Scribd
Upload
12 views

2

This document discusses convolutions and their properties. It proves that if f and g are Schwartz functions, then their convolution f * g is also a Schwartz function. It also shows that the Fourier transform converts convolution into ordinary pointwise multiplication, so that the Fourier transform of f * g is the product of the Fourier transforms of f and g. Finally, it introduces approximate identities and proves that convolving a continuous function or Lp function with an approximate identity kernel yields the original function in the limit.

Uploaded by

Ola
Copyright
© © All Rights Reserved
Available Formats
Download as PDF, TXT or read online on Scribd
12 views

2

This document discusses convolutions and their properties. It proves that if f and g are Schwartz functions, then their convolution f * g is also a Schwartz function. It also shows that the Fourier transform converts convolution into ordinary pointwise multiplication, so that the Fourier transform of f * g is the product of the Fourier transforms of f and g. Finally, it introduces approximate identities and proves that convolving a continuous function or Lp function with an approximate identity kernel yields the original function in the limit.

Uploaded by

Ola
Copyright
© © All Rights Reserved
Available Formats
Download as PDF, TXT or read online on Scribd
You are on page 1/ 1

Using (23) and commutativity of convolution, we can rewrite this as

Z
1
d(h) = (φ(x + hej − y) − φ(x − y))f (y)dy.
h
The quotients
1
Ah (y) = (φ(x + hej − y) − φ(x − y))
h
are bounded by k ∂x
∂φ
j
k∞ by the mean value theorem. For fixed x and for |h| ≤ 1, the
support of Ah is contained in the fixed compact set E = D(x, 1)+ supp φ. Thus the inte-
grands Ah f are dominated by the L1 function k ∂x
∂φ
j
k∞ χE |f |. The dominated convergence
theorem implies
Z Z
∂φ
lim d(h) = f (y) lim Ah (y)dy = f (y) (x − y)dy.
h→0 h→0 ∂xj

This proves (26) (when |α| = 1). The continuity of the partials then follows from 2.
above. 

Corollary 3.1’ If f, g ∈ S then f ∗ g ∈ S.

Proof By Lemma 3.1 it suffices to show that if (1 + |x|)N f (x) and (1 + |x|)N g(x)
are bounded for every N then so is (1 + |x|)N f ∗ g(x). This follows by writing out the
definitions and using (15); the details are left to the reader. 

Convolution interacts with the Fourier transform as follows: Fourier transform con-
verts convolution to ordinary pointwise multiplication. Thus we have the following for-
mulas:
f[∗ g = fˆĝ, f, g ∈ L1 , (27)
fcg = fˆ ∗ ĝ, f, g ∈ S. (28)
(27) follows from Fubini’s theorem and is in many textbooks; the proof is left to the
reader. (28) then follows easily from the inversion theorem, so we defer the proof until
after Theorem 3.4. R
Let φ ∈ S, and assume that φ = 1. Define φ (x) = −nRφ(−1 x). The family of
functions {φ } is called an approximate identity. Notice that φ = 1 for all . Thus
one can regard the φ as roughly convergent to the Dirac mass δ 0 as  → 0. Indeed, the
following fact is basic but quite standard; see any reasonable book on real analysis for the
proof.
R
Lemma 3.2 Let φ ∈ S and φ = 1. Then:
1. If f is a continuous function which goes to zero at ∞ then φ ∗ f → f uniformly as
 → 0.
2. If f ∈ Lp , 1 ≤ p < ∞ then φ ∗ f → f in Lp as  → 0.

14

You might also like