From 5195c14025cf87fcc29e8d107260ad2b1a70605e Mon Sep 17 00:00:00 2001 From: Frank Schultz Date: Tue, 23 Apr 2024 19:58:30 +0200 Subject: [PATCH] mods for ex02 - [] -> () - better task for Signaloperationen: Zeitverschiebung und Negative Zeitskalierung --- random_hex.txt | 1 - tutorial_latex_deu/sig_sys_ex_02.tex | 767 +++++++++++++++++++-------- 2 files changed, 535 insertions(+), 233 deletions(-) diff --git a/random_hex.txt b/random_hex.txt index 35f3be0..49efd20 100644 --- a/random_hex.txt +++ b/random_hex.txt @@ -1,4 +1,3 @@ -E984089B66 41C1D0D586 3355CAE61A 27F5623A06 diff --git a/tutorial_latex_deu/sig_sys_ex_02.tex b/tutorial_latex_deu/sig_sys_ex_02.tex index ed21dbf..26d77fa 100644 --- a/tutorial_latex_deu/sig_sys_ex_02.tex +++ b/tutorial_latex_deu/sig_sys_ex_02.tex @@ -4,7 +4,7 @@ \section{UE 2: Elementarsignale- und Operationen, LTI-System Eigenschaften, Falt \label{sec:ue2_faltung} Wir müssen uns zunächst mit den zeitkontinuierlichen Elementarsignalen vertraut -machen, siehe \verb|fundamental_signals_CT.pdf| im StudIP unter Formelsammlung. +machen, siehe \verb|fundamental_signals_CT.pdf| im StudIP. Diese sind \begin{itemize} @@ -35,7 +35,7 @@ \subsection{Signaloperationen: Verschiebung und Skalierung} die Operationen Zeitverschiebung, Zeitskalierung und die Amplitudenskalierung, also -$A \cdot f(\left[t-\tau\right]\cdot \theta)$ mit $A,\tau,\theta\in\mathbb{R}$ +$A \cdot f(\left(t-\tau\right)\cdot \theta)$ mit $A,\tau,\theta\in\mathbb{R}$ klarmachen. % Wie man so schön sagt, müssen wir das für SigSys im Schlaf beherrschen. @@ -46,24 +46,25 @@ \subsection{Signaloperationen: Verschiebung und Skalierung} \begin{Werkzeug} Für \begin{equation} -A \cdot f(\left[t-\tau\right]\cdot \theta) \qquad A,\tau,\theta\in\mathbb{R} +A \cdot f(\left(t-\tau\right)\cdot \theta) \qquad A,\tau,\theta\in\mathbb{R} \end{equation} gilt \begin{itemize} \item Zeitverschiebung mit - $\tau$: für $\tau>0$ verzögert, rechtsverschoben; für $\tau<0$ vorauseilend, linksverschoben - \item Zeitskalierung mit $\theta$: für $\theta>1$ Stauchung; für $0<\theta<1$ Streckung - \item Amplitudenskalierung mit $A$: für $0<|A|<1$ Stauchung, für $|A|>1$ Streckung + $\tau$: für $\tau>0$ verzögert, d.h. rechtsverschoben; für $\tau<0$ vorauseilend, d.h. linksverschoben + \item Zeitskalierung mit $\theta$: für $\theta>1$ Stauchung (vgl. für $\theta\rightarrow \omega$ 'schneller schwingend'); für $0<\theta<1$ Streckung ('langsamer schwingend') + \item Amplitudenskalierung mit $A$: für $0<|A|<1$ Stauchung ('kleinere Amplitude'), für $|A|>1$ Streckung ('größere Amplitude') \item (für $A<0$ zusätzlich noch Invertieren der Signalpolarität) \item (für $\theta<0$ zusätzlich noch Invertieren des Zeitverlaufs) \end{itemize} \end{Werkzeug} \begin{Ansatz} -Es ist sinnvoll, das Signal in die Darstellung +Es ist zunächst sinnvoll, das Signal in die Darstellung \begin{equation} -x(t) = \frac{3}{4} \cdot \mathrm{rect}(\left[t-\frac{3}{2}\right]\cdot \frac{1}{3}) +x(t) = \frac{3}{4} \cdot \mathrm{rect}(\left(t-\frac{3}{2}\right)\cdot \frac{1}{3}) \end{equation} -umzuformen, weil dann $t$ ohne Vorfaktor und so wie obige Formel. +umzuformen, weil dann $t$ in der großen Klammer ohne Vorfaktor und exakt so wie +obige Werkzeug-Formel. Dann lässt sich Zeitverschiebung und Zeitskalierung getrennt und nacheinander abarbeiten. Hier also zunächst Zeitverzögerung um $\frac{3}{2}$ und dann zeitliche Streckung (das Signal wird 'langsamer' bzw. @@ -73,7 +74,7 @@ \subsection{Signaloperationen: Verschiebung und Skalierung} % Es ist empfehlenswert, sich das kochrezeptartig anzugewöhnen, bis