Is the Fourier transform of a continuous function necessarily in $L^{infty}$?












4














This is a followup to this earlier question. Given a (let's say continuous) function $f:mathbb{R} to mathbb{C}$ such that the Fourier Transform
$$widehat{f}(y)=int_{mathbb{R}} f(x)e^{-2pi i xy} dx$$
exists for all $y in mathbb{R}$, does it follow that $widehat{f} in L^{infty}$ i.e. there is a constant $C$ (depending on $f$) such that $vert widehat{f}(y)vert le C$ for all $y$?



This would trivially be true if $f in L^1$, but we are not assuming that. This of course also means that the integrals should better not be absolutely convergent. Instead we require convergence only in the sense that both $lim_{Tto infty} int_0^T$ and $lim_{T to infty} int_{-T}^0$ exist.










share|cite|improve this question


















  • 1




    I think the answer is negative. By the closed graph theorems if every continuous function $f$ would have $hat{f}$ in $L^infty$ that would mean that the Fourier transform is a bounded map between $CB(mathbb R)$, the bounded continuous functions and $L^infty(mathbb R)$. This will imply that $| hat{f} |_infty leq C , | f |_infty$. You can verify that there is no such $C$ by taking Gaussian $e^{t x^2}$, whose Fourier transforms have norm $t^{-1/2}$ despite the original function being bounded.
    – Adrián González-Pérez
    Nov 29 '18 at 12:37












  • I guess that the largest class of Function that satisfy what you require are those $g$ such that $hat{g} in H^infty(mathbb{R})$, the real hardy space. But I am not sure.
    – Adrián González-Pérez
    Nov 29 '18 at 12:40












  • Choose $h$ increasing to $1$ and satisfying $int_0^infty ysin(e^y)h(y)dy=0$. Then look at $f(x) =int_0^infty e^{2i pi xy} y sin(e^y)h(y)$ and try applying the Fourier inversion theorem.
    – reuns
    Nov 29 '18 at 20:18


















4














This is a followup to this earlier question. Given a (let's say continuous) function $f:mathbb{R} to mathbb{C}$ such that the Fourier Transform
$$widehat{f}(y)=int_{mathbb{R}} f(x)e^{-2pi i xy} dx$$
exists for all $y in mathbb{R}$, does it follow that $widehat{f} in L^{infty}$ i.e. there is a constant $C$ (depending on $f$) such that $vert widehat{f}(y)vert le C$ for all $y$?



This would trivially be true if $f in L^1$, but we are not assuming that. This of course also means that the integrals should better not be absolutely convergent. Instead we require convergence only in the sense that both $lim_{Tto infty} int_0^T$ and $lim_{T to infty} int_{-T}^0$ exist.










share|cite|improve this question


















  • 1




    I think the answer is negative. By the closed graph theorems if every continuous function $f$ would have $hat{f}$ in $L^infty$ that would mean that the Fourier transform is a bounded map between $CB(mathbb R)$, the bounded continuous functions and $L^infty(mathbb R)$. This will imply that $| hat{f} |_infty leq C , | f |_infty$. You can verify that there is no such $C$ by taking Gaussian $e^{t x^2}$, whose Fourier transforms have norm $t^{-1/2}$ despite the original function being bounded.
    – Adrián González-Pérez
    Nov 29 '18 at 12:37












  • I guess that the largest class of Function that satisfy what you require are those $g$ such that $hat{g} in H^infty(mathbb{R})$, the real hardy space. But I am not sure.
    – Adrián González-Pérez
    Nov 29 '18 at 12:40












  • Choose $h$ increasing to $1$ and satisfying $int_0^infty ysin(e^y)h(y)dy=0$. Then look at $f(x) =int_0^infty e^{2i pi xy} y sin(e^y)h(y)$ and try applying the Fourier inversion theorem.
    – reuns
    Nov 29 '18 at 20:18
















4












4








4


2





This is a followup to this earlier question. Given a (let's say continuous) function $f:mathbb{R} to mathbb{C}$ such that the Fourier Transform
$$widehat{f}(y)=int_{mathbb{R}} f(x)e^{-2pi i xy} dx$$
exists for all $y in mathbb{R}$, does it follow that $widehat{f} in L^{infty}$ i.e. there is a constant $C$ (depending on $f$) such that $vert widehat{f}(y)vert le C$ for all $y$?



