Skip to Main Content

Transactions of the American Mathematical Society

Published by the American Mathematical Society since 1900, Transactions of the American Mathematical Society is devoted to longer research articles in all areas of pure and applied mathematics.

ISSN 1088-6850 (online) ISSN 0002-9947 (print)

The 2020 MCQ for Transactions of the American Mathematical Society is 1.48.

What is MCQ? The Mathematical Citation Quotient (MCQ) measures journal impact by looking at citations over a five-year period. Subscribers to MathSciNet may click through for more detailed information.

 

The approximate functional formula for the theta function and Diophantine Gauss sums
HTML articles powered by AMS MathViewer

by E. A. Coutsias and N. D. Kazarinoff PDF
Trans. Amer. Math. Soc. 350 (1998), 615-641 Request permission

Abstract:

We consider the polygonal lines in the complex plane $\Bbb {C}$ whose $N$-th vertex is defined by $S_N = \sum _{n=0}^{N ’} \exp (i\omega \pi n^2)$ (with $\omega \in \Bbb {R}$), where the prime means that the first and last terms in the sum are halved. By introducing the discrete curvature of the polygonal line, and by exploiting the similarity of segments of the line, for small $\omega$, to Cornu spirals (C-spirals), we prove the precise renormalization formula \begin{equation} \begin {split} &\left | \sum _{k=0}^{N} ’ \exp (i\omega \pi k^2) -\frac {\exp (sgn(\omega )i\pi /4)}{\sqrt {|\omega |}} \sum _{k=0}^n ’ \exp (-i\frac {\pi }{\omega } k^2)\right |\\ &\qquad \leq C \left |\frac {\omega N - n}{\omega }\right |, 0<|\omega | <1, \end{split} \end{equation} where $N=[[n/\omega ]]$, the nearest integer to $n/\omega$ and $1<C<3.14$ . This formula, which sharpens Hardy and Littlewood’s approximate functional formula for the theta function, generalizes to irrationals, as a Diophantine inequality, the well-known sum formula of Gauss. The geometrical meaning of the relation between the two limits is that the first sum is taken to a point of inflection of the corresponding C-spirals. The second sum replaces whole C-spirals of the first by unit vectors times scale and phase factors. The block renormalization procedure implied by this replacement is governed by the circle map \begin{equation} \omega \rightarrow -\frac {1}{\omega } \pmod 2 , \omega \in ]-1,+1[ \setminus \{0\}, \end{equation} whose orbits are analyzed by expressing $\omega$ as an even continued fraction.
References
Similar Articles
Additional Information
  • E. A. Coutsias
  • Affiliation: Department of Mathematics and Statistics, University of New Mexico, Albuquerque, New Mexico 87131
  • Email: vageli@math.unm.edu
  • N. D. Kazarinoff
  • Affiliation: Department of Mathematics and Statistics, University of New Mexico, Albuquerque, New Mexico 87131
  • Received by editor(s): January 25, 1995
  • Received by editor(s) in revised form: December 4, 1995
  • © Copyright 1998 American Mathematical Society
  • Journal: Trans. Amer. Math. Soc. 350 (1998), 615-641
  • MSC (1991): Primary 11G10; Secondary 11L05, 11L07, 11J25, 11J70
  • DOI: https://doi.org/10.1090/S0002-9947-98-02024-8
  • MathSciNet review: 1443869