Lemma 13.5.6.label Let $[a, b] \subset \real$, $E$ be a separated locally convex space over $K \in \RC$, $F$ be a Banach space over $K$, $H$ be a complete locally convex space over $K$, all over $K \in \RC$, $E \times F \to H$ with $(x, y) \mapsto xy$ be a continuous bilinear map, $\gamma \in C([a, b]; F)$ be a piecewise $C^{1}$ curve that is constant on $[a, a + \eps)$ and $(b - \eps, b]$, and $U \in \cn_{F}(\gamma([a, b]))$.

Extend $\gamma$ to $\real$ by

\[\ol \gamma : \real \to U \quad x \mapsto \begin{cases}\gamma(a) &x \le a \\ \gamma(x) &x \in [a, b] \\ \gamma(b) &x \ge b\end{cases}\]

For each $\varphi \in C_{c}^{\infty}(\real; \real)$ with $\int_{\real} \varphi = 1$ and $t > 0$, let

\[\gamma_{t}: [a, b] \to F \quad x \mapsto \frac{1}{t}\int_{\real}\ol \gamma(y) \varphi\braks{\frac{x - y}{t}}dy\]

then

  1. (1)

    For each $t > 0$, $\gamma_{t} \in C^{\infty}([a, b]; F)$.

  2. (2)

    There exists $t > 0$ such that for any $s \in (0, t)$, $\gamma_{s}(a) = \gamma(a)$ and $\gamma_{s}(b) = \gamma(b)$.

  3. (3)

    For any $f \in C(U; E)$,

    \[\int_{\gamma} f = \lim_{t \downto 0}\int_{\gamma_t}f\]

Proof. (1): By the Mean Value Theorem, for each $x, y \in [a, b]$,

\[\norm{\frac{\varphi(x) - \varphi(y)}{x - y}}_{F} \le \sup_{z \in \real}\norm{D\varphi(z)}_{F}\]

By the Dominated Convergence Theorem, $\gamma_{t} \in C^{\infty}([a, b]; F)$.

(2): For sufficiently small $t$, $\supp{\varphi}\subset (-\eps, \eps)$. In which case, by assumption, $\gamma_{t}(a) = \gamma(a)$ and $\gamma_{t}(b) = \gamma(b)$.

(3): Since $\gamma$ is piecewise $C^{1}$ and $\gamma_{t} \in C^{\infty}([a, b]; F)$,

\[\int_{\gamma} f = \int_{a}^{b} f(t) D\gamma(t)dt = \lim_{t \downto 0}\int_{a}^{b} f(t) D\gamma_{t}(t) dt = \lim_{t \downto 0}\int_{\gamma_t}f\]

by the Dominated Convergence Theorem.$\square$

Post a Comment

Name:Email:
Please enter the tag of the current page (T3) to post the comment.
Tag: