Alphabeta Math
PropositionStatement: Literature-sourcedProof: AI-generated
How statement and proof provenance work

The first chip identifies the source of the statement or construction; the second identifies the source of its local proof or verification.

  • Literature-sourced: the exact statement appears in a cited source; only wording and notation differ.
  • AI-adapted: a semantically identical restatement of literature-sourced material, modulo indexing, notation, and boundary cases adopted by the library.
  • AI-generated: a genuinely novel statement formulated by AI, with no source for the claim itself.

These labels describe origin, not correctness: citations and verification chips remain separate evidence.

Coercivity and bounded trajectory control operator

PDF p. 19: Proof of Theorem 1.4, coercivity and conclusion

Statement

Assume the Axiom of Choice and hypotheses (H), (W), and (UCPw) of Proposition 4.6: conditional adjoint observability. In particular, T>0, Ω is the connected exterior of a smooth compact nontrapping obstacle, w is the specified reference trajectory, and the stated energy-level unique continuation property holds. Let s∈{1,2}, σ=−s, E=HDs(Ω) and F=HDσ(Ω). Regard these complex spectral Hilbert spaces as real Hilbert spaces. Let U(t) be the real-linear adjoint evolution, φ the real control cutoff, and Sw:F→E the backward-control map of Real HUM operator for the trajectory linearization. The pivot pairing ⟨⋅,⋅⟩s,−s is linear in its first variable.

For every v0∈F,

Re⁡⟨Swv0,v0⟩s,−s=∫0T∥φU(t)v0∥F2 dt≥C−1∥v0∥F2,

where C is the observability constant for σ. The bounded real-linear map Sw:F→E is a bijection with bounded inverse. Consequently the map Gw:E×E→C([0,T];E) defined in Real HUM operator for the trajectory linearization is bounded and real-linear and selects a control steering the linearized system between any prescribed endpoints in E.

Facts & Assumptions

[F1]

The construction in Real HUM operator for the trajectory linearization sets A=I−ΔD, Λs=As:E→F, ha(t)=−iΛs−1(φU(t)a), and Swa=za(0), where za(T)=0 and izt+ΔDz+V1z+V2z‾=φha. It supplies bounded real-linear Sw and a bounded path-valued map a↦ha. Once Sw has bounded inverse, its displayed formula for Gw gives the claimed steering and boundedness.

[F2]

By Scale Riesz map and free HUM operator and Positive spectral Hilbert scale: duality and interpolation, Λs−1:F→E is an onto isometry and Re⁡⟨Λs−1f,g⟩s,−s=(f,g)F for the real part of the F inner product. The common spectral core is dense in F. Realification preserves the Hilbert norms by Realification of Hilbert pairings and conjugate-linear coefficients.

[F3]

The forward and adjoint evolutions are uniformly bounded and strongly continuous on the relevant scale levels: for each fixed datum, its orbit is norm-continuous in that scale space Forward and backward real-linear propagators. The potential operators are norm-continuous bounded operators on those levels, with negative-level actions defined by pivot duality Time-dependent potential and conjugation bounds on the scale. The real cutoff is bounded on both E and F and its negative-level action is the pivot transpose of its positive-level action Cutoff multipliers through order two and by duality. The mild solutions obey the oriented Duhamel formula of Bounded real-linear perturbations preserve solution estimates.

[F4]

For strong L2 solutions of the linearized and adjoint systems, Real pairing cancellation for the linearized-adjoint pair gives ddtRe⁡⟨z(t),v(t)⟩2=Re⁡⟨−iφh(t),v(t)⟩2.

[F5]

For σ∈{−1,−2}, Proposition 4.6: conditional adjoint observability gives ∫0T∥φU(t)a∥F2dt≥C−1∥a∥F2. Under Countable Choice, the observation Gramian of a bounded real-linear observation with this lower bound is a bounded isomorphism Observation lower bounds give Hilbert control right inverses. Full Choice is assumed The Axiom of Choice.

[F6]

The time-space definition Continuous and mixed time-space function spaces gives Z=L2(0,T;F) as the real vector space of almost-everywhere classes of strongly measurable F-valued functions with norm ∥u∥Z=(∫0T∥u(t)∥F2dt)1/2. Sums and scalar multiples are strongly measurable and their norms are measurable. Under Countable Choice, Bochner Lp is complete for a Banach value space makes this norm complete for every real Banach value space, with p=2; no separability hypothesis is required. The assumed Axiom of Choice supplies Countable Choice.

