754 CHAPTER 37. SOME FUNDAMENTAL FUNCTIONS AND TRANSFORMS

37.3 Fourier TransformDefinition 37.3.1 The Fourier transform is defined as follows for f ∈ L1 (R) meaning that

limR→∞

∫ R

−R| f (t)|dt < ∞

f is piecewise continuous on [−R,R] . Then the Fourier transform is given by

F f (t)≡ 1√2π

∫∞

−∞

e−itx f (x)dx

The inverse Fourier transform is defined the same way except you delete the minus sign inthe complex exponential.

F−1 f (t)≡ 1√2π

∫∞

−∞

eitx f (x)dx

Does it deserve to be called the “inverse” Fourier transform? This question will beexplored somewhat below.

There is a very important improper integral involving sin(x)/x. You can show with alittle estimating that x→ sin(x)/x is not in L1 (0,∞) . Nevertheless, a lot can be said aboutimproper integrals involving this function.

Theorem 37.3.2 The following hold

1.∫

0sinu

u du = π

2

2. limr→∞

∫∞

δ

sin(ru)u du = 0 whenever δ > 0.

3. If f ∈L1 (R) , then limr→∞

∫R sin(ru) f (u)du= 0. This is called the Riemann Lebesgue

lemma.

Proof: The first claim follows from Problem 6 on Page 720 above.Now consider

∫∞

δ

sin(ru)u du. It equals

∫∞

0sin(ru)

u du−∫

δ

0sin(ru)

u du which can be seen fromthe definition of what the improper integral means. Also, you can change the variable. Letru = t so rdu = dt and the above reduces to∫

0

sin(t)t

r1r

dt−∫ rδ

0

sin(t)t

dt =∫

δ

sin(ru)u

du

Thusπ

2−∫ rδ

0

sin(t)t

dt =∫

δ

sin(ru)u

du

and so limr→∞

∫∞

δ

sin(ru)u du = 0 from the first part.

Now consider the third claim, the Riemann Lebesgue lemma. For I an interval let

XI (t)≡

{1 if t ∈ I0 if t /∈ I

754 CHAPTER 37. SOME FUNDAMENTAL FUNCTIONS AND TRANSFORMS37.3. Fourier TransformDefinition 37.3.1 The Fourier transform is defined as follows for f € L' (IR) meaning thatRlim |f (t)|dt < oxRoo J __Rf is piecewise continuous on [—R,R]. Then the Fourier transform is given by— ix x) dxFf (t) — |e~ anThe inverse Fourier transform is defined the same way except you delete the minus sign inthe complex exponential.Fo f (t) = |e itx f(x~ inDoes it deserve to be called the “inverse” Fourier transform? This question will beexplored somewhat below.There is a very important improper integral involving sin (x) /x. You can show with alittle estimating that x > sin (x) /x is not in L' (0,00). Nevertheless, a lot can be said aboutimproper integrals involving this function.Theorem 37.3.2 The following hold1. fg Bau = 32. limo fs sinr4) 4 — ( whenever & > 0.3. If f EL (R), then lim, fg sin (ru) f (u) du =0. This is called the Riemann Lebesguelemma.Proof: The first claim follows from Problem 6 on Page 720 above.Now consider [5° sin(ri) du. It equals [jy sine) dy — Ke sin(ru) du which can be seen fromthe definition of what the i improper integral means. Also, you can change the variable. Letru =t so rdu = dt and the above reduces toco . ro . co ./ val Par f nO ay = | sin(r)0 t r 0 t 6 urd of OO gt*—f nO ay f sin (ru) 5,JO J uThus2 tco sin sin(ru)and so lim, [5 du = 0 from the first part.Now consider the third claim, the Riemann Lebesgue lemma. For / an interval letlifrelAy(th=i() { onegs