Bounds for incomplete hyper-Kloosterman sums (Q1283042)

From MaRDI portal
Revision as of 11:16, 31 January 2024 by Import240129110113 (talk | contribs) (Added link to MaRDI item.)
scientific article
Language Label Description Also known as
English
Bounds for incomplete hyper-Kloosterman sums
scientific article

    Statements

    Bounds for incomplete hyper-Kloosterman sums (English)
    0 references
    0 references
    12 October 1999
    0 references
    For a prime \(p \geq 3\), the ``complete'' hyper-Kloosterman sum is \[ Kl _m(p)=\sum_{d_1=1}^{p-1}\cdots \sum_{d_m=1}^{p-1}e\left (\frac{d_1+\dotsb +d_m+\overline{d_1\dotsb d_m}}{p}\right), \] where the bar indicates multiplicative inverse \(\pmod{p}\). In the corresponding ``incomplete'' sum \(Kl_m(p, x_1, \dotsc ,x_m)\), each variable \(d_i\) runs, respectively, over the interval \([1, x_i]\). The following bound is obtained for the latter sum: \[ Kl _m(p, x_1, \dotsc ,x_m) \ll x_1\dotsb x_m/p + p^{1/2}(x_1\dotsb x_m)^{1-1/r}(p^{1/4(r-1)}\log ^2p)^m \] for any fixed integer \(r\geq 2\). The proof is based on estimates of \textit{D. A. Burgess} [Bull. Lond. Math. Soc. 20, 589-592 (1988; Zbl 0667.10024)] for incomplete Gauss sums.
    0 references
    0 references
    0 references
    incomplete Kloosterman sums
    0 references
    incomplete Gauss sums
    0 references