Number Theory

Raj Jain
Washington University in Saint Louis
Saint Louis, MO 63130
[email protected]
Audio/Video recordings of this lecture are available at:
http://www.cse.wustl.edu/~jain/cse571-14/
Washington University in St. Louis CSE571S ©2014 Raj Jain
8-1
 Overview

1. Prime numbers
2. Fermat’s and Euler’s Theorems
3. Testing for primality
4. The Chinese Remainder Theorm
5. Discrete Logarithms
These slides are partly based on Lawrie Brown’s slides supplied with William Stallings’s
book “Cryptography and Network Security: Principles and Practice,” 6th Ed, 2013.
Washington University in St. Louis CSE571S ©2014 Raj Jain
8-2
 Fermat's Little Theorem
 Given a prime number p:
ap-1 = 1 (mod p)
For all integers a≠p
Or
ap = a (mod p)
 Example:
 14 mod 5=1
 24 mod 5=1
 34 mod 5=1
 44 mod 5=1

Washington University in St. Louis CSE571S ©2014 Raj Jain

8-3
 Euler Totient Function ø(n)
 When doing arithmetic modulo n complete set of residues is:
0..n-1
 Reduced set of residues is those residues which are relatively
prime to n, e.g., for n=10,
complete set of residues is {0,1,2,3,4,5,6,7,8,9}
reduced set of residues is {1,3,7,9}
 Number of elements in reduced set of residues is called the
Euler Totient Function ø(n)
 In general need prime factorization, but
 for p (p prime) ø(p)=p-1
 for p.q (p,q prime) ø(p.q)=(p-1)x(q-1)
 Examples: ø(37) = 36
ø(21) = (3–1)x(7–1) = 2x6 = 12
Washington University in St. Louis CSE571S ©2014 Raj Jain
8-4
 Euler's Theorem
 A generalisation of Fermat's Theorem
 aø(n) = 1 (mod n)
 for any a, n where gcd(a,n)=1
 Example:
a=3; n=10; ø(10)=4;
hence 34 = 81 = 1 mod 10
a=2; n=11; ø(11)=10;
hence 210 = 1024 = 1 mod 11
 Also have: aø(n)+1 = a (mod n)

Washington University in St. Louis CSE571S ©2014 Raj Jain

8-5
 Miller Rabin Algorithm for Primality
 A test for large primes based on
Fermat’s Theorem Y
 TEST (n) is: aq mod n = 1?
1. Find integers k, q, k > 0, q odd, Y
so that (n–1)=2kq
aq mod n = n-1?
2. Select a random integer a,
1<a<n–1

Inconclusive
3. if aq mod n = 1 then return Y
(aq)2 mod n = n-1?
(“inconclusive");
4. for j = 0 to k – 1 do
j Y
5. if (a2 q mod n = n-1) (aq)4 mod n = n-1?
then return(“inconclusive")
6. return (“composite")
 If inconclusive after t tests with
different a’s: k−1
Y
Probability (n is Prime after t tests)
(aq )2 mod n = n-1?
= 1- 4-t
 E.g., for t=10 this probability is > Composite
0.99999
Washington University in St. Louis CSE571S ©2014 Raj Jain
8-6
 Miller Rabin Algorithm Example
 Test 29 for primality
 29-1 = 28 = 227 = 2kq  k=2, q=7

 Let a = 10

 10 mod 29 = 17
7

 17 mod 29 = 28  Inconclusive
2

 Test 221 for primality

 221-1=220=22 55

 Let a=5

 5
55 mod 221 =112

 112 mod 221 =168  Composite

2

Washington University in St. Louis CSE571S ©2014 Raj Jain

8-7
 Prime Distribution
 Prime numbers: 1 2 3 5 7 11 13 17 19 23 29 31
 Prime number theorem states that primes occur roughly every
(ln n) integers
 But can immediately ignore even numbers
 So in practice need only test 0.5 ln(n) numbers of size n to
locate a prime
 Note this is only the “average”

