Documentation

LeanPool.PhaseRetrieval.DimdPoly.Internal.Hermite1Dimd.ImportedAnalyticInputs

ImportedAnalyticInputs #

oneDimLift: one Dim Lift.

Equations
Instances For

    ImportedAnalyticInputs #

    Statement-only scaffold for the one-variable and abstract analytic inputs. Scaffolding notes: ScaffoldingNotes/Imported/analytic_inputs.md.

    Imported one-variable basis orthonormality at fixed true-Hermite level.

    theorem Hermite1DimdLEAN.oneVariableFiniteParseval (k : ℕ) (G : FiniteHermiteSum 1) :
    hermiteNormSq (fun (x : Fin 1) => k) G = ∑ n ∈ G.support, ‖G.coeff n‖ ^ 2

    Imported one-variable Parseval statement for finite coefficient families.

    theorem Hermite1DimdLEAN.oneVariableLocalization (k : ℕ) :
    ∃ (C : ℝ) (c : ℝ), 0 < C ∧ 0 < c ∧ ∀ (n j : ℕ), 1 ≤ n → annulusMass (fun (x : Fin 1) => j) (oneDimLift (oneDimPhi k n)) ≤ C * Real.exp (-c * max (|↑j - √↑n| - ↑(k + 4)) 0 ^ 2)

    Imported one-variable localization estimate for n ≥ 1.

    Imported positive-frequency circle estimate with frozen constant 144.

    theorem Hermite1DimdLEAN.highFrequencyBandEstimate (N L : ℕ) (c : Fin L → ℂ) (hgap : 1343 * L ^ 2 ≤ N ^ 2) (hP : HasBandlimitedSupport (bandLimitedPolynomial N L c) N L) :

    Imported high-frequency band estimate with frozen constants 32 and 1343.

    theorem Hermite1DimdLEAN.localizationIncludingZero (k : ℕ) :
    ∃ (C : ℝ) (c : ℝ), 0 < C ∧ 0 < c ∧ ∀ (n j : ℕ), annulusMass (fun (x : Fin 1) => j) (oneDimLift (oneDimPhi k n)) ≤ C * Real.exp (-c * max (|↑j - √↑n| - ↑(k + 4)) 0 ^ 2)

    Project-owned upgrade of localization that now includes the zero mode.

    theorem Hermite1DimdLEAN.scaledPositiveFrequencyCircleEstimate (a : ℂ) (E : Finset ℕ) (P : Circle → ℂ) (hpos : ∀ n ∈ E, 0 < n) (hP : HasPositiveFrequencySupport P E) :
    circleL2NormSq P ≤ 144 * ↑E.card * circleL2NormSq fun (t : Circle) => rho a (P t)

    Scalar-rescaled positive-frequency estimate for an arbitrary background a.

    theorem Hermite1DimdLEAN.scaledHighFrequencyBandEstimate (a : ℂ) (N L : ℕ) (P : Circle → ℂ) (hgap : 1343 * L ^ 2 ≤ N ^ 2) (hP : HasBandlimitedSupport P N L) :
    circleL2NormSq P ≤ 32 * circleL2NormSq fun (t : Circle) => rho a (P t)

    Scalar-rescaled high-frequency estimate for an arbitrary background a.

    theorem Hermite1DimdLEAN.oneVariableAngularFactorization (k n : ℕ) :
    ∃ (radial : Polynomial ℝ), ∀ (r t : ℝ), oneDimPhi k n (↑r * Complex.exp (Complex.I * ↑t)) = Complex.exp (Complex.I * ((↑↑n - ↑↑k) * ↑t)) * Polynomial.eval₂ (algebraMap ℝ ℂ) (↑r) radial
    theorem Hermite1DimdLEAN.tensorGaussianFactorization (d : ℕ) (F G : Fin d → ℂ → ℂ) (hFG : ∀ (q : Fin d), MeasureTheory.Integrable (fun (z : CSpace 1) => F q (z 0) * (starRingEnd ℂ) (G q (z 0))) (gaussianMeasure 1)) :
    MeasureTheory.Integrable (fun (z : CSpace d) => ∏ q : Fin d, F q (z q) * (starRingEnd ℂ) (G q (z q))) (gaussianMeasure d) ∧ (gaussianInner (fun (z : CSpace d) => ∏ q : Fin d, F q (z q)) fun (z : CSpace d) => ∏ q : Fin d, G q (z q)) = ∏ q : Fin d, gaussianInner (fun (z : CSpace 1) => F q (z 0)) fun (z : CSpace 1) => G q (z 0)

    Tensor-product factorization over the Gaussian product measure.