Enigma probable word attack

From: Aquila (braempje_at_netscape.net)
Date: 03/22/05 

Date: Tue, 22 Mar 2005 16:18:54 +0100

I'm working out a probable word attack on the machine I described earlier. I
have been able to find four independent loops in my crib so it should work.
I'm trying to attack as described on page 23 in
<http://www.esat.kuleuven.ac.be/~abiryuko/Cryptan/enigma_lecture.ps>.
 After trying all combinations for my 5 rotors (only 3 are used) I get 74
"fixpoints": letters that are identical when decoded by all loops (in fact
by enigma's started at the loops positions in the crib).
How should I go on now, I don't really understand the paper from then on:
"thus there remain only three possible letters", why?
Say all my loops are going trough a common letter X, should I then try to
find a letter that eventually get's transformed into X and is that the
other part of the plug with X? Or should I try to find a unique traversal
of all letters and is the starting letter the one I am looking for?

Thanks for your help 

-- 
"May the source be with you"

Relevant Pages
Quantcast