CSE 311: Foundations of Computing. Lecture 12: Two s Complement, Primes, GCD

Transcription

1 CSE 311: Foundations of Computing Lecture 12: Two s Complement, Primes, GCD

2 n-bit Unsigned Integer Representation Represent integer as sum of powers of 2: If 2 where each {0,1} then representation is b n-1...b 2 b 1 b 0 99 = = For n = 8: 99: :

3 Sign-Magnitude Integer Representation -bit signed integers Suppose that 2 < < 2 First bit as the sign, 1 bits for the value 99 = = For n = 8: 99: : Any problems with this representation?

4 Two s Complement Representation bit signed integers, first bit will still be the sign bit Suppose that 0 < 2, is represented by the binary representation of Suppose that 0 < 2, is represented by the binary representation of 2 Key property: Twos complement representation of any number is equivalent to so arithmetic works 99 = = For n = 8: 99: :

5 Sign-Magnitude vs. Two s Complement Sign-bit Two s complement

6 Two s Complement Representation For 0 < 2, is represented by the binary representation of 2 That is, the two s complement representation of any number has the same value as modulo 2. To compute this: Flip the bits of then add 1: All 1 s string is 2 1, so Flip the bits of replace by 2 1 Then add 1 to get 2

7 Basic Applications of mod Hashing Pseudo random number generation Simple cipher

8 Hashing Scenario: Map a small number of data values from a large domain 0,1,, into a small set of locations 0,1,, 1 so one can quickly check if some value is present hash = mod ' for 'a prime close to or hash = ()+) mod ' Depends on all of the bits of the data helps avoid collisions due to similar values need to manage them if they occur

9 Pseudo-Random Number Generation Linear Congruential method, = ) +- mod. Choose random,), -,.and produce a long sequence of s

10 Simple Ciphers Caesar cipher, A = 1, B = 2,... HELLO WORLD Shift cipher f(p) = (p + k) mod 26 f -1 (p) = (p k) mod 26 More general f(p) = (ap + b) mod 26

11 Primality An integer p greater than 1 is called prime if the only positive factors of p are 1 and p. A positive integer that is greater than 1 and is not prime is called composite.

12 Fundamental Theorem of Arithmetic Every positive integer greater than 1 has a unique prime factorization 48 = = ,523 = 45, ,950 = ,234,567,890 = ,607 3,803

13 Euclid s Theorem There are an infinite number of primes. Proof by contradiction: Suppose that there are only a finite number of primes and call the full list ',' /,,'.

14 Euclid s Theorem There are an infinite number of primes. Proof by contradiction: Suppose that there are only a finite number of primes and call the full list ',' /,,'. Define the number 0 =' ' / ' 2 ' and let 4 =0+1.

15 Euclid s Theorem There are an infinite number of primes. Proof by contradiction: Suppose that there are only a finite number of primes and call the full list ',' /,,'. Define the number 0 =' ' / ' 2 ' and let 4 =0+1. Case 1: 4 is prime: Then 4is a prime different from all of ',' /,,' since it is bigger than all of them. Case 2: 4 > 1is not prime: Then 4has some prime factor '(which must be in the list). Therefore ' 0 and ' 4so ' 4 0 which means that ' 1. Both cases are contradictions so the assumption is false.

16 Famous Algorithmic Problems Primality Testing Given an integer, determine if is prime Factoring Given an integer, determine the prime factorization of

17 Factoring Factor the following 232 digit number [RSA768]:

18

19 Greatest Common Divisor GCD(a, b): Largest integer 8such that 8 )and 8 GCD(100, 125) = GCD(17, 49) = GCD(11, 66) = GCD(13, 0) = GCD(180, 252) =

20 GCD and Factoring a = = 46,200 b = = 204,750 GCD(a, b) = 2 min(3,1) 3 min(1,2) 5 min(2,3) 7 min(1,1) 11 min(1,0) 13 min(0,1) Factoring is expensive! Can we compute GCD(a,b) without factoring?

21 Useful GCD Fact If a and b are positive integers, then gcd(a,b)=gcd(b, a modb)

22 Useful GCD Fact If a and b are positive integers, then gcd(a,b) = gcd(b, a modb) Proof: By definition of mod, ) = :+() mod ) for some integer : = ) div. Let 8 = gcd (),). Then 8 ) and 8 so ) =?8 and for some integers? Therefore () mod ) = ) : =?8 :@8 = (? :@)8. So, 8 () mod ) and since 8 we must have 8 gcd (,) mod ). Now, let A = gcd (,) mod ). Then A and A () mod ) so =.A and () mod )=A for some integers. and. Therefore ) = :+().B8 ) = :.A+ A = (:.+)A. So, A ) and since A we must have A gcd (),). It follows that gcd (),)=gcd (,) mod ).

