Benguria–Levitin–Parnovski conjecture

For n≥2n\ge 2, let K⊂RnK\subset\mathbb{R}^n be a centrally symmetric convex body, meaning K=−KK=-K, and let B⊂RnB\subset\mathbb{R}^n be a Euclidean ball satisfying ∣B∣=∣K∣\lvert B\rvert=\lvert K\rvert. Define 1K^(ξ)=∫Ke−i⟨x,ξ⟩ dx\widehat{\mathbf{1}_K}(\xi)=\int_K e^{-i\langle x,\xi\rangle}\,dx and d(K)=inf⁡{∣ξ∣:1K^(ξ)=0}d(K)=\inf\{\lvert\xi\rvert:\widehat{\mathbf{1}_K}(\xi)=0\}. The conjecture asserts that d(K)≤d(B)d(K)\le d(B) for every such KK; equivalently, among centrally symmetric convex bodies of fixed volume, the Euclidean ball maximizes the distance from the origin to the zero set of the Fourier transform of the characteristic function.

References

Progress summary

Refreshed
Claimed solved

A recent unrefereed preprint claims the conjecture is false in every dimension above one, with no maximizing shape in dimensions three and higher.

The conjecture proposed the ball as the extremizing shape for the Fourier-zero isoperimetric problem.

September 2026 preprint

A preprint claims that centrally symmetric polygons with at least twelve sides outperform the disk in the plane, while no maximizer exists in dimensions at least three. If correct, this overturns the proposed ball extremizer in every dimension greater than one.

Current status (as of September 2026): The conjecture is claimed false by an unrefereed preprint; its planar counterexamples and higher-dimensional nonexistence claim remain unverified.

Sources

Solutions 0

No solutions have been posted yet.