Lecture 6

Cryptographic
Hash
Functions

Purpose
Ø One of the most important tools in
modern cryptography and security

Ø In crypto, instantiates a Random Oracle

Ø In security, used in a variety of

authentication and integrity applications

Ø Not the same as hashing used in DB or

CRCs in communications 2

1
 Cryptographic HASH Functions
Ø  Purpose: produce a fixed-size “fingerprint” or digest of
arbitrarily long input data

Ø  Why? To guarantee integrity

Ø  Properties of a “good” cryptographic HASH function H():

1.  Takes on input of any size
2.  Produces fixed-length output
3.  Easy to compute (efficient)
4.  Given any h, computationally infeasible to find any x such that
H(x) = h
5.  For a given x, computationally infeasible to find y such that
H(y) = H(x) and y<>x
6.  Computationally infeasible to find any (x, y) such that
H(x) = H(y) and x<>y
3

Same properties re-stated:

v  Cryptographic properties of a “good” HASH
function:
v  One-way-ness (#4)
v  Weak Collision-Resistance (#5)
v  Strong Collision-Resistance (#6)

v  Non-cryptographic properties of a “good”

HASH function
v  Efficiency (#3)
v  Fixed output (#1)
v  Arbitrary-length input (#2)
4

2
 Construction
Ø  A hash function is typically based on an internal compression
function f() that works on fixed-size input blocks (Mi)

M1 M2 Mn

h1 h2 hn-1
IV f f … f h

Ø  Sort of like a Chained Block Cipher

v  Produces a hash value for each fixed-size block based on (1) its
content and (2) hash value for the previous block

v  “Avalanche” effect: 1-bit change in input produces

“catastrophic” and unpredictable changes in output
5

Simple Hash Functions

Ø  Bitwise-XOR

Ø  Not secure, e.g., for English text (ASCII<128) the high-order bit is
almost always zero
Ø  Can be improved by rotating the hash code after each block is
XOR-ed into it
Ø  If message itself is not encrypted, it is easy to modify the message
and append one block that would set the hash code as needed
6
Ø  Another weak hash example: IP Header CRC

3
 Another example
Ø  IPv4 header checksum
Ø  One’s complement of the ones' complement sum of the IP
header's 16-bit words

The Birthday Paradox

v  Example hash function: y=H(x) where: x=person and H() is Bday()
v  y ranges over set Y=[1…365], let n = size of Y, i.e., number of
distinct values in the range of H()
v  How many people do we need to ‘hash’ to have a collision?
v  Or: what is the probability of selecting at random k DISTINCT
numbers from Y?

v  P0=1*(1-1/n)*(1-2/n)*…*(1-(k-1)/n)) == e(k(1-k)/2n)

v  P1=1-P0 ---> probability of at least one collision
v  Set P1 to be at least 0.5 and solve for k
v  k == 1.17 * SQRT(n)
v  k = 22.3 for n=365

So, what’s the point? 8

4
 The Birthday Paradox

m = log(n) = size of H ()
2 m = 2 m/2 trials must
be computationally
infeasible!

How long should a hash be?

Ø Many input messages yield the same hash

v e.g., 1024-bit message, 128-bit hash
v On average, 2896 messages map into one hash
Ø With m-bit hash, it takes about 2m/2 trials
to find a collision (with >=50% probability)
Ø When m=64, it takes 232 trials to find a
collision (doable in very little time)
Ø Today, need at least m=160, requiring
about 280 trials
10

5
 Hash Function Examples
SHA-1 (or MD5 RIPEMD-160
SHA-160) (defunct) (unloved) J
Digest length 160 bits 128 bits 160 bits
Block size 512 bits 512 bits 512 bits

# of steps 80 64 160
(4 rounds of (4 rounds (5 paired
20) of 16) rounds of
16)

Max message 264-1 bits ∞ ∞

size
Other (stronger) variants of SHA are SHA-256 and SHA-512
See: http://en.wikipedia.org/wiki/SHA_hash_functions 11

