Mathematics Stack Exchange is a question and answer site for people studying math at any level and professionals in related fields. Join them; it only takes a minute:

Sign up
Here's how it works:
  1. Anybody can ask a question
  2. Anybody can answer
  3. The best answers are voted up and rise to the top

While debugging an NTT implementation I've noticed something. Looks like if I have a primitive $(n = 2^k)$-th root of unity $\omega$ in a $GF(p)$, then

$\omega ^0 = -\omega ^{n/2} + p,$

$\omega ^1 = -\omega ^{n/2 + 1} + p,$


$\omega ^{n/2 - 1} = -\omega ^{n-1} + p.$

Or, maybe, I could rewrite it as

$\omega ^i \equiv \omega ^ {n/2 + i} \mod p \quad \forall i=1..n/2-1$.

Does it always hold? Why? Does it hold under stronger conditions (e.g. for all even $n$, not just $n = 2^k$)?

Thank you very much in advance!

share|cite|improve this question
up vote 2 down vote accepted

In $GF(p)$ we have characteristic $p$ so the $+p$ appended at the end of each equation is superfluous.

Otherwise notice that each of the subsequent equations follow by multiplying the first equation by a power of $\omega$. It suffices therefore to demonstrate the first equation.

Since $\omega^n=1$ and $n$ is even, factoring yields $(\omega^{n/2}+1)(\omega^{n/2}-1)=0$. Because this is a field and so has no zero divisors, we find that $\omega^{n/2}=\pm1$, but $\omega^{n/2}=+1$ would contradict the hypothesis that $\omega$ is a primitive root of unity, so we must have $\omega^{n/2}=-1$. Note this works for any even $n$.

share|cite|improve this answer
A brilliant clarification. Thanks a lot! – wh1t3cat1k Apr 22 '12 at 9:33

Your Answer


By posting your answer, you agree to the privacy policy and terms of service.

Not the answer you're looking for? Browse other questions tagged or ask your own question.