Information security

Nguyen Phi Le
About this course

n Materials
q slides
n Evaluation
q Project + exercises: (30%)
q final exam: writing exam (70%)
n Contents
q Cryptography foundation
n Ciphers
n Cryptography protocols
q Security applications
n Digital signature
n Network security
Cryptography I

General concepts and some classical

ciphers
Outline

n Basic concepts
n Attack models
n Classic ciphers: mono-alphabetic
n Vigenere cipher
n One-time-pad cipher
Security Goals

n Confidentiality (secrecy, privacy)

q Assure that data is accessible to only one who are
authorized to know
n Integrity
q Assure that data is only modified by authorized
parties and in authorized ways
n Availability
q Assure that resource is available for authorized
users
General tools

n Cryptography
n Software controls
n Hardware controls
n Policies and procedures
n Physical controls
What is Crypto?

n Constructing and analyzing cryptographic protocols

which enable parties to achieve security objectives
q Under the present of adversaries.
n A protocol (or a scheme) is a suite of procedures
that tell each party what to do
q usually, computer algorithms
n Cryptographers devise and analyze protocols under
Attack model
q assumptions about the resources and actions available to
the adversary
n So, you need to think as an adversary
Terms

n Cryptography: the study of mathematical

techniques for providing information security
services.
n Cryptanalysis: the study of mathematical
techniques for attempting to get security services
breakdown.
n Cryptology: the study of cryptography and
cryptanalysis.
Terms
n plaintexts
n ciphertexts
n keys
n encryption
n decryption

Y=EZ(X) Y X=DZ’(Y)
Sender Alice Receiver Bob

Key Z Enemy/Adversary Eve Key Z’

Secret-key cryptography

n Also called: symmetric cryptography

n Use the same key for both encryption & decryption
(Z=Z’)
n Key must be kept secret
n Key distribution – how to share a secret between A
and B very difficult
Public-key cryptography

n Also called: asymmetric cryptography

n Encryption key different from decryption key and
q It is not possible to derive decryption key from encryption
key
n Higher cost than symmetric cryptography
Breaking ciphers …

n There are different methods of breaking a

cipher, depending on:
q the type of information available to the attacker
q the interaction with the cipher machine
q the computational power available to the attacker
Breaking ciphers …

n Ciphertext-only attack:
q The cryptanalyst knows only the ciphertext.
q Goal: to find the plaintext and the key.
q NOTE: such vulnerable is seen completely insecure
n Known-plaintext attack:
q The cryptanalyst knows one or several pairs of
ciphertext and the corresponding plaintext.
q Goal: to find the key used to encrypt these messages
n or a way to decrypt any new messages that use the same key
(although may not know the key).
Breaking ciphers …

n Chosen-plaintext attack
q The cryptanalyst can choose a number of messages and
obtain the ciphertexts for them
q Goal: deduce the key used in the other encrypted
messages or decrypt any new messages (using that key).
n Chosen-ciphertext attack
q Similar to above, but the cryptanalyst can choose a
number of ciphertexts and obtain the plaintexts.
n Both can be adaptive
q The choice of ciphertext may depend on the plaintext
received from previous requests.
Classic ciphers
Shift cipher (additive cipher)
n Key Space: [1 .. 25]
n Encryption given a key K:
q each letter in the plaintext P is replaced with the K’th letter
following corresponding number (shift right):
q Another way: Y=X Å K è additive cipher

n Decryption given K:
q shift left

ABCDEFGHIJKLMNOPQRSTUVWXYZ
0 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25
P = CRYPTOGRAPHYISFUN
K = 11
C = NCJAVZRCLASJTDQFY
Shift Cipher: Cryptanalysis

n Easy, just do exhaustive search

q key space is small (<= 26 possible keys).
q once K is found, very easy to decrypt
General Mono-alphabetical Substitution
Cipher
n The key space: all permutations of Σ = {A, B, C, …, Z}
n Encryption given a key π:
q each letter X in the plaintext P is replaced with π(X)

n Decryption given a key π :

q each letter Y in the cipherext P is replaced with π-1(Y)

