UWB uses every frequency there is, and uses them all at the same time, which means the data-carrying capacity of UWB is enormous.

UWB requires ultra-low power, often one ten thousandth as much as a cellphone, and because of that low power it is undetectable by conventional radios, looking to them just like very quiet noise. This is one reason why the FCC is considering allowing UWB as an unlicensed service; it is hard to regulate a signal that can't be detected and doesn't interfere. Besides, UWB has important friends — companies like Intel, IBM, Motorola and Texas Instruments — that are committed to making UWB chipsets or have invested in UWB startups.

A UWB phone uses so little power it can remain on for weeks without recharging. And UWB will ultimately be cheaper to make than conventional radios since it is built entirely of commercial grade computer chips and requires no tuning. There is an irony here that UWB was impossibly expensive to build until computer chips came along, and now, it is going to be impossibly cheap. The U.S. Navy, for example, plans to put a UWB location marker on almost everything it ships overseas, just to keep track of all the stuff and keep it from being stolen.

UWB products will probably begin to hit the market in the next 18 to 24 months. In addition to radios, these products will include radar and electronic positioning devices. For soldiers entering a strange building, UWB radar can show literally where all the bodies are, right through walls, ceilings and floors. As an electronic measuring device, UWB is accurate to within 10 centimeters — much better than competing technologies like Global Positioning System satellites, and UWB can be used indoors, while GPS cannot.

The first commercial UWB radios will probably be use for wireless computer networks (GE is working on that) and portable phones. UWB computer networks being developed now will initially operate at 40 to 60 megabits-per-second, or four to six times as fast as the most common networking technologies now in use. Researchers think that UWB networks may eventually run at speeds up to a gigabit-per-second, and therefore, should be able handle all the phone, television, and Internet traffic for any home or business. And since UWB radios don't interfere with each other, every home and business in the neighborhood can have its own gigabit.

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.

But one kilometer is about the range of PCS mobile phones, and those are successful. UWB can also solve the "last mile" problem that presently allows high-speed Internet services to be everywhere except in most of our homes, which are served by slow telephone wire.

Now back to the 100 mile-per-gallon carburetor metaphor. UWB is going to make a lot of trouble for existing communication businesses as it comes to offer a cheaper, better alternative for almost every way of getting in touch. Local phone companies, cable TV companies, mobile phone companies, and Internet Service Providers all look vulnerable. Even if they adopt UWB in order to compete, the value of their old infrastructure will drop to zero, which can't be good for business. Maybe this is why the big business-loving administration of George W. Bush is suddenly digging in its heels, delaying FCC approval of UWB.

Or perhaps, as my engineer correspondents have claimed, the FCC is just finally realizing at the last moment how bad UWB really is. Only time will tell.

UWB is even worse for those companies that are regulated communication monopolies or have exclusive licenses to certain parts of the radio spectrum. Having given billions to the FCC for those licenses, are they now going to be worth nothing? That is a possibility. Oh, the phone and cable and cell and Internet companies can all switch to UWB and compete, but their cost structure won't make it easy. And because UWB is unlicensed, there can suddenly be a hundred cable companies in town, not just one, with all the new cable companies having a distinct cost advantage.

The biggest losers, though, would appear to be the radio astronomers. Just as the light pollution from street lamps made work harder for astronomers with optical telescopes, UWB will raise the noise threshold for the radio astronomers.

The winners in this new market will be those best positioned to leverage UWB, like PCS vendor Sprint Communications, if it is willing to add new UWB radios to its thousands of short range cell sites. Optical backbone providers will also win, because UWB will increase the demand for long-haul bandwidth, yet can't compete in that sector because of UWB range limitations. And content providers like TV networks and movie studios will win because there will be an even broader market for their programs.

"Seinfeld" will still be "Seinfeld," even over UWB.

Comments from the Tribe

Status: [CLOSED] read all comments (0)