Estimates of sums related to the Nyman-Beurling criterion for the Riemann hypothesis (Q1708070)

The authors give an estimate for sums appearing in the Nyman-Beurling criterion for the Riemann hypothesis (see [\textit{L. Báez-Duarte} et al., Adv. Math. 149, No. 1, 130--144 (2000; Zbl 1008.11032)]) Let \(0<\delta<D/2 ,\; k^{2 \delta}\leq B \leq k^D ,\) where \(k^{-\delta}\leq \eta \leq 1.\) Then there is a positive constant \(\beta\) depending only \(\delta\) and \(D,\) such that \[ \sum\limits_{Bk \leq n <(1+ \eta )Bk }\mu(n) g\left(\frac{n}{k}\right)= \mathcal{O} \left( (\eta B k) ^{1- \beta}\right), \] where \(\mu\) denotes the Möbius function and \(g(x) = \sum\limits_{l\geq 1}\frac{1-2\{ lx\}}{l}.\) This estimate is remarkably sharp in comparison to the existing estimates of other sums containing the Möbius function. In the proof the authors intensively use the theory of continued fractions and the theory of Fourier series.