To help explain what an adaptive mirror field is, let's first explain what a typical mirror field is.
A mirror field is a square grid that contains randomly placed mirrors in a diagonal orientation. Characters that can be encrypted are placed around the perimeter of the mirror field. Encryption occurres by standing on the character you wish to encrypt, looking straight into the field, and observing the character that the mirrors reflect back.
Mirror Field Example:
A B C D E
-----------
K: :/: :/: :F
-----------
L: : :\: :/:G
Input ----------- Output
------- ==> M:\: : :/: :H ==> -------
"ABCDE" ----------- "FKEMC"
N:\:/: : : :I
-----------
O: : :/: :\:J
-----------
P Q R S T
In practice, using a typical mirror field is a weak form of encryption. Given the same plaintext character as input, the same cyphertext character is produced as output, creating a pattern and providing hints that an attacker can use to derive the key.
A B C D E
-----------
K: :/: :/: :F
-----------
L: : :\: :/:G
Input ----------- Output
------- ==> M:\: : :/: :H ==> -------
"CCCCC" ----------- "EEEEE"
N:\:/: : : :I
-----------
O: : :/: :\:J
-----------
P Q R S T
In an adaptive mirror field, the mirror orientations and position of the perimeter characters are dynamic.
For every character that is processed:
- The character's path through the mirror field reorients the mirrors it contacts, creating a unique mirror field permutation for the next character.
- The plaintext and cyphertext characters are moved to a new perimeter position.
These two features eliminate identifiable patterns in the output, most importantly when given repeating characters as input.
MrrCrypt encrypts data 1 byte at a time. A byte has 256 possible values. The size of the mirror field is 64x64 to accomodate a 256 character alphabet.
In this visualization, characters are represented in hexidecimal.
