On Segre's theorem on asymmetric diophantine approximation (Q1378248): Difference between revisions

From MaRDI portal
Added link to MaRDI item.
ReferenceBot (talk | contribs)
Changed an Item
 
(3 intermediate revisions by 2 users not shown)
Property / author
 
Property / author: Horst Alzer / rank
Normal rank
 
Property / reviewed by
 
Property / reviewed by: Peter Bundschuh / rank
Normal rank
 
Property / author
 
Property / author: Horst Alzer / rank
 
Normal rank
Property / reviewed by
 
Property / reviewed by: Peter Bundschuh / rank
 
Normal rank
Property / MaRDI profile type
 
Property / MaRDI profile type: MaRDI publication profile / rank
 
Normal rank
Property / cites work
 
Property / cites work: Q3241280 / rank
 
Normal rank
Property / cites work
 
Property / cites work: Q3243354 / rank
 
Normal rank
Property / cites work
 
Property / cites work: On asymmetric approximations / rank
 
Normal rank
Property / cites work
 
Property / cites work: Lattice points in infinite domains and asymmetric diophantine approximations / rank
 
Normal rank
Property / cites work
 
Property / cites work: On a theorem of Segre / rank
 
Normal rank
Property / cites work
 
Property / cites work: Segre's theorem on asymmetric diophantine approximation / rank
 
Normal rank
Property / cites work
 
Property / cites work: A conjecture of Segre on Diophantine approximation / rank
 
Normal rank

Latest revision as of 10:52, 28 May 2024

scientific article
Language Label Description Also known as
English
On Segre's theorem on asymmetric diophantine approximation
scientific article

    Statements

    On Segre's theorem on asymmetric diophantine approximation (English)
    0 references
    0 references
    31 August 1998
    0 references
    Im folgenden seien \(\alpha, t\) reelle Zahlen mit \(\alpha\not\in{\mathbb{Q}}, t\in[0,1[\). \textit{B. Segre} [Duke Math. J. 12, 337-365 (1945; Zbl 0060.11807)] bewies die Existenz unendlich vieler \((p,q)\in{\mathbb{Z}}\times{\mathbb{N}}\), die der Ungleichung \[ -(1+4t)^{-1/2} < (p-\alpha q)q < t(1+4t)^{-1/2} \tag \(*\) \] genügen. Segre betrachtete auch die Frage, inwieweit sein zitiertes Resultat verschärft werden könnte, und er zeigte, daß \((\ast)\) jedenfalls für \(t = 0\) oder \((1/t)\in{\mathbb{Z}}\) bestmöglich ist. Er vermutete, daß sich \((\ast)\) für alle anderen \(t\) verbessern ließe. Nach Vorarbeiten von \textit{P. Szüsz} [Acta Arith. 23, 371-377 (1973; Zbl 0262.10019)] in dieser Richtung konnte \textit{J. Tong} [Monatsh. Math. 112, 141-147 (1991; Zbl 0736.11033)] Segres Vermutung voll bestätigen und zwar in quantitativer Weise, indem er die Existenz unendlich vieler \((p,q)\) mit \[ -((1+4t)^{1/2}+\delta(t))^{-1} < (p-\alpha q)q < ((1+4t)^{1/2}t^{-1}+ \delta(t))^{-1} \] bewies, wobei \(\delta(t) := {1\over 5} t^2\{{1\over t}\} (1-\{{1\over t}\})\) gesetzt ist, \(\{x\} := x-[x]\) der gebrochene Anteil einer reellen Zahl \(x\). In der vorliegenden Arbeit wird nun in enger Anlehnung an die Methode von Tong gezeigt, daß der Faktor \(1/5\) im Ausdruck für \(\delta(t)\) auf \(2/\sqrt 5\) verbessert werden kann und daß dies bestmöglich ist.
    0 references
    0 references
    Segre's theorem
    0 references
    asymmetric diophantine approximation
    0 references
    best possible estimates
    0 references
    0 references