This would trivially be true if $f in L^1$, but we are not assuming that. This of course also means that the integrals should better not be absolutely convergent. Instead we require convergence only in the sense that both $lim_{Tto infty} int_0^T$ and $lim_{T to infty} int_{-T}^0$ exist.










share|cite|improve this question













This is a followup to this earlier question. Given a (let's say continuous) function $f:mathbb{R} to mathbb{C}$ such that the Fourier Transform
$$widehat{f}(y)=int_{mathbb{R}} f(x)e^{-2pi i xy} dx$$
exists for all $y in mathbb{R}$, does it follow that $widehat{f} in L^{infty}$ i.e. there is a constant $C$ (depending on $f$) such that $vert widehat{f}(y)vert le C$ for all $y$?



This would trivially be true if $f in L^1$, but we are not assuming that. This of course also means that the integrals should better not be absolutely convergent. Instead we require convergence only in the sense that both $lim_{Tto infty} int_0^T$ and $lim_{T to infty} int_{-T}^0$ exist.







fourier-analysis fourier-transform






share|cite|improve this question













share|cite|improve this question











share|cite|improve this question




share|cite|improve this question










asked Nov 28 '18 at 21:00









Tintarn

2,668514




2,668514








  • 1




    I think the answer is negative. By the closed graph theorems if every continuous function $f$ would have $hat{f}$ in $L^infty$ that would mean that the Fourier transform is a bounded map between $CB(mathbb R)$, the bounded continuous functions and $L^infty(mathbb R)$. This will imply that $| hat{f} |_infty leq C , | f |_infty$. You can verify that there is no such $C$ by taking Gaussian $e^{t x^2}$, whose Fourier transforms have norm $t^{-1/2}$ despite the original function being bounded.
    – Adrián González-Pérez
    Nov 29 '18 at 12:37












  • I guess that the largest class of Function that satisfy what you require are those $g$ such that $hat{g} in H^infty(mathbb{R})$, the real hardy space. But I am not sure.
    – Adrián González-Pérez
    Nov 29 '18 at 12:40












  • Choose $h$ increasing to $1$ and satisfying $int_0^infty ysin(e^y)h(y)dy=0$. Then look at $f(x) =int_0^infty e^{2i pi xy} y sin(e^y)h(y)$ and try applying the Fourier inversion theorem.
    – reuns
    Nov 29 '18 at 20:18
















  • 1




    I think the answer is negative. By the closed graph theorems if every continuous function $f$ would have $hat{f}$ in $L^infty$ that would mean that the Fourier transform is a bounded map between $CB(mathbb R)$, the bounded continuous functions and $L^infty(mathbb R)$. This will imply that $| hat{f} |_infty leq C , | f |_infty$. You can verify that there is no such $C$ by taking Gaussian $e^{t x^2}$, whose Fourier transforms have norm $t^{-1/2}$ despite the original function being bounded.
    – Adrián González-Pérez
    Nov 29 '18 at 12:37












  • I guess that the largest class of Function that satisfy what you require are those $g$ such that $hat{g} in H^infty(mathbb{R})$, the real hardy space. But I am not sure.
    – Adrián González-Pérez
    Nov 29 '18 at 12:40












  • Choose $h$ increasing to $1$ and satisfying $int_0^infty ysin(e^y)h(y)dy=0$. Then look at $f(x) =int_0^infty e^{2i pi xy} y sin(e^y)h(y)$ and try applying the Fourier inversion theorem.
    – reuns
    Nov 29 '18 at 20:18