es sitzt. -Bei einem cos()-Signal leuchtet uns die zeitliche Streckung, Stauchung vielleicht +Bei einem cos()-Signal leuchtet uns die zeitliche Streckung bzw. Stauchung vielleicht intuitiv ein: ausgehend von $\cos(t)$ schwingt $\cos(2\,t)$ schneller, das ist eine zeitliche Stauchung. % @@ -81,7 +82,7 @@ \subsection{Signaloperationen: Verschiebung und Skalierung} langsamer, das ist dann eine zeitliche Streckung. Machen wir uns klar, dass wir auch eine Verschiebung des Signals bezüglich der -Ordinate vornehmen können. Das heißt wir manipulieren den Gleichanteil. +Ordinate vornehmen können. Das heißt, wir manipulieren den Gleichanteil. % Signale mit Gleichanteil werden in den meisten signallastigen Ingenieursdisziplinen vermieden, weil man den Gleichanteil schlecht @@ -107,7 +108,7 @@ \subsection{Signaloperationen: Verschiebung und Skalierung} und die Dreieckfunktion mit \verb|tri(t)|, damit können wir eigene einfache Experimente zur Signalskalierung und -verschiebung machen. % -Das sollten wir in jedem Fall mal machen in einer Mußestunde! +Das sollten wir in jedem Fall mal machen in einer Mußestunde, bis es sitzt! %Mit UE 1.1 aus dem SS2019 können wir noch mehr üben. \end{Loesung} @@ -177,7 +178,7 @@ \subsection{Signaloperationen: Verschiebung und Skalierung} legend pos=outer north east, xlabel = {t}, %ylabel = {x(t)}, -title = {$x(t) = \mathrm{rect}(\left[t-\frac{3}{2}\right]\cdot \frac{1}{3}) = \mathrm{rect}( \frac{t}{3} - \frac{1}{2})$}, +title = {$x(t) = \mathrm{rect}(\left(t-\frac{3}{2}\right)\cdot \frac{1}{3}) = \mathrm{rect}( \frac{t}{3} - \frac{1}{2})$}, xmin=-4, xmax=4, ymin=-0.1, ymax=1.1, xtick={-4,-3,-2,-1,0,1,2,3,4}, @@ -203,7 +204,7 @@ \subsection{Signaloperationen: Verschiebung und Skalierung} legend pos=outer north east, xlabel = {t}, %ylabel = {x(t)}, -title = {$x(t) = \frac{3}{4} \cdot \mathrm{rect}(\left[t-\frac{3}{2}\right]\cdot \frac{1}{3})$}, +title = {$x(t) = \frac{3}{4} \cdot \mathrm{rect}(\left(t-\frac{3}{2}\right)\cdot \frac{1}{3})$}, xmin=-4, xmax=4, ymin=-0.1, ymax=1.1, xtick={-4,-3,-2,-1,0,1,2,3,4}, @@ -255,6 +256,7 @@ \subsection{Ableitung eines Cosinus-Signalausschnitts} \draw (0.05,-1) -- (-0.05,-1); \draw (0.05,1) -- (-0.05,1) node[left]{\small $1$}; \draw(2,0) node[below]{\small $2$}; +\draw(-2,0) node[below]{\small $-2$}; \begin{scope} \draw[C0, ultra thick, domain=-2:2,variable=\t,samples=100,smooth] plot(\t,{ cos (4*pi*\t r)}); \draw[C0, ultra thick] (-4,0) -- (-2,0); @@ -277,12 +279,12 @@ \subsection{Ableitung eines Cosinus-Signalausschnitts} Multiplikationseigenschaft des Dirac-Impulses \begin{align} &\delta(t-\tau) f(t) = \delta(t-\tau) f(\tau)\\ -&\delta(t) f(t) = \delta(t) f(0) \qquad \text{speziell für} \qquad \tau=0 +&\delta(t - 0) f(t) = \delta(t - 0) f(0) \qquad \text{speziell für} \qquad \tau=0 \end{align} \end{Werkzeug} \begin{Ansatz} -Typische SigSys-Denke: Zeitlich begrenzten Cosinus herstellen indem wir mit der +Typische SigSys-Denke: Zeitlich begrenzten Cosinus herstellen, indem wir mit der Rechteckfunktion einen Teil ausschneiden. Dann Mathematik betreiben. \end{Ansatz} \begin{ExCalc} @@ -298,7 +300,7 @@ \subsection{Ableitung eines Cosinus-Signalausschnitts} Breite von $\mathrm{rect}(t)$ ist 1 (hat auch Fläche 1!), wir brauchen aber Breite 4, daher $\mathrm{rect}(t\cdot \frac{1}{4})$ (zur Erinnerung: die Funktion soll sich bei Streckung langsamer über der Zeit ändern, daher -Skalierungsfaktor kleiner 1) +Skalierungsfaktor kleiner 1, also hier $\frac{1}{4}$). Daher kann das Signal als \begin{align} @@ -313,9 +315,11 @@ \subsection{Ableitung eines Cosinus-Signalausschnitts} \frac{\fsd }{\fsd t} \cos(4\pi t) \cdot \mathrm{rect}(\frac{t}{4})\\ \end{align} benötigen wir nun die Produkt- und Kettenregel, sowie die Ableitung des Rechtecksignals. +% -Wir können das Rechtecksignal als Überlagerung von zwei Sprungfunktionen darstellen, -nämlich +Letzteres ist auf direktem Wege nicht besonders gut zu handhaben, aber mit einem eleganten, sehr SigSys-typischem Umweg. +% +Wir können nämlich das Rechtecksignal als Überlagerung von zwei Sprungfunktionen darstellen als \begin{equation} \mathrm{rect}(\frac{t}{4}) = \epsilon(t+2) -\epsilon(t-2), \end{equation} @@ -352,11 +356,12 @@ \subsection{Ableitung eines Cosinus-Signalausschnitts} \end{tikzpicture} \end{center} -Nun brauchen wir folgende wichtige Erkenntnis (die nur indirekt in der Formelsammlung +Nun brauchen wir folgende fundamentale (Need-To-Know!) Signalrelation (die deswegen +absichtlich auch nur indirekt in der Formelsammlung steht, nämlich bei den Korrespondenzen zur Laplace Transformation) und in \fig{fig:114F06AFAA} veranschaulicht ist \begin{equation} - \epsilon(t)' = \delta(t) + \epsilon(t)' = \delta(t). \end{equation} Damit können wir jetzt ableiten. @@ -430,7 +435,7 @@ \subsection{Ableitung eines Cosinus-Signalausschnitts} coordinates {(-2,0)(0,0)(4,4)}; \end{axis} \end{tikzpicture} -\caption{$t \cdot \epsilon(t)$.} +\caption{Rampe $t \cdot \epsilon(t)$.} %\label{fig:} \end{subfigure} % @@ -509,7 +514,7 @@ \subsection{Ableitung eines Cosinus-Signalausschnitts} \begin{Loesung} \begin{align} -x(t) = \cos(4\pi t) \cdot \mathrm{rect}(\frac{t}{4})\\ +x(t) = \cos(4\pi t) \cdot \mathrm{rect}(\frac{t}{4}) \end{align} \begin{align}x'(t)= @@ -551,23 +556,18 @@ \subsection{Ableitung eines Cosinus-Signalausschnitts} $+\delta(t+2)$ hat Gewicht $+1$ und ist um Zeit 2 zeitlich vorauseilend. \end{Loesung} - - - - \newpage \subsection{Signaloperationen: Zeitverschiebung und Negative Zeitskalierung} -\label{sec:FBE36B0684} +\label{sec:E984089B66} \begin{Ziel} Es ist wichtig, dass wir uns Zeitverschiebung im Verbund mit negativer Zeitskalierung als Konzept der Zeitspiegelung bezüglich einer Achse zu einem speziellen Zeitpunkt -klarmachen. Diese Signaloperation ist fundamental wichtig in der Faltung. +klarmachen. Diese Signaloperation ist fundamental wichtig für die Faltung. \end{Ziel} -\textbf{Aufgabe} {\tiny FBE36B0684}: Stellen Sie die Signale -$x_1(t) = \mathrm{e}^{-t-1} \cdot \epsilon(t+1)$ und -$x_2(t) = \mathrm{e}^{t-2} \cdot \epsilon(-t+2)$ -grafisch dar und machen Sie deutlich wie aus Signal $x_1(t)$ mittels Zeitverschiebung -und negativer Zeitskalierung das Signal $x_2(t)$ resultiert. +\textbf{Aufgabe} {\tiny E984089B66}: Geben Sie analytische Ausdrücke für die in +\fig{fig:E984089B66_1} und \fig{fig:E984089B66_2} dargestellten Signale an. +Prüfen Sie mit Formeln den Übergang von $x(t) \rightarrow x_2(t)$ und +$x(t) \rightarrow x_4(t)$ mittels Zeitverschiebung und negativer Zeitskalierung. \begin{Werkzeug} Wir brauchen die Konzepte (a) der Zeitverschiebung, hier speziell Zeitverzögerung, also Verschiebung des Signals nach rechts und (b) der Umkehr des Funktionsarguments, @@ -580,7 +580,7 @@ \subsection{Signaloperationen: Zeitverschiebung und Negative Zeitskalierung} mit Wolfram Alpha als Referenzlösung anschauen. \end{Werkzeug} \begin{Ansatz} -Es gibt zwei Varianten um von $x_1(t)$ zu $x_2(t)$ zu gelangen: +Es gibt zwei Varianten um von $x(t)$ zu $x_2(t)=x_4(t)$ zu gelangen. Variante I: erst Zeitskalierung $\theta=-1$, dann Zeitverschiebung um $\tau$ @@ -591,220 +591,523 @@ \subsection{Signaloperationen: Zeitverschiebung und Negative Zeitskalierung} Mit