to encrypt the letters of the clear text message. Again, the cipher alphabet

can consist of any type of symbol or set of symbols as long a specific cleartext

unit will always encode to the same cipher text unit and each cipher text unit

will always decode to the same clear text unit. I say unit, as an abstraction,

because monoalphabetic substitution ciphers can be classified as one of two

methods of encryption: uniliteral ciphers and multiliteral ciphers. This will

be explained in depth later but for now just assume that a unit means a single

symbol [clear or cipher ].

1. Standard Sequence

2. Systematically Mixed Sequences

3. Random Sequences

order or Z-A order in reverse standard sequences]. Both are equally "relatively"

secure - by relatively secure I mean that if the cryptanalyst knows the type

of cipher and is attacking the key then neither is more secure than the other

because both only have 25 possible keys [25 possible shifts or rotations, as in

the Caesar cipher, which makes these a special sub-class of a monoalphabetic

substitution cipher known as a displacement ciphers].

Clear Alphabet

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

Cipher Alphabet

Z

Y

X

W

V

U

T

S

R

Q

P

O

N

M

L

K

J

I

H

G

F

E

D

C

B

A

were encoded by using the letter that directly maps beneath it [in the Cipher

Alphabet] the encrypted message would be: GSRH RH Z HVXIVG NVHHZTV.

cipher alphabet because the alphabet has been reversed but it is still in standard

order [the form follows from A - Z].

start of the clear text alphabet, then, omitting repetitive letters, the rest of

the alphabet is written out. If the keyword is E V E R C R A C K it is reduced to

E V R C A K and the below clear text to clear text alphabet results. Again, the

message: THIS IS A SECRET MESSAGE

Clear Alphabet

E

V

R

C

A

K

B

D

F

G

H

I

J

L

M

N

O

P

Q

S

T

U

W

X

Y

Z

Cipher Alphabet

Z

Y

X

W

V

U

T

S

R

Q

P

O

N

M

L

K

J

I

H

G

F

E

D

C

B

A

encrypts to: FPOG OG V GZWXZF LZGGVQZ.

but is really no different - a single cleartext unit maps to a single ciphertext

unit. If the cryptanalyst attacks the key, the solution space is much greater

than the previous method. If the cryptanalyst is using a generalized solution

[that applies to all uniliteral, monoalphabetic substitution ciphers], like

EverCrack, then the methods are equally "difficult" - a generalized solution

covers all (26! - 1) possible random cipher alphabets [which includes all of

the above].

space [possible alphabets used] (26! - 1). This is a fairly high number and even

a brute force attack on modern computers would still take a while. A brute force

attack would equate to a permutation attack against the alphabet itself. On my

500 MHz laptop a permutation of 11 letters [39, 916, 800 permutations] took

approximately 25 minutes. For a permutation of 26 letters [403, 291, 461, 126

605, 635, 584, 000 ,000 permutations] is roughly 10 pentillion times greater

than the permutation of 11 letters it would be suffice to say that it would

take 480, 560, 378, 380, 273 years to solve to perform a permutation attack

on the cipher alphabet.

analysis. Frequency analysis relies on the fact that specific letters in

specific languages occur in fairly regular frequencies or percentage of a text.

This is because languages tend to be redundant [and some phonemes and letter

combinations easier to verbalize are used more frequently]. By redundant I

mean that most words [especially longer words] tend to have repetitive letters

[e.g., "repetitive" has three letters, 'e', 't', and 'i', that occur more than

once].

Letter	Percentage	Letter	Percentage
E	12.77	C	2.96
T	8.55	M	2.88
O	8.07	P	2.23
A	7.78	Y	1.96
N	6.86	W	1.76
I	6.67	G	1.74
R	6.51	B	1.41
S	6.22	V	1.12
H	5.95	K	.74
D	4.02	J	.51
L	3.72	X	.27
U	3.08	Z	.17
F	2.97	Q	.8

This distribution of frequencies tends to vary as the size of text

EverCrack does not use frequency analysis. It uses a comparison and reduction

approach based on the patterns of the cipher words. The full description of how

it works EverCrack Kernel.

and the online tool is EverCrack Tool

single cipher symbol - this is the simplest form of a monoalphabetic sub-

stitution cipher - there is always a one-to-one correspondence between cleartext

units and cipher text units. Here the difference between a unit and a symbol

emerges - in multiliteral ciphers the single cleartext symbol is replaced by more

than one ciphertext symbol [a polygraph].

cipher symbols tend to exhibit the same frequency distribution as the underlying

clear text message.

Specific Uniliteral Monoalphabetic Substitution Ciphers

with two or more cipher symbols. If two cipher symbols are used to rep-

resent a single clear symbol, then using the following plain and cipher alphabets:

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

ZY

XW

VU

TS

RQ

PO

NM

LK

JI

HG

FE

DC

BA

AB

CD

EF

GH

IJ

KL

MN

OP

QR

ST

UV

WX

YZ

clear text message: T H I S I S A S E C R E T M E S S A G E

cipher text message: MNLKJIKL JIKL ZY KLRQVUIJRQMN BARQKLKLZYNMRQ

if you know that it is a multiliteral cipher it is as trivial to crack as a

uniliteral cipher - just a little extra manual work. It should be obvious that

it is not a uiliteral cipher because it would then be an anomolous message

consisting of five words, three of which are quite lengthy [words of specific

lengths also tend to have specific frequency distributions].

distribution of the underlying cipher text. In the previous section, it was

discovered that every language has a characteristic pattern to the distribution

of the letters that comprise that language. That property of language is a ref-

erence that can be used to perform cryptanalysis. In this case, to make the

cipher more difficult to solve, that property of the language has been trivially

disguised.

to-Uniliteral Cipher Convertor can simply replace the multiliteral polygraphs

with uniliteral monographs [the conversion is arbitrary] to then feed into the

EverCrack Monoalphabetic Substutition Cipher Cracker to crack the message.

[Specific Multiliteral Monoalphabetic Substitution Ciphers]

symbol at a time [whether this encrypts into one or more cipher symbols does

not make a difference]. All of the previous plain-cipher alphabets have used

this method. The important factor resulting from this method is, if only

alphabetical characters are encrypted [in contrast to punctuation and numbers],

there will always be a pool of 26 [or less] unique clear symbols and cipher

symbols.

ciphers is that the cipher symbols still exhibit the characteristic frequency

distribution of the clear letters comprising the clear message.

[Specific Monographic Monoalphabetic Substitution Ciphers]

plain symbol at a time [in this case, the number of cipher symbols the clear

symbols encrypt to does matter]. Although, since we are still dealing with

monoalphabetic substitution ciphers at this point, the correspondence between

clear text units and cipher text units is still one-to-one, this type of

cipher is considerably more difficult to crack than a monographic cipher. Out of

the 26 letters, a pool of 676 digraphs can be constructed [although many di-

graphs, BX, never actually appear in clear text]. Thus, there must be a

few hundred cipher digraphs to account for the clear digraphs [that are en-

crypted] depending on the size of the message.

we have approximately 9 unique digraphs to encrypt [if the message does not

contain an equal number of letters a "cipher pad" can be used]:

Clear Digraphs	TH	IS	AS	EC	RE	TM	ES	SA	GE
Cipher Digraphs	AB	CD	EF	GH	IJ	KL	MN	OP	QR

with a resulting cipher text of: AB CD CD EF GH IJ KL MN OP QR

[Return to Evercrack Main]