Note on a sum involving the divisor function

From MaRDI portal





The paper under review belongs to a collection of numerous articles dealing with sums of the type \(\sum_{n \leqslant x} f \lfloor x/n \rfloor\), where \(f\) is an arithmetic function and \(\lfloor t \rfloor\) is the integer part of \(t\), whose pioneering article in this topic seems to be [the reviewer et al., J. Number Theory 202, 278--297 (2019; Zbl 1454.11009)]. In this work, the author investigates the sum \N\[\N\sum_{n \leqslant x^{1/c}} \tau \left( \left \lfloor \frac{x}{n^{c}} \right \rfloor\right),\N\]\Nwhere \(\tau\) is the usual Dirichlet divisor function and \(c > 0\), and proves that, for all \(\varepsilon > 0\) \N\[\N\sum_{n \leqslant x^{1/c}} \tau \left( \left \lfloor \frac{x}{n^{c}} \right \rfloor\right) = d_c x^{1/c} + O \left( x^{\theta_c + \varepsilon} \right),\N\]\Nwhere \(d_c\) is a constant depending on \(c\) explicitly given by the author and \(\theta_c := \max \left( \frac{2c+2}{2c^2+5c+2},\frac{5}{5c+6}\right)\), improving a previous result by Feng. The proof uses an estimate of [\textit{M. Jutila}, Lectures on a method in the theory of exponential sums. Berlin etc.: Springer-Verlag (1987; Zbl 0671.10031)] dealing with exponential sums twisted by the divisor function.











This page was built for publication: Note on a sum involving the divisor function

Report a bug (only for logged in users!)Click here to report a bug for this page (MaRDI item Q6934077)