Wolfram Alpha \url{https://www.wolframalpha.com} können wir uns bei Unsicherheit Klarheit verschaffen. % -Ansonsten ist das eine exzellente Gelegenheit für Stift und Papier. +Ansonsten ist das eine exzellente Gelegenheit für Stift und Papier, +Zeichnen und ein bisschen Rechnen. \end{ExCalc} \begin{Loesung} -Für Variante I siehe \fig{fig:FBE36B0684_1}, für Variante II siehe -\ref{fig:FBE36B0684_2}. - -Das Faltungsintegral -\begin{equation} -y(t) -= \int\limits_{-\infty}^{+\infty} x(-\tau+t) h(\tau) \, \fsd \tau -= \int\limits_{-\infty}^{+\infty} x(\tau) h(-\tau+t) \, \fsd \tau -\end{equation} -werden wir in den nächsten Aufgaben benutzen und verstehen. +Für Variante I siehe \fig{fig:E984089B66_1}, für Variante II siehe +\fig{fig:E984089B66_2}. % -Wir sehen hier, dass genau die hier -diskutierte Zeitverschiebung und negative Zeitskalierung benutzt werden muss, um -die Signaloperationen $x(\tau) \rightarrow x(-\tau+t)$ -oder $h(\tau) \rightarrow h(-\tau+t)$ abzubilden. -\textbf{Wir beachten aber}, dass im Faltungsintegral die temporäre Zeitvariable -$\tau$ (das ist keine RC-Glied Zeitkonstante o.ä.) ist und die Zeit $t$ -im Ergebnissignal $y(t)$ die Verschiebungsvariable in der Faltung entspricht. +\begin{itemize} +\setlength\itemsep{-0.5em} +\item Variante I: erst Zeitskalierung $\theta=-1$, dann Zeitverschiebung um $\tau$ +\item für $x(-t+\tau)$ ergibt $\tau > 0$ eine Rechtsverschiebung von $x(-t)$, Beispiel unten +\item für $x(-t+\tau)$ ergibt $\tau < 0$ eine Linksverschiebung von $x(-t)$ +\item $x_2(t)$ ergibt sich aus der Spiegelung von $x(t)$ an der vertikalen Achse bei $t=\frac{\tau}{2}$ +\end{itemize} +% +\begin{itemize} +\setlength\itemsep{-0.5em} +\item Variante II: erst Zeitverschiebung um $\tau$, dann Zeitskalierung $\theta=-1$ +\item für $x(t-\tau)$ ergibt $\tau > 0$ eine Rechtsverschiebung von $x(t)$, Beispiel unten +\item für $x(t-\tau)$ ergibt $\tau < 0$ eine Linksverschiebung von $x(t)$ +\item $x_4(t)$ ergibt sich aus der Spiegelung von $x_3(t)$ (nicht $x(t)$!) an der vertikalen Achse bei $t=\tau$ +\end{itemize} \end{Loesung} -\begin{figure*}[h] -\centering -\begin{subfigure}{0.45\textwidth} -\begin{tikzpicture} -\begin{axis}[ -width=1\textwidth, -height=0.45\textwidth, -domain=-1:4, -samples=50, -legend pos=outer north east, -xlabel = {t}, -%ylabel = {x(t)}, -title = {$x_1(t) = \mathrm{e}^{-\left[t+1\right]} \cdot \epsilon(\left[t+1\right])$}, -xmin=-4, xmax=4, -ymin=-0.1, ymax=1.1, -xtick={-4,-3,-2,-1,0,1,2,3,4}, -ytick={0,0.5,1,1.5,2}, -ymajorgrids=true, -xmajorgrids=true -] -\addplot[mark=None, color=C0, ultra thick] -{exp(-x-1)}; -\addplot[mark=None, color=C0, ultra thick] -coordinates {(-4,0)(-1,0)(-1,1)}; -\end{axis} +\begin{figure}[h!] +\begin{center} +\begin{tikzpicture}[scale=0.8] +\def \xmin {-6} +\def \xmax {+7.25} +\def \ymin {0} +\def \ymax {1.25} +\def \tauv {1/2} % > 0 !!! +\draw[->] (\xmin-0.1 ,0) -- (\xmax+0.1,0) node[right]{$t$}; +\draw[->] (0,\ymin-0.1) -- (0,\ymax+0.1) node[above]{$x(t)$}; +\foreach \x in {\xmin,-5.5,...,\xmax}{\draw (\x,0.05) -- (\x,-0.05);} +\draw (0.05,1) -- (-0.05,1) node[left]{\small $1$}; +\draw(0,0) node[below]{\small $0$}; +\draw(1,0) node[below]{\small $\frac{2}{2}$}; +\draw(2,0) node[below]{\small $\frac{4}{2}$}; +\draw(3,0) node[below]{\small $\frac{6}{2}$}; +\draw(4,0) node[below]{\small $\frac{8}{2}$}; +\draw(5,0) node[below]{\small $\frac{10}{2}$}; +\draw(6,0) node[below]{\small $\frac{12}{2}$}; +\draw(-1.125,0) node[below]{\small $-\frac{2}{2}$}; +\draw(-2.125,0) node[below]{\small $-\frac{4}{2}$}; +\draw(-3.125,0) node[below]{\small $-\frac{6}{2}$}; +\draw(-4.125,0) node[below]{\small $-\frac{8}{2}$}; +\draw(-5.125,0) node[below]{\small $-\frac{10}{2}$}; +\draw(-6.125,0) node[below]{\small $-\frac{12}{2}$}; +\draw[C0, ultra thick] (1/2,0) -- (1/2,1) -- (5/2,1) -- (5/2,0) -- (7/2,1) -- (9/2,0); +\draw[C0, ultra thick,->] (6,0) -- (6,1) node[right]{$(1)$}; +\draw[C2, thick] (\tauv/2,1) -- (\tauv/2,-1); \end{tikzpicture} -\caption{Ausgangspunkt: abfallende, vorauseilende Exponential-Funktion.} -%\label{fig:} -\end{subfigure} +\end{center} % -\begin{subfigure}{0.45\textwidth} -\begin{tikzpicture} -\begin{axis}[ -width=1\textwidth, -height=0.45\textwidth, -domain=-4:-1, -samples=50, -legend pos=outer north east, -xlabel = {t}, -%ylabel = {x(t)}, -title = {$\mathrm{e}^{+\left[t+1\right]} \cdot \epsilon(-\left[t+1\right])$}, -xmin=-4, xmax=4, -ymin=-0.1, ymax=1.1, -xtick={-4,-3,-2,-1,0,1,2,3,4}, -ytick={0,0.5,1,1.5,2}, -ymajorgrids=true, -xmajorgrids=true -] -\addplot[mark=None, color=C0, ultra thick] -{exp(x+1)}; -\addplot[mark=None, color=C0, ultra thick] -coordinates {(-1,1)(-1,0)(4,0)}; -\end{axis} -\end{tikzpicture} -\caption{Umkehr der Funktionsargumente von (a) ist Spiegelung an der $t=-1$ Achse.} -%\label{fig:} -\end{subfigure} - -\begin{subfigure}{0.45\textwidth} -\begin{tikzpicture} -\begin{axis}[ -width=1\textwidth, -height=0.45\textwidth, -domain=-4:2, -samples=50, -legend pos=outer north east, -xlabel = {t}, -%ylabel = {x(t)}, -title = {$x_2(t) = \mathrm{e}^{(+\left[t+1-3\right])} \cdot \epsilon(-\left[t+1-3\right])$}, -xmin=-4, xmax=4, -ymin=-0.1, ymax=1.1, -xtick={-4,-3,-2,-1,0,1,2,3,4}, -ytick={0,0.5,1,1.5,2}, -ymajorgrids=true, -xmajorgrids=true -] -\addplot[mark=None, color=C0, ultra thick] -{exp(+(x-2))}; -\addplot[mark=None, color=C0, ultra thick] -coordinates {(2,1)(2,0)(4,0)}; -\end{axis} +\begin{center} +\begin{tikzpicture}[scale=0.8] +\def \xmin {-6} +\def \xmax {+7.25} +\def \ymin {0} +\def \ymax {1.25} +\draw[->] (\xmin-0.1 ,0) -- (\xmax+0.1,0) node[right]{$t$}; +\draw[->] (0,\ymin-0.1) -- (0,\ymax+0.1) node[above]{$x_1(t) = x(\theta t) = x(-t)$}; +\foreach \x in {\xmin,-5.5,...,\xmax}{\draw (\x,0.05) -- (\x,-0.05);} +\draw (0.05,1) -- (-0.05,1) node[left]{\small $1$}; +\draw(0,0) node[below]{\small $0$}; +\draw(1,0) node[below]{\small $\frac{2}{2}$}; +\draw(2,0) node[below]{\small $\frac{4}{2}$}; +\draw(3,0) node[below]{\small $\frac{6}{2}$}; +\draw(4,0) node[below]{\small $\frac{8}{2}$}; +\draw(5,0) node[below]{\small $\frac{10}{2}$}; +\draw(6,0) node[below]{\small $\frac{12}{2}$}; +\draw(-1.125,0) node[below]{\small $-\frac{2}{2}$}; +\draw(-2.125,0) node[below]{\small $-\frac{4}{2}$}; +\draw(-3.125,0) node[below]{\small $-\frac{6}{2}$}; +\draw(-4.125,0) node[below]{\small $-\frac{8}{2}$}; +\draw(-5.125,0) node[below]{\small $-\frac{10}{2}$}; +\draw(-6.125,0) node[below]{\small $-\frac{12}{2}$}; +\draw[C0, ultra thick] (-1/2,0) -- (-1/2,1) -- (-5/2,1) -- (-5/2,0) -- (-7/2,1) -- (-9/2,0); +\draw[C0, ultra thick,->] (-6,0) -- (-6,1) node[right]{$(1)$}; \end{tikzpicture} -\caption{Erst Umkehr der Funktionsargumente von (a) und dann Zeitverzögerung von -(b) um $3$.