Solution.
First, we have the following identity:
∫ R ∣ T f ( x ) ∣ 2 d x = ∫ R ∫ 0 ∣ T f ( x ) ∣ 2 λ d λ d x = ∫ R ∫ 0 ∞ 2 λ χ { 0 < λ < ∣ T f ( x ) ∣ } d λ d x = ∫ 0 ∞ 2 λ ∫ R χ { 0 < λ < ∣ T f ( x ) ∣ } d x d λ = ∫ 0 ∞ 2 λ m ( { ∣ T f ( x ) ∣ > λ } ) d λ . \begin{aligned}
\int_\R \abs{Tf\p{x}}^2 \,\diff{x}
&= \int_\R \int_0^{\abs{Tf\p{x}}} 2\lambda \,\diff\lambda \,\diff{x} \\
&= \int_\R \int_0^\infty 2\lambda \chi_{\set{0 < \lambda < \abs{Tf\p{x}}}} \,\diff\lambda \,\diff{x} \\
&= \int_0^\infty 2\lambda \int_\R \chi_{\set{0 < \lambda < \abs{Tf\p{x}}}} \,\diff{x} \,\diff\lambda \\
&= \int_0^\infty 2\lambda m\p{\set{\abs{Tf\p{x}} > \lambda}} \,\diff\lambda.
\end{aligned} ∫ R ∣ T f ( x ) ∣ 2 d x = ∫ R ∫ 0 ∣ T f ( x ) ∣ 2 λ d λ d x = ∫ R ∫ 0 ∞ 2 λ χ { 0 < λ < ∣ T f ( x ) ∣ } d λ d x = ∫ 0 ∞ 2 λ ∫ R χ { 0 < λ < ∣ T f ( x ) ∣ } d x d λ = ∫ 0 ∞ 2 λm ( { ∣ T f ( x ) ∣ > λ } ) d λ .
In the above, we switched the order of integration by applying Fubini-Tonelli, which we can do since everything is non-negative. We will first show the result for non-negative f f f and extend it to a general f f f . Consider the decomposition of f ≥ 0 f \geq 0 f ≥ 0 via a cutoff:
g = f χ { f < λ 2 } + λ 2 χ { f ≥ λ 2 } and h = f − g . g = f\chi_{\set{f < \frac{\lambda}{2}}} + \frac{\lambda}{2}\chi_{\set{f \geq \frac{\lambda}{2}}}
\quad\text{and}\quad
h = f - g. g = f χ { f < 2 λ } + 2 λ χ { f ≥ 2 λ } and h = f − g .
It's clear that g g g is continuous on { f ≠ λ 2 } \set{f \neq \frac{\lambda}{2}} { f = 2 λ } , since f f f is continuous. For x ∈ { f ( x ) = λ 2 } x \in \set{f\p{x} = \frac{\lambda}{2}} x ∈ { f ( x ) = 2 λ } , let { x n } n ⊆ R \set{x_n}_n \subseteq \R { x n } n ⊆ R be a sequence which converges to x x x . If x n ∈ { f ≥ λ 2 } x_n \in \set{f \geq \frac{\lambda}{2}} x n ∈ { f ≥ 2 λ } , then g ( x n ) = λ 2 g\p{x_n} = \frac{\lambda}{2} g ( x n ) = 2 λ . The remaining elements form a subsequence { x n k } k \set{x_{n_k}}_k { x n k } k and satisfies g ( x n k ) = f ( x n k ) g\p{x_{n_k}} = f\p{x_{n_k}} g ( x n k ) = f ( x n k ) . By continuity of f f f , g ( x n k ) → k → ∞ f ( x ) = λ 2 g\p{x_{n_k}} \xrightarrow{k\to\infty} f\p{x} = \frac{\lambda}{2} g ( x n k ) k → ∞ f ( x ) = 2 λ , so g g g is continuous. Since h h h is a difference of two continuous functions, h h h is also continuous. Because f f f had compact support, it follows that g g g and hence h h h have compact support as well.
