Positivity of the singular series from local solubility
ProvedCircleMethod.singSeries_lower_bound_of_localcircle-methodnumber-theorysingular-serieswaring
Assume , and LocalSolubility k s. Then there is a constant ,
independent of , with
and is real. The proof combines the Euler product with the two-sided control of its factors: the finitely many primes contribute at least by local solubility, while for the estimate and the inequality keep the tail of the product above .
Preamble
/- # Positivity of the singular series in Waring's problem `singSeries k s n = ∑' q, locDens k s q n` is the singular series `𝔖(n)`. This file establishes its Euler product `𝔖(n) = ∏_p σ_p(n)`, `σ_p(n) = ∑_j A(p^j, n)`, bounds the tail of the product, and assembles a uniform lower bound `𝔖(n) ≥ c(k,s) > 0` from the local-solubility statement `LocalSolubility k s` (Vaughan, Lemmas 2.13–2.15). -/ import Definitions.Def_CircleMethod_singpos import Theorems.Thm_CircleMethod_locDens_mul import Theorems.Thm_CircleMethod_locDens_one import Theorems.Thm_CircleMethod_locDens_zero_modulus import Theorems.Thm_CircleMethod_locDens_bound import Theorems.Thm_CircleMethod_locDens_exponent_lt_neg_one import Theorems.Thm_CircleMethod_sing_series_summable_norm import Theorems.Thm_CircleMethod_e_zero import Theorems.Thm_CircleMethod_e_add import Theorems.Thm_CircleMethod_sum_e_div_orthogonality import Theorems.Thm_Vino_e_div_congr import Mathlib.NumberTheory.EulerProduct.Basic import Mathlib.NumberTheory.SumPrimeReciprocals import Mathlib.NumberTheory.Padics.Hensel import Mathlib.NumberTheory.Padics.RingHoms import Mathlib.Combinatorics.Additive.CauchyDavenport open Finset Filter Topology open Finset Filter Topology
Formal statement
namespace CircleMethod
theorem singSeries_lower_bound_of_local {k s : ℕ} (hk : 2 ≤ k) (hs : 2 ^ k < s)
(hloc : LocalSolubility k s) :
∃ c : ℝ, 0 < c ∧ ∀ n : ℤ, ∃ r : ℝ, c ≤ r ∧ singSeries k s n = r := by sorry
end CircleMethodSource
R. C. Vaughan, The Hardy-Littlewood Method, 2nd ed., Cambridge Tracts in Mathematics 125, Cambridge University Press, 1997, Section 2.6, Theorem 2.4 (eq. 2.32)