MD5
Ø  Author: R. Rivest, 1992
Ø  128-bit hash
based on earlier, weaker MD4 (1990)
Ø Collision resistance (B-day attack resistance)
only 64-bit    
Ø  Output size not long enough today  (due to various attacks)

12

6
MD5: Message Digest Version 5

Input message

Output: 128-bit digest

13

Overview of MD5

14

7
 MD5 Padding

Ø  Given original message M, add padding bits “100…” such

that resulting length is 64 bits less than a multiple of
512 bits.
Ø  Append original length in bits to the padded message
Ø  Final message chopped into 512-bit blocks
15

MD5: Padding

1 2 3 4

input Message

512 bit block Padding

Initial Value
MD5 Transformation block by block

Final Output
Output: 128-bit digest
16

8
 MD5 Blocks

512: B1

512:B2
MD5
512: B3
MD5
512: B4
MD5

MD5

Result
17

MD5 Box
512-bit message chunks (16 words)

Initial F: (x ∧ y) ∨ (~x ∧ z)
128-bit vector G: (x ∧ z) ∨ (y ∧~ z)
H: x ⊕ y ⊕ z
I: y ⊕ (x ∧ ~z)

x↵y: x left rotate y bits

128-bit result
18

9
 MD5 Process
Ø As many stages as the number of 512-bit blocks
in the final padded message

Ø Digest: 4 32-bit words: MD=A|B|C|D

Ø Every message block contains 16 32-bit words:

m0|m1|m2…|m15
v Digest MD0 initialized to:
A=01234567,B=89abcdef,C=fedcba98, D=76543210
v Every stage consists of 4 passes over the message
block, each modifying MD; each pass involves
different operation
19

Processing of Block mi - 4 Passes

mi MDi

ABCD=fF(ABCD,mi,T[1..16])
A B C D
ABCD=fG(ABCD,mi,T[17..32])

ABCD=fH(ABCD,mi,T[33..48])

Convention:
ABCD=fI(ABCD,mi,T[49..64])
A – d0 ; B – d1
C – d2 ; B – d3
Ti :diff. constant + + + +
MD i+1 20

10
 Different Passes...

Ø Different functions and constants

Ø Different set of mi-s

Ø Different sets of shifts

21

Functions and Random Numbers

Ø F(x,y,z) == (x∧y)∨(~x ∧ z)
Ø G(x,y,z) == (x ∧ z) ∨(y ∧~ z)
Ø H(x,y,z) == x⊕y⊕ z
Ø I(x,y,z) == y⊕(x ∧ ~z)
Ø Ti = int(232 * abs(sin(i))), 0<i<65

22

11
 Secure Hash Algorithm (SHA)
Ø  SHA-0 was published by NIST in 1993

Ø  Revised in 1995 as SHA-1

v  Input: Up to 264 bits
v  Output: 160 bit digest
v  80-bit collision resistance
Ø  Pad with at least 64 bits to
resist padding attack
v  1000…0 || <message length>

Ø  Processes 512-bit block

v  Initiate 5x32bit MD registers
v  Apply compression function
Ø  4 rounds of 20 steps each
Ø  each round uses different
non-linear function
Ø  registers are shifted and
switched

23

Digest Generation with SHA-1

24

12
 SHA-1 of a 512-Bit Block

25

General Logic

Ø Input message must be < 264 bits

v  not a realistic limitation
Ø Message processed in 512-bit blocks
sequentially
Ø Message digest (hash) is 160 bits
Ø SHA design is similar to MD5, but a lot
stronger

26

13
 Basic Steps

Step1: Padding
Step2: Appending length as 64-bit unsigned
Step3: Initialize MD buffer: 5 32-bit
words: A|B|C|D|E
A = 67452301
B = efcdab89
C = 98badcfe
D = 10325476
E = c3d2e1f0
27

Basic Steps...

