50
Chap. 4: Finite Fields Jen-Chang Liu, 2004 Adapted from lecture slides by Lawrie Brown

Chap. 4: Finite Fields

Embed Size (px)

DESCRIPTION

Chap. 4: Finite Fields. Jen-Chang Liu, 2004 Adapted from l ecture slides by Lawrie Brown. - PowerPoint PPT Presentation

Citation preview

Page 1: Chap. 4: Finite Fields

Chap. 4: Finite Fields

Jen-Chang Liu, 2004

Adapted from lecture slides by Lawrie Brown

Page 2: Chap. 4: Finite Fields

The next morning at daybreak, Star flew indoors, seemingly keen for a lesson. I said, "Tap eight." She did a brilliant exhibition, first tapping it in 4, 4, then giving me a hasty glance and doing it in 2, 2, 2, 2, before coming for her nut. It is astonishing that Star learned to count up to 8 with no difficulty, and of her own accord discovered that each number could be given with various different divisions, this leaving no doubt that she was consciously thinking each number. In fact, she did mental arithmetic, although unable, like humans, to name the numbers. But she learned to recognize their spoken names almost immediately and was able to remember the sounds of the names. Star is unique as a wild bird, who of her own free will pursued the science of numbers with keen interest and astonishing intelligence.

— Living with Birds, Len Howard

Page 3: Chap. 4: Finite Fields

Len Howard British naturalist, who studied bird

behaviors Publications:

Birds as Individuals (1953) Living with Birds (1956).

Page 4: Chap. 4: Finite Fields

Introduction Finite fields: increasing importance in

cryptography AES, Elliptic Curve, IDEA, Public Key

Goal: mathematics for GF(2n)

01100…0100

Plaintext: 2n

11010…0111

Ciphertext: 2n

T(Plaintext, Key)

1. Arbitrary mapping => large keys2. Mathematical form, ex. Hill cipher

Page 5: Chap. 4: Finite Fields

Outline Group, rings, and fields Modular arithmetic Finite fields of the form GF(p) Euclid’s algorithm (find GCD) Polynomial arithmetic Finite fields of the form GF(2n)

Page 6: Chap. 4: Finite Fields

Motivation 在封閉且有限的數集合中,可操作的運算種

類? (+, -, x, /)

Group

Ring

Field

+,-

+,-,x

+,-,x,/

Page 7: Chap. 4: Finite Fields

Group Group: {G, •}

G: a set of elements •: binary operation to each pair (a,b) in G

obeys: closure: a•b is also in G associative law: (a•b)•c = a•(b•c) has identity e: e•a = a•e = a has inverses a-1: a•a-1 = e

if commutative a•b = b•a then forms an abelian group

Page 8: Chap. 4: Finite Fields

Example: group

1. Set of integers under addition=> Abelian group

2. N symbols : Nn={1, 2, …, n} Set Sn: all permutations over Nn

Binary operation •: permutation according to the first element

=> Group

{{1, 2, …, n}, {2, 1 …, n}, …, {n, n-1, …, 1}}

{3,2,1}•{1,3,2}={2,3,1}

Page 9: Chap. 4: Finite Fields

Cyclic Group define exponentiation as repeated

application of operator example: a3 = a•a•a

and let identity be: e=a0

a group is cyclic if every element is a power of some fixed element i.e. b = ak for some a and every b in group a is said to be a generator of the group

EX. integers with addition: 1 as the generator

Page 10: Chap. 4: Finite Fields

Ring Ring: {R, +, x}

R: a set of “numbers” +, x: two operations (addition and

multiplication) Ring obeys

{R,+}: an abelian group multiplication:

has closure is associative: ax(bxc)=(axb)xc distributive over addition: ax(b+c) = axb + axc

if multiplication operation is commutative, it forms a commutative ring axb = bxa

Page 11: Chap. 4: Finite Fields

Example: Ring The set of even integers (pos, neg, and

0) Commutative ring

Page 12: Chap. 4: Finite Fields

Field Field: {F, +, x}

F: a set of numbers +, x: two operations (addition and

multiplication) Field obeys:

Ring {F, +}: abelian group for addition {F\0, x}: abelian group for multiplication

(ignoring 0) i.e. with multiplication inverse => division is

possible {Z, +, x} ?