} -%\label{fig:} -\end{subfigure} -% +\end{center} % +\begin{center} +\begin{tikzpicture}[scale=0.8] +\def \xmin {-6} +\def \xmax {+7.25} +\def \ymin {0} +\def \ymax {1.25} +\def \tauv {1/2} % > 0 !!! +\draw[->] (\xmin-0.1 ,0) -- (\xmax+0.1,0) node[right]{$t$}; +\draw[->] (0,\ymin-0.1) -- (0,\ymax+0.1) node[above]{$x_2(t) = x(\theta t + \tau) = x(-t+\tau) = x(-(t-\tau)),\quad \tau=\tauv$}; +\foreach \x in {\xmin,-5.5,...,\xmax}{\draw (\x,0.05) -- (\x,-0.05);} +\draw (0.05,1) -- (-0.05,1) node[left]{\small $1$}; +\draw(0,0) node[below]{\small $0$}; +\draw(1,0) node[below]{\small $\frac{2}{2}$}; +\draw(2,0) node[below]{\small $\frac{4}{2}$}; +\draw(3,0) node[below]{\small $\frac{6}{2}$}; +\draw(4,0) node[below]{\small $\frac{8}{2}$}; +\draw(5,0) node[below]{\small $\frac{10}{2}$}; +\draw(6,0) node[below]{\small $\frac{12}{2}$}; +\draw(-1.125,0) node[below]{\small $-\frac{2}{2}$}; +\draw(-2.125,0) node[below]{\small $-\frac{4}{2}$}; +\draw(-3.125,0) node[below]{\small $-\frac{6}{2}$}; +\draw(-4.125,0) node[below]{\small $-\frac{8}{2}$}; +\draw(-5.125,0) node[below]{\small $-\frac{10}{2}$}; +\draw(-6.125,0) node[below]{\small $-\frac{12}{2}$}; +\draw[C0, ultra thick] (-1/2+\tauv,0) -- (-1/2+\tauv,1) -- (-5/2+\tauv,1) -- (-5/2+\tauv,0) -- (-7/2+\tauv,1) -- (-9/2+\tauv,0); +\draw[C0, ultra thick,->] (-6+\tauv,0) -- (-6+\tauv,1) node[right]{$(1)$}; +\draw[C2, thick] (\tauv/2,1) -- (\tauv/2,-1); +\end{tikzpicture} +\end{center} +\caption{Skizzenfahrplan für Variante I: \textbf{zuerst Zeitskalierung $\theta t=-1 \cdot t$, dann Zeitverschiebung $\theta t + \tau$}.} +\label{fig:E984089B66_1} +\end{figure} % -\caption{Skizzenfahrplan für Aufgabe \ref{sec:FBE36B0684}, -\textbf{zuerst Zeitskalierung $\theta=-1$, dann Zeitverschiebung}. -Vgl. \texttt{TimeMirrorShift\_FBE36B0684.ipynb}} -\label{fig:FBE36B0684_1} -\end{figure*} % % +\begin{figure}[h!] +\begin{center} +\begin{tikzpicture}[scale=0.8] +\def \xmin {-6} +\def \xmax {+7.25} +\def \ymin {0} +\def \ymax {1.25} +\draw[->] (\xmin-0.1 ,0) -- (\xmax+0.1,0) node[right]{$t$}; +\draw[->] (0,\ymin-0.1) -- (0,\ymax+0.1) node[above]{$x(t)$}; +\foreach \x in {\xmin,-5.5,...,\xmax}{\draw (\x,0.05) -- (\x,-0.05);} +\draw (0.05,1) -- (-0.05,1) node[left]{\small $1$}; +\draw(0,0) node[below]{\small $0$}; +\draw(1,0) node[below]{\small $\frac{2}{2}$}; +\draw(2,0) node[below]{\small $\frac{4}{2}$}; +\draw(3,0) node[below]{\small $\frac{6}{2}$}; +\draw(4,0) node[below]{\small $\frac{8}{2}$}; +\draw(5,0) node[below]{\small $\frac{10}{2}$}; +\draw(6,0) node[below]{\small $\frac{12}{2}$}; +\draw(-1.125,0) node[below]{\small $-\frac{2}{2}$}; +\draw(-2.125,0) node[below]{\small $-\frac{4}{2}$}; +\draw(-3.125,0) node[below]{\small $-\frac{6}{2}$}; +\draw(-4.125,0) node[below]{\small $-\frac{8}{2}$}; +\draw(-5.125,0) node[below]{\small $-\frac{10}{2}$}; +\draw(-6.125,0) node[below]{\small $-\frac{12}{2}$}; +\draw[C0, ultra thick] (1/2,0) -- (1/2,1) -- (5/2,1) -- (5/2,0) -- (7/2,1) -- (9/2,0); +\draw[C0, ultra thick,->] (6,0) -- (6,1) node[right]{$(1)$}; +\end{tikzpicture} +\end{center} % -\begin{figure*}[h] -\centering -\begin{subfigure}{0.45\textwidth} -\begin{tikzpicture} -\begin{axis}[ -width=1\textwidth, -height=0.45\textwidth, -domain=-1:4, -samples=50, -legend pos=outer north east, -xlabel = {t}, -%ylabel = {x(t)}, -title = {$x_1(t) = \mathrm{e}^{-\left[t+1\right]} \cdot \epsilon(\left[t+1\right])$}, -xmin=-4, xmax=4, -ymin=-0.1, ymax=1.1, -xtick={-4,-3,-2,-1,0,1,2,3,4}, -ytick={0,0.5,1,1.5,2}, -ymajorgrids=true, -xmajorgrids=true -] -\addplot[mark=None, color=C0, ultra thick] -{exp(-(x+1))}; -\addplot[mark=None, color=C0, ultra thick] -coordinates {(-4,0)(-1,0)(-1,1)}; -\end{axis} +\begin{center} +\begin{tikzpicture}[scale=0.8] +\def \xmin {-6} +\def \xmax {+7.25} +\def \ymin {0} +\def \ymax {1.25} +\def \tauv {1/2} % > 0 !!! +\draw[->] (\xmin-0.1 ,0) -- (\xmax+0.1,0) node[right]{$t$}; +\draw[->] (0,\ymin-0.1) -- (0,\ymax+0.1) node[above]{$x_3(t') = x(t-\tau),\quad \tau = \tauv$}; +\foreach \x in {\xmin,-5.5,...,\xmax}{\draw (\x,0.05) -- (\x,-0.05);} +\draw (0.05,1) -- (-0.05,1) node[left]{\small $1$}; +\draw(0,0) node[below]{\small $0$}; +\draw(1,0) node[below]{\small $\frac{2}{2}$}; +\draw(2,0) node[below]{\small $\frac{4}{2}$}; +\draw(3,0) node[below]{\small $\frac{6}{2}$}; +\draw(4,0) node[below]{\small $\frac{8}{2}$}; +\draw(5,0) node[below]{\small $\frac{10}{2}$}; +\draw(6,0) node[below]{\small $\frac{12}{2}$}; +\draw(-1.125,0) node[below]{\small $-\frac{2}{2}$}; +\draw(-2.125,0) node[below]{\small $-\frac{4}{2}$}; +\draw(-3.125,0) node[below]{\small $-\frac{6}{2}$}; +\draw(-4.125,0) node[below]{\small $-\frac{8}{2}$}; +\draw(-5.125,0) node[below]{\small $-\frac{10}{2}$}; +\draw(-6.125,0) node[below]{\small $-\frac{12}{2}$}; +\draw[C0, ultra thick] (1/2+\tauv,0) -- (1/2+\tauv,1) -- (5/2+\tauv,1) -- (5/2+\tauv,0) -- (7/2+\tauv,1) -- (9/2+\tauv,0); +\draw[C0, ultra thick,->] (6+\tauv,0) -- (6+\tauv,1) node[right]{$(1)$}; +\draw[C2, thick] (\tauv,1) -- (\tauv,-1); \end{tikzpicture} -\caption{Ausgangspunkt: abfallende, vorauseilende Exponential-Funktion.} -%\label{fig:} -\end{subfigure} +\end{center} % -\begin{subfigure}{0.45\textwidth} -\begin{tikzpicture} -\begin{axis}[ -width=1\textwidth, -height=0.45\textwidth, -domain=2:4, -samples=50, -legend pos=outer north east, -xlabel = {t}, -%ylabel = {x(t)}, -title = {$\mathrm{e}^{(-\left[t+1-3\right])} \cdot \epsilon(\left[t+1-3\right])$}, -xmin=-4, xmax=4, -ymin=-0.1, ymax=1.1, -xtick={-4,-3,-2,-1,0,1,2,3,4}, -ytick={0,0.5,1,1.5,2}, -ymajorgrids=true, -xmajorgrids=true -] -\addplot[mark=None, color=C0, ultra thick] -{exp(-(x-2))}; -\addplot[mark=None, color=C0, ultra thick] -coordinates {(-4,0)(2,0)(2,1)}; -\end{axis} +\begin{center} +\begin{tikzpicture}[scale=0.8] +\def \xmin {-6} +\def \xmax {+7.25} +\def \ymin {0} +\def \ymax {1.25} +\def \tauv {1/2} % > 0 !!! +\draw[->] (\xmin-0.1 ,0) -- (\xmax+0.1,0) node[right]{$t$}; +\draw[->] (0,\ymin-0.1) -- (0,\ymax+0.1) node[above]{$x_4(t) = x_2(t) = x_3(\theta t') = x(-(t-\tau))=x(-t+\tau),\quad \tau = \tauv$}; +\foreach \x in {\xmin,-5.5,...,\xmax}{\draw (\x,0.05) -- (\x,-0.05);} +\draw (0.05,1) -- (-0.05,1) node[left]{\small $1$}; +\draw(0,0) node[below]{\small $0$}; +\draw(1,0) node[below]{\small $\frac{2}{2}$}; +\draw(2,0) node[below]{\small $\frac{4}{2}$}; +\draw(3,0) node[below]{\small $\frac{6}{2}$}; +\draw(4,0) node[below]{\small $\frac{8}{2}$}; +\draw(5,0) node[below]{\small $\frac{10}{2}$}; +\draw(6,0) node[below]{\small $\frac{12}{2}$}; +\draw(-1.125,0) node[below]{\small $-\frac{2}{2}$}; +\draw(-2.125,0) node[below]{\small $-\frac{4}{2}$}; +\draw(-3.125,0) node[below]{\small $-\frac{6}{2}$}; +\draw(-4.125,0) node[below]{\small $-\frac{8}{2}$}; +\draw(-5.125,0) node[below]{\small $-\frac{10}{2}$}; +\draw(-6.125,0) node[below]{\small $-\frac{12}{2}$}; +\draw[C0, ultra thick] (-1/2+\tauv,0) -- (-1/2+\tauv,1) -- (-5/2+\tauv,1) -- (-5/2+\tauv,0) -- (-7/2+\tauv,1) -- (-9/2+\tauv,0); +\draw[C0, ultra thick,->] (-6+\tauv,0) -- (-6+\tauv,1) node[right]{$(1)$}; +\draw[C2, thick] (\tauv,1) -- (\tauv,-1); \end{tikzpicture} -\caption{Zeitverzögerung von (a) um $3$.} -%\label{fig:} -\end{subfigure} +\end{center} +\caption{Skizzenfahrplan für Variante II: +\textbf{zuerst Zeitverschiebung $t' = t-\tau$, dann Zeitskalierung $\theta t'=-1 \cdot (t-\tau)$}.