1




1




I think the answer is negative. By the closed graph theorems if every continuous function $f$ would have $hat{f}$ in $L^infty$ that would mean that the Fourier transform is a bounded map between $CB(mathbb R)$, the bounded continuous functions and $L^infty(mathbb R)$. This will imply that $| hat{f} |_infty leq C , | f |_infty$. You can verify that there is no such $C$ by taking Gaussian $e^{t x^2}$, whose Fourier transforms have norm $t^{-1/2}$ despite the original function being bounded.
– Adrián González-Pérez
Nov 29 '18 at 12:37






I think the answer is negative. By the closed graph theorems if every continuous function $f$ would have $hat{f}$ in $L^infty$ that would mean that the Fourier transform is a bounded map between $CB(mathbb R)$, the bounded continuous functions and $L^infty(mathbb R)$. This will imply that $| hat{f} |_infty leq C , | f |_infty$. You can verify that there is no such $C$ by taking Gaussian $e^{t x^2}$, whose Fourier transforms have norm $t^{-1/2}$ despite the original function being bounded.
– Adrián González-Pérez
Nov 29 '18 at 12:37














I guess that the largest class of Function that satisfy what you require are those $g$ such that $hat{g} in H^infty(mathbb{R})$, the real hardy space. But I am not sure.
– Adrián González-Pérez
Nov 29 '18 at 12:40






I guess that the largest class of Function that satisfy what you require are those $g$ such that $hat{g} in H^infty(mathbb{R})$, the real hardy space. But I am not sure.
– Adrián González-Pérez
Nov 29 '18 at 12:40














Choose $h$ increasing to $1$ and satisfying $int_0^infty ysin(e^y)h(y)dy=0$. Then look at $f(x) =int_0^infty e^{2i pi xy} y sin(e^y)h(y)$ and try applying the Fourier inversion theorem.
– reuns
Nov 29 '18 at 20:18






Choose $h$ increasing to $1$ and satisfying $int_0^infty ysin(e^y)h(y)dy=0$. Then look at $f(x) =int_0^infty e^{2i pi xy} y sin(e^y)h(y)$ and try applying the Fourier inversion theorem.
– reuns
Nov 29 '18 at 20:18












1 Answer
1






active

oldest

votes


















1














It is not necessary that $hat{f} in L^infty$.



Consider the Fourier transform defined as $hat{f}(y) = int_{-infty}^infty f(x) e^{iyx} , dx$. We can construct a counterexample where $hat{f}(m) to infty$ as $m (in mathbb{N}) to infty$, using the function



$$f(x) = sum_{n=1}^infty f_n(x) = sum_{n=1}^inftyfrac{n e^{-inx}}{x}sinleft[(2n^3)^{-1}x right] chi_{[-n^4,n^4]}(x)$$



The series converges uniformly since $|f_n(x)| leqslant frac{1}{2n^2}$ and we can switch sum and integral to get



$$hat{f}(m) = sum_{n=1}^infty nint_{-n^4}^{n^4}frac{e^{i(m-n)x}sin[(2n^3)^{-1}x]}{x} , dx$$



Note that



$$frac{e^{i(m-n)x}sin[(2n^3)^{-1}x]}{x} = frac{cos[(m-n)x] sin[(2n^3)^{-1}x]}{x}+ i frac{sin[(m-n)x]sin[(2n^3)^{-1}x]}{x}$$



The second term on the RHS is an odd function whose integral over $[-n^4,n^4]$ vanishes and the first term reduces to



$$frac{cos[(m-n)x] sin[(2n^3)^{-1}x]}{x} = frac{sin[(|m-n| + (2n^3)^{-1})x] - sin[(|m-n| - (2n^3)^{-1})x]}{2x}$$



