An interesting article from Cringley on UWB as a potential secure communications medium. He speculates that, because UWB offered the promise of “completely secure”* communications (you have to know what nanosecond to be listening) the NSA did some covert mojo and ensured the FCC regulated UWB down to the point where it was practically unusable.
* In theory, at least.
I’ve covered UWB a few times here before. It’s indeed a very interesting technology. A UWB-transmitter equipped bug, for example, would not appear on most TSCM teams’ spectrum analyzers. (The US actually regulates specturm analyzers capable of detecting short pulses such as UWB uses, and prevents them from being exported without a license.)
Of course, non-selective receivers (like the “Sinitsa” I’ve also covered) detect them fine.
More interesting to me is the underlying physics of UWB work. The math gets very hairy very fast, but it’s one of the few places I’ve seen the beginnings of an explanation for military communications technologies that manage to transmit non-optical signals on optical fibres, and other “dielectric” mode phenomena. (Essentially using the fibers as a waveguide. This was mentioned in passing in a Reddit engineering discussion I once saw, for example.)
In any case, given that there’s evidently lots of interest in really secure communications… how come nobody’s put together a “one time pad kit” for sale via the usual hacker conference tables and privacy outlets?
 Tesla was a big fan of working with “dielectric” phenomena, I think. Two books I’d someday like to get around to reading on the subject —