Dual of Modulation Spaces with Variable Smoothness and Integrability

Zhu, Hua

doi:https://doi.org/10.1155/2022/4480849

Journal of Function Spaces

On this page

Abstract Introduction Preliminaries Data Availability Conflicts of Interest Acknowledgments References Copyright Related Articles

Research Article | Open Access

Volume 2022 | Article ID 4480849 | https://doi.org/10.1155/2022/4480849

Dual of Modulation Spaces with Variable Smoothness and Integrability

Hua Zhu¹

Academic Editor: Ozgur Ege

Received08 Apr 2022

Accepted06 May 2022

Published29 May 2022

Abstract

In this article, we first give a proof for the denseness of the Schwartz class in the modulation spaces with variable smoothness and integrability. Then, we study the dual spaces of such modulation spaces.

1. Introduction

The modulation spaces were introduced by Feichtinger [1] on a locally compact Abelian group in 1983 through short-time Fourier transform. His original motivation for modulation spaces was to introduce a new theory of function spaces and to offer an alternative to the class of Besov spaces. In recent years, it is gradually recognized that the modulation spaces are very useful for studying time-frequency behavior of functions. Therefore, the modulation spaces, -modulation spaces, and their applications have received a lot of attention and research, such as [2–10] and the references therein. Particularly, in [11–14], Wang and other authors showed that from PDE point of view, the combination of frequency-uniform decomposition operators and Banach function spaces is important in making nonlinear estimates, where is a Banach function space defined on .

On the other hand, function spaces with variable exponents have received extensive attention recently. Even though the study on variable Lebesgue spaces can be traced back to [15, 16] by Orlicz, the modern development started from the article [17] by Kováčik and Rákosník in 1991. In [18], Fan and Zhao obtained the results in [17] again through the method of Musielak-Orlicz spaces. Thereafter, variable Lebesgue and Sobolev spaces have been widely studied (see, for example, [19–23]). In addition, function spaces with variable exponents have a wealth of applications in many fields, such as in fluid dynamics [24], image processing [25], and partial differential equations [26].

The function spaces with variable smoothness and variable integrability were firstly introduced by Diening et al. in [27], where they studied Triebel-Lizorkin spaces with variable exponents . Then, Almeida and Hästö introduced the Besov space with variable smoothness and integrability in [28]. Since then, many articles about these function spaces appeared, such as [29, 30]. In the past few years, many function spaces with variable exponents have appeared, such as Besov-type spaces with variable exponent, Bessel potential spaces with variable exponent, and Hardy spaces with variable exponent (see [31–35]). Recently, we studied the modulation spaces with variable smoothness and integrability and gave some properties about these spaces in [36]. Since the modulation spaces and the function spaces with variable exponents have rich applications, we believe that the modulation spaces with variable exponents will also have many application areas, and we will continue to explore these application areas, especially in partial differential equations and time frequency analysis.

The dual is an important content when we study function spaces; for example, Triebel [37] has obtained duality of the usual Besov spaces and applied it to real interpolation and Sobolev embedding, Izuki [38] has given the duality of Herz spaces with variable exponent and applied it to characterize the above spaces by wavelet expansions. In [8, 12], the dual of modulation spaces was studied, respectively. In [39], Izuki and Noi were concerned with the dual of Triebel-Lizorkin spaces and Besov spaces with variable exponents. In this paper, we will study the dual of modulation spaces with variable smoothness and integrability.

The paper is organized as follows. In Section 2, we review some notions and notations about semimodular spaces and function spaces with variable exponents. In the theories of function spaces, the research on denseness of the Schwartz class has always been an important topic, by which we can obtain many conclusions such as duality of function spaces and boundedness of some operators. Therefore, in Section 3, we study the denseness of the Schwartz class in the modulation spaces with variable smoothness and integrability. In Section 4, we give the dual of modulation spaces with variable exponents.

2. Preliminaries

In this section, we review some notions and conventions and state some basic results. Throughout this article, we let denote constants that are independent of the main parameters involved but whose value may differ from line to line. By , we mean that there exists a positive constant such that . The symbol means that . The symbol for denotes the maximal integer not more than . We also set and . We write and for . It is easy to see that . For any multi-index , we denote , and for , we denote . We also denote the sequence Lebesgue space by and Lebesgue space by for which the norm is written by .

Let be the Schwartz function space and be its strongly topological dual space which is also known as the space of all tempered distributions. For , we define the Fourier transform and the inverse Fourier transform , respectively, by

2.1. Modular Spaces

In what follows, let be a vector space over or . The function spaces studied in this paper fit into the framework semimodular spaces, and we refer to monograph [23] for a detailed exposition of these concepts.

