[CroCivic91]

I have a text crypted with Cesar's Key Word encryption algorithm. It is a single-letter substitution algorithm which uses a key in the form of ( A, B ) where A is a word or phrase and B is a number. For example: if the key is ( "Hello world", 5 ), that means that the "table" of encryption looks like:

A B C D E F G H I J K L M N O P Q R S T U V W X Y Z <- plain text
V X Y Z H E L O W R D A B C F G I J K P Q R S T U <- crypted text

So the "Hello world" is written from the 5th letter on, and when the phrase is written (one letter one time only) then the alphabet is added to it (again, looking out not to have the same letter written twice).

I know that basic technique to decrypt this would be to take a crypted text and count the frequency of crypted letters in the text, and then "guess" which letter is which using frequency of letters in a plain text (depending on which language the original text was in). Now there are generally 26! Of possibilities to make this kind of encryption tables (a bit less, but it doesn't really matter), which is a too big number to go about solving it with brute force algorithms.

What can I do to decrypt this text? I got this homework, and statistics of frequency of letters is just messed up on purpose and it is very annoying to try solving the problem this way, because 6 most frequent letters are all messed up, so I get rubbish each time I try something, and there are 720 ways to mix those 6 letters. Any smarter way to go about this?

Thanks a lot,

Kresho

[AnupamBhatia]

Try genetic search techniques

[ebaines]

You can look at the 1-letter and 2-letter words to get some clues. The only 1-letter words are "A" and "I," so if there are any occurrences of these you have a good head start. For 2-letter words, at least one of them must be a vowel, and the list of common 2-letter words is pretty short (am, an, as, at, be, by, go, he, hi, if, in, is, it, me, my, of, on, or, so, to, up, we). Also look for occurrences of "the" - it's the most common 3-letter word. All words must contain at least one vowel (or "y"), so you may be able to find a set of 6 letters that occur in every word. Finally, you may be able to glean something from any 2-letter combinations that occur frequently - they could be common diphthongs such as "th," "sh," "ck", etc. Good luck!

How do you know that the frequency of the most common letters is "messed up"? As you probably know, "e" is the most common letter - did the author of the puzzle tell you that this isn't the case?