Since this is an even function, the integral over $[-n^4,n^4]$ is twice the integral over $[0,n^4]$, and we have with $alpha_{m,n} = |m-n|$ and $beta_n = (2n^3)^{-1}$,



$$tag{*}hat{f}(m) = sum_{n=1}^infty nint_0^{n^4}frac{sin[(alpha_{m.n} + beta_n)x] - sin[(alpha_{m,n} - beta_n)x]}{x} ,,, dx \ = 2m int_0^{m^4} frac{sin(beta_mx)}{x} , dx + sum_{nneq m} nint_0^{n^4}frac{sin[(alpha_{m,n} + beta_n)x] - sin[(alpha_{m,n} - beta_n)x]}{x} , dx $$



We have the well-known results (where $c > 0$):



$$tag{**}int_0^z frac{sin (cx)}{x} , dx = frac{pi}{2} - int_z^infty frac{sin (cx)}{x} , dx , \ - frac{2}{cz} leqslant int_z^infty frac{sin(cx)}{x} , dx leqslant frac{2}{cz}$$



Since $alpha_{m,n} - beta_n = |m-n| - beta_n geqslant 1 - 1/2 > 0$ if $m neq n$, we can apply (**) to (*) and obtain



$$hat{f}(m) > pi m - 8 - sum_{n neq m} frac{2n}{n^4}left(frac{1}{alpha_{m,n}+beta_n} + frac{1}{alpha_{m,n} - beta_n} right)$$



Since



$$- frac{1}{alpha_{m,n} + beta_n} - frac{1}{alpha_{m,n} - beta_n}geqslant - frac{1}{1 + 0}- frac{1}{1 - 1/2} = -3,$$



we have



$$hat{f}(m) > pi m - 8 - 6sum_{n=1}^infty frac{1}{n^3}, $$



and $hat{f}(m) to infty$ as $m to infty$.






share|cite|improve this answer





















    Your Answer





    StackExchange.ifUsing("editor", function () {
    return StackExchange.using("mathjaxEditing", function () {
    StackExchange.MarkdownEditor.creationCallbacks.add(function (editor, postfix) {
    StackExchange.mathjaxEditing.prepareWmdForMathJax(editor, postfix, [["$", "$"], ["\\(","\\)"]]);
    });
    });
    }, "mathjax-editing");

    StackExchange.ready(function() {
    var channelOptions = {
    tags: "".split(" "),
    id: "69"
    };
    initTagRenderer("".split(" "), "".split(" "), channelOptions);

    StackExchange.using("externalEditor", function() {
    // Have to fire editor after snippets, if snippets enabled
    if (StackExchange.settings.snippets.snippetsEnabled) {
    StackExchange.using("snippets", function() {
    createEditor();
    });
    }
    else {
    createEditor();
    }
    });

    function createEditor() {
    StackExchange.prepareEditor({
    heartbeatType: 'answer',
    autoActivateHeartbeat: false,
    convertImagesToLinks: true,
    noModals: true,
    showLowRepImageUploadWarning: true,
    reputationToPostImages: 10,
    bindNavPrevention: true,
    postfix: "",
    imageUploader: {
    brandingHtml: "Powered by u003ca class="icon-imgur-white" href="https://imgur.com/"u003eu003c/au003e",
    contentPolicyHtml: "User contributions licensed under u003ca href="https://creativecommons.org/licenses/by-sa/3.0/"u003ecc by-sa 3.0 with attribution requiredu003c/au003e u003ca href="https://stackoverflow.com/legal/content-policy"u003e(content policy)u003c/au003e",
    allowUrls: true
    },
    noCode: true, onDemand: true,
    discardSelector: ".discard-answer"
    ,immediatelyShowMarkdownHelp:true
    });


    }
    });














    draft saved

    draft discarded


















    StackExchange.ready(
    function () {
    StackExchange.openid.initPostLogin('.new-post-login', 'https%3a%2f%2fmath.stackexchange.com%2fquestions%2f3017704%2fis-the-fourier-transform-of-a-continuous-function-necessarily-in-l-infty%23new-answer', 'question_page');
    }
    );

    Post as a guest















    Required, but never shown

























    1 Answer
    1






    active

    oldest

    votes








    1 Answer
    1






    active

    oldest

    votes









    active

    oldest

    votes






    active

    oldest

    votes









    1














    It is not necessary that $hat{f} in L^infty$.