Definition 1. A function is called a semimodular on if it satisfies (i) for all , with (ii) for all implies (iii) is left-continuous on for every

A semimodular is called a modular if implies , and it is called continuous if the mapping is continuous on for every . A semimodular can also be qualified by the term (quasi)convex; that is, for all and , there exists such that where in the convex case and in the quasiconvex case. By semimodular, we can obtain a normed space as follows:

Definition 2. If is a (semi)modular on , then is called a (semi)modular space.

In [11], the authors have proven that the is a (quasi)normed space with the Luxemburg (quasi)norm , where the infimum of the empty set is infinity by definition. The following conclusion can be found in [23], and we omit the proof here.

Theorem 3 (norm-modular unit ball property). Let be a semimodular on and . Then, if and only if . If is continuous, then and are equivalent, so are and ϱ.

2.2. Function Spaces with Variable Exponents

A measurable function is called a variable exponent function if it is bounded away from zero; namely, the range of the is for some . For a measurable function and a measurable set , let and For simplicity, we abbreviate and .

We denote by for the set of all measurable functions such that and denote by for the set of all measurable functions such that .

In order to make the Hardy-Littlewood maximal function bounded in the variable exponent Lebesgue spaces, one need to add some conditions to the variable exponent function, that is, so-called log-Hölder continuity, which was first introduced in [40].

Definition 4. Let . (i)If there exists such thatfor all , then is called locally log-Hölder continuous, abbreviated as . (ii)If is locally log-Hölder continuous and there exists such thatfor all , then is called globally log-Hölder continuous, abbreviated as .

If a variable exponent satisfies , we say that it belongs to the class . The class is defined similarly.

Remark 5. (i)One can notice that all functions always belong to (ii)Let , then if and only if . If satisfies (5), then (iii)We define the conjugate exponent function by the formula If is in , then is also in

We define and we adopt the convention in order that is left-continuous. The variable exponent modular of a measurable function on is defined by

According to Definition 2, one can define the corresponding semimodular space, namely, the variable exponent Lebesgue space which is denoted by , and the Luxemburg (quasi)norm of the is defined by

Now let us recall the mixed Lebesgue sequence space which was introduced by Almeida and Hästö in [28].

Definition 6. Let and be a measurable subset of . The mixed Lebesgue sequence space is the collection of all sequences of -functions such that where with the convention for all .

Remark 7. Let . (i)If , then , and we use the notation(ii)By Proposition 3.3 of [28], if is constant, then we have (iii)In [28], Almeida and Hästö proved that is a quasinorm for all , and is a norm when pointwise or is a constant. In [30], Kempka and Vybíral proved that is a norm if satisfy either for almost every or , and is a constant almost everywhere

To define modulation space with variable exponents, we need some general definitions from the constant exponent case. For , let be the unit cube with the center at ; then, constitutes a decomposition of . Let and be a smooth function satisfying for and for . Let be a translation of : Then, we see that in and for all . If we denote , for , then we have

We denote is nonempty, and for every sequence , one can construct an operator sequence as follows:

are said to be frequency-uniform decomposition operators. Let and ; the modulation space can be defined as

Further details about the frequency-uniform decomposition techniques and their applications to PDE can be found in the book [13] and articles [11, 12, 14].

Definition 8. Let and be the corresponding frequency-uniform decomposition operators. For and , the modulation space with variable smoothness and integrability is defined to be the set of all distributions such that

For above modulation space, we can define the following modular: which can be used to define the norm. In [36], we have shown that the space given by Definition 8 is independent of the choice of and the corresponding frequency-uniform decomposition operators. Thus, we can choose according to our requirements, and we will omit in the notation of the norm and modular.

3. Density

In [28], the authors showed that the maximal function is not a good tool in the variable exponent space ; hence, they used so-called -functions which were also used in [27]. Similarly, in our article, we define the so-called -functions on by with , , and . Note that when and that is independent of . These functions are different from the -functions since we use the uniform decomposition of rather than the dyadic decomposition.

Now let us review some useful results about -functions which have been proven in [36].

Lemma 9 (see [36]). Let and ; then, there exists a positive constant such that for all and , where is the constant from (4) for .

Remark 10. By Lemma 9, we can move the term inside the convolution as follows: which helps us to treat the variable smoothness in many cases.

Lemma 11 (see [36]). Let , for and every sequence of -functions, there exists a constant such that

Remark 12. In some cases, although we need to require that , we can weaken this condition by the following identity:

In [36], we have proven that , and in this section, we will prove that is also dense in . For this purpose, we need the following conclusions as in [41]. The first one is the generalization of Lemma A.6 in [27] and Lemma 3.4 in [36].

