Operator-valued Fourier multipliers of bounded s s -variation

Chenxi Deng, Emiel Lorist, Mark Veraar · 2026

Read Operator-valued Fourier multipliers of bounded s s -variation by Chenxi Deng, Emiel Lorist, Mark Veraar (2026) in the Tower of Babel, a free online library of public-domain works preserved in clean, searchable text. No sign-up. The full text is served as a plain-text file at a predictable URL, where text is present: https://antarctic-labs.com/tower-of-babel/arxiv-2601-04803.txt.

Introduction

Chenxi Deng, Emiel Lorist, and Mark Veraar prove a Fourier multiplier theorem for operator-valued functions in a 2026 mathematics paper. Fourier multipliers reshape a signal frequency by frequency, and the operator-valued kind acts on quantities living in Banach spaces, a broad class of measured spaces. Their result works across weighted Lebesgue, Besov, and Triebel-Lizorkin spaces.

The theorem assumes the multiplier's range is well controlled and that its bounded variation over dyadic intervals, doubling frequency bands, meets a summability condition. As the main tool, they prove a weighted variational Carleson inequality and an estimate of Littlewood-Paley-Rubio de Francia type. Such theorems underpin work on equations where solutions live in function spaces.

Catalog record

CREATORChenxi Deng, Emiel Lorist, Mark Veraar
YEAR2026
COLLECTIONPAPERS
FORMATTXT
FILE SIZE70 KB
VERSION1.0
LICENSEhttp://creativecommons.org/licenses/by/4.0/
RIGHTSCC BY
SOURCEarXiv

Source: https://arxiv.org/abs/2601-04803

Browse the full library · About the Tower of Babel

Full text

The opening of the work, from the library's hosted text:

Operator-valued Fourier multipliers of bounded s s -variation Thanks: The second author is supported by the Dutch Research Council (NWO) on the project “The sparse revolution for stochastic partial differential equations” with project number VI.Veni.242.057 . The third author is supported by the VICI subsidy VI.C.212.027 of the Netherlands Organisation for Scientific Research (NWO) Chenxi Deng, Emiel Lorist, Mark Veraar Abstract . In this paper, we establish an operator-valued Fourier multiplier theorem in weighted Lebesgue spaces, Besov and Triebel–Lizorkin spaces, assuming the multiplier has ℛ \mathcal{R} -bounded range and satisfies an ℓ r \ell^{r} -summability condition on its bounded s s -variation seminorms over dyadic intervals. The exponents r r and s s reflect the relationship between the geometric properties of the underlying Banach spaces (type and cotype) and the boundedness of Fourier multiplier operators. As our main tool we prove a weighted vector-valued variational Carleson inequality and deduce an estimate of Littlewood–Paley–Rubio de Francia type. 2020 Mathematics Subject Classification Primary: 42A45, Secondary: 42B25, 42B35, 46B09, 46B20, 46E35, 46E40 1. Introduction Fourier multiplier operators are a central tool in harmonic analysis, with far-reaching applications to partial differential equations, function space theory, numerical analysis and analytic number theory. In this paper we are concerned with an operator-valued version of this theory. Before turning to this setting, we briefly recall some classical results from the scalar-valued case. In the scalar-valued case, a Fourier multiplier operator T_{m} for m\colon{\mathbb{R}}\to{\mathbb{C}} is given by T_{m}f:=\mathcal{F}^{-1}(m\cdot\mathcal{F}(f)),\quad f\in\mathcal{S}({\mathbb{R}}), where \mathcal{F} denotes the Fourier transform. Sufficient conditions for the boundedness of a Fourier multiplier operator T_{m} on L^{p}({\mathbb{R}}) are provided by the classical Fourier multiplier theorems due to Marcinkiewicz [ 51 ] , Mihlin [ 54 , 55 ] and Hörmander [ 31 ] . For the current paper the formulation of Marcinkiewicz’ multiplier theorem is the most natural starting point. Here and below \Delta is the dyadic interval partition of {\mathbb{R}} . Theorem 1.1 (Marcinkiewicz [ 51 ] ) . Let p\in(1,\infty) and let m\colon{\mathbb{R}}\to{\mathbb{C}} such that for J\in\Delta , m|_{J}\in V^{1}(J) and \displaystyle\lVert m\rVert_{\ell^{\infty}(V^{1}(\Delta))} \displaystyle:=\sup_{J\in\Delta}\lVert m|_{J}\rVert_{V^{1}(J)}<\infty, where V^{1}(J) is the space of functions of bounded variation. Then T_{m} is bounded on L^{p}({\mathbb{R}}) . In [ 15 ] , Coifman, Rubio de Francia and Semmes relaxed the assumption of bounded variation in Theorem 1.1 to bounded s -variation for some s>1 (see Section 2.5 below). Bounded s -variation is weaker for larger s and is implied by \frac{1}{s} -Hölder smoothness. Coifman et al. proved their result for the boundedness of T_{m} on L^{p}({\mathbb{R}}) for m\in\ell^{\infty}(V^{s}(\Delta)) in three steps: (1) First, they considered the case p\in[2,\infty) and s\in[1,2) , using the extension of the Littlewood–Paley inequality of Rubio de Francia to arbitrary intervals of [ 18 ] : \displaystyle\Big\|\Big(\sum_{I\in\mathcal{I}}|S_{I}f|^{2}\Big)^{1/2}\Big\|_{L^{p}({\mathbb{R}})}\leq C_{p}\|f\|_{L^{p}({\mathbb{R}})}, (1.1) where p\in[2,\infty) , \mathcal{I} is a family of disjoint intervals in {\mathbb{R}} and S_{I}=T_{m} for m:={{\bf 1}}_{I} . (2) Secondly, the case p\in(1,2) and s\in[1,2) follows by duality. (3) Finally, note that the map (m,f)\mapsto T_{m}f is bounded from L^{\infty}({\mathbb{R}})\times L^{2}({\mathbb{R}}) to L^{2}({\mathbb{R}}) by the Plancherel theorem. Hence, using bilinear interpolation they concluded that T_{m} is bounded on L^{p}({\mathbb{R}}) if m\in\ell^{\infty}(V^{s}(\Delta)) with \frac{1}{s}>\lvert\frac{1}{p}-\frac{1}{2}\rvert for p,s\in(1,\infty) . Related multiplier results have been studied in [ 7 , 44 , 45 ] . 1.1. Fourier multipliers in the vector-valued setting Now let X and Y be Banach spaces. An operator-valued Fourier multiplier operator T_{m} is of the form T_{m}f=\mathcal{F}^{-1}(m\cdot\mathcal{F}(f)),\quad f\in\mathcal{S}({\mathbb{R}};X), where m\colon{\mathbb{R}}\to\mathcal{L}(X,Y) , the space \mathcal{S}({\mathbb{R}};X) denotes the X -valued Schwartz functions and \mathcal{F} denotes the Fourier transform. The classical multiplier theorems due to Marcinkiewicz and Mihlin were extended to \UMD spaces in two stages. The case X=Y with scalar-valued multipliers m was first addressed in the 1980s by McConnell [ 52 ] and Bourgain [ 9 ] , and extended by Zimmermann [ 71 ] (see also [ 37 ] ). Subsequently, in the early 2000s, the more general setting with X\neq Y and operator-valued multipliers m was developed and was motivated by applications to evolution equations such as maximal L^{p} -regularity and stability of semigroups (see [ 68 ] ). An operator-valued analogue of the Mihlin multiplier theorem was proven by Weis [ 69 ] and applied to maximal L^{p} -regularity, and shortly after an analogue of the Marcinkiewicz multiplier theorem by Štrkalj and Weis [ 64 ] . These theorems require a so-called \mathcal{R} -boundedness condition (see [ 39 , Chapter 8] ) on the range of m , a strengthening of uniform boundedness that was shown to be necessary by Clément and Prüss [ 14 ] . The reader is referred to the Notes of Chapter 8 in [ 39 ] for further references. This naturally leads to the question whether the aforementioned result of Coifman et al. [ 15 ] admits an extension to the operator-valued setting. This programme was initiated by Hytönen and Potapov [ 35 ] , where the three main steps mentioned above are followed. (1) The first step in the operator-valued setting builds on two hypotheses: the range of m is \mathcal{R} -bounded and the availability of an X -valued version of the Littlewood–Paley–Rubio de Francia estimate ( 1.1 ), introduced by Berkson, Gillespie and Torrea [ 6 ] and called the \LPR_{p} -property of X . This estimate was further studied in [ 3 , 26 , 42 , 58 ] . At present, the \LPR_{p} -property is only known to hold for Banach function spaces satisfying additional geometric requirements, such as the requirement that the 2 -concavification X^{2} is UMD (see [ 3 , 58 ] ). It is an open problem whether, e.g., the Schatten class S^{t} with t\in[2,\infty) has the \LPR_{p} -property. (2) In the second step one can still employ a duality argument. (3) In the third step, Hytönen and Potapov [ 35 ] additionally require that X is a complex interpolation space between a Banach space X_{0} with the \LPR_{p} -property and a Hilbert space H . Furthermore, the \mathcal{R} -boundedness condition has to be replaced by a more intricate condition: the existence of an \mathcal{R} -bounded family \mathcal{T}\subseteq{\mathcal{L}}(X_{0}\cap H) such that \Ran(m)\subseteq\mathrm{span}(\mathcal{T}) and m has bounded s -variation in the Minkowski functional with respect to \mathcal{T} . This requirement can be quite delicate to check. A further generalization of this approach for Banach function spaces was obtained in [ 2 ] . 1.2. Special case of our main result The goal of this paper is to prove a weighted operator-valued Fourier multiplier theorem for symbols m of bounded s -variation with s>1 , where s is related to the geometry of the underlying Banach spaces. We aim for assumptions on the underlying Banach spaces that can be verified beyond the class of Banach function spaces and, at the same time, avoid the intricate operator-theoretic conditions arising from the interpolation argument sketched above. To do so, we will introduce a different type of Rubio de Francia property. A simplified version of our main result reads as follows. A space X is called \theta -intermediate \UMD if it can be written as X=[X_{0},H]_{\theta} for some \theta\in(0,1] . We will use the class A_{p} of Muckenhoupt weights for p\in(1,\infty) . Theorem 1.2 . Let \theta\in(0,1] , t\in(1,2] , q\in[2,\infty) and set \frac{1}{r}:=\frac{1}{t}-\frac{1}{q} . Let X be

Read the complete text