Euclid's algorithm is a way to find the GCD of any Euclidian domain, in this case, the natural numbers.
Euclid's Algorithm: Let a and b be two positive integers with a <= b. Let d be their greatest common divisor(GCD). Then there are two lists of positive integers q(i) and r(i) such that
b > r(1) > r(2) > r(3) ... > r(k-1) > r(k) > r(k+1) = 0 and
a = bq(1) + r(1)
b = r(1)q(2) + r(2)
r(1) = r(2)q(3) + r(3)
r(2) = r(3)q(2) + r(4)
.
.
.
r(k-2) = r(k-1)q(k) + r(k)
r(k-1) = r(k)q(k+1) (that is, r(k+1) = 0 )
and therefore d = r(k)
Proof
Since a < = b then there exists a q(1) and r(1) such that a = bq(1) + r(1) where 0 < = r(1) < b , and therefore, since r(1) < b there exists a q(2) and r(2) such that b=r(1)q(2) + r(2) and so one for r(2)< r(1). If we keep going down the line of positive integers we reach a point were r(k+1)=0 and so it is evident that:
b > r(1) > r(2) > r(3) ... > r(k-1) > r(k) > r(k+1) = 0
Now, to show that d (the GCD) is equal to r(k) we have to show that r(k) < = d and r(k) > = d.
First we show that r(k) < = d.
We first notice that r(k) divides r(k-1) (because r(k-1) = r(k)q(k+1)), we also notice that r(k) divides r(k-2) because r(k-2) = r(k-1)q(k) + r(k), and so r(k) divides both r(k) and r(k-1) by the proof we did two days ago. (When another proof is used in a proof, it is called a lemma). Now, if we keep going up the line, we see that every number has a component of the previous integer, and so r(k) is a common divisor for all the numbers including a and b. Since d is the GCD for a and b, then r(k) has to be < = d by the proof we did yesterday.
Now we show that r(k) > = d.
We note that d divides r(1) since r(1) = a - bq(1), and since d divides r(1) then by the proof we did two days ago, d also divides r(2) since r(2) = b - r(1)q(2), continuing this pattern down the line we see that d divides r(k-1) and r(k) and therefore r(k) > = d.
Now we can say that d= r(k) which is equal to 1, and 1 is the GCD of all positive integers, or the natural numbers.
Showing posts with label eulcid. Show all posts
Showing posts with label eulcid. Show all posts
Monday, November 24, 2008
Subscribe to:
Posts (Atom)