n Example:
ABCDEFGHIJKLMNOPQRSTUVWXYZ
π=BADCZHWYGOQXSVTRNMSKJIPFEU

BECAUSE → AZDBJSZ
Looks secure, early days

n Exhaustive search is infeasible

q key space size is 26! ≈ 4*1026
n Dominates the art of secret writing throughout
the first millennium A.D.
n Thought to be unbreakable by many back
then
Cryptanalysis of Substitution Ciphers:
Frequency Analysis
n Each language has certain features:
q frequency of letters, or of groups of two or more letters.
n Substitution ciphers preserve the mentioned
language features è vulnerable to frequency
analysis attacks
General Mono-alphabetical Substitution
Cipher
n Observations:
q A cipher system should not allow statistical properties of
plaintext to pass to the ciphertext.
q The ciphertext ginerated by a "good" cipher systim should
be satistically indistinguishable form random text.
n Idea for a stronger cipher (1460’s by Alberti)
q use more than one cipher alphabet, and switch between
them when encrypting different letters è Polyalphabetic
Substitution Ciphers
q Developed into a practical cipher by Vigenère (published in
1586)
Polyalphabetic Substitution Ciphers
n Polyalphabetic Substitution Ciphers
(Vigenère cipher - published in 1586)
q Definition:
n Given m, a positive integer, P = C = (Z26)n, and K = (k1, k2, …, km) a
key, we define:
q Encryption:
n ek (p1, p2… pm) = (p1+k1, p2+k2…pm+km) (mod 26)
q Decryption:
n dk (c1, c2… cm) = (c1-k1, c2-k2 … cm- km) (mod 26)
q Example:
Plaintext: CRYPTOGRAPHY
Key: LUCKLUCKLUCK
Ciphertext: NLAZE I I B LJ J I
Vigenère cipher
n Cryptanalysis
Plain text 𝑃 Cipher text 𝐶

𝑝! 𝑝" …… 𝑝# 𝑐! 𝑐" …… 𝑐#
𝑝#$! 𝑝#$" …… 𝑝"# 𝑐#$! 𝑐#$" …… 𝑐"#
. .
. .
. .
𝑝%#$! 𝑝%#$" …… 𝑐%#$! 𝑐%#$" ……

+𝑘!
+𝑘!

Same statistical properties

Can be broken by the statistical method once the key length is determined
Vigenère cipher
n How to determine the key length
q The frequency of letters in 𝑝! , 𝑝"#! , … , 𝑝$"#! is
approximately the same as that in the plain text 𝑃
q The frequency of letters in 𝑐! , 𝑐"#! , … , 𝑐$"#! is
the same as that in 𝑝! , 𝑝"#! , … , 𝑝$"#!
n The index of coincidence (IC)
q Suppose 𝑥 = 𝑥% 𝑥& … . 𝑥' is a string of alphabetic
characters à 𝐼𝐶 𝑥 is the probability that two
random elements of 𝑥 are identical
 Vigenère cipher
n The index of coincidence (IC)
q Suppose the frequencies of 𝐴, 𝐵, … , 𝑍 in 𝑥 are
𝑓( , 𝑓% … , 𝑓&)
+&
∑)*
&'( ) -& -& .%
q 𝐼𝐶 𝑥 = , = ∑&)
+,( ≈ ∑&)
+,( 𝑝+
&
) ' '.%
𝑝" : the frequency of the i-th letter
letter probability
A .082
B .015
C .028 For an English text For a totally random string
𝐼𝐶 𝑥 ≈ 0.065 $% 1
D .043 𝐼𝐶 𝑥 ≈ % = 0.038
E .127 !"# 26
F .022
…
Z .001
 Vigenère cipher
n The index of coincidence (IC)
q Let 𝑃& = 𝑝& , 𝑝'(& , … , 𝑝)'(& ; 𝐶& = 𝑐& , 𝑐'(& , … , 𝑐)'(&
n 𝐼𝐶 𝐶& = 𝐼𝐶 𝑃& ≈ 0.065
n Cryptanalysis algorithm
1. Set 𝑚 = 1
2. Check if 𝑚 is indeed the key length
n Divide the cipher into 𝑚 letter group and compute the IC of each
n If they are quite the same and approximately equals to 0.065 then
𝑚 is the key length
n If they are quite different and smaller than 0.065, then the key length
should be greater
3. Increase 𝑚 by 1 and go to step 1
 Vigenère cipher

