Harmonic Analysis Lecture Notes University of Illinois at Urbana{Champaign Richard S. Laugesen * January 9, 2009 *Copyright °c 2009, Richard S. Laugesen ([email protected]). This work is licensed under the Creative Commons Attribution{Noncommercial{Share Alike 3.0 Unported License. To view a copy of this license, visit http://creativecommons. org/licenses/by-nc-sa/3.0/. 2 Preface A textbook presents more than any professor can cover in class. In contrast, these lecture notes present exactly* what I covered in Harmonic Analysis (Math 545) at the University of Illinois, Urbana{Champaign, in Fall 2008. The ¯rst part of the course emphasizes Fourier series, since so many aspects of harmonic analysis arise already in that classical context. The Hilbert transform is treated on the circle, for example, where it is used to prove Lp convergence of Fourier series. Maximal functions and Calder¶on{ Zygmund decompositions are treated in Rd, so that they can be applied again in the second part of the course, where the Fourier transform is studied. Real methods are used throughout. In particular, complex methods such as Poisson integrals and conjugate functions are not used to prove bounded- ness of the Hilbert transform. Distribution functions and interpolation are covered in the Appendices. I inserted these topics at the appropriate places in my lectures (after Chap- ters 4 and 12, respectively). The references at the beginning of each chapter provide guidance to stu- dents who wish to delve more deeply, or roam more widely, in the subject. Those references do not necessarily contain all the material in the chapter. Finally, a word on personal taste: while I appreciate a good counterex- ample, I prefer spending class time on positive results. Thus I do not supply proofs of some prominent counterexamples (such as Kolmogorov's integrable function whose Fourier series diverges at every point). I am grateful to Noel DeJarnette, Eunmi Kim, Aleksandra Kwiatkowska, Kostya Slutsky, Khang Tran and Ping Xu for TEXing parts of the document. Please email me with corrections, and with suggested improvements of any kind. Richard S. Laugesen Email: [email protected] Department of Mathematics University of Illinois Urbana, IL 61801 U.S.A. *modulo some improvements after the fact 3 Introduction Harmonic analysis began with Fourier's e®ort to analyze (extract informa- tion from) and synthesize (reconstruct) the solutions of the heat and wave equations, in terms of harmonics. Speci¯cally, the computation of Fourier coe±cients is analysis, while writing down the Fourier series is synthesis, and the harmonics in one dimension are sin(nt) and cos(nt). Immediately one asks: does the Fourier series converge? to the original function? In what sense does it converge: pointwise? mean-square? Lp? Do analogous results hold on Rd for the Fourier transform? We will answer these classical qualitative questions (and more!) using modern quantitative estimates, involving tools such as summability meth- ods (convolution), maximal operators, singular integrals and interpolation. These topics, which we address for both Fourier series and transforms, con- stitute the theoretical core of the course. We further cover the sampling theorem, Poisson summation formula and uncertainty principles. This graduate course is theoretical in nature. Students who are intrigued by the fascinating applications of Fourier series and transforms are advised to browse [Dym and McKean], [KÄorner]and [Stein and Shakarchi], which are all wonderfully engaging books. If more time (or a second semester) were available, I might cover ad- ditional topics such as: Littlewood{Paley theory for Fourier series and in- tegrals, Fourier analysis on locally compact abelian groups [Rudin] (espe- cially Bochner's theorem on Fourier transforms of nonnegative functions), short-time Fourier transforms [GrÄochenig], discrete Fourier transforms, the Schwartz class and tempered distributions and applications in Fourier analy- sis [Strichartz], Fourier integral operators (including solutions of the wave and SchrÄodingerequations), Radon transforms, and some topics related to signal processing, such as maximum entropy, spectral estimation and predic- tion [Benedetto]. I might also cover multiplier theorems, ergodic theorems, and almost periodic functions. 