1. CS 312: Algorithm Analysis Lecture #4: Primality Testing, GCD This work is licensed under a Creative Commons Attribution-Share Alike 3.0 Unported License.Creative Commons Attribution-Share Alike 3.0 Unported License Slides by: Eric Ringger, with contributions from Mike Jones, Eric Mercer, Sean Warnick

2. Announcements  HW #2 Due Now  Homework: Required to show your work  Project #1  Today we’ll work through the rest of the math  Early: this Wednesday, 1/16  Due: Friday, 1/18  Bonus help session on Proof by Induction  Tuesday, 4pm, 1066 TMCB

3. Practice Key points: Represent exponent in binary Break up the problem into factors (one per binary digit) Use the substitution rule

4. Practice Key points: Represent exponent in binary Break up the problem into factors (one per binary digit) Use the substitution rule

5. Objectives  Part 1:  Introduce Fermat’s Little Theorem  Understand and analyze the Fermat primality tester  Part 2:  Discuss GCD and Multiplicative Inverses, modulo N  Prepare to Introduce Public Key Cryptography  This adds up to a lot of ideas!

6. Part 1: Primality Testing

7. Fermat’s Little Theorem p = 3, a = 2 p = 7, a = 4 How do you wish you could use this theorem?

8. Fermat’s Little Theorem p = 3, a = 2 p = 7, a = 4 How do you wish you could use this theorem?

9. Logic Review a  b (a implies b) Which is equivalent to the above statement?  b  a  ~a  ~b  ~b  ~a

10. Logic Review a  b (a implies b) Which is equivalent to the above statement?  b  aThe Converse  ~a  ~bThe Inverse  ~b  ~aThe Contrapositive

11. Contrapositive of Fermat’s Little Theorem

12. First Prime Number Test

13. False Witnesses  If primality(N) returns “possibly prime”, then N might or might not be prime, as the answer indicates  Consider 15:  4 14 mod 15 = 1  but 15 clearly isn’t prime!  4 is called a false witness of 15  Given a non-prime N, we call a number a where a n-1 mod N = 1 a “false witness” (to the claim that N is prime)

14. Relatively Prime  Two numbers, a and N, are relatively prime if their greatest common divisor is 1.  3 and 5?  4 and 8?  4 and 9?  Consider the Carmichael numbers:  They pass the test (i.e., a n-1 mod N = 1) for all a relatively prime to N.

15. False Witnesses  Ignoring Carmichael numbers,  How common are false witnesses?  Lemma: If a n-1 mod n = 1 for some a relatively prime to n, then it must hold for at least half the choices of a < n

16. State of Affairs  Summary:  If n is prime, then a n-1 mod n = 1 for all a < n  If n is not prime, then a n-1 mod n = 1 for at most half the values of a < n  Allows us to put a bound on how often our primality() function is wrong.

17. False Witnesses  Notes:  Less than 3.3% of the possible witnesses for n < 1000 are false.  Consider 651,693,055,693,681.  Have 99.9965% chance of picking a false witness!

18. Correctness  Question #1: Is the “Fermat test” correct?  No  Question #1’: How correct is the Fermat test?  The algorithm is ½-correct with one-sided error.  The algorithm has 0.5 probability of saying “yes N is prime” when N is not prime.  But when the algorithm says “no N is not prime”, then N must not be prime (by contrapositive of Fermat's Little Theorem)

19. Amplification  Repeat the test  Decrease the probability of error: C = Composite; P = Prime  Amplification of stochastic advantage 1 st run2 nd runError? Cn/a PC PP

20. Amplification  Repeat the test  Decrease the probability of error: C = Composite; P = Prime  Amplification of stochastic advantage 1 st run2 nd runError? Cn/a PC PP

21. P(Correct)  k trials gives 1/(2 k ) probability of being incorrect when the answer is "prime“ P(Correct) =

22. Modified Primality Test function primality2(N) Input: Positive integer N Output: yes/no for i = 1 to k do: a = uniform(1..N-1) if (modexp(a, N-1, N) == 1): // possibly prime; do nothing else: return “not prime” return yes

23. 2. Greatest Common Divisor

24. Greatest Common Divisor  Euclid’s rule:  gcd(x, y) = gcd (x mod y, y) = gcd (y, x mod y)  Can compute gcd(x,y) for large x, y by modular reduction until we reach the base case! function Euclid (a,b) Input: Two integers a and b with a  b  0 (n-bit integers) Output: gcd(a,b) if b=0: return a return Euclid(b, a mod b)

25. Example  gcd(25, 11)

26. Example  gcd(25, 11)

27. 3 Questions  1. Is it Correct?  2. How long does it take?  3. Can we do better?

28. Analysis function Euclid (a,b) Input: Two integers a and b with a  b  0 (n-bit integers) Output: gcd(a,b) if b=0: return a return Euclid(b, a mod b)

29. Bezout’s Identity

30. Extended Euclid Algorithm function extended-Euclid (a, b) Input: Two positive integers a & b with a  b  0 (n-bits) Output: Integers x, y, d such that d = gcd(a, b) and ax + by = d if b=0: return (1,0,a) (x’, y’, d) = extended-Euclid(b, a mod b) return (y’, x’ – floor(a/b)y’, d)

31. Example  Note: there’s a great worked example of how to use the extended-Euclid algorithm on Wikipedia here: http://en.wikipedia.org/wiki/Extended_Euclidean _algorithm  And another linked from the reading column on the schedule for today

32. Multiplicative Inverses  In the rationals, what’s the multiplicative inverse of a?  In modular arithmetic (modulo N), what is the multiplicative inverse?

33. Multiplicative Inverses  In the rationals, what’s the multiplicative inverse of a?  In modular arithmetic (modulo N), what is the multiplicative inverse?

34. Finding Multiplicative Inverses (Modulo N)

35. Multiplicative Inverses

36. Next  RSA

37. Assignment  Read and Understand: Section 1.4  HW #3: 1.9, 1.18, 1.20  Finish your project #1 early!  Submit on Wednesday by 5pm