Lemma 13. Let , , and . Then, for all and with , there exists a constant such that

Proof. As in the proof of Lemma 3.4 of [36], for , there exists such that , which implies . Then, for and a fixed dyadic cube , when , we have In addition, for and , we have when is large enough, which implies . Since we get Hence, for , we have where depends only on , , and . For any , there exists a such that . Then, we get the desired conclusion.

Definition 14. (i)Let be a compact subset of ; then, we denote the space of all elements with by (ii)Let and be a sequence of compact subsets of ; then, we denote by the space of all sequences in such that and for

Lemma 15. Let , , and be a sequence of compact subsets of such that . If and , then for all , there exists a constant such that

Proof. Let and ; then, for any by Lemmas 9 and 13, we have Therefore, for and , by Lemma 11, we obtain which completes the proof.

For a real number , we denote

Proposition 16. Let , , and be a sequence of compact subsets of such that . If , then for all and , there exists a constant such that

Proof. According to Lemma 15, by the similar argument in the proof of Theorem 4.15 of [41], we have Since then for and , by the same argument in 1.6.3 of [42], we have Thus, In addition, since then together with above inequality, (33) and Lemma 15, we can get the conclusion.

Remark 17. In fact, in the conclusion of the above lemma, the “” in can be replaced by .

Theorem 18 (density). Let and , then is dense in .

Proof. Let , for and , we put where for . Then, we have . In fact, by Proposition 16, we obtain Consequently, when , in which the last limit can be deduced by Lemma 2.2 of [43]. Hence, in when .
Next, we should approximate by some functions in for . Let satisfy and . Then, for any and , we have and Since , we have by Lemma 3.2.8 of [23]. Now, we prove that is an approximation of in . By (38), we have Let , then and . For each with , let ; then, , where . By Remark 17 and the embedding properties of , we get Then, combining (39) and , we obtain Therefore, is dense in .

4. Dual Spaces of

For a quasi-Banach space , we denote the dual space of by . In this section, we show that for and .

Lemma 19. Let , , and be sequences of locally Lebesgue integrable functions satisfying and . Then, we have

The above lemma has been proven in [39]; hence, we omit the proof here.

Proposition 20. Let and . We denote by the collection of all satisfying that there exists such that and If we define then , and is an equivalent norm on .

Proof. Let and write ; then, and which implies . On the other hand, for any , by Proposition 16, we have by which we can deduce the conclusion of the proposition.

Let us define where

Then, we have the following proposition about dual spaces.

Proposition 21. Let and . Then, Moreover, is equivalent to for all , in which

Proof. Firstly, by Lemma 19, we have and for all ,
On the other hand, for any , let us define by where . It follows that , whence Therefore, We assume that for all . Then, for any , let us define as follows: when , we put ; when , we set , where , and set Then, by Proposition 2.21 of [22], we have which implies Thus, by the definition of , we have from which we can get By Proposition 3.5 of [28], we know that is continuous. Therefore, by , we have Then, it is easy to see that Hence, for any , by (61), we have which implies and .

Theorem 22 (dual space). Let and ; then, we have

Proof. Firstly, we prove that For any and , by Lemma 19, we have where . Since is dense in , we obtain .
Now, let us prove that It is easy to see that, for , is an isometric mapping from into a subspace of . Hence, for any , we can regard it as a continuous functional on , which can be extended onto with the same norm. Then, by Proposition 21, for any , we have where and Since for any , we have which implies . Thus, by Proposition 20, we obtain by which we can get

Data Availability

No data were used to support this study.

Conflicts of Interest

The author declares that he has no conflicts of interest.

Acknowledgments

The author is grateful to Professor Jingshi Xu for his valuable discussions about function spaces with variable exponents, and the author would also like to express great thanks to Takahiro Noi for his instructive discussion and for sending his paper [41]. The research is supported by the “Young top-notch talent” Program concerning teaching faculty development of universities affiliated to Beijing Municipal Government (2018-2020).