    Consider the Fourier transform defined as $hat{f}(y) = int_{-infty}^infty f(x) e^{iyx} , dx$. We can construct a counterexample where $hat{f}(m) to infty$ as $m (in mathbb{N}) to infty$, using the function



    $$f(x) = sum_{n=1}^infty f_n(x) = sum_{n=1}^inftyfrac{n e^{-inx}}{x}sinleft[(2n^3)^{-1}x right] chi_{[-n^4,n^4]}(x)$$



    The series converges uniformly since $|f_n(x)| leqslant frac{1}{2n^2}$ and we can switch sum and integral to get



    $$hat{f}(m) = sum_{n=1}^infty nint_{-n^4}^{n^4}frac{e^{i(m-n)x}sin[(2n^3)^{-1}x]}{x} , dx$$



    Note that



    $$frac{e^{i(m-n)x}sin[(2n^3)^{-1}x]}{x} = frac{cos[(m-n)x] sin[(2n^3)^{-1}x]}{x}+ i frac{sin[(m-n)x]sin[(2n^3)^{-1}x]}{x}$$



    The second term on the RHS is an odd function whose integral over $[-n^4,n^4]$ vanishes and the first term reduces to



    $$frac{cos[(m-n)x] sin[(2n^3)^{-1}x]}{x} = frac{sin[(|m-n| + (2n^3)^{-1})x] - sin[(|m-n| - (2n^3)^{-1})x]}{2x}$$



    Since this is an even function, the integral over $[-n^4,n^4]$ is twice the integral over $[0,n^4]$, and we have with $alpha_{m,n} = |m-n|$ and $beta_n = (2n^3)^{-1}$,



    $$tag{*}hat{f}(m) = sum_{n=1}^infty nint_0^{n^4}frac{sin[(alpha_{m.n} + beta_n)x] - sin[(alpha_{m,n} - beta_n)x]}{x} ,,, dx \ = 2m int_0^{m^4} frac{sin(beta_mx)}{x} , dx + sum_{nneq m} nint_0^{n^4}frac{sin[(alpha_{m,n} + beta_n)x] - sin[(alpha_{m,n} - beta_n)x]}{x} , dx $$



    We have the well-known results (where $c > 0$):



    $$tag{**}int_0^z frac{sin (cx)}{x} , dx = frac{pi}{2} - int_z^infty frac{sin (cx)}{x} , dx , \ - frac{2}{cz} leqslant int_z^infty frac{sin(cx)}{x} , dx leqslant frac{2}{cz}$$



    Since $alpha_{m,n} - beta_n = |m-n| - beta_n geqslant 1 - 1/2 > 0$ if $m neq n$, we can apply (**) to (*) and obtain



    $$hat{f}(m) > pi m - 8 - sum_{n neq m} frac{2n}{n^4}left(frac{1}{alpha_{m,n}+beta_n} + frac{1}{alpha_{m,n} - beta_n} right)$$



    Since



    $$- frac{1}{alpha_{m,n} + beta_n} - frac{1}{alpha_{m,n} - beta_n}geqslant - frac{1}{1 + 0}- frac{1}{1 - 1/2} = -3,$$



    we have



    $$hat{f}(m) > pi m - 8 - 6sum_{n=1}^infty frac{1}{n^3}, $$



    and $hat{f}(m) to infty$ as $m to infty$.






    share|cite|improve this answer


























      1














      It is not necessary that $hat{f} in L^infty$.