Notice that by linearity of T T T , ∣ T f ( x ) ∣ ≤ ∣ T g ( x ) ∣ + ∣ T h ( x ) ∣ \abs{Tf\p{x}} \leq \abs{Tg\p{x}} + \abs{Th\p{x}} ∣ T f ( x ) ∣ ≤ ∣ T g ( x ) ∣ + ∣ T h ( x ) ∣ , and so
{ ∣ T f ( x ) ∣ > λ } ⊆ { ∣ T g ( x ) ∣ > λ 2 } ∪ { ∣ T h ( x ) ∣ > λ 2 } . \set{\abs{Tf\p{x}} > \lambda} \subseteq \set{\abs{Tg\p{x}} > \frac{\lambda}{2}} \cup \set{\abs{Th\p{x}} > \frac{\lambda}{2}}. { ∣ T f ( x ) ∣ > λ } ⊆ { ∣ T g ( x ) ∣ > 2 λ } ∪ { ∣ T h ( x ) ∣ > 2 λ } .
Otherwise, if both ∣ T g ( x ) ∣ , ∣ T h ( x ) ∣ ≤ λ 2 \abs{Tg\p{x}}, \abs{Th\p{x}} \leq \frac{\lambda}{2} ∣ T g ( x ) ∣ , ∣ T h ( x ) ∣ ≤ 2 λ , then ∣ T f ( x ) ∣ ≤ λ \abs{Tf\p{x}} \leq \lambda ∣ T f ( x ) ∣ ≤ λ . By construction, ∥ g ∥ L ∞ = λ 2 \norm{g}_{L^\infty} = \frac{\lambda}{2} ∥ g ∥ L ∞ = 2 λ , so by the first property of T T T , ∥ T g ∥ L ∞ ≤ ∥ T g ∥ L ∞ = λ 2 \norm{Tg}_{L^\infty} \leq \norm{Tg}_{L^\infty} = \frac{\lambda}{2} ∥ T g ∥ L ∞ ≤ ∥ T g ∥ L ∞ = 2 λ , and so { ∣ T g ( x ) ∣ > λ 2 } = ∅ \set{\abs{Tg\p{x}} > \frac{\lambda}{2}} = \emptyset { ∣ T g ( x ) ∣ > 2 λ } = ∅ , which gives
{ ∣ T f ( x ) ∣ > λ } ⊆ { ∣ T h ( x ) ∣ > λ 2 } . \set{\abs{Tf\p{x}} > \lambda} \subseteq \set{\abs{Th\p{x}} > \frac{\lambda}{2}}. { ∣ T f ( x ) ∣ > λ } ⊆ { ∣ T h ( x ) ∣ > 2 λ } .
Hence,
∫ R ∣ T f ( x ) ∣ 2 d x = ∫ 0 ∞ 2 λ m ( { ∣ T f ( x ) ∣ > λ } ) d λ ≤ ∫ 0 ∞ 2 λ ⋅ ∥ h ∥ L 1 λ / 2 d λ = 4 ∫ 0 ∞ ∫ 0 ∞ ∣ h ( x ) ∣ d x d λ = 4 ∫ 0 ∞ ∫ 0 ∞ ( f ( x ) − λ 2 ) χ { f ≥ λ 2 } d x d λ = 4 ∫ 0 ∞ ∫ { f ≥ λ 2 } f ( x ) − λ 2 d x d λ ≤ 4 ∫ 0 ∞ ∫ { f ≥ λ 2 } f ( x ) d x d λ = 4 ∫ 0 ∞ f ( x ) ∫ 0 2 f ( x ) d λ d x = 8 ∫ 0 ∞ ∣ f ( x ) ∣ 2 d x . \begin{aligned}
\int_\R \abs{Tf\p{x}}^2 \,\diff{x}
&= \int_0^\infty 2\lambda m\p{\set{\abs{Tf\p{x}} > \lambda}} \,\diff\lambda \\
&\leq \int_0^\infty 2\lambda \cdot \frac{\norm{h}_{L^1}}{\lambda/2} \,\diff\lambda \\
&= 4 \int_0^\infty \int_0^\infty \abs{h\p{x}} \,\diff{x} \,\diff\lambda \\
&= 4 \int_0^\infty \int_0^\infty \p{f\p{x} - \frac{\lambda}{2}}\chi_{\set{f \geq \frac{\lambda}{2}}} \,\diff{x} \,\diff\lambda \\
&= 4 \int_0^\infty \int_{\set{f \geq \frac{\lambda}{2}}} f\p{x} - \frac{\lambda}{2} \,\diff{x} \,\diff\lambda \\
&\leq 4 \int_0^\infty \int_{\set{f \geq \frac{\lambda}{2}}} f\p{x} \,\diff{x} \,\diff\lambda \\
&= 4 \int_0^\infty f\p{x} \int_0^{2f\p{x}} \,\diff\lambda \,\diff{x} \\
&= 8 \int_0^\infty \abs{f\p{x}}^2 \,\diff{x}.