Page 13: Chap. 4: Finite Fields

Field for n-bit block?

01100…0100

Plaintext: 2n

11010…0111

Ciphertext: 2n

+: additionx: multiplication

?

Problems: 1. the set of plaintext (and ciphertext) finite 2. how to define +,-,x,/ operations

Page 14: Chap. 4: Finite Fields

Outline Group, rings, and Fields Modular arithmetic Finite fields of the form GF(p) Euclid’s algorithm (find GCD) Polynomial arithmetic Finite fields of the form GF(2n)

Page 15: Chap. 4: Finite Fields

Modulo operator modulo operator: a mod n to be the

remainder (>0) when a is divided by n

) mod (/ nannaa , a and n are integers

Page 16: Chap. 4: Finite Fields

Modulo operator (cont.) Integers a and b are congruence

modulo n a ≡ b mod n when divided by n, a & b have same

remainder eg. 100 = 34 mod 11

b is called the residue of a: b= a mod n integers can always write: a = qn + b usually have 0 <= b <= n-1 -12 mod 7 ≡ -5 mod 7 ≡ 2 mod 7 ≡ 9 mod 7

Page 17: Chap. 4: Finite Fields

Integers Modulo 7 Example

... -21 -20 -19 -18 -17 -16 -15 -14 -13 -12 -11 -10 -9 -8 -7 -6 -5 -4 -3 -2 -1 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 26 27 28 29 30 31 32 33 34 ...

congruence

Z7: residue classmodulo n

Page 18: Chap. 4: Finite Fields

(mod n) maps all integers into the set Zn={0, 1, 2, …, n-1} Can we perform arithmetic (+,-,x,/)

with this set?

Page 19: Chap. 4: Finite Fields

Modular Arithmetic We can do modular arithmetic with any

set of integers: Zn={0, 1, … , n-1} Under normal arithmetic (+,-,x) Properties:

1. [(a mod n)+(b mod n)] mod n = (a+b) mod n

2. [(a mod n)- (b mod n)] mod n = (a-b) mod n3. [(a mod n)x (b mod n)] mod n = (axb) mod

n

Page 20: Chap. 4: Finite Fields

Modulo 8(=23) Example

FindAdditiveinverse

=> group

Page 21: Chap. 4: Finite Fields

Modulo 8(=23) Example

Not all haveMulti. Inverse Not a field

Does notproduce allelements inZ8

Page 22: Chap. 4: Finite Fields

Modular Arithmetic (cont.) Peculiarities compared with ordinary

arith. if (a+b)≡(a+c) mod n then b≡c mod n Existence of (-a):

if (ab)≡(ac) mod n then b≡c mod n ? Existence of (a-1) ?

((-a)+a+b)≡((-a)+a+c) mod n

((a-1) ab)≡((a-1) ac) mod n

Not always true, eg. n=8Multiplicative inverse exists iff. a is relatively prime to n

Page 23: Chap. 4: Finite Fields

Question Do we have a finite field within Zn ?

Page 24: Chap. 4: Finite Fields

Modulo 7 example: addition

Yes foradditive inverse

Page 25: Chap. 4: Finite Fields

Modulo 7 Example: multiply

Yes formulti. inverse

Page 26: Chap. 4: Finite Fields

Question Do we have a finite field within Zn ? Galois field

GF(p): ZP ,the modulo p is a prime number Under ordinary arithmetic

GF(pn): , p is a prime number 23 does not form a field under normal

arithmetic Under polynomial arithmetic

npZ

Page 27: Chap. 4: Finite Fields

Outline Group, rings, and Fields Modular arithmetic Finite fields of the form GF(p) Euclid’s algorithm (find GCD) Polynomial arithmetic Finite fields of the form GF(2n)