Step 4: the 80-step processing of 512-bit

blocks: 4 rounds, 20 steps each
Each step t (0 <= t <= 79):
v Input:
Ø Wt – 32-bit word from the message
Ø Kt – constant
Ø ABCDE: current MD
v Output:
Ø ABCDE: new MD

28

14
 Basic Steps...

Ø Only 4 per-round distinctive additive

constants:
0 <= t <= 19 Kt = 5A827999
20<=t<=39 Kt = 6ED9EBA1
40<=t<=59 Kt = 8F1BBCDC
60<=t<=79 Kt = CA62C1D6

29

Basic Steps – Zooming in

A B C D E

+
ft
+
CLS5
Wt
+

CLS30 Kt
+

A B C D E 30

15
 Basic Logic Functions

Ø Only 3 different functions

Round Function ft(B,C,D)

0 <=t<= 19 (B∧C)∨(~B ∧D)
20<=t<=39 B⊕C⊕D
40<=t<=59 (B∧C)∨(B∧D)∨(C∧D)
60<=t<=79 B⊕C⊕D

31

Twist With Wt’s

Ø Additional mixing used with input

message 512-bit block
W0|W1|…|W15 = m0|m1|m2…|m15
For 15 < t <80:
Wt = Wt-16 ⊕Wt-14 ⊕Wt-8 ⊕Wt-3

Ø XOR is a very efficient operation, but

with multilevel shifting, it produces very
extensive and random mixing!

32

16
 SHA Versus MD5

Ø SHA is a stronger algorithm:

v A birthday attack requires on the order of
280 operations, in contrast to 264 for MD5

Ø SHA has 80 steps and yields a 160-bit

hash (vs. 128) - involves more
computation

33

Summary:
What are hash functions
good for?

34

17
 Message Authentication Using a Hash Function

Use symmetric encryption such as AES or 3-DES

•  Generate H(M) of same size as E() block

•  Use EK(H(M)) as the MAC (instead of, say, DES MAC)

•  Alice sends EK(H(M)) , M

•  Bob receives C,M’ decrypts C with k, hashes result
H(DK(C)) =?= H(M’)

Collision è MAC forgery! 35

Using Hash for Authentication

Ø Alice to Bob: random challenge rA

Ø Bob to Alice: H(KAB||rA)
Ø Bob to Alice: random challenge rB
Ø Alice to Bob: H(KAB||rB)
Ø Only need to compare H() results

36

18
 Using Hash to Compute MAC: integrity
Ø Cannot just compute and append H(m)
Ø Need “Keyed Hash”:
v  Prefix:
Ø  MAC: H(KAB | m), almost works, but…
Ø  Allows concatenation with arbitrary message:
H( KAB | m | m’ )
v  Suffix:
Ø  MAC: H(m | KAB), works better, but what if m’ is
found such that H(m)=H(m’)?
v  HMAC:
Ø  H ( KAB | H (KAB | m) )
37

Hash Function MAC (HMAC)

Ø  Main Idea: Use a MAC derived from any cryptographic
hash function
v Note that hash functions do not use a key, and therefore
cannot serve directly as a MAC
Ø  Motivations for HMAC:
v Cryptographic hash functions execute faster in software than
encryption algorithms such as DES
v No need for the reverseability of encryption
v No US government export restrictions (was important in the
past)
Ø  Status: designated as mandatory for IP security
v Also used in Transport Layer Security (TLS), which will
replace SSL, and in SET

38

19
 HMAC Algorithm
Ø  Compute H1 = H() of the
concatenation of M and K1
Ø  To prevent an “additional
block” attack, compute
again H2= H() of the
concatenation of H1 and K2
Ø  K1 and K2 each use half
the bits of K
Ø  Notation:
v  K+ = K padded with 0’s
v  ipad=00110110 x b/8
v  opad=01011100 x b/8
Ø  Execution:
v  Same as H(M), plus 2
blocks

39

Just for fun…

Using a Hash to Encrypt
Ø (Almost) One-time pad: similar to OFB
v compute bit streams using H(), K, and IV
Ø b1=H(KAB | IV) , …, bi=H(KAB | bi-1), …
Ø c1= p1 ⊕ b1 , … , ci= pi ⊕ bi , …

Ø Or, mix in the plaintext

v similar to cipher feedback mode (CFB)
Ø b1=H(KAB | IV), …, bi=H(KAB | ci-1), …
Ø c1= p1 ⊕ b1 , … , ci= pi ⊕ bi , …

40

20