Theorem in harmonic analysis
In mathematics , the Plancherel theorem (sometimes called the Parseval–Plancherel identity ) is a result in harmonic analysis , proven by Michel Plancherel in 1910. It is a generalization of Parseval's theorem ; often used in the fields of science and engineering, proving the unitarity of the Fourier transform .
The theorem states that the integral of a function's squared modulus is equal to the integral of the squared modulus of its frequency spectrum . That is, if
f
(
x
)
{\displaystyle f(x)}
is a function on the real line, and
f
^
(
ξ
)
{\displaystyle {\widehat {f}}(\xi )}
is its frequency spectrum, then
∫
−
∞
∞
|
f
(
x
)
|
2
d
x
=
∫
−
∞
∞
|
f
^
(
ξ
)
|
2
d
ξ
{\displaystyle \int _{-\infty }^{\infty }|f(x)|^{2}\,dx=\int _{-\infty }^{\infty }|{\widehat {f}}(\xi )|^{2}\,d\xi }
A more precise formulation is that if a function is in both L p spaces
L
1
(
R
)
{\displaystyle L^{1}(\mathbb {R} )}
and
L
2
(
R
)
{\displaystyle L^{2}(\mathbb {R} )}
, then its Fourier transform is in
L
2
(
R
)
{\displaystyle L^{2}(\mathbb {R} )}
and the Fourier transform is an isometry with respect to the L 2 norm. This implies that the Fourier transform restricted to
L
1
(
R
)
∩
L
2
(
R
)
{\displaystyle L^{1}(\mathbb {R} )\cap L^{2}(\mathbb {R} )}
has a unique extension to a linear isometric map
L
2
(
R
)
↦
L
2
(
R
)
{\displaystyle L^{2}(\mathbb {R} )\mapsto L^{2}(\mathbb {R} )}
, sometimes called the Plancherel transform. This isometry is actually a unitary map. In effect, this makes it possible to speak of Fourier transforms of quadratically integrable functions .
Plancherel's theorem remains valid as stated on n -dimensional Euclidean space
R
n
{\displaystyle \mathbb {R} ^{n}}
. The theorem also holds more generally in locally compact abelian groups . There is also a version of the Plancherel theorem which makes sense for non-commutative locally compact groups satisfying certain technical assumptions. This is the subject of non-commutative harmonic analysis .
Due to the polarization identity , one can also apply Plancherel's theorem to the
L
2
(
R
)
{\displaystyle L^{2}(\mathbb {R} )}
inner product of two functions. That is, if
f
(
x
)
{\displaystyle f(x)}
and
g
(
x
)
{\displaystyle g(x)}
are two
L
2
(
R
)
{\displaystyle L^{2}(\mathbb {R} )}
functions, and
P
{\displaystyle {\mathcal {P}}}
denotes the Plancherel transform, then
∫
−
∞
∞
f
(
x
)
g
(
x
)
¯
d
x
=
∫
−
∞
∞
(
P
f
)
(
ξ
)
(
P
g
)
(
ξ
)
¯
d
ξ
,
{\displaystyle \int _{-\infty }^{\infty }f(x){\overline {g(x)}}\,dx=\int _{-\infty }^{\infty }({\mathcal {P}}f)(\xi ){\overline {({\mathcal {P}}g)(\xi )}}\,d\xi ,}
and if
f
(
x
)
{\displaystyle f(x)}
and
g
(
x
)
{\displaystyle g(x)}
are furthermore
L
1
(
R
)
{\displaystyle L^{1}(\mathbb {R} )}
functions, then
(
P
f
)
(
ξ
)
=
f
^
(
ξ
)
=
∫
−
∞
∞
f
(
x
)
e
−
2
π
i
ξ
x
d
x
,
{\displaystyle ({\mathcal {P}}f)(\xi )={\widehat {f}}(\xi )=\int _{-\infty }^{\infty }f(x)e^{-2\pi i\xi x}\,dx,}
and
(
P
g
)
(
ξ
)
=
g
^
(
ξ
)
=
∫
−
∞
∞
g
(
x
)
e
−
2
π
i
ξ
x
d
x
,
{\displaystyle ({\mathcal {P}}g)(\xi )={\widehat {g}}(\xi )=\int _{-\infty }^{\infty }g(x)e^{-2\pi i\xi x}\,dx,}
so
∫
−
∞
∞
f
(
x
)
g
(
x
)
¯
d
x
=
∫
−
∞
∞
f
^
(
ξ
)
g
^
(
ξ
)
¯
d
ξ
.