Page 28: Chap. 4: Finite Fields

Galois Fields: GF(p) Theorem: Zn={0,1,…,n-1} is a commutative

ring. Any integer aZn in has a multiplicative inverse iff a is relatively prime to n

In other words, If n is a prime number, then all nonzero

integers in Zn are relatively prime to n

=> all nonzero integers in Zn have multiplicative

inverses=> Zn is a finite field=> GF(p), p is a prime

Page 29: Chap. 4: Finite Fields

Galois Fields: GF(p) GF(p) is the set of integers {0,1, … , p-

1} with arithmetic operations modulo prime p

these form a finite field since have multiplicative inverses

hence arithmetic is “well-behaved” and can do addition, subtraction, multiplication, and division without leaving the field GF(p)

Page 30: Chap. 4: Finite Fields

Example: GF(2)

+ 0 1

01

0 11 0

x 0 1

01

0 00 1

w -w w-1

01

0 x1 1

XOR AND

Page 31: Chap. 4: Finite Fields

Q: How to find the multiplicative inverse?

For GF(7), it is easy to build a table

For GF(1759), how to find the multiplicative inverse of 550?

Page 32: Chap. 4: Finite Fields

Outline Group, rings, and Fields Modular arithmetic Finite fields of the form GF(p) Euclid’s algorithm (find GCD) and

extended Euclid’s algorithm (find multiplicative inverse)

Polynomial arithmetic Finite fields of the form GF(2n)

Page 33: Chap. 4: Finite Fields

Greatest Common Divisor (GCD)

a common problem in number theory GCD (a,b) of a and b is the largest

number that divides evenly into both a and b eg. GCD(60,24) = 12

GCD(.,.)=1: no common factors (except 1) and hence numbers are relatively prime eg. GCD(8,15) = 1 hence 8 & 15 are relatively prime

Page 34: Chap. 4: Finite Fields

Euclid's GCD Algorithm uses theorem that: (a>b>0)

GCD(a,b) = GCD(b, a mod b)

To prove: The set of common divisors of a and b = The set of common divisors of a and (a mod b)

Page 35: Chap. 4: Finite Fields

Example GCD(1970,1066)1970 = 1 x 1066 + 904 gcd(1066, 904)1066 = 1 x 904 + 162 gcd(904, 162)904 = 5 x 162 + 94 gcd(162, 94)162 = 1 x 94 + 68 gcd(94, 68)94 = 1 x 68 + 26 gcd(68, 26)68 = 2 x 26 + 16 gcd(26, 16)26 = 1 x 16 + 10 gcd(16, 10)16 = 1 x 10 + 6 gcd(10, 6)10 = 1 x 6 + 4 gcd(6, 4)6 = 1 x 4 + 2 gcd(4, 2)4 = 2 x 2 + 0 gcd(2, 0)

Page 36: Chap. 4: Finite Fields

Finding Inverses can extend Euclid’s algorithm:

EXTENDED EUCLID(m, b)1.(A1, A2, A3)=(1, 0, m);

(B1, B2, B3)=(0, 1, b)2. if B3 = 0

return A3 = gcd(m, b); no inverse3. if B3 = 1

return B3 = gcd(m, b); B2 = b–1 mod m4. Q = A3 div B35. (T1, T2, T3)=(A1 – Q B1, A2 – Q B2, A3 – Q B3)6. (A1, A2, A3)=(B1, B2, B3)7. (B1, B2, B3)=(T1, T2, T3)8. goto 2

Page 37: Chap. 4: Finite Fields

Inverse of 550 in GF(1759)

( 商 )

4 1 0

B2=A2-Q*B2=0-3*1=-3

inverse

Page 38: Chap. 4: Finite Fields

Recall: inverse matrix in Hill cipher

Find inverse matrix for

Note that 26 is not a prime, not every element in {0,1,2,…,25} has multiplicative inverse

