 h a l f b a k e r y Veni, vedi, fish velocipede
idea:
add, search, annotate, link, view, overview, recent, by name, best, random
meta:
news, help, about, links, report a problem
account:
Browse anonymously,
or get an account
and write.
or Create a new account.
|
|
|
Sudocrypto
Those puzzles might be ubiquitous for a non-obvious reason | |
A big problem with sending encrypted messages is managing the keys required to decrypt them. There are various solutions to this problem, heres a fun one.
A solved sudoku puzzle consists of 81 decimal digits. Why not use this as a crypto key? You could even send the unsolved puzzle right along
with the encrypted message. The recipient solves the puzzle, punches the digits into a program and voila.
A big limitation, of course, is anyone who can solve the puzzle can read the message. As long as your arch-enemy is stupid or lazy, this might well be enough.
Can we improve this? Sure
First, the number of possible sudoku puzzles is roughly equivalent to the key space of a 72 bit crypto key. That is not good at all. Instead of using the solution as a key directly, we could use it as a numeric index into a large database of pre-calculated keys. The keys themselves could be as large as necessary to resist direct cryptographic attack. The key database could be transmitted ahead of time, perhaps on a portable disk, and thus acts somewhat like a one time pad.
Second, dont transmit the puzzle with the message. Pre-arrange to find the correct puzzle in a known public location. Perhaps if we could get a series of these published somewhere, we could use a different one for each day you transmit a message...
...hey, wait a sec
.I wonder if this is baked? WIkipedia: Mathematics of Sudoku
http://en.wikipedia...thematics_of_Sudoku [jutta, May 02 2007]
[link]
|
| |
I'm not sure I get your point, bigsleep. I'm not proposing anything new cryptographically, just a novel way to manage keys that involves puzzles. This method would work equally well (if at all) for any chosen crypto algorithm. |
|
| |