{\displaystyle \int _{-\infty }^{\infty }f(x){\overline {g(x)}}\,dx=\int _{-\infty }^{\infty }{\widehat {f}}(\xi ){\overline {{\widehat {g}}(\xi )}}\,d\xi .}
Proof
Assumption.
f
∈
L
1
∩
L
2
{\displaystyle f\in L^{1}\cap L^{2}}
, i.e.
∫
|
f
(
x
)
|
d
x
,
∫
|
f
(
x
)
|
2
d
x
<
∞
{\displaystyle \int |f(x)|dx,\int |f(x)|^{2}dx<\infty }
Step 1. The equality holds if f is differentiable and f' is bounded
Let
f
⋆
(
y
)
=
f
¯
(
−
y
)
,
ϕ
(
x
)
=
(
f
∗
f
⋆
)
(
x
)
=
∫
f
(
x
−
y
)
f
⋆
(
y
)
d
y
=
∫
f
(
x
−
y
)
f
¯
(
−
y
)
d
y
=
∫
f
(
x
+
t
)
f
¯
(
t
)
d
t
{\displaystyle f^{\star }(y)={\bar {f}}(-y),\phi (x)=(f\ast f^{\star })(x)=\int f(x-y)f^{\star }(y)dy=\int f(x-y){\bar {f}}(-y)dy=\int f(x+t){\bar {f}}(t)dt}
, then
|
∂
[
f
(
x
+
t
)
f
¯
(
t
)
]
∂
x
|
=
|
f
′
(
x
+
t
)
f
¯
(
t
)
|
≤
C
|
f
(
t
)
|
{\displaystyle |{\frac {\partial [f(x+t){\bar {f}}(t)]}{\partial x}}|=|f'(x+t){\bar {f}}(t)|\leq C|f(t)|}
, and the Dominated Convergence Theorem implies the interchangibility of differentiation and integration, thus
ϕ
′
(
x
)
=
∫
f
′
(
x
+
t
)
f
¯
(
t
)
d
t
{\displaystyle \phi '(x)=\int f'(x+t){\bar {f}}(t)dt}
,
ϕ
{\displaystyle \phi }
is differentiable, hence by Fourier inversion theorem ,
∫
|
f
(
x
)
|
2
d
x
=
ϕ
(
0
)
=
lim
L
→
∞
∫
−
L
L
F
(
ϕ
)
(
ξ
)
e
x
p
(
2
π
i
⋅
0
⋅
ξ
)
d
ξ
=
lim
L
→
∞
∫
−
L
L
F
(
ϕ
)
(
ξ
)
d
ξ
{\displaystyle \int |f(x)|^{2}dx=\phi (0)=\lim \limits _{L\rightarrow \infty }\int _{-L}^{L}{\mathcal {F}}(\phi )(\xi )exp(2\pi i\cdot 0\cdot \xi )d\xi =\lim \limits _{L\rightarrow \infty }\int _{-L}^{L}{\mathcal {F}}(\phi )(\xi )d\xi }
By convolution theorem of Fourier transform,
F
(
ϕ
)
=
F
(
f
)
F
(
f
⋆
)
=
|
F
(
f
)
|
2
=
|
f
^
|
2
{\displaystyle {\mathcal {F}}(\phi )={\mathcal {F}}(f){\mathcal {F}}(f^{\star })=|{\mathcal {F}}(f)|^{2}=|{\hat {f}}|^{2}}
,
lim
L
→
∞
∫
−
L
L
|
f
^
(
ξ
)
|
2
d
ξ
=
∫
|
f
^
(
ξ
)
|
2
d
ξ
{\displaystyle \lim \limits _{L\rightarrow \infty }\int _{-L}^{L}|{\hat {f}}(\xi )|^{2}d\xi =\int |{\hat {f}}(\xi )|^{2}d\xi }
by Monotone Convergence Theorem , hence
∫
|
f
(
x
)
|
2
d
x
=
∫
|
f
^
(
ξ
)
|
2
d
ξ
{\displaystyle \int |f(x)|^{2}dx=\int |{\hat {f}}(\xi )|^{2}d\xi }
