The Fejér kernel concentrates at the origin
ProvedFejer.fejerKernel_tendsto_zero_away_from_originFor every , as .
import Mathlib
import Definitions.Def_Fejer_fejerKernel
namespace Fejer
open MeasureTheory
/-- **The Fejér kernel's mass away from the origin vanishes.** For every `δ ∈ (0, π)`,
`∫_{δ ≤ |θ| ≤ π} F_N(θ) dθ → 0` as `N → ∞`. -/
theorem fejerKernel_tendsto_zero_away_from_origin
(δ : ℝ) (hδ0 : 0 < δ) (hδπ : δ < Real.pi) :
Filter.Tendsto
(fun N : ℕ => ∫ θ in (Set.Icc (-Real.pi) Real.pi) ∩ {θ : ℝ | δ ≤ |θ|}, fejerKernel N θ)
Filter.atTop (nhds 0) := by
sorry
end Fejer
Read-back
What the Lean code literally says, in plain math · claude-sonnet-5
The statement fixes a real number subject to exactly two hypotheses: (strict) and (strict), i.e. ranges over the open interval , with both endpoints excluded. Under these hypotheses, the integration domain
is, in plain terms, — the closed interval with the open middle portion removed. The theorem's claim is that the function of given by
(the Lebesgue integral of restricted to that two-piece domain) tends to as : precisely, as ranges over the natural numbers under the "eventually large" (atTop) filter, this sequence of real numbers converges, in the standard topology on , to . No claim is made about the rate of convergence, about behavior for or , or about the integral over any other subset of .
Confirmed by the mission captain (proposal self-audit).