Diamond Maximal Operators: Classification, Moduli, and Oscillation Envelopes
Mohammadreza Mohajer
Source abstract
We study maximal averages generated by bounded cancellative kernels taking values in finite-dimensional real normed spaces, on Euclidean balls and on balls in a non-Archimedean local field with a fixed uniformizer. For angular kernels, pointwise domination of the full maximal operators is equivalent to factorization of their angular averaging maps, with the same optimal constant. Nonzero scalar angular comparison classes are therefore projective kernel classes. For angular kernels constant on a fixed finite measurable partition of the direction space, symmetric convex bodies classify exact norm measurements, Grassmannians parametrize nonzero classes under mutual domination, and positive semidefinite Gram matrices classify Hilbert-valued measurements. For general kernels, the closed convex hull of the averaging densities obtained from all scales and the dual unit ball is a complete invariant of the full maximal operator. A dilation example exhibits identifications of residue kernels beyond scaling and unit-coordinate changes. For nonzero radial residue kernels on bounded inputs, mutual domination is classified by the unit-circle zeros of filter polynomials, counted with multiplicity, with optimal constants given by the norms of convolution quotients. We identify the envelope over all normalized bounded cancellative kernels with the best local approximation by constants and prove quantitative convergence of residue envelopes as the resolution increases. Finally, the weights of the finite Fourier transform classify the one-level energy and the associated Carleson functional. This functional is equivalent to the seminorm precisely when every nontrivial additive character of the residue field is detected, and the weights determine the optimal comparison constants.
Evidence graph
No public relationships recorded yet.
Integrity note: This page is a factual metadata record created by deterministic ingestion. It is not a claim that the work moves a mathematical frontier or has been independently verified.