Hermite–Biehler and Hurwitz criteria

The Hermite–Biehler theorem characterizes half-plane stability through interlacing of the even and odd parts. The Hurwitz criterion characterizes weak Hurwitz stability by total nonnegativity of the classical Hurwitz matrix.

References

O. Holtz, “Hermite–Biehler, Routh–Hurwitz, and total positivity,” Linear Algebra and its Applications 372 (2003), 105–110. See also the Hermite–Biehler theorem and Hurwitz criterion on symmetricfunctions.com.

Theorems

  • Hermite–Biehler: interlacing gives stability

    RealRooted.Challenges.HermiteBiehlerHurwitz.hermiteBiehler_forwardsource
    theorem hermiteBiehler_forward :
        ∀ {f g : ℝ[X]},
          HasPosLeadingCoeff f →
          HasPosLeadingCoeff g →
          StrictInterl g f →
          IsUpperHalfPlaneStable (hermiteBiehlerPolynomial f g)
  • Hermite–Biehler: stability gives interlacing

    RealRooted.Challenges.HermiteBiehlerHurwitz.hermiteBiehler_conversesource
    theorem hermiteBiehler_converse :
        ∀ ⦃f g : ℝ[X]⦄,
          HasPosLeadingCoeff f →
          HasPosLeadingCoeff g →
          IsUpperHalfPlaneStable (hermiteBiehlerPolynomial f g) →
          StrictInterl g f ∨ StrictInterl f g
  • Hurwitz criterion via total nonnegativity

    RealRooted.Challenges.HermiteBiehlerHurwitz.classicalHurwitzCriterionsource
    theorem classicalHurwitzCriterion {p : ℝ[X]} (hp : p ≠ 0) :
        RealRooted.IsHurwitzStable p ↔
          (Matrix.hurwitz p.coeff).IsTotallyNonneg

Lean source RealRooted/Challenges/HermiteBiehlerHurwitz.lean at revision 5827550b.