      Consider the Fourier transform defined as $hat{f}(y) = int_{-infty}^infty f(x) e^{iyx} , dx$. We can construct a counterexample where $hat{f}(m) to infty$ as $m (in mathbb{N}) to infty$, using the function



      $$f(x) = sum_{n=1}^infty f_n(x) = sum_{n=1}^inftyfrac{n e^{-inx}}{x}sinleft[(2n^3)^{-1}x right] chi_{[-n^4,n^4]}(x)$$



      The series converges uniformly since $|f_n(x)| leqslant frac{1}{2n^2}$ and we can switch sum and integral to get



      $$hat{f}(m) = sum_{n=1}^infty nint_{-n^4}^{n^4}frac{e^{i(m-n)x}sin[(2n^3)^{-1}x]}{x} , dx$$



      Note that



      $$frac{e^{i(m-n)x}sin[(2n^3)^{-1}x]}{x} = frac{cos[(m-n)x] sin[(2n^3)^{-1}x]}{x}+ i frac{sin[(m-n)x]sin[(2n^3)^{-1}x]}{x}$$



      The second term on the RHS is an odd function whose integral over $[-n^4,n^4]$ vanishes and the first term reduces to



      $$frac{cos[(m-n)x] sin[(2n^3)^{-1}x]}{x} = frac{sin[(|m-n| + (2n^3)^{-1})x] - sin[(|m-n| - (2n^3)^{-1})x]}{2x}$$



      Since this is an even function, the integral over $[-n^4,n^4]$ is twice the integral over $[0,n^4]$, and we have with $alpha_{m,n} = |m-n|$ and $beta_n = (2n^3)^{-1}$,



      $$tag{*}hat{f}(m) = sum_{n=1}^infty nint_0^{n^4}frac{sin[(alpha_{m.n} + beta_n)x] - sin[(alpha_{m,n} - beta_n)x]}{x} ,,, dx \ = 2m int_0^{m^4} frac{sin(beta_mx)}{x} , dx + sum_{nneq m} nint_0^{n^4}frac{sin[(alpha_{m,n} + beta_n)x] - sin[(alpha_{m,n} - beta_n)x]}{x} , dx $$



      We have the well-known results (where $c > 0$):



      $$tag{**}int_0^z frac{sin (cx)}{x} , dx = frac{pi}{2} - int_z^infty frac{sin (cx)}{x} , dx , \ - frac{2}{cz} leqslant int_z^infty frac{sin(cx)}{x} , dx leqslant frac{2}{cz}$$



      Since $alpha_{m,n} - beta_n = |m-n| - beta_n geqslant 1 - 1/2 > 0$ if $m neq n$, we can apply (**) to (*) and obtain



      $$hat{f}(m) > pi m - 8 - sum_{n neq m} frac{2n}{n^4}left(frac{1}{alpha_{m,n}+beta_n} + frac{1}{alpha_{m,n} - beta_n} right)$$



      Since



      $$- frac{1}{alpha_{m,n} + beta_n} - frac{1}{alpha_{m,n} - beta_n}geqslant - frac{1}{1 + 0}- frac{1}{1 - 1/2} = -3,$$



      we have



      $$hat{f}(m) > pi m - 8 - 6sum_{n=1}^infty frac{1}{n^3}, $$



      and $hat{f}(m) to infty$ as $m to infty$.






      share|cite|improve this answer
























        1












        1








        1






        It is not necessary that $hat{f} in L^infty$.



        Consider the Fourier transform defined as $hat{f}(y) = int_{-infty}^infty f(x) e^{iyx} , dx$. We can construct a counterexample where $hat{f}(m) to infty$ as $m (in mathbb{N}) to infty$, using the function



        $$f(x) = sum_{n=1}^infty f_n(x) = sum_{n=1}^inftyfrac{n e^{-inx}}{x}sinleft[(2n^3)^{-1}x right] chi_{[-n^4,n^4]}(x)$$