References

H. G. Feichtinger, “Modulation spaces on locally compact Abelian group, Technical report, University Vienna, 1983,” in Proceedings of International Conference on Wavelet and Applications, pp. 99–140, New Delhi Allied Publishers, New Delhi, 2003.
View at: Google Scholar
Á. Bényi and K. A. Okoudjou, Modulation Spaces, Springer, New York, 2020.
View at: Publisher Site
L. Borup, “Pseudodifferential operators on α-modulation spaces,” Journal of Function Spaces and Applications, vol. 2, no. 2, pp. 107–123, 2004.
View at: Publisher Site | Google Scholar
K. Gröchenig, Foundations of Time-Frequency Analysis, Birkhauser, Boston, 2001.
View at: Publisher Site
W. Guo, D. Fan, H. Wu, and G. Zhao, “Sharpness of some properties of weighted modulation spaces,” Science China. Mathematics, vol. 59, no. 1, pp. 169–190, 2016.
View at: Publisher Site | Google Scholar
A. T. Gürkanli, Ö. Kulak, and A. Sandikçi, “The spaces of bilinear multipliers of weighted Lorentz type modulation spaces,” Georgian Mathematical Journal, vol. 23, no. 3, pp. 351–362, 2016.
View at: Publisher Site | Google Scholar
M. Kobayashi, “Modulation spaces for ,” Journal of function spaces and applications, vol. 4, article 409840, pp. 329–341, 2006.
View at: Google Scholar
M. Kobayashi, “Dual of modulation spaces,” Journal of Function Spaces and Applications, vol. 5, no. 1, pp. 1–8, 2007.
View at: Publisher Site | Google Scholar
M. Sugimoto and N. Tomita, “The dilation property of modulation spaces and their inclusion relation with Besov spaces,” Journal of Functional Analysis, vol. 248, no. 1, pp. 79–106, 2007.
View at: Publisher Site | Google Scholar
J. Toft and P. Wahlberg, “Embeddings of -modulation spaces,” Pliska Studia Mathematica, vol. 21, pp. 25–46, 2012.
View at: Google Scholar
B. Wang and C. Huang, “Frequency-uniform decomposition method for the generalized BO, KdV and NLS equations,” Journal of Difference Equations, vol. 239, no. 1, 250 pages, 2007.
View at: Publisher Site | Google Scholar
B. Wang and H. Hudzik, “The global Cauchy problem for the NLS and NLKG with small rough data,” Journal of Difference Equations, vol. 232, no. 1, 73 pages, 2007.
View at: Publisher Site | Google Scholar
B. Wang, Z. Huo, C. Hao, and Z. Guo, Harmonic Analysis Method for Nonlinear Evolution Equations, I, World Scientific Publishing, Singapore, 2011.
View at: Publisher Site
B. Wang, L. Zhao, and B. Guo, “Isometric decomposition operators, function spaces and applications to nonlinear evolution equations,” Journal of Functional Analysis, vol. 233, no. 1, pp. 1–39, 2006.
View at: Publisher Site | Google Scholar
W. Orlicz, “Über konjugierte Exponentenfolgen,” Studia Mathematica, vol. 3, no. 1, pp. 200–211, 1931.
View at: Publisher Site | Google Scholar
W. Orlicz, “Über eine gewisse Klasse von Räumen vom Typus B,” Bull. Int. Acad. Pol. Ser. A, vol. 8, pp. 207–220, 1932.
View at: Google Scholar
O. Kováčik and J. Rákosník, “On spaces and ,” Czechoslovak mathematical journal, vol. 41, no. 4, pp. 592–618, 1991.
View at: Google Scholar
X. L. Fan and D. Zhao, “On the spaces and ,” Journal of mathematical analysis and applications, vol. 263, no. 2, pp. 424–446, 2001.
View at: Publisher Site | Google Scholar
R. A. Bandaliev, “On an inequality in Lebesgue space with mixed norm and with variable summability exponent,” Mathematical Notes, vol. 84, no. 3-4, pp. 303–313, 2008.
View at: Publisher Site | Google Scholar
O. V. Besov, “Interpolation, embedding, and extension of spaces of functions of variable smooth-ness,” Proceedings of the Steklov Institute of Mathematics, vol. 248, pp. 47–58, 2005.
View at: Google Scholar
D. Cruz-Uribe, L. Diening, and P. Hästö, “The maximal operator on weighted variable Lebesgue spaces,” Applicable Analysis, vol. 14, no. 3, pp. 361–374, 2011.
View at: Publisher Site | Google Scholar
D. Cruz-Uribe and A. Fiorenza, “Variable Lebesgue spaces: foundations and harmonic analysis,” Applied and Numerical Harmonic Analysis, Springer, (Birkhäuser), Heidelberg, 2013.
View at: Publisher Site | Google Scholar
L. Diening, P. Harjulehto, P. Hästö, and M. Ruzicka, Lebesgue and Sobolev Spaces with Variable Exponents, Springer, Heidelberg, vol. 2017, 2011.