23 Another simple GCD fact If a is a positive integer, gcd(a,0)=a.

24 Euclid s Algorithm gcd(a, b) = gcd(b, a mod b) int gcd(int a, int b){ /* a >= b, b > 0 */ if (b == 0) { return a; } else { return gcd(b, a % b); } Example: GCD(660, 126)

25 Euclid s Algorithm Repeatedly use gcd ), = gcd,) mod to reduce numbers until you get gcd (C,0) =C. gcd(660,126) =

26 Euclid s Algorithm Repeatedly use gcd ), = gcd,) mod to reduce numbers until you get gcd (C,0) =C. gcd(660,126) = gcd(126, 660 mod 126) = gcd(126, 30) = gcd(30, 126 mod 30) = gcd(30, 6) = gcd(6, 30 mod 6) = gcd(6, 0) = 6

27 Bézout s theorem If a and b are positive integers, then there exist integers s and t such that gcd(a,b) = sa + tb.

28 Extended Euclidean algorithm Can use Euclid s Algorithm to find D,Esuch that gcd ), =D)+E

29 Extended Euclidean algorithm Can use Euclid s Algorithm to find D,Esuch that gcd ), =D)+E Step 1 (Compute GCD & Keep Intermediary Information): a b b a mod b = r b r a = q * b + r gcd(35, 27) = gcd(27, 35 mod 27) = gcd(27, 8) (35 = 1 * )

30 Extended Euclidean algorithm Can use Euclid s Algorithm to find D,Esuch that gcd ), =D)+E Step 1 (Compute GCD & Keep Intermediary Information): a b b a mod b = r b r a = q * b + r gcd(35, 27) = gcd(27, 35 mod 27) = gcd(27, 8) (35 = 1 * ) = gcd(8, 27 mod 8) = gcd(8, 3) (27 = 3 * 8 + 3) = gcd(3, 8 mod 3) = gcd(3, 2) (8 = 2 * 3 + 2) = gcd(2, 3 mod 2) = gcd(2, 1) (3 = 1 * 2 + 1) = gcd(1, 2 mod 1) = gcd(1, 0)

31 Extended Euclidean algorithm Can use Euclid s Algorithm to find D,Esuch that gcd ), =D)+E Step 2 (Solve the equations for r): a = q * b + r 35 = 1 * = 3 * = 2 * = 1 * r = a -- q * b 8 = 35 1 * 27

32 Extended Euclidean algorithm Can use Euclid s Algorithm to find D,Esuch that gcd ), =D)+E Step 2 (Solve the equations for r): a = q * b + r 35 = 1 * = 3 * = 2 * = 1 * = 2 * r = a -- q * b 8 = 35 1 * 27 3 = 27 3 * 8 2 = 8 2 * 3 1 = 3 1 * 2

33 Extended Euclidean algorithm Can use Euclid s Algorithm to find D,Esuch that gcd ), =D)+E Step 3 (Backward Substitute Equations): 8 = 35 1 * 27 3 = 27 3 * 8 2 = 8 2 * 3 1 = 3 1 * 2 1 = 3 1*(8 2* 3) = * 3 = ( 1)* * 3 Plug in the defof 2 Re-arrange into 3 s and 8 s

34 Extended Euclidean algorithm Can use Euclid s Algorithm to find D,Esuch that gcd ), =D)+E Step 3 (Backward Substitute Equations): 8 = 35 1 * 27 3 = 27 3 * 8 2 = 8 2 * 3 1 = 3 1 * 2 Re-arrange into 27 s and 35 s 1 = 3 1*(8 2* 3) = * 3 = ( 1)* * 3 Plug in the defof 2 Re-arrange into 3 s and 8 s Plug in the defof 3 = ( 1)* *(27 3* 8) = ( 1)* * 27 + ( 9)* 8 = 3 * 27 + ( 10)* 8 Re-arrange into 8 s and 27 s = 3 * 27 + ( 10) * (35 1 * 27) = 3 * 27 + ( 10) * * 27 = 13 * 27 + ( 10) * 35

35 multiplicative inverse mod. Suppose GCD ),. = 1 By Bézout s Theorem, there exist integers D and E such that D)+E. = 1. D mod.is the multiplicative inverse of ): 1 = D)+E. mod. = D) mod.

36 Example Solve: 7 1 (mod 26)

37 Example Solve: 7 1 (mod 26) gcd (26,7) = gcd (7,5) = gcd (5,2) = gcd (2,1) = 1 26 = = = = = = = 5 2 (7 5 1) = ( 7) = (26 7 3) = Multiplicative inverse of 7 mod 26 Now ( 11) mod 26 =15. So, =15+26? for? Z.

38 Example of a more general equation Now solve: 7 3 (mod 26) We already computed that 15is the multiplicative inverse of 7modulo 26: That is, (mod 26) By the multiplicative property of mod we have (mod 26) So any 15 3 mod 26 is a solution. That is, = 19+26?for any integer?is a solution.

39 Math mod a prime is especially nice gcd (),.)=1if.is prime and 0 < ) <. so can always solve these equations mod a prime X mod 7