Koshliakov zeta functions I: modular relations (Q2054258): Difference between revisions

The paper is devoted to the study of functions \(\zeta_p(s)\) and \(\eta_p(s)\) defined as \[ \zeta_p(s):=\sum_{j=1}^{\infty}\frac{p^2+\lambda_j^2}{p\big ( p+{1}/{\pi}\big )+\lambda_j^2 }\cdot \frac{1}{\lambda_j^s}, \quad \Re(s)>1, \] and \[ \eta_p(s):=\sum_{k=1}^{\infty}\frac{\big(s,2\pi p k\big)_k}{k^s}, \quad \Re (s)>1, \] respectively. Here \(\lambda_1,\lambda_2,\ldots\) are the positive roots of the equation \(p \sin(\pi \lambda)+\lambda \cos(\pi \lambda)=0\) (\(p >0)\), and \(\big(s,\nu k\big)_k:={1}/{\Gamma(s)}\cdot \int_{0}^{\infty}e^{-x}\big(({k\nu-x})/({k\nu+x})\big)^k x^{s-1}dx\). The function \(\zeta_p(s)\) has an analytic continuation to the entire complex plane except for a simple pole at \(s=1\) with residue 1, while the function \(\eta_p(s)\) can be analytically continued in the whole \(s\)-plane except for a simple pole at \(s=1\) with residue \(1/(1+1/(\pi p))\). The functions satisfy as well the functional equation of the form \[ \zeta_p(s)=\frac{2 \cos (\pi s / 2)\Gamma(s)}{(2 \pi)^s}\eta_p(s). \] The functions under investigation were introduced and firstly studied by the Russian mathematician Nikolai Sergeevich Koshliakov. It was done in his unpublished manuscript and in [\textit{N. S. Sergeev}, Untersuchung einer Klasse von transzendenten Funktionen, die durch die verallgemeinerte Riemannsche Gleichung bestimmt werden (Russian). Akad. Nauk SSSR, Mat. Inst. Steklov. Moskau-Leningrad: Verlag der Akad. d. Wiss. d. UdSSR. 154 S. (1949; Zbl 0045.18401)] (note that manuscript was written under Koshliakov's patronymic name N.S.~Sergeev). Koshliakov's manuscript has never been examined deeply in details until 2010 when the first author, namely A.~Dixit, found out about it. Moreover, the authors of this paper named the functions \(\zeta_p(s)\) and \(\eta_p(s)\) as Koshliakov zeta functions. The authors develop further Koshliakov's results on \(\zeta_p(s)\) and \(\eta_p(s)\) investigating two different kinds of modular relations. The first of them is Ramanujan's formula for odd zeta values. The second kind is the study of an integral containing in its integrand the Riemann's function \(\Xi(t)\) and its generalization \( \Xi_p(t):=\xi_p\big({1}/{2}+it\big) \) with \(\xi_p(s):=s(s-1)/2 \cdot \pi^{-s/2}\Gamma(s/2)\big(\zeta_p(s)+\eta_p(s)\big)/2\). Independently from new results presented in this paper, the authors gave a very good description and explanation of contents and results from unpublished manuscripts of N.S.~Koshliakov. This paper plays a very important role for developing of newly re-opened Koshliakov zeta functions theory as a part of the wide theory of zeta functions, and can be interesting to the mathematicians in two senses: historical and pure scientific.