# Analytic continuation of twisted Hecke $L$-function

Let $K$ be a (real quadratic) number field and $\chi$ be a Hecke character on it. We can think about Hecke $L$-function $$L(\chi, s):= \sum_{0\neq \mathfrak{a}\subseteq \mathcal{O}_{K}} \frac{\chi(\mathfrak{a})}{(N\mathfrak{a})^{s}} = \sum_{n\geq 1} \frac{a(n)}{n^{s}}$$ which has analytic continuation on $\mathbb{C}$ and admits a functional equation.

Now I'm interested in the following "twisted" Hecke $L$-function: $$L(\chi, \beta, s):= \sum_{n\geq 1} \frac{\beta^{n}\cdot a(n)}{n^{s}}$$ where $\beta$ is a root of unity. Does this $L$-function have analytic continuation and functional equation? I discovered some identities related to these $L$-values at negative integers, and the identities are completely nonsense if $L(\chi,\beta, s)$ doesn't have analytic continuation.

More precisely, I'm interested in the case when Hecke character $\chi$ of $\mathbb{Q}(\sqrt{2})$ defined as $$\chi(\mathfrak{a}) = \begin{cases} 1 & N\mathfrak{a}\equiv \pm 1\mod{16} \\ -1 & N\mathfrak{a}\equiv \pm 7 \mod{16} \\ 0 & otherwise \end{cases}$$ for any $\mathfrak{a}\subset \mathbb{Z}[\sqrt{2}]$, and $\beta =\zeta_{8p}$ for any odd prime $p$. I hope there is a formula for the value $L(\chi, \zeta_{8p}^{j}, -n)$ for $n>0$ and $j=0, \dots, p-1$.

## 1 Answer

Yes, this has an analytic continuation and functional equation. Basically, the crucial point is that the sequence $\{a(n)\}$ is equal to the sequence of Hecke eigenvalues of an automorphic form. Then the twist by a root of unity of the associated $L$-function appears in the Voronoi summation formula (sometimes this twisted $L$-function is called an Estermann $L$-function).

The functional equation is quite complicated; it depends highly on the level of the associated automorphic form, the data appearing in the gamma factors of the untwisted $L$-function, and the exact choice of root of unity. A starting point is Appendix A of the paper of this paper.

• Thank you very much! I have one more question - is there any result about computation of some special values of such Estermann L-functions? Also, I can't find any other references for Estermann L-functions, could you recommend other references to read? – Seewoo Lee Mar 15 '18 at 14:38
• You should look up the Voronoi summation formula. This twisted $L$-function is a generalisation of the Estermann zeta function, which involves a divisor function instead of Hecke eigenvalues (so corresponds to an Eisenstein series instead of a cusp form). As for special values, I don't know off the top of my head (but I can't imagine why special values might be interesting). – Peter Humphries Mar 15 '18 at 14:40
• I don't know it is important or not, but if I know some exact values of such L-functions, I may get other interesting identities from the identity I discovered. I'll add the specific example that I'm interested in. – Seewoo Lee Mar 15 '18 at 14:49