Viewing Single Post From: Brute Forcing The Adfgvx Cipher

Viewing Single Post From: Brute Forcing The Adfgvx Cipher
jdege	Apr 29 2008, 03:42 PM
Elite member Posts: 263 Group: Admin Member #246 Joined: December 7, 2006	Revelation Apr 28 2008, 01:09 PM That's what I suggested in my second and third post and that's what my application would do: first sweeping the pattern of a crib through the text and then checking the frequency of the bigrams (btw, I haven't started writing it yet) And you were a step or two ahead of us. I was just figuring it out. Revelation Apr 28 2008, 01:09 PM Your other sequence is a pain, especially if you use two different squares in step 1 and 3. I'm not sure how to solve that. When I think about it, I'm always driven back to the same question - how can we recognize when we've properly reversed one step? Fractionate+transpose+unfractionate. Is there something we can recognize in a fractionate+transpose, that would allow us to determine that we have properly reversed the unfractionate? If so, does a not-quite-there attempt at unfractionating look more like a fully-there attempt at unfractionating? If so, then the heuristic methods should work - shotgun hill climbing, simulated annealing, genetic algorithms. But it'd take an in-depth look at the properties of fractionate+transpose to determine whether that was the case. One thing's for certain - no exhaustive search of a Polybius square keyspace is going to work. Exhaustive search of single columnar transposition is doable, assuming moderate key lengths. 8!, 9!, or 10! can be handled. So using exhaustive search to remove the transposition is feasible, particularly since we think we may have a good test for recognizing the correct result. But 25! is a whole different world.
	When cryptography is outlawed, bayl bhgynjf jvyy unir cevinpl.

Brute Forcing The Adfgvx Cipher · General