Inverse formula:

75

49K

921

227

43

1

95

47

4579

11K

Page 39: Chap. 4: Finite Fields

Inverse of Hill cipher

Q A3 B3 26 171 17 91 9 81 8 18 1 0

A2 B2 0 1 1 -1-1 2 2 -3

2515

125

207483

506161

921

22723

921

227

17

11K

Multiplicative inverse

Page 40: Chap. 4: Finite Fields

Outline Group, rings, and Fields Modular arithmetic Finite fields of the form GF(p) Euclid’s algorithm (find GCD) Polynomial arithmetic Finite fields of the form GF(2n)

Page 41: Chap. 4: Finite Fields

Motivation for GF(2n) For a 8-bit block

Z256 ={0,1,…,255} is not a field => no division

The largest prime < 256 is 251 Z251 ={0,1,…,250} is a field => 251,

…,255 are wasted Is that possible to find a field for Z256 ?

Yes. Define new arithmetic operations for Z256

Page 42: Chap. 4: Finite Fields

Mapping from GF(2n) to polynomials

Ex. 8-bit block

10010111

01234567 11101001 xxxxxxxx =>

• To build the field for {0,1,2,…,2n-1}=> Require modular polynomial arithmetic

Page 43: Chap. 4: Finite Fields

Polynomial Arithmetic Polynomials

3 polynomial arithmetic available: ordinary polynomial arithmetic Polynomial with coefficients in Zp; arithmetic

on the coefficients is performed modulo p Polynomial with coefficients in Zp, and the

polynomials are defined modulo a polynomial m(x); arithmetic on the coefficients is performed modulo p

Page 44: Chap. 4: Finite Fields

1. Ordinary Polynomial Arithmetic

Add(+) or subtract(-) corresponding coefficients

Multiply(x) all terms by each other eg. 1101, 0111

f(x) = x3 + x2 + 1 and g(x) = x2 + x + 1f(x) + g(x) = x3 + 2x2 + x + 2f(x) – g(x) = x3 - x f(x) x g(x) = x5+2x4+2x3+2x2+x+1

GF(24) ?1. Not in {0,1} 2. degree>3

Page 45: Chap. 4: Finite Fields

2. Polynomial Arithmetic with Modulo Coefficients

Polynomial coefficients are in a field F={p,+,x}

We are most interested in coefficients in GF(2) eg. 1101, 0111 let f(x) = x3 + x2 + 1 and g(x) = x2 + x + 1f(x) + g(x) = x3 + x f(x) x g(x) = x5 + x + 1GF(24) ?

degree > 3

Page 46: Chap. 4: Finite Fields

3. Modular Polynomial Arithmetic for GF(2n)

Polynomial coefficients are in a field F={p,+,x} Ex. coefficients are in GF(2)

If multiplication results in poly. of degree > n-1 Reduce it by modulo some irreducible poly.

m(x) f(x) = q(x) m(x) + r(x)=> r(x) = f(x) mod m(x)

if m(x) has no divisors other than itself & 1 say it is irreducible (or prime) polynomial

Page 47: Chap. 4: Finite Fields

GF(23) =m(x)

Page 48: Chap. 4: Finite Fields

Modular Polynomial Arithmetic

Computation in field GF(2n) polynomials with coefficients modulo 2 whose degree is less than n hence must reduce modulo an irreducible

poly of degree n (for multiplication only) Can always find an inverse

Extended Euclid’s Inverse algorithm to find

Page 49: Chap. 4: Finite Fields

Computational Considerations

since coefficients are 0 or 1, can represent any such polynomial as a bit string

addition becomes XOR of these bit strings

multiplication is shift & XOR modulo reduction done by repeatedly

substituting highest power with remainder of irreducible poly (also shift & XOR) – see textbook for details

Page 50: Chap. 4: Finite Fields

Summary

01100…0100

Plaintext: 2n

11010…0111

Ciphertext: 2n

Mathematicswithin GF(2n)