n Kasiski method: a hint to find the key length

q Observation: two identical segments of plaintext
will be encrypted to the same cipher text wherever
their occurrence in the plain text is 𝛿 position
apart, 𝛿 ≡ 0(𝑚𝑜𝑑 𝑚)
𝑝! … 𝑝" 𝑝"#! … 𝑝"#$ …… 𝑝% 𝑐! … 𝑐" 𝑐"#! … 𝑐"#$ …… 𝑐%
. .
. .
. .
𝑝&%#! … 𝑝&%#" 𝑝&%#"#! … 𝑝&%"#$ …… 𝑝&%#% 𝑐&%#! … 𝑐&%#" 𝑐&%#"#! … 𝑐&%"#$ …… 𝑐&%#%

If these are the same Then, these will be the same

Vigenère cipher

n Kasiski method
q Search the cipher text for pairs of identical
segments and record the distance between their
starting positions
n Suppose the obtained distances are 𝛿! , … , 𝛿%
q Then, 𝑚 should divides the greatest common
divisor of 𝛿% , … , 𝛿$
Vigenère cipher

n Example
Vigenère cipher

n Example

Kasiski method: CHR’s occurrence positions: 1, 166, 236, 276 and 286
à Distances: 165, 235, 275 and 285
à Gcd(165, 235, 275, 285) = 5
à The key length should divides 5
Vigenère cipher

n Example

Confirmation of Kasiski method

M = 1 à IC = 0.045
M = 2 à ICs = 0.046 and 0.041
M = 3 à ICs = 0.043, 0.050, 0.047
M = 4 à ICs = 0.042, 0.039, 0.046, 0.040
M = 5 à ICs = 0.063, 0.068, 0.069, 0.061 and 0.072
Vigenère cipher

𝑝" 𝑓"(#
&'
𝑀# = ,
"$% 𝑛′

If 𝑔 ≠ 𝑘" , then 𝑀# ≪ 0.065

The almond tree was in tentative blossom. The days were longer,
often ending with magnificent evenings of corrugated pink skies.
𝐾 = 9, 0, 13, 4, 19 = 𝐽𝐴𝑁𝐸𝑇 The hunting season was over, with hounds and guns put away for
six months. The vineyards were busy again as the well-organized
farmers treated their vines and the more lackadaisical neighbors
hurried to do the pruning they should have done in November.
Exercises
n Decode the following cipher texts
q Encrypted by shift cipher:
n JBCRCLQRWCRVNBJENBWRWN
q Encrypted by substitution cipher:
n Pjmu mu b amtjfo rfsr. Mr jbu cffi fiaowtrfg cw rjf uvcurmrvrmqi amtjfo. Wqv
bof xfow nvahw. Rjf amtjfo jbu cffi coqhfi
n YIFQFMZRWQFYVECFMDZPCVMRZWNMDZVEJBTXCDDUMJ
NDIFEFMDZCDMQZKCEYFCJMYRNCWJCSZREXCHZUNMXZ
NZUCDRJXYYSMRTMEYIFZWDYVZVYFZUMRZCRWNZDZJJ
XZWGCHSMRNMDHNCMFQCHZJMXJZWIEJYUCFWDJNZDIR
n Hints:
q The letters in the English alphabet can be divided into 5 groups of similar frequencies
§ e
§ t,a,o,i,n,s,h,r
§ d,l
§ c,u,m,w,f,g,y,p,b
§ v,k,j,x,q,z
q Some frequently appearing bigrams or trigrams
§ Th, he, in, an, re, ed, on, es, st, en at, to
§ The, ing, and, hex, ent, tha, nth, was eth, for, dth.
Exercises
n Decode the following cipher texts
q Encrypted by substitution cipher:
YIFQFMZRWQFYVECFMDZPCVMRZWNMDZVEJBTXCDDUMJ
NDIFEFMDZCDMQZKCEYFCJMYRNCWJCSZREXCHZUNMXZ
NZUCDRJXYYSMRTMEYIFZWDYVZVYFZUMRZCRWNZDZJJ
XZWGCHSMRNMDHNCMFQCHZJMXJZWIEJYUCFWDJNZDIR

DZ and ZW: four times each

NZ and ZU: three times each
RZ, HZ, YZ, FZ, ZR, ZV, ZC, ZD, ZJ: twice each