Answers are grounded in this paper and cite their sources.
For the Schrödinger operator −Δ+q in ℝ3 we consider a mixed reconstruction problem: the Fourier transform of the potential at frequencies |ξ| ≤ 2K is taken from the scattering amplitude at the single energy K2, and the frequencies |ξ| > 2K are taken from the potential itself. We prove explicit, dimensionally homogeneous bounds for the error in terms of regularity constants of q: of order K−1/4 for q ∈ C4 and of order K−1 for q ∈ C5 with decay (1+|x|)−3−η, under the condition V(q) = supx (4π)−1∫|q(y)||x−y|−1dy < 1. The power K−1 is attained. We show that the imaginary part of the reconstruction is given exactly by the unitarity defect of the data, provided the parametrisation of frequencies respects the symmetry ξ ↦ −ξ. Numerically, the real part then converges like K−2: for a strong radial potential (exact phase-shift computation) and, to second order in q, for a non-radial potential; with a parametrisation violating the symmetry the real part converges only like K−1. The bounds are uniform along families of potentials with bounded regularity constants, which is the setting needed for potentials generated along trajectories of evolution equations.
No takes yet. Share an insight, caveat, or question.
Asset A. Durmagambetov (2026) studied this question.
Synapse has enriched 8 closely related papers on similar clinical questions. Consider them for comparative context: