CSE565: Computer Security

Lectures 2 & 3
Basic Encryption &
Decryption
Shambhu Upadhyaya
Computer Science &
Eng.
University at Buffalo
Buffalo, New York,
14260
CSE565: S.
9/03/02 UB Fall 2002 Upadhyaya
Lec 3.1
 Overvie
 Lecture 3 w
 Terminology (5 minutes)
 Encryption by Substitutions &
Transpositions (70 minutes)
 Lecture 4
 Any leftover material from Lecture 3
 Examples
 Tracking a hacker (55 minutes video
presentation)

CSE565: S.
9/03/02 UB Fall 2002 Upadhyaya
Lec 3.2
 Terminolog
 Encryption y
 A process of encoding a message
 Decryption
Original
 It is the reverse process
Plaintext Plaintext
Ciphertext
Encryption Decryption
P=(p1,p2, …pn) P=D(C)
C=(c1,c2, … cm) =D(E(P))
C=E(P)

 Encryption Algorithms
 A key K is generally used
P  D ( K , E ( K , P ))
 Symmetric encryption:
P  D( KD, E ( KE , P ))
 Asymmetric encryption: CSE565: S.
9/03/02 UB Fall 2002 Upadhyaya
Lec 3.3
 Cryptography &
 Cryptography
Cryptanalysis
 Hidden writing -- encryption to conceal
text
 Cryptanalysis
 Break an encryption
 Attempt to break a single message
 Attempt to recognize patterns in encrypted
messages
 Attempt to find general weaknesses
 Breakable Encryption
 Given enough time and data, encryption
can be broken
 Practicality is an issue
 Risky to pronounce an algorithm secure!
CSE565: S.
9/03/02 UB Fall 2002 Upadhyaya
Lec 3.4
 Substitutions &
Transpositions
 Substitutions
 Monoalphabetic Ciphers
 Polyalphabetic Ciphers
 These are called Stream Ciphers
 Transpositions
 Permutation
 These are Block Ciphers

CSE565: S.
9/03/02 UB Fall 2002 Upadhyaya
Lec 3.5
 Monoalphabetic

 Caesar Cipher
Ciphers
 Formula: ci  E ( pi )  pi  3
 Treaty Impossible --> wuhdwb lpsrvvleoh
 Advantages & Disadvantages
 Easy to perform
 Simple, so easy to predict the pattern
 Repeat characters give clue to break
 Permutation Cipher
 Use a key
 ABCDE ………
 k e ya b……….
 Since key is short, most plaintext letters are only
one or two positions off
 Time complexity of these algorithms -- linear in n
where n is the no. of characters in plaintext
CSE565: S.
9/03/02 UB Fall 2002 Upadhyaya
Lec 3.6
 Polyalphabetic
Substitution
 Main problem withCiphers
monoalphabetic ciphers is
the non-flat frequency distribution
 A flat distribution gives no information to
cryptanalyst
 Flattening by combining distributions
 Two or more separate encryption alphabets for
the same character of the plaintext
 Eg., one encryption for characters in odd
positions and another for at even positions
 Reduces the repeat symbols in the ciphertext
 Makes it hard to break
 (Discuss an example from the Book) CSE565: S.
9/03/02 UB Fall 2002 Upadhyaya
Lec 3.7
 Vigenere
 Need to Tableaux
smooth out peaks and valleys of
frequency distribution
 Can be done by extending no. of permutations
 Extreme case -- extend to 26 permutations
 A 26 X 26 table is used (example from Book)
 Key word of length 3 is sufficient to smooth
out the distribution
 Cryptanalysis of Polyalphabetic Substitution
 Reading exercise
 Focus on Kasiski method and Index of
Coincidence method
CSE565: S.
9/03/02 UB Fall 2002 Upadhyaya
Lec 3.8
 Perfect Substitution

 Cipher
Requires an infinite non-repeating sequence of
alphabets
 This will confound the cryptanalyst
 Vernam Cipher
 Uses an arbitrarily long sequence of numbers
 Formula: ci  ri  pi mod n
 Random Numbers
 Middle digits of phone numbers in a residential
directory
 Pseudo random numbers -- linear congruential
random number generator
 Formula: ri  1  ( a * ri  b) mod n CSE565: S.
9/03/02 UB Fall 2002 Upadhyaya
Lec 3.9
 Transposition
 sthe message are rearranged
It is a method where letters of
 Goal here is diffusion rather than confusion
 Information is spread widely across the ciphertext
 Columnar transposition is an easy one
 Characters are rearranged into columns
 Study an example from Book
 Encipherment/Decipherment Complexity
 Algorithm is constant in the amount of work per character
 Time is proportional to length of message
 Space required is directly proportional to message length
 Output characters cannot be produced until all characters
are read in
 Delay depends on the length of the message
 Not appropriate for long messages CSE565: S.
9/03/02 UB Fall 2002 Upadhyaya
Lec 3.10
 Cryptanalysi
s
 Digrams, trigrams & other patterns used to break
encryption
 Digram -- letter pairs such as -re- -th- -en- -ed-
etc.
 Trigram -- groups such as ent, and, ing, thi etc.
 Cryptanalysis by trial & error & using a moving
window for comparison
 This is a reading exercise! (Sec. 2.4)
 Double transposition algorithm makes breaking
more difficult
 It uses transposition of a transposed text
 See example from the Book CSE565: S.
9/03/02 UB Fall 2002 Upadhyaya
Lec 3.11
 Summar
 Stream Ciphers y
 Mono and poly-alphabetic encryptions
 Since they convert one symbol of plaintext
immediately into a symbol of ciphertext
 Features include: a) speed of transformation, b) low
error propagation
 Disadvantages are: a) low diffusion, b) susceptible
to malicious insertions
 Block Ciphers
 Columnar transposition algorithm and fractionated
Morse
 Since a group of plaintext symbols are encrypted as
one block
 Features include: a) diffusion, b) immunity to
insertions
 Disadvantages are: a) slowness of encryption, b)
CSE565: S.
9/03/02
error propagation UB Fall 2002 Upadhyaya
Lec 3.12