Step 2. the General Case
Let
ρ
ϵ
{\displaystyle \rho _{\epsilon }}
be a family of mollifiers ,
f
ϵ
=
f
∗
ρ
ϵ
{\displaystyle f_{\epsilon }=f\ast \rho _{\epsilon }}
, then for each ε,
f
ϵ
′
=
f
∗
ρ
ϵ
′
{\displaystyle f_{\epsilon }'=f\ast \rho _{\epsilon }'}
,
|
f
ϵ
′
|
=
|
f
∗
ρ
ϵ
′
|
≤
‖
f
‖
L
2
‖
ρ
ϵ
′
‖
L
2
{\displaystyle |f_{\epsilon }'|=|f\ast \rho _{\epsilon }'|\leq \|f\|_{L^{2}}\|\rho _{\epsilon }'\|_{L^{2}}}
by Hölder's inequality , hence
f
ϵ
{\displaystyle f_{\epsilon }}
is differentiable and has a bounded derivative. By Step 1 ,
∫
|
f
ϵ
(
x
)
|
2
d
x
=
∫
|
f
ϵ
^
(
ξ
)
|
2
d
ξ
{\displaystyle \int |f_{\epsilon }(x)|^{2}dx=\int |{\hat {f_{\epsilon }}}(\xi )|^{2}d\xi }
. By the property of mollification, the left hand side converges to
‖
f
‖
L
2
2
{\displaystyle \|f\|_{L^{2}}^{2}}
as
ϵ
→
0
{\displaystyle \epsilon \rightarrow 0}
, and by convolution theorem ,
|
f
ϵ
^
|
=
|
f
^
|
|
ρ
ϵ
^
|
→
|
f
^
|
as
ϵ
→
0
{\displaystyle |{\hat {f_{\epsilon }}}|=|{\hat {f}}||{\hat {\rho _{\epsilon }}}|\rightarrow |{\hat {f}}|{\text{ as }}\epsilon \rightarrow 0}
, hence by Fatou' lemma , we have
∫
|
f
^
|
2
d
ξ
≤
lim inf
ϵ
→
0
∫
|
f
ϵ
^
|
2
d
ξ
=
lim inf
ϵ
→
0
∫
|
f
ϵ
|
2
d
x
=
∫
|
f
|
2
d
x
{\displaystyle \int |{\hat {f}}|^{2}d\xi \leq \liminf \limits _{\epsilon \rightarrow 0}\int |{\hat {f_{\epsilon }}}|^{2}d\xi =\liminf \limits _{\epsilon \rightarrow 0}\int |f_{\epsilon }|^{2}dx=\int |f|^{2}dx}
, thus
|
f
^
|
2
{\displaystyle |{\hat {f}}|^{2}}
is integrable. Thus the right hand side converges to
‖
f
^
‖
L
2
2
{\displaystyle \|{\hat {f}}\|_{L^{2}}^{2}}
as
ϵ
→
0
{\displaystyle \epsilon \rightarrow 0}
by Dominated Convergence Theorem . Q.E.D.
See also
References
Plancherel, Michel (1910), "Contribution à l'étude de la représentation d'une fonction arbitraire par des intégrales définies", Rendiconti del Circolo Matematico di Palermo , 30 (1): 289–335, doi :10.1007/BF03014877 , S2CID 122509369 .
Dixmier, J. (1969), Les C*-algèbres et leurs Représentations , Gauthier Villars .
Yosida, K. (1968), Functional Analysis , Springer Verlag .
External links
Spaces
Theorems Operators Algebras Open problems Applications Advanced topics