MedVision ad

n roots of unity (1 Viewer)

OLDMAN

Member
Joined
Feb 20, 2003
Messages
251
Location
Mudgee
Gender
Undisclosed
HSC
N/A
Here's a retake of an otherwise useful question.

If w, w^2, ...,w^n are the n complex roots of 1, with w being of smallest positive argument, what is the condition that need to be satisfied for W=w^k, so that W, W^2, W^3,...,W^n are also the n roots of unity; and why?
 

OLDMAN

Member
Joined
Feb 20, 2003
Messages
251
Location
Mudgee
Gender
Undisclosed
HSC
N/A
aw c'mon give it a break will you. Lets help the students.
 

Archman

Member
Joined
Jul 29, 2003
Messages
337
Gender
Undisclosed
HSC
N/A
gcd(k,n)=1
gcd means greatest common divisor
 

Grey Council

Legend
Joined
Oct 14, 2003
Messages
1,426
Gender
Male
HSC
2004
I dunno why, but Archman is doing the HSC this year...

wtf, how do we even stand a chance against him. :confused:
 

OLDMAN

Member
Joined
Feb 20, 2003
Messages
251
Location
Mudgee
Gender
Undisclosed
HSC
N/A
____________________________________________________
Grey Council: I dunno why, but Archman is doing the HSC this year...

wtf, how do we even stand a chance against him.
____________________________________________________

Don't worry there's at least 6 mths to go, to build up your strength to tackle question 8 types. Invariably these questions rely on a series of quite basic steps strung together.

___________________________________________________
If w, w^2, ...,w^n are the n complex roots of 1, with w being of smallest positive argument, what is the condition that need to be satisfied for W=w^k, so that W, W^2, W^3,...,W^n are also the n roots of unity; and why?
___________________________________________________

Here's why gcd(k,n)=1

1) By the factor theorem, you can't have more than n roots of z^n=1.
2) Easy to show that each W^j is also a root.
3) Must prove that for n>=j1>j2>=1 W^j1 not.=W^j2

Assume false, ie. W^j1 =W^j2
then cis(j1*k2pi/n)=cis(j2*k2pi/n)
hence j1*k2pi/n - j2*k2pi/n = m2pi where m is an integer
therefore k(j1-j2) = mn
thus n divides k(j1-j2)
but gcd(k,n)=1 hence n divides (j1-j2) * contradiction*.

Therefore W, W^2, W^3,...,W^n are the same n roots of unity.

Now this should end the nth root of unity story. I say this tongue in cheek with apologies to Francis Fukuyama's seminal work End of History when the Berlin Wall was torn down: history doesn't end, it moves on. No doubt the examiners will tell and retell the story starting with the phrase "Question 8"

Really tried to stir the discussion to above, but unfortunately it became a brawl.
 
Last edited:

enter~space~cap

{Enter-Space-Capsule}
Joined
Feb 19, 2003
Messages
153
hmmphhh.....i dont get it one bit:S:S:S

can someone who understand this please draw it on a paper and upload it, it sounds interesting.
 

OLDMAN

Member
Joined
Feb 20, 2003
Messages
251
Location
Mudgee
Gender
Undisclosed
HSC
N/A
_________________________________________________
then w^(j1*k2pi/n)=w^(j2*k2pi/n)
_________________________________________________
oops. Just realized I've been mixing up my exponents and cis.

Please change statement to
cis(j1*k2pi/n)=cis(j2*k2pi/n)

I have edited the original accordingly.
 

Users Who Are Viewing This Thread (Users: 0, Guests: 1)

Top