[F7]

Pointwise Cauchy–Schwarz bounds ∣(x,y)F∣ by ∥x∥F∥y∥F Cauchy–Schwarz: |⟨ x,y⟩|≤‖x‖ ‖y‖, with equality exactly for dependent pairs. Scalar Hölder with exponents 2,2 bounds ∫0Tfg by ∥f∥2∥g∥2 for nonnegative measurable f,g∈L2(0,T) Holder's inequality for integrals, including the endpoint cases.

Proof

Given: The hypotheses and constructions in the Statement, with s∈{1,2}.

Goal: Establish the pairing identity and coercivity, invert Sw, and obtain the bounded endpoint-control selection.

1.1F1F2F3F4

First take a,b in the common spectral core and use the control ha in the backward equation for za. The data and control are in HD2: a∈HD2, multiplication by φ preserves that space, and Λs−1 increases spectral order. The Duhamel equations, the free group on D(ΔD)=HD2, and norm-continuity of the potential operators show that za and U(⋅)b are continuous in HD2 and continuously differentiable in L2. Indeed the Duhamel integrands are continuous in HD2; differentiating the free-group convolution in L2 gives the displayed differential equations. Thus the strong pairing identity applies.

1.2F1F2F3F4step 1.1

Integrating that identity from 0 to T, using za(T)=0, yields Re⁡⟨Swa,b⟩2=∫0TRe⁡⟨iφha(t),U(t)b⟩2dt. Since iha=Λs−1(φU(t)a) and φ is real, the cutoff transpose identity and the scale Riesz identity turn the integrand into (φU(t)a,φU(t)b)F. Both sides are continuous real-bilinear functions of (a,b)∈F×F: use the boundedness of Sw, the pivot pairing, the uniform evolution and cutoff bounds, and the finite length of the time interval. Density of the spectral core therefore extends the identity to all a,b∈F: Re⁡⟨Swa,b⟩s,−s=∫0T(φU(t)a,φU(t)b)F dt.

1.3F2F5F6F7

Give Z=L2(0,T;F) the real pairing (u,v)Z=∫0T(u(t),v(t))F dt. Since F is a real Hilbert space, it is a real Banach space, so the completeness theorem applies to F, p=2, and the finite interval (0,T) under the assumed Choice. For strongly measurable u,v, the real polarization identity (u(t),v(t))F=(∥u(t)+v(t)∥F2−∥u(t)−v(t)∥F2)/4 makes the integrand measurable. Cauchy–Schwarz in F and scalar Hölder show ∫0T∣(u(t),v(t))F∣dt≤∥u∥Z∥v∥Z, so the pairing is finite and independent of representatives. Pointwise bilinearity and symmetry pass to the integral; its diagonal equals ∥u∥Z2, which vanishes exactly for the zero class because this is the norm in F6. Thus this pairing is an inner product inducing the complete Bochner norm, and Z is a real Hilbert space.

2.1F2F3F5F6step 1.2step 1.3

Define O:F→Z by (Oa)(t)=φU(t)a. Each observation orbit is norm-continuous by F3. The finite-range step functions taking its value at the left endpoint of each interval of the partition into n equal pieces converge pointwise to it as n→∞, so the orbit is strongly measurable. If M=sup⁡0≤t≤T∥U(t)∥F→F and K=∥φ∥F→F, then ∥Oa∥Z≤TKM∥a∥F. Thus O is bounded and real-linear between the real Hilbert spaces F and Z, and observability gives ∥Oa∥≥C−1/2∥a∥F. Let Q=O∗O. The Gramian result makes Q a bounded isomorphism of F. By step 1.2 and the scale pairing identity, ΛsSw=Q: their real inner products with every b∈F coincide. Since Λs:E→F is a bounded isomorphism, Sw=Λs−1Q is a bounded isomorphism from F to E. Taking b=a in step 1.2 gives the displayed identity and coercivity.

3.1F1F3step 2.1∎

The inverse just obtained satisfies the additional hypothesis in the definition of Gw. For endpoints x0,xT∈E, that definition takes a=Sw−1(x0−P(0,T)xT) and Gw(x0,xT)=ha. Its backward controlled solution starts at x0−P(0,T)xT and ends at zero; adding the homogeneous solution with terminal value xT gives initial value x0 and terminal value xT. The boundedness and real linearity of the inverse, propagator and path-valued control map give the corresponding properties of Gw. This proves all conclusions.

Source notes

The original article is cited for attribution of the HUM construction and conclusion. The proof above uses the stated mathematical suppliers; the article's assertion is not used as its own premise. The Choice assumption records the hypotheses of the supplied functional-analytic results.

Depends on

Used by · 2

Dependency tree · 2 levels

Each result is drawn at its shortest distance from this item. Arrows run from prerequisites to the items that use them; this item has a heavier outline. External source nodes link directly to the cited text. The view shows up to two dependency steps; open a prerequisite to continue.

Sources