Let $k\in\mathbb{N}$ be odd and $N\in\mathbb{N}$. You may assume that $N>k^2/4$ although I don't think that is relevant.
Let $\zeta:=\exp(2\pi i/k)$ and $\alpha_v:=\zeta^v+\zeta^{-v}+\zeta^{-1}$.
As it comes from the trace of a positive matrix I know that the following is real:
$$\sum_{v=1}^{\frac{k-1}{2}}\sec^2\left(\frac{2\pi v}{k}\right)\frac{(\overline{\alpha_v}^N+\alpha_v^N\zeta^{2N})(\zeta^{2v}-1)^2}{\zeta^N\zeta^{2v}}.$$
I am guessing, and numerical evidence suggests, that in fact
$$\frac{(\overline{\alpha_v}^N+\alpha_v^N\zeta^{2N})(\zeta^{2v}-1)^2}{\zeta^N\zeta^{2v}}$$
is real for (at least) each $v=1...(k-1)/2$. Therefore I am assuming that there is some nice simplification of it.
Can anyone simplify this expression?
Summing or even bounding the series would go above and beyond.
Context
I need to calculate or rather bound traces to calculate a distance to random for the convolution powers of a $\nu\in M_p(\mathbb{G}_k)$ for $\mathbb{G}_k$ a series of quantum groups of order $2k^2$ ($k$ odd).
Update
Following mercio's answer below I am now dealing with:
$$\frac{2}{4^{2N+1}}\sum_{v=1}^{\frac{k-1}{2}}\sec^2\left(\frac{2\pi v}{k}\right)\left(8+8|\alpha_v|^{2N}-8\sin^2\left(\frac{2\pi v}{k}\right)\Re\left((\alpha_v\zeta)^N\right)\right).$$
I can handle the first term (it is $2(k^2-1)/4^{2N}$) and am now looking at the other two terms.