Mollifier existence, rescale technicalities, and Bridge Lemma 1 #
Smooth compactly supported infrastructure on ℝⁿ:
Main contents #
mollifier_exists— there existsψ ∈ C^∞_c(ℝⁿ; ℝ)withψ ≥ 0,∫ ψ = 1, andsupp(ψ) ⊂ (-π, π)ⁿ.rescale_hasCompactSupport,rescale_contDiff,rescale_support_in_cube—rescale n φ εpreserves compact support, smoothness, and the cube-support condition.ftRn_at_zero—ftRn n φ 0 = ∫ φ.cinfty_rapidDecay(Bridge Lemma 1) — forφ ∈ C^∞_c(ℝⁿ; ℂ),FTRapidDecay n φholds.convDistrib_memSobolevDistrib—convDistrib n φ u ∈ H^s_*wheneverφis integrable andu ∈ H^s_*.
Mollifier existence #
Rescale technicalities #
theorem
NashEmbedding.Sobolev.rescale_hasCompactSupport
{n : ℕ}
{φ : (Fin n → ℝ) → ℂ}
(hsupp : HasCompactSupport φ)
{ε : ℝ}
(hε : 0 < ε)
:
HasCompactSupport (rescale n φ ε)
Bridge Lemma 1: C^∞_c ⟹ FTRapidDecay #
The statement and proof now live in NashEmbedding/Sobolev/IntegrationByParts.lean
alongside the ftRn-side IBP infrastructure it depends on. Downstream
callers still see NashEmbedding.Sobolev.cinfty_rapidDecay through this file's
transitive re-export.
MemSobolevDistrib closure for convDistrib #
theorem
NashEmbedding.Sobolev.convDistrib_memSobolevDistrib
{n : ℕ}
{φ : (Fin n → ℝ) → ℂ}
(hφ : MeasureTheory.Integrable φ MeasureTheory.volume)
{s : ℝ}
{u : TrigPolyDual n}
(hu : MemSobolevDistrib n s u)
:
MemSobolevDistrib n s (convDistrib n φ u)