Search arXivSearch

arXiv · 2404.03616

On algebras of Dirichlet series invariant under permutations of coefficients

Abstract

Let $\mathscr O_u$ be the algebra of holomorphic functions on ${\bf C}_+:=\{s\in{\bf C}:\text{Re }s>0\}$ that are limits of Dirichlet series $D=\sum_{n=1}^\infty a_n n^{-s}$, $s\in \bf{C}_+$, that converge uniformly on proper half-planes of $\bf{C}_+$. We study algebraic-topological properties of subalgebras of $\mathscr O_u$: the Banach algebras $\mathscr W, \mathscr A, \mathscr H^\infty$ and the Frechet algebra $\mathscr O_b$. Here $\mathscr W$ consists of functions in $\mathscr O_u$ of absolutely convergent Dirichlet series on the closure of $\bf{C}_+$, $\mathscr A$ is the uniform closure of $\mathscr W$, $\mathscr H^\infty$ is the algebra of all bounded functions in $\mathscr O_u$, and $\mathscr O_b$ is set of all $f(s)=\sum_{n=1}^\infty a_n n^{-s}$ in $\mathscr O_u$ so that $f_r\in \mathscr H^\infty$, $r\in (0,1)$, where $f_r(s):=\sum_{n=1}^\infty a_n r^{Ω(n)} n^{-s}$ and $Ω(n)$ is the number of prime factors of $n$. Let $S_\bf{N}$ be the group of permutations of $\bf{N}$. Each $σ\in S_\bf{N}$ determines a permutation $\hatσ\in S_\bf{N}$ (i.e., such that $\hatσ(mn)=\hatσ(n)\hatσ(m)$ for all $m,n\in \bf{N}$) via the fundamental theorem of arithmetic. For a Dirichlet series $D=\sum_{n=1}^\infty a_n n^{-s}$, and $σ\in S_\bf{N}$, $S_σ(D)=\sum_{n=1}^\infty a_{\hatσ^{-1}(n)} n^{-s}$ determines an action of $S_\bf{N}$ on the set of all Dirichlet series. It is shown that each of the algebras above is invariant with respect to this action. Given a subgroup $G$ of $S_\bf{N}$, the set of $G$-invariant subalgebras of these algebras are studied, and their maximal ideal spaces are described, and used to characterise groups of units and of invertible elements having logarithms, find the stable rank, show projective freeness, and describe when the special linear group is generated by elementary matrices, with bounds on the number of factors.

Explore related subjects

Keep this discovery

Explore connections, maps & timelines

BibTeXRIS

Alexander Brudnyi, Amol Sasane. 2024-04-08. On algebras of Dirichlet series invariant under permutations of coefficients. https://arxiv.org/abs/2404.03616

Cite the original work for its findings. Save a collection to share your selection of sources.

KEEP EXPLORING

Related papers

Pre-Schwarzian and Schwarzian norm estimates for harmonic functions with fixed analytic part

In the present article, we discuss about the estimate of the pre-Schwarzian and Schwarzian norms for locally univalent harmonic functions $f=h+\overline{g}$ in the unit disk $\mathbb{D}:=\{z\in\mathbb{C}:\, |z|<1\}$. First, we prove a general result for the estimate of the pre-Schwarzian norm which rectify few earlier flawed results. We also consider a new class $\mathcal{F}_0$ consisting of all harmonic functions $f=h+\overline{g}$ in the unit disk $\mathbb{D}$ such that ${\rm Re\,}\left(1+z\frac{h''(z)}{h'(z)}\right)>0$ for $z\in\mathbb{D}$ with dilatation $ω_f(z)\in Aut(\mathbb{D})$ and obtain best possible estimates of the pre-Schwarzian and Schwarzian norms for functions in the class $\mathcal{F}_0$. Moreover, we obtain the distortion and coefficient estimates of the co-analytic function $g$ when $f=h+\overline{g}\in\mathcal{F}_0$.

math.CV

The Reciprocal Problem on Weighted Bergman Spaces

The reciprocal problem on weighted Bergman spaces has been posed as an open problem. In this paper, we establish several sufficient conditions for the reciprocal property and clarify the parameter ranges in which the available methods are applicable. In particular, we prove that functions in $A_α^p\cap H^\infty$ enjoy the reciprocal property in the parameter ranges where the required analytic Besov composition theorem is available. In addition, using Hardy boundary estimates, we solve the reciprocal problem in the Drury--Arveson space $H_d^2$ when the dimension is $d=3$, and give an equivalent condition for the reciprocal problem in the four-dimensional Drury--Arveson space.

math.CV

Solving non-oscillatory solutions of the Hill equation via the Tumura--Clunie method

We consider the Hill equation $f''-(\sum_{i=-\mathbf{l}}^{\mathbf{k}}b_{i}e^{iz})f=0$ ($†$), where $\mathbf{k}\geq 1$ and $\mathbf{l}\geq 0$ are integers and $b_{-\mathbf{l}}$, $\cdots$, $b_{\mathbf{k}}$ are constants such that $b_{\mathbf{k}}\not=0$. We point out that there is a full correspondence between the class of non-oscillatory solutions such that $λ(f)<\infty$ of equation ($†$) and the class of Liouvillian solutions of equation $x^2u''-(\sum_{i=-\mathbf{l}}^{\mathbf{k}}b_{i}x^{i})u=0$ ($‡$). Then this paper has twofold purposes. First, parallel to Kovacic's algorithms to find the Liouvillian solutions of equation ($‡$), we develop the Tumura--Clunie method to find the non-oscillatory solutions of a higher order version of the Hill equation. In this part, we first determine the form of entire solutions of a general Tumura--Clunie type differential equation. Second, for the particular Hill equation $f''-(b_{\mathbf{k}}e^{\mathbf{k}z}+b_{\mathbf{s}}e^{\mathbf{s}z}+b_0)f=0$, where $\mathbf{k}>\mathbf{s}\geq 1$ are integers and $b_{\mathbf{k}}b_{\mathbf{s}}\not=0$, we use the Tumura--Clunie method to determine the non-oscillatory solution $f$ with an additional zero property.

math.CV