Research notes on every problem, and a library of the papers behind them. Built from the open erdos repository.
Updated
Statement
Lemma (printed p. 380, unnumbered). Let , let be a sequence of complex numbers with period , and write
Then for all integers ,
There is no restriction on the length in the lemma itself.
Inequality (1) (printed p. 381). For an interval of length at most and an integer , write , where and the terms with are omitted, and let
be the Kloosterman sum, with the same omission. The article states that the lemma, applied to under this length assumption, gives
Source. D. R. Heath-Brown, Arithmetic applications of Kloosterman sums, Nieuw Arch. Wiskd. (5) 1 (2000), no. 4, 380–384; the lemma on printed p. 380, inequality (1) and the definition of on printed p. 381. The edition is identified on the source card.
Read depth. Claims checked: the lemma and inequality (1) were read clause by clause against the print. The article gives no proof of either; nothing here is independently reviewed.
Proof pointer
The article states the lemma without proof, as the standard device for converting an incomplete sum into complete ones. The usual argument expands by Fourier inversion, ; the frequency gives the main term, and each other frequency contributes a geometric series over , bounded by , whose sum over , divided by , is at most . For (1), the sequence equal to on residues prime to and elsewhere has ; the main term is at most one complete sum in size when the interval has length at most , which accounts for the in .
Dependencies
None in the article.
Bears on
- Problem 445: the lemma, with (1) and Weil's bound, is the input to the origin-rectangle estimate on p. 382. The lemma holds for every interval , but the article applies it only to the origin rectangle and states no translated-interval result.