4 Contents I Fourier series 7 1 Fourier coe±cients: basic properties 9 2 Fourier series: summability in norm 15 3 Fourier series: summability at a point 25 4 Fourier coe±cients in `1(Z) (or, f 2 A(T)) 27 5 Fourier coe±cients in `2(Z) (or, f 2 L2(T)) 31 6 Maximal functions 35 7 Fourier summability pointwise a.e. 43 8 Fourier series: convergence at a point 47 9 Fourier series: norm convergence 53 10 Hilbert transform on L2(T) 57 11 Calder¶on{Zygmund decompositions 61 12 Hilbert transform on Lp(T) 67 13 Applications of interpolation 71 5 6 CONTENTS II Fourier integrals 75 14 Fourier transforms: basic properties 79 15 Fourier integrals: summability in norm 87 16 Fourier inversion when fb 2 L1(Rd) 95 17 Fourier transforms in L2(Rd) 97 18 Fourier integrals: summability a.e. 101 19 Fourier integrals: norm convergence 107 20 Hilbert and Riesz transforms on L2(Rd) 113 21 Hilbert and Riesz transforms on Lp(Rd) 123 III Fourier series and integrals 127 22 Band limited functions 129 23 Periodization and Poisson summation 135 24 Uncertainty principles 141 IV Problems 147 V Appendices 159 A Minkowski's integral inequality 161 B Lp norms and the distribution function 163 C Interpolation 165 Part I Fourier series 7 Chapter 1 Fourier coe±cients: basic properties Goal Derive basic properties of Fourier coe±cients Reference [Katznelson] Section I.1 Notation T = R=2¼Z is the one dimensional torus Lp(T) = fcomplex-valued, p-th power integrable, 2¼-periodic functionsg ¡ R ¢ R 1 p 1=p kfkLp(T) = 2¼ T jf(t)j dt where T can be taken over any interval of length 2¼ Nesting of Lp-spaces: L1(T) ½ L2(T) ½ L1(T) C(T) = fcomplex-valued, continuous, 2¼-periodic functionsg, Banach space with norm k¢kL1(T) PN int Trigonometric polynomial P (t) = n=¡N ane Translation f¿ (t) = f(t ¡ ¿) 9 10 CHAPTER 1. FOURIER COEFFICIENTS: BASIC PROPERTIES De¯nition 1.1. For f 2 L1(T) and n 2 Z, de¯ne fb(n) = n-th Fourier coe±cient of f Z 1 = f(t)e¡int dt: (1.1) 2¼ T P The formal series S[f] = fb(n)eint is the Fourier series of f. R 2 b int 1 Aside. For f 2 L (T), note f(n) = hf; e i where hf; gi = 2¼ T f(t)g(t) dt is that L2 inner product. Thus fb(n) =amplitude of f in direction of eint. See Chapter 5. Theorem 1.2 (Basic properties). Let f; g 2 L1(T); j; n 2 Z; c 2 C;¿ 2 T. Linearity (\f + g)(n) = fb(n) + gb(n) and (dcf)(n) = cfb(n) Conjugation fb(n) = fb(¡n) P N int b Trigonometric polynomial P (t) = n=¡N ane has P (n) = an for jnj · N and Pb(n) = 0 for jnj > N c ¡in¿ b b takes translation to modulation, f¿ (n) = e f(n) b takes modulation to translation, [f(t)eijt]b(n) = fb(n ¡ j) 1 1 b b : L (T) ! ` (Z) is bounded, with jf(n)j · kfkL1(T) 1 c b Hence if fm ! f in L (T) then fm(n) ! f(n) (uniformly in n) as m ! 1. Proof. Exercise. Lemma 1.3 (Di®erence formula). For n 6= 0, Z 1 fb(n) = [f(t) ¡ f(t ¡ ¼=n)] e¡int dt: 4¼ T Proof. Z 1 fb(n) = ¡ f(t)e¡in(t+¼=n) dt since e¡i¼ = ¡1 2¼ ZT 1 = ¡ f(t ¡ ¼=n)e¡int dt (1.2) 2¼ T by t 7! t ¡ ¼=n and periodicity. By (1.2) and the de¯nition (1.1), Z 1 1 1 fb(n) = fb(n) + fb(n) = [f(t) ¡ f(t ¡ ¼=n)] e¡int dt: 2 2 4¼ T 11 Lemma 1.4 (Continuity of translation). Fix f 2 Lp(T); 1 · p < 1. The map Á : T ! Lp(T) ¿ 7! f¿ is continuous. Proof. Let ¿0 2 T. Take g 2 C(T) and observe p p p p kf¿ ¡ f¿0 kL (T) · kf¿ ¡ g¿ kL (T) + kg¿ ¡ g¿0 kL (T) + kg¿0 ¡ f¿0 kL (T) p p = 2kf ¡ gkL (T) + kg¿ ¡ g¿0 kL (T) ! 2kf ¡ gkLp(T) as ¿ ! ¿0, by uniform continuity of g. By density of continuous functions in Lp(T); 1 · p < 1, the di®erence f ¡ g can be made arbitrarily small. Hence p lim sup¿!¿0 kf¿ ¡ f¿0 kL (T) = 0, as desired. Corollary 1.5 (Riemann{Lebesgue lemma). fb(n) ! 0 as jnj ! 1. Proof. Lemma 1.3 implies 1 jfb(n)j · kf ¡ f k 1 ; 2 ¼=n L (T) which tends to zero as jnj ! 1 by the L1-continuity of translation in Lemma 1.4, since f = f0. Smoothness and decay The Riemann{Lebesgue lemma says fb(n) = o(1), with fb(n) = O(1) explicitly by Theorem 1.2. We show the smoother f is, the faster its Fourier coe±cients decay. Theorem 1.6 (Less than one derivative). If f 2 C®(T); 0 < ® · 1, then fb(n) = O(1=jnj®). Here C®(T) denotes the HÄoldercontinuous functions: f 2 C®(T) if f 2 C(T) and there exists A > 0 such that jf(t) ¡ f(¿)j · Ajt ¡ ¿j® whenever jt ¡ ¿j · 2¼. 12 CHAPTER 1. FOURIER COEFFICIENTS: BASIC PROPERTIES Proof. Z 1 fb(n) = [f(t) ¡ f(t ¡ ¼=n)]e¡int dt 4¼ T by the Di®erence Formula in Lemma 1.3. Therefore ¯ ¯ 1 ¯¼ ¯® const. jfb(n)j · A ¯ ¯ 2¼ = : 4¼ n jnj® Theorem 1.7 (One derivative).