\end{aligned} ∫ R ∣ T f ( x ) ∣ 2 d x = ∫ 0 ∞ 2 λm ( { ∣ T f ( x ) ∣ > λ } ) d λ ≤ ∫ 0 ∞ 2 λ ⋅ λ /2 ∥ h ∥ L 1 d λ = 4 ∫ 0 ∞ ∫ 0 ∞ ∣ h ( x ) ∣ d x d λ = 4 ∫ 0 ∞ ∫ 0 ∞ ( f ( x ) − 2 λ ) χ { f ≥ 2 λ } d x d λ = 4 ∫ 0 ∞ ∫ { f ≥ 2 λ } f ( x ) − 2 λ d x d λ ≤ 4 ∫ 0 ∞ ∫ { f ≥ 2 λ } f ( x ) d x d λ = 4 ∫ 0 ∞ f ( x ) ∫ 0 2 f ( x ) d λ d x = 8 ∫ 0 ∞ ∣ f ( x ) ∣ 2 d x .
Here, we were able to apply Fubini-Tonelli since f ≥ 0 f \geq 0 f ≥ 0 . For a general f f f , we may decompose f = f + − f − f = f^+ - f^- f = f + − f − where f + = max { 0 , f } f^+ = \max\,\set{0, f} f + = max { 0 , f } and f − = max { 0 , − f } f^- = \max\,\set{0, -f} f − = max { 0 , − f } , which are non-negative. Thus,
∫ ∣ T f ( x ) ∣ 2 d x = ∫ { f ≥ 0 } ∣ T f + ( x ) ∣ 2 d x + ∫ { f < 0 } ∣ T f − ( x ) ∣ 2 d x ≤ 8 ∫ ∣ f + ( x ) ∣ 2 d x + 8 ∫ ∣ f − ( x ) ∣ 2 d x ≤ 16 ∫ ∣ f ( x ) ∣ 2 d x , \begin{aligned}
\int \abs{Tf\p{x}}^2 \,\diff{x}
&= \int_{\set{f \geq 0}} \abs{Tf^+\p{x}}^2 \,\diff{x} + \int_{\set{f < 0}} \abs{Tf^-\p{x}}^2 \,\diff{x} \\
&\leq 8\int \abs{f^+\p{x}}^2 \,\diff{x} + 8\int \abs{f^-\p{x}}^2 \,\diff{x} \\
&\leq 16 \int \abs{f\p{x}}^2 \,\diff{x},
\end{aligned} ∫ ∣ T f ( x ) ∣ 2 d x = ∫ { f ≥ 0 } ∣ ∣ T f + ( x ) ∣ ∣ 2 d x + ∫ { f < 0 } ∣ ∣ T f − ( x ) ∣ ∣ 2 d x ≤ 8 ∫ ∣ ∣ f + ( x ) ∣ ∣ 2 d x + 8 ∫ ∣ ∣ f − ( x ) ∣ ∣ 2 d x ≤ 16 ∫ ∣ f ( x ) ∣ 2 d x ,
which completes the proof.