M. Ruzicka, “Electrorheological fluids: modeling and mathematical theory,” Lecture Notes in Mathematics, Springer-Verlag, Berlin, vol. 1748, 2000.
View at: Google Scholar
Y. Chen, S. Levine, and R. Rao, “Variable exponent, linear growth functionals in image restoration,” SIAM Journal on Applied Mathematics, vol. 66, no. 4, pp. 1383–1406, 2006.
View at: Publisher Site | Google Scholar
Y. Fu and X. Zhang, Variable Exponent Function Spaces and Their Appications in Partial Differential Equations, Scientific Press, Beijing, 2011.
L. Diening, P. Hästö, and S. Roudenko, “Function spaces of variable smoothness and integrability,” Journal of Functional Analysis, vol. 256, no. 6, pp. 1731–1768, 2009.
View at: Publisher Site | Google Scholar
A. Almeida and P. Hästö, “Besov spaces with variable smoothness and integrability,” Journal of Functional Analysis, vol. 258, no. 5, pp. 1628–1655, 2010.
View at: Publisher Site | Google Scholar
D. Drihem, “Atomic decomposition of Besov spaces with variable smoothness and integrability,” Journal of Mathematical Analysis and Applications, vol. 389, no. 1, pp. 15–31, 2012.
View at: Publisher Site | Google Scholar
H. Kempka and J. Vybíral, “A note on the spaces of variable integrability and summability of Almeida and Hästö,” Proceedings of the American Mathematical Society, vol. 141, no. 9, pp. 3207–3212, 2013.
View at: Publisher Site | Google Scholar
J. Liu, D. Yang, and W. Yuan, “Anisotropic variable Hardy-Lorentz spaces and their real interpolation,” Journal of Mathematical Analysis and Applications, vol. 456, no. 1, pp. 356–393, 2017.
View at: Publisher Site | Google Scholar
E. Nakai and Y. Sawano, “Hardy spaces with variable exponents and generalized Campanato spaces,” Journal of Functional Analysis, vol. 262, no. 9, pp. 3665–3748, 2012.
View at: Publisher Site | Google Scholar
J. Xu, “An atomic decomposition of variable Besov and Triebel-Lizorkin spaces,” Armenian Journal of Mathematics, vol. 2, no. 1, pp. 1–12, 2009.
View at: Google Scholar
X. Yan, D. Yang, W. Yuan, and C. Zhuo, “Variable weak Hardy spaces and their applications,” Journal of Functional Analysis, vol. 271, no. 10, pp. 2822–2887, 2016.
View at: Publisher Site | Google Scholar
D. Yang, C. Zhuo, and W. Yuan, “Besov-type spaces with variable smoothness and integrability,” Journal of Functional Analysis, vol. 269, no. 6, pp. 1840–1898, 2015.
View at: Publisher Site | Google Scholar
H. Zhu and L. Tang, “Modulation spaces with variable smoothness and integrability, submitted”.
View at: Google Scholar
H. Triebel, “Spaces of distributions of Besov type on Euclidean n-space. Duality, interpolation,” Arkiv för Matematik, vol. 11, no. 1-2, pp. 13–64, 1973.
View at: Publisher Site | Google Scholar
M. Izuki, “Herz and amalgam spaces with variable exponent, the Haar wavelets and greediness of the wavelet system,” East Journal on Approximations, vol. 15, pp. 87–109, 2009.
View at: Google Scholar
M. Izuki and T. Noi, “Duality of Besov, Triebel-Lizorkin and Herz spaces with variable exponents,” Rendiconti del Circolo Matematico di Palermo, vol. 63, no. 2, pp. 221–245, 2014.
View at: Publisher Site | Google Scholar
I. Sharapudinov, “On the basis property of the Haar system in the space and the principle of localization in the mean,” Matematicheskii Sbornik, vol. 58, no. 1, pp. 279–287, 1987.
View at: Google Scholar
T. Noi, “Fourier multiplier theorems on Triebel-Lizorkin and Besov spaces with variable exponents,” Mathematical Inequalities & Applications, vol. 17, no. 1, pp. 49–74, 2014.
View at: Google Scholar
H. Triebel, Theory of Function Spaces, Birkhäuser, Basel, Boston, 1983.
View at: Publisher Site
A. Almeida and A. Caetano, “Atomic and molecular decompositions in variable exponent 2-microlocal spaces and applications,” Journal of Functional Analysis, vol. 270, no. 5, pp. 1888–1921, 2016.
View at: Publisher Site | Google Scholar

Copyright

Copyright © 2022 Hua Zhu. This is an open access article distributed under the Creative Commons Attribution License, which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly cited.

PDF Download Citation

Download other formats

Order printed copies

Views

209

Downloads

316

Citations