        The series converges uniformly since $|f_n(x)| leqslant frac{1}{2n^2}$ and we can switch sum and integral to get



        $$hat{f}(m) = sum_{n=1}^infty nint_{-n^4}^{n^4}frac{e^{i(m-n)x}sin[(2n^3)^{-1}x]}{x} , dx$$



        Note that



        $$frac{e^{i(m-n)x}sin[(2n^3)^{-1}x]}{x} = frac{cos[(m-n)x] sin[(2n^3)^{-1}x]}{x}+ i frac{sin[(m-n)x]sin[(2n^3)^{-1}x]}{x}$$



        The second term on the RHS is an odd function whose integral over $[-n^4,n^4]$ vanishes and the first term reduces to



        $$frac{cos[(m-n)x] sin[(2n^3)^{-1}x]}{x} = frac{sin[(|m-n| + (2n^3)^{-1})x] - sin[(|m-n| - (2n^3)^{-1})x]}{2x}$$



        Since this is an even function, the integral over $[-n^4,n^4]$ is twice the integral over $[0,n^4]$, and we have with $alpha_{m,n} = |m-n|$ and $beta_n = (2n^3)^{-1}$,



        $$tag{*}hat{f}(m) = sum_{n=1}^infty nint_0^{n^4}frac{sin[(alpha_{m.n} + beta_n)x] - sin[(alpha_{m,n} - beta_n)x]}{x} ,,, dx \ = 2m int_0^{m^4} frac{sin(beta_mx)}{x} , dx + sum_{nneq m} nint_0^{n^4}frac{sin[(alpha_{m,n} + beta_n)x] - sin[(alpha_{m,n} - beta_n)x]}{x} , dx $$



        We have the well-known results (where $c > 0$):



        $$tag{**}int_0^z frac{sin (cx)}{x} , dx = frac{pi}{2} - int_z^infty frac{sin (cx)}{x} , dx , \ - frac{2}{cz} leqslant int_z^infty frac{sin(cx)}{x} , dx leqslant frac{2}{cz}$$



        Since $alpha_{m,n} - beta_n = |m-n| - beta_n geqslant 1 - 1/2 > 0$ if $m neq n$, we can apply (**) to (*) and obtain



        $$hat{f}(m) > pi m - 8 - sum_{n neq m} frac{2n}{n^4}left(frac{1}{alpha_{m,n}+beta_n} + frac{1}{alpha_{m,n} - beta_n} right)$$



        Since



        $$- frac{1}{alpha_{m,n} + beta_n} - frac{1}{alpha_{m,n} - beta_n}geqslant - frac{1}{1 + 0}- frac{1}{1 - 1/2} = -3,$$



        we have



        $$hat{f}(m) > pi m - 8 - 6sum_{n=1}^infty frac{1}{n^3}, $$



        and $hat{f}(m) to infty$ as $m to infty$.






        share|cite|improve this answer












        It is not necessary that $hat{f} in L^infty$.



        Consider the Fourier transform defined as $hat{f}(y) = int_{-infty}^infty f(x) e^{iyx} , dx$. We can construct a counterexample where $hat{f}(m) to infty$ as $m (in mathbb{N}) to infty$, using the function



        $$f(x) = sum_{n=1}^infty f_n(x) = sum_{n=1}^inftyfrac{n e^{-inx}}{x}sinleft[(2n^3)^{-1}x right] chi_{[-n^4,n^4]}(x)$$



        The series converges uniformly since $|f_n(x)| leqslant frac{1}{2n^2}$ and we can switch sum and integral to get



        $$hat{f}(m) = sum_{n=1}^infty nint_{-n^4}^{n^4}frac{e^{i(m-n)x}sin[(2n^3)^{-1}x]}{x} , dx$$



        Note that



        $$frac{e^{i(m-n)x}sin[(2n^3)^{-1}x]}{x} = frac{cos[(m-n)x] sin[(2n^3)^{-1}x]}{x}+ i frac{sin[(m-n)x]sin[(2n^3)^{-1}x]}{x}$$



