×

The variance of the error function in the shifted circle problem is a wild function of the shift. (English) Zbl 0808.11058

The authors study the lattice rest of a convex planar domain \({\mathcal B}\) (with distance function \(f\)), “blown up” by a large real parameter \(R\) and shifted by some vector \({\mathbf a}= (a_ 1, a_ 2)\in \mathbb{R}^ 2\). To be precise, they assume that \(f^ 2\) is \(C^ \infty\) on \(\mathbb{R}^ 2- \{(0,0)\}\) and put \[ F_{\mathcal B} (R,{\mathbf a})= {{\#\{{\mathbf m}\in \mathbb{Z}^ 2:\;f({\mathbf m}-{\mathbf a})\leq R\}- \text{area}\{{\mathbf x}\in \mathbb{R}^ 2:\;f({\mathbf x})\leq 1\}R^ 2} \over {\sqrt{R}}}. \] In an earlier paper [Duke Math. J. 67, 461-481 (1992; Zbl 0762.11031)], the first named author had proved that the limit \[ D_{\mathcal B} ({\mathbf a})= \lim_{T\to\infty} \Biggl( {\textstyle {1\over T}} \int_ 1^ T (F_{\mathcal B} (R,{\mathbf a}))^ 2 dR\Biggr) \] always exists and is equal to \(D_{\mathcal B}({\mathbf a})= {1\over {2\pi^ 2}} \sum_{n=1}^ \infty | u_{\mathbf a}(n) |^ 2\) where \[ u_{\mathbf a}(n)=\sum _{\substack{ (k,l)\in \mathbb{Z}^ 2\\ (k,l)\neq (0,0)}} (k^ 2+ l^ 2)^{-3/4} \sqrt{\kappa (k,l)} e^{2\pi i(ka_ 1+ la_ 2)} \] and \(\kappa(k,l)\) denotes the curvature of the boundary of \({\mathcal B}\) at that point where the outward normal is collinear to \((k,l)\).
In the present article the authors show that the function \(D_{\mathcal B}({\mathbf a})\) is “wild”, i.e., continuous but nondifferentiable on a dense set of the plane: In every rational point \({\mathbf a}'\in \mathbb{Q}^ 2\), \[ \lim_{{\mathbf a}\to {\mathbf a}'} \Biggl( {{D_{\mathcal B} ({\mathbf a})- D_{\mathcal B}({\mathbf a}')} \over {|{\mathbf a}-{\mathbf a}'|}} {1\over {\log|{\mathbf a}-{\mathbf a}'|}} \Biggr)=- C_{\mathcal B}({\mathbf a}')<0. \]
Reviewer: W.G.Nowak (Wien)

MSC:

11P21 Lattice points in specified regions

Citations:

Zbl 0762.11031
PDFBibTeX XMLCite
Full Text: DOI

References:

[1] [B] Bleher, P.M.: On the distribution of the number of lattice points inside a family of convex ovals. Duke Math. J.67, 461–481 (1992) · Zbl 0762.11031 · doi:10.1215/S0012-7094-92-06718-4
[2] [BCDL] Bleher, P.M., Cheng, Zh., Dyson, F.J., Lebowitz, J.L.: Distribution of the error term for the number of lattice points inside a shifted circle. Commun. Math. Phys.154, 433–469 (1993) · Zbl 0781.11038 · doi:10.1007/BF02102104
[3] [BD] Bleher, P.M., Dyson, F.J.: Mean square value of exponential sums related to representation of integers as sum of two squares. Preprint IASSNS-HEP-92/84, Institute for Advanced Study, Princeton, 1992
[4] [C] Cramér, H.: Über zwei Sätze von Herrn G.H. Hardy. Math. Zeit.15, 201–210 (1922) · JFM 48.0184.01 · doi:10.1007/BF01494394
[5] [H-B] Heath-Brown, D.R.: The distribution and moments of the error term in the Dirichlet divisor problem. Acta Arithmetica60, 389–415 (1992) · Zbl 0725.11045
[6] [K] Kendall, D.G.: On the number of lattice points inside a random oval. Quart. J. Math. (Oxford)19, 1–26 (1948) · Zbl 0031.11201 · doi:10.1093/qmath/os-19.1.1
This reference list is based on information provided by the publisher or from digital mathematics libraries. Its items are heuristically matched to zbMATH identifiers and may contain data conversion errors. In some cases that data have been complemented/enhanced by data from zbMATH Open. This attempts to reflect the references listed in the original paper as accurately as possible without claiming completeness or a perfect matching.