Criptoanalisys Outwit using Context Sensitive Grammars

This paper presents the use of context sensitive grammars in order to outwit cryptanalysis based on language character frequency. The main goal is to take a plain text written in an idiom and transform it into a new text whose character frequency mimics a different idiom. Once created this new text, it can be submitted to any cryptographic algorithm. The proposed approach uses theory of computation techniques combined with linguistics appended to traditional encode and decode algorithms. Therefore, it was possible to obtain cryptograms more resistant to this kind of attack.