        The second term on the RHS is an odd function whose integral over $[-n^4,n^4]$ vanishes and the first term reduces to



        $$frac{cos[(m-n)x] sin[(2n^3)^{-1}x]}{x} = frac{sin[(|m-n| + (2n^3)^{-1})x] - sin[(|m-n| - (2n^3)^{-1})x]}{2x}$$



        Since this is an even function, the integral over $[-n^4,n^4]$ is twice the integral over $[0,n^4]$, and we have with $alpha_{m,n} = |m-n|$ and $beta_n = (2n^3)^{-1}$,



        $$tag{*}hat{f}(m) = sum_{n=1}^infty nint_0^{n^4}frac{sin[(alpha_{m.n} + beta_n)x] - sin[(alpha_{m,n} - beta_n)x]}{x} ,,, dx \ = 2m int_0^{m^4} frac{sin(beta_mx)}{x} , dx + sum_{nneq m} nint_0^{n^4}frac{sin[(alpha_{m,n} + beta_n)x] - sin[(alpha_{m,n} - beta_n)x]}{x} , dx $$



        We have the well-known results (where $c > 0$):



        $$tag{**}int_0^z frac{sin (cx)}{x} , dx = frac{pi}{2} - int_z^infty frac{sin (cx)}{x} , dx , \ - frac{2}{cz} leqslant int_z^infty frac{sin(cx)}{x} , dx leqslant frac{2}{cz}$$



        Since $alpha_{m,n} - beta_n = |m-n| - beta_n geqslant 1 - 1/2 > 0$ if $m neq n$, we can apply (**) to (*) and obtain



        $$hat{f}(m) > pi m - 8 - sum_{n neq m} frac{2n}{n^4}left(frac{1}{alpha_{m,n}+beta_n} + frac{1}{alpha_{m,n} - beta_n} right)$$



        Since



        $$- frac{1}{alpha_{m,n} + beta_n} - frac{1}{alpha_{m,n} - beta_n}geqslant - frac{1}{1 + 0}- frac{1}{1 - 1/2} = -3,$$



        we have



        $$hat{f}(m) > pi m - 8 - 6sum_{n=1}^infty frac{1}{n^3}, $$



        and $hat{f}(m) to infty$ as $m to infty$.







        share|cite|improve this answer












        share|cite|improve this answer



        share|cite|improve this answer










        answered Nov 29 '18 at 18:20









        RRL

        49.2k42573




        49.2k42573






























            draft saved

            draft discarded




















































            Thanks for contributing an answer to Mathematics Stack Exchange!


            • Please be sure to answer the question. Provide details and share your research!

            But avoid



            • Asking for help, clarification, or responding to other answers.

            • Making statements based on opinion; back them up with references or personal experience.


            Use MathJax to format equations. MathJax reference.


            To learn more, see our tips on writing great answers.





            Some of your past answers have not been well-received, and you're in danger of being blocked from answering.


            Please pay close attention to the following guidance:


            • Please be sure to answer the question. Provide details and share your research!

            But avoid



            • Asking for help, clarification, or responding to other answers.

            • Making statements based on opinion; back them up with references or personal experience.


            To learn more, see our tips on writing great answers.




            draft saved


            draft discarded














            StackExchange.ready(
            function () {
            StackExchange.openid.initPostLogin('.new-post-login', 'https%3a%2f%2fmath.stackexchange.com%2fquestions%2f3017704%2fis-the-fourier-transform-of-a-continuous-function-necessarily-in-l-infty%23new-answer', 'question_page');
            }
            );

            Post as a guest















            Required, but never shown





















































            Required, but never shown














            Required, but never shown












            Required, but never shown







            Required, but never shown

































            Required, but never shown














            Required, but never shown












            Required, but never shown







            Required, but never shown







            Popular posts from this blog

            Bundesstraße 106

            Verónica Boquete

            Ida-Boy-Ed-Garten