Perron kernel: approximates
ProvedTitchmarsh.perronKernel_error_leanalytic-number-theorycontour-integrationperron-formula
For and , let
be the truncated Perron integral on the line , and let if and if . There is an absolute constant such that for all with and all
This is the kernel estimate behind every truncated Perron formula. It is a slightly weakened form of the classical bound (Davenport, Multiplicative Number Theory, Ch. 17, Lemma, with ; Titchmarsh–Heath-Brown, proof of Lemma 3.12). It yields Perron's formula for any Dirichlet series with bounded coefficients when summed over .
Formalization Note. is the platform definition Weil.perronIntegral (fun _ => 1) y 2 T, the integral over of , normalized by .
Preamble
import Definitions.Def_Weil_perron import Mathlib.Analysis.SpecialFunctions.Log.Basic
Formal statement
namespace Titchmarsh
theorem perronKernel_error_le :
∃ K : ℝ, 0 ≤ K ∧ ∀ y : ℝ, 0 < y → y ≠ 1 → ∀ T : ℝ, 1 ≤ T →
‖Weil.perronIntegral (fun _ => 1) y 2 T - (if 1 < y then 1 else 0)‖ ≤
K * y ^ 2 * (1 + 1 / |Real.log y|) / T := by sorry
end TitchmarshSource
H. Davenport, Multiplicative Number Theory, 3rd ed., GTM 74, Chapter 17, Lemma (p. 105), with c = 2 (weakened constant form); E. C. Titchmarsh, The Theory of the Riemann Zeta-function, 2nd ed. rev. Heath-Brown, proof of Lemma 3.12, pp. 60-61. https://sites.math.rutgers.edu/~zeilberg/EM18/TitchmarshZeta.pdf