Normalized Goldbach kernel comparison for frequencies at least eight
ProvedGoldbachKernel_normalized_complex_laplace_tail_eightcomplex-analysisgoldbachharmonic-analysisnumber-theory
Let
For all real with , and ,
The normalizers are strictly positive. This supporting kernel estimate covers the large-frequency portion of the normalized-transform comparison used in the Goldbach density argument. It leaves the middle-frequency range separate and does not imply strong Goldbach. The threshold improves the earlier local formal bound of 16; no mathematical novelty claim is made.
Preamble
import Mathlib.Analysis.SpecialFunctions.Integrals.Basic import Mathlib.Analysis.Complex.RealDeriv import Mathlib.Analysis.Complex.PhragmenLindelof import Mathlib.MeasureTheory.Integral.DominatedConvergence import Mathlib.Analysis.Complex.ExponentialBounds import Mathlib.Tactic open MeasureTheory set_option autoImplicit false
Formal statement
theorem GoldbachKernel_normalized_complex_laplace_tail_eight
(a b t : ℝ) (ha : 0 ≤ a) (hb_lower : (1/2:ℝ) ≤ b)
(hb_upper : b ≤ 1) (ht : 8 ≤ |t|) :
((∫ u in (0:ℝ)..2, ((((2-u)^3*(4+6*u+u^2)/30):ℝ):ℂ)*
Complex.exp (-(((-b:ℝ):ℂ)+(t:ℂ)*Complex.I)*(u:ℂ))) : ℂ).re /
(∫ u in (0:ℝ)..2, ((2-u)^3*(4+6*u+u^2)/30)*Real.exp (-(-b)*u)) ≤
((∫ u in (0:ℝ)..2, ((((2-u)^3*(4+6*u+u^2)/30):ℝ):ℂ)*
Complex.exp (-((a:ℂ)+(t:ℂ)*Complex.I)*(u:ℂ))) : ℂ).re /
(∫ u in (0:ℝ)..2, ((2-u)^3*(4+6*u+u^2)/30)*Real.exp (-a*u)) := by sorrySource
Supporting polynomial-transform estimate for Pintz, arXiv:1804.09084v2, Conditions 1-2 and Lemmas 1-5, pp. 28-30, https://arxiv.org/pdf/1804.09084v2#page=28. Independently formalized analytic tail refinement to frequency eight; no novelty claim and no middle-frequency result. Formal half-plane positivity uses Mathlib PhragmenLindelof.right_half_plane_of_bounded_on_real at revision 777aaa61dcd2a1258d2b4962dbe983ede4d23b2e.