 Sometimes primes are close together

 Other times are quite far apart

Washington University in St. Louis CSE571S ©2014 Raj Jain

8-8
 Chinese Remainder Theorem
 If working modulo a product of numbers
 E.g., mod M = m1m2..mk
 Chinese Remainder theorem lets us work in each moduli mi
separately
 Since computational cost is proportional
à to size, this is faster
!
Xk · ¸ −1
M M
A mod M = (A mod mi ) mod mi
mi mi
i=1
35-1 =x mod 3
 Example: 452 mod 105 35x=1 mod 3  x=2
= (452 mod 3)(105/3){(105/3)-1 mod 3} 21x=1 mod 5  x=1
+(452 mod 5)(105/5){(105/5)-1 mod 5}
+(452 mod 7)(105/7){(105/7)-1 mod 7} 15x=1 mod 7  x=1
= 235(35-1 mod 3) +2x21(21-1 mod 5) +415(15-1 mod 7)
= 2352 +2211 +4151
= (140+42+60) mod 105 = 242 mod 105 = 32
Washington University in St. Louis CSE571S ©2014 Raj Jain
8-9
 Chinese Remainder Theorem
 Alternately, the solution to the following equations:
x = a1 mod m1
x = a2 mod m2
x = ak mod mk
where m1, m2, ... , mk are relatively prime is found as follows:
M = m1 m2 ... Mk then

k
÷ ¸−1 !
X M M
x= ai mod mi
mi mi
i=1

Washington University in St. Louis CSE571S ©2014 Raj Jain

8-10
 Chinese Remainder Theorem Example
 For a parade, marchers are arranged in columns of seven, but one person is
left out. In columns of eight, two people are left out. With columns of nine,
three people are left out. How many marchers are there?
x = 1 mod 7
x = 2 mod 8
x = 3 mod 9
N = 7³ × 8 × 9 = 504
504
£ 504 ¤−1 504
£ 504 ¤−1
x = 1× 7 × 7 7 +2× 8 × 8 8
£ ¤ ´
−1
+3 × 504
9 × 504
9 9 mod 7 × 8 × 9
= (1 × 72 × (72−1 mod 7) + 2 × 63 × (63−1 mod 8)
+3 × 56 × (56−1 mod 9)) mod 504
= (1 × 72 × 4 + 2 × 63 × 7 + 3 × 56 × 5) mod 504
= (288 + 882 + 840) mod 504
= 2010 mod 504
= 498
Ref: http://demonstrations.wolfram.com/ChineseRemainderTheorem/
Washington University in St. Louis CSE571S ©2014 Raj Jain
8-11
 Primitive Roots
 From Euler’s theorem have aø(n)mod n=1
 Consider am = 1 (mod n), GCD(a,n)=1
 For some a’s, m can smaller than ø(n)

 If the smallest m is ø(n) then a is called a primitive root

 If n is prime, then successive powers of a "generate" the group
mod n
 These are useful but relatively hard to find

Washington University in St. Louis CSE571S ©2014 Raj Jain

8-12
 Powers mod 19

 2, 3, 10, 13, 14, 15 are primitive roots of 19

Washington University in St. Louis CSE571S ©2014 Raj Jain
8-13
 Discrete Logarithms
 The inverse problem to exponentiation is to find the discrete
logarithm of a number modulo p
 That is to find i such that b = ai (mod p)
 This is written as i = dloga b (mod p)
 If a is a primitive root then it always exists, otherwise it may
not, e.g.,
x = log3 4 mod 13 has no answer
x = log2 3 mod 13 = 4 by trying successive powers
 While exponentiation is relatively easy, finding discrete
logarithms is generally a hard problem

Washington University in St. Louis CSE571S ©2014 Raj Jain

8-14
 Discrete Logarithms mod 19

Washington University in St. Louis CSE571S ©2014 Raj Jain

8-15
 Summary

1. Fermat’s little theorem: ap-1=1 mod p

2. Euler’s Totient Function ø(p) = # of a<p relative prime to p
3. Euler’s Theorem: aø(p) =1 mod p
Primality Testing: n-1=2kq, aq=1, a2q=n-1, …, (aq )2 =n-1
k−1
4.
5. Chinese Remainder Theorem: x=ai mod mi, i=1,…,k, then you
can calculate x by computing inverse of Mi mod mi
6. Primitive Roots: Minimum m such that am=1 mod p is m=p-1
7. Discrete Logarithms: ai=b mod p  i=dlogb,p(a)
Washington University in St. Louis CSE571S ©2014 Raj Jain
8-16
 Homework 8
a. Use Fermat’s theorem to find a number x between 0
and 22, such that x111 is congruent to 8 modulo 23.
Do not use bruteforce searching.
b. Use Miller Rabin test to test 19 for primality
c. X = 2 mod 3 = 3 mod 5 = 5 mod 7, what is x?
d. Find all primitive roots of 11
e. Find discrete log of 17 base 2 mod 29

Washington University in St. Louis CSE571S ©2014 Raj Jain

8-17