A new mathematical result says that, for a class of wave equations with a critically singular potential at the boundary, the initial energy of an eligible Dirichlet solution can be estimated from the Neumann boundary trace integrated over the observed boundary and time interval. In ordinary language, the theorem links information at the edge of a domain to the amount of mathematical energy present at the start. The central statement is an observability inequality: the Neumann boundary trace can be used to bound that starting energy when the equation, domain and solution meet the paper's conditions. The result is not a claim that the system can be controlled from the boundary.
The work targets boundary observability for critically singular wave equations on general domains and allows general lower-order coefficients, including time dependence that need not be analytic. The question is whether the boundary trace can bound information encoded in the solution's initial energy under those assumptions.
A theorem with a sharply defined scope
The formal setting uses a bounded open domain with a C4 boundary, meaning the boundary has the four levels of regularity specified in the paper's notation. The spatial dimension n is at least 1, and the equation is considered over a finite time interval centered at zero. These conditions define the class of domains and equations to which the estimate applies.
At the heart of the model is a critically singular potential at the boundary. Its strength is represented by a unique parameter q between -1 and 0, and the paper links q to the singular strength sigma by assigning sigma the value q(2-q) divided by four. The established theorem stays in the negative-sigma regime. The opposite-sign regime is discussed as a separate question, not as a result established here.
That singular boundary behavior is handled through specialized Dirichlet and Neumann traces. In the paper's formulation, solution behavior near the boundary is expressed through powers of the distance to the boundary, and the Neumann trace is the boundary quantity used in the observation.
The machinery behind the estimate
The main proof tool is a global Carleman estimate, a weighted inequality for the singular wave operator that is designed to capture the Neumann boundary trace on the admissible domains. The paper derives an integrated version for boundary-admissible functions supported away from the temporal endpoints, using a convex boundary-defining pair and sufficiently large Carleman parameters.
The weight has two exponential layers and two independent large parameters. The authors describe added positivity for derivatives in the y direction as an advantage of the construction. A later two-foliation step addresses negative terms by applying the pointwise estimate to two different convex boundary-defining functions and summing their contributions.
The geometric assumption is more specific than smoothness alone. The theorem requires a convex boundary-defining function that is positive inside the domain, agrees with boundary distance near the edge, has one critical point, and is convex in directions tangent to its level sets. This is the global convexity condition built into the result.
The geometry sets the clock
The proof also sets a lower bound for the observation-time parameter. It requires T to exceed mu times b raised to the power 1+q, divided by 1+q. Here, mu is a Carleman parameter, b is the supremum of the defining function over the domain, and q is the singularity parameter. The threshold is therefore tied to the proof's parameters and to the size and geometry of the domain, rather than given as one universal numerical value.
Under these assumptions, the observability theorem says that the initial mathematical energy of an eligible Dirichlet solution is bounded by the squared Neumann trace integrated over the observed boundary-time cylinder. The constant in the inequality depends on the spatial dimension, the observation time, the domain, and the coefficients denoted in the paper by X and V. The conclusion is conditional on those assumptions and on the stated class of solutions.
Where the claim stops
The authors explicitly draw a line between observability and boundary controllability. They say the estimate does not imply boundary controllability. A duality route would additionally require a hidden-regularity estimate, which they expect to be false in this setting. The theorem therefore stops at an energy-to-boundary-trace inequality.
That limitation also narrows the geometric and parameter claims. The paper does not establish observability for arbitrary bounded domains without the convexity condition, and it does not settle the opposite-sign singularity regime. The time requirement remains construction-dependent, with no universal numerical calibration reported. As a proof-based result, it reports a conditional inequality rather than a statistical estimate.
By the authors' account, and to their best knowledge, the work is the first boundary observability estimate for this equation on general domains and in all dimensions when general time-dependent lower-order terms are allowed. That statement is the authors' interpretation of the work's novelty.
The document is identified as arXiv:2608.28132v1 in math.AP and dated 28 Aug 2026. It is a preprint, and no journal venue is reported in the supplied metadata.
Paper data and sources
Original title: Observability for Wave Equations with Critically Singular Potentials
Authors: Vaibhav Kumar Jena, Arick Shao
Journal/Repository: arXiv
Status: Preprint, not yet peer-reviewed
First online: 2026-08-28
DOI: Not available
Original paper · Full text