So what’s the deal with number stations? The shortwave stations that play series of numbers on loops continuously?
Are they really the government’s way of communicating with it’s spies abroad, or simply some goofballs that want to hear looped series of numbers while they sleep?
I admit, I didn’t know about them until one turned up on the tv show LOST.
Been here done this many times on here.
If you don’t like clicking links, nobody who does know has come forward yet, and everyone else is simply speculating.
It seems likely that they are information beacons for field operatives (either governmental or drug-financed) with one-time pads, but they could also be a ruse or a joke.
Interestingly, the one-time pad has been proven absolutely secure as long as the implementation is ideal. That is, no key can ever be reused. It is damned hard to accomplish that in practice, however.
Of course, this proof assumes that the pad is secret (only known to the encryptor and decryptor), which is sort of a strange assumption when you’re broadcasting it in cleartext.
Uh, what?
Presumably, it works like this: The pad is a list of numbers to subtract from the encoded numbers being sent out on the radio frequency. There are precisely two copies of this pad in existence anywhere: The sending station has one and the agent has one. As the station sends out numbers for that operative (at a specified time and frequency) the agent applies the pad and gets the cleartext, which can then be decoded. The pad must be destroyed at both ends once it is used, and that sequence of numbers can never be reused.
It is known that this is absolutely secure. Claude Shannon proved it decades ago. It is, however, usually impractical, because the key must be at least as long as all the cryptext ever sent on that channel.
Oh, sorry, I misunderstood, braindead today. I thought you meant that they were sending out new pads, not using existing ones to send information (stupid idea, I agree; I don’t know why I read you that way). I know the OTP is secure when correctly used.
To be fair, they could be sending out new keys using that method: The field agent is told to tune in to a specific frequency and use his pad to decrypt the stream of numbers he hears, which then becomes a 128-bit key for the symmetric-key cipher he’s using on his laptop.
But that’s Rube Goldberg-esque.
Using a one-time pad to send a key is a reliable yet fallible procedure. If anybody has got hold of the one-time pad, you’re shot, whether your communication is a new key or “attack at dawn”. (If the new key is encrypted by a less secure method, you’ve added another layer of confusion, but not something that can defeat cryptoanalysis. Unless the new key relies on its own one-time pad, etc.)
Precisely. One-time pads aren’t used much in cryptography because they are horribly impractical to deploy on a large scale, and there are good algorithms that aren’t as theoretically secure but are more than secure enough and worlds more practical.