The discrete Fourier-transform feasibility conjecture

About 6 years old · traced to

Let s∈{+,−}s\in\{+,-\}, let q⩾1q\geqslant 1, and let Asdisc(q)\mathbb{A}_s^{\textup{disc}}(q) be the discrete feasibility constant defined for even functions on Z2q+1\mathbb{Z}_{2q+1}. A pair (k,q)(k,q) is ss-feasible when there is an admissible function with ksf⩽kk_{sf}\leqslant k.

Discrete Fourier-transform feasibility conjecture. If (k,q)(k,q) is ss-feasible, then (k+1,q)(k+1,q) and (k,q−1)(k,q-1) are ss-feasible. The function q↦Asdisc(q)q\mapsto\mathbb{A}_s^{\textup{disc}}(q) is non-decreasing, its range contains all integers k⩾2k\geqslant2 for s=+1s=+1 and all integers k⩾3k\geqslant3 for s=−1s=-1, and

lim⁡q→∞Asdisc(q)2q+1=As(1),\lim_{q\to\infty}\frac{\mathbb{A}_s^{\textup{disc}}(q)}{\sqrt{2q+1}}=\mathbb{A}_s(1),

where As(1)\mathbb{A}_s(1) is the corresponding continuous one-dimensional constant.

This conjecture asserts that the discrete feasibility patterns converge to the continuous sign uncertainty principles. The source reports numerical evidence but no proof.

References

Primary source

Felipe Gonçalves, Diogo Oliveira e Silva and João P. G. Ramos, “New Sign Uncertainty Principles”, arXiv:2003.10771 (2023).

Progress summary

Never refreshed

Nothing recorded yet. Refresh searches the literature and the public web for attempts on this problem, and writes the first summary here.

Solutions 0

No solutions have been posted yet.