Type II bilinear sums for the von Mangoldt decomposition
ProvedVaughan.typeII_boundcircle-methodexponential-sumsnumber-theoryvinogradov
Type II estimate for the third term of Vaughan's identity. There is an absolute constant such that for , , , , ,
Both coefficient sequences here are genuinely arithmetic, so there is no smooth variable to exploit; instead the -range is cut into dyadic blocks, the block estimate is applied to each, and the divisor second moment controls the resulting coefficient sums. The terms and reflect the truncation levels: the coefficients vanish unless and , which is what makes both variables long enough for Cauchy--Schwarz to pay.
Preamble
import Definitions.Def_Vino_vaughan import Mathlib.NumberTheory.ArithmeticFunction.VonMangoldt import Mathlib.NumberTheory.ArithmeticFunction.Moebius import Mathlib.Analysis.SpecialFunctions.Log.Basic import Mathlib.Analysis.SpecialFunctions.Pow.Real import Mathlib.NumberTheory.ArithmeticFunction.Misc import Mathlib.Algebra.BigOperators.Module import Mathlib.Algebra.Order.Interval.Finset.SuccPred import Mathlib.Algebra.Order.BigOperators.Ring.Finset import Mathlib.Algebra.BigOperators.Field import Mathlib.NumberTheory.Harmonic.Bounds import Theorems.Thm_CircleMethod_e_neg import Theorems.Thm_Vino_psi_le import Theorems.Thm_CircleMethod_aux_sum_min_le import Theorems.Thm_CircleMethod_nrm_le import Theorems.Thm_CircleMethod_nrm_triangle import Theorems.Thm_CircleMethod_nrm_neg import Theorems.Thm_CircleMethod_nrm_nonneg import Theorems.Thm_CircleMethod_nrm_eq_zero_iff import Theorems.Thm_CircleMethod_nrm_pos_of_not_int import Theorems.Thm_CircleMethod_norm_sum_e_le_inv_nrm import Theorems.Thm_CircleMethod_norm_sum_e_le import Theorems.Thm_CircleMethod_e_add import Theorems.Thm_CircleMethod_norm_e open Finset ArithmeticFunction open CircleMethod
Formal statement
namespace Vaughan
theorem typeII_bound :
∃ C : ℝ, 0 < C ∧ ∀ (N U V q : ℕ) (a : ℤ) (α : ℝ), 2 ≤ N → 1 ≤ U → 1 ≤ V → 0 < q → q ≤ N →
Nat.Coprime a.natAbs q → |α - (a : ℝ) / (q : ℝ)| ≤ 1 / (q : ℝ) ^ 2 →
‖bilin (vonMangoldt - lamT U) (((moebius : ArithmeticFunction ℝ) - muT V) * zeta) α N‖
≤ C * ((N : ℝ) / Real.sqrt q + (N : ℝ) / Real.sqrt U + (N : ℝ) / Real.sqrt V
+ Real.sqrt ((N : ℝ) * q)) * Real.log N ^ 4 := by sorry
end VaughanSource
R. C. Vaughan, The Hardy-Littlewood Method, 2nd ed., Cambridge Tracts in Mathematics 125, Cambridge University Press, 1997, Section 3.1, Lemma 3.2 combined with the dyadic decomposition of Section 3.2