} +\label{fig:E984089B66_2} +\end{figure} -\begin{subfigure}{0.45\textwidth} -\begin{tikzpicture} -\begin{axis}[ -width=1\textwidth, -height=0.45\textwidth, -domain=-4:2, -samples=50, -legend pos=outer north east, -xlabel = {t}, -%ylabel = {x(t)}, -title = {$x_2(t) = \mathrm{e}^{(+\left[t+1-3\right])} \cdot \epsilon(-\left[t+1-3\right])$}, -xmin=-4, xmax=4, -ymin=-0.1, ymax=1.1, -xtick={-4,-3,-2,-1,0,1,2,3,4}, -ytick={0,0.5,1,1.5,2}, -ymajorgrids=true, -xmajorgrids=true -] -\addplot[mark=None, color=C0, ultra thick] -{exp(+(x-2))}; -\addplot[mark=None, color=C0, ultra thick] -coordinates {(2,1)(2,0)(4,0)}; -\end{axis} -\end{tikzpicture} -\caption{Erst Zeitverzögerung von (a) um $3$ und danach Umkehr der Funktionsargumente von (b). -Daher Spiegelung des Signals von (b) an der $t=2$ Achse.} -%\label{fig:} -\end{subfigure} -% +\begin{Loesung} +\textbf{Variante I} folgend: +\begin{align} +x(t) = &\mathrm{rect}(\left(t-\frac{3}{2}\right)\cdot \frac{1}{2}) + \Lambda(\left(t-\frac{7}{2}\right)\cdot 1) + \delta(t-6)\\ +x_1(t) = x(-t) = &\mathrm{rect}(\left(-t-\frac{3}{2}\right)\cdot \frac{1}{2}) + \Lambda(\left(-t-\frac{7}{2}\right)\cdot 1) + \delta(-t-6)\\ +x_2(t) = x(-t+\tau) = &\mathrm{rect}(\left(-t+\tau-\frac{3}{2}\right)\cdot \frac{1}{2}) + \Lambda(\left(-t+\tau-\frac{7}{2}\right)\cdot 1) + \delta(-t+\tau-6) +\end{align} +mit speziellem $\tau = \frac{1}{2}$ +\begin{align} +x_2(t) = &\mathrm{rect}(\left(-t+\frac{1}{2}-\frac{3}{2}\right)\cdot \frac{1}{2}) + \Lambda(\left(-t+\frac{1}{2}-\frac{7}{2}\right)\cdot 1) + \delta(-t+\frac{1}{2}-6)\\ +x_2(t) = &\mathrm{rect}(\left(-t-\frac{2}{2}\right)\cdot \frac{1}{2}) + \Lambda(\left(-t-\frac{6}{2}\right)\cdot 1) + \delta(-t-\frac{11}{2})\\ +x_2(t) = &\mathrm{rect}(-\left(t+\frac{2}{2}\right)\cdot \frac{1}{2}) + \Lambda(-\left(t+\frac{6}{2}\right)\cdot 1) + \delta(-(t+\frac{11}{2})) +\end{align} +Da $\mathrm{rect}(-t)=\mathrm{rect}(t)$, $\Lambda(-t)=\Lambda(t)$ und $\delta(-t) = \delta(t)$, also alles axialsymmetrische Signale +\begin{align} +x_2(t) = &\mathrm{rect}(\left(t+\frac{2}{2}\right)\cdot \frac{1}{2}) + \Lambda(\left(t+\frac{6}{2}\right)\cdot 1) + \delta(t+\frac{11}{2}). +\end{align} +Dies ist konsistent mit dem Signal $x_2(t)$, was wir direkt aus der Grafik aufstellen können +\begin{align} +x_2(t) = &\mathrm{rect}(\left(t+\frac{2}{2}\right)\cdot \frac{1}{2}) + \Lambda(\left(t+\frac{6}{2}\right)\cdot 1) + \delta(t+\frac{11}{2}) +\end{align} % +\textbf{Variante II} folgend für $\tau=\frac{1}{2}$: +\begin{align} +x(t) = &\mathrm{rect}(\left(t-\frac{3}{2}\right)\cdot \frac{1}{2}) + \Lambda(\left(t-\frac{7}{2}\right)\cdot 1) + \delta(t-6)\\ +x_3(t') = x(t-\tau) = &\mathrm{rect}(\left(t-\tau-\frac{3}{2}\right)\cdot \frac{1}{2}) + \Lambda(\left(t-\tau-\frac{7}{2}\right)\cdot 1) + \delta(t-\tau-6)\\ +x_4(t) = x_3(-t') = x(-(t-\tau)) = &\mathrm{rect}(\left(-(t-\tau)-\frac{3}{2}\right)\cdot \frac{1}{2}) + \Lambda(\left(-(t-\tau)-\frac{7}{2}\right)\cdot 1) + \delta(-(t-\tau)-6)\\ +x_4(t) = &\mathrm{rect}(-\left(t-\tau+\frac{3}{2}\right)\cdot \frac{1}{2}) + \Lambda(-\left(t-\tau+\frac{7}{2}\right)\cdot 1) + \delta(-(t-\tau+6))\\ +x_4(t) = &\mathrm{rect}(-\left(t+\frac{2}{2}\right)\cdot \frac{1}{2}) + \Lambda(-\left(t+\frac{6}{2}\right)\cdot 1) + \delta(-(t+\frac{11}{2}))\\ +\end{align} +Da $\mathrm{rect}(-t)=\mathrm{rect}(t)$, $\Lambda(-t)=\Lambda(t)$ und $\delta(-t) = \delta(t)$, also alles axialsymmetrische Signale +\begin{align} +x_4(t) = &\mathrm{rect}(\left(t+\frac{2}{2}\right)\cdot \frac{1}{2}) + \Lambda(\left(t+\frac{6}{2}\right)\cdot 1) + \delta(t+\frac{11}{2}) +\end{align} +was wir auch bekommen würden, wenn wir $x_4(t)$ direkt aus der Grafik herauslesen würden. Zudem haben wir gezeigt, dass wie erwartet $x_4(t)=x_2(t)$, d.h. die Varianten I und II konsistente Lösungswege sind. + +\end{Loesung} + +\begin{Werkzeug} +Das Faltungsintegral +\begin{equation} +y(t) += \int\limits_{-\infty}^{+\infty} x(-\tau+t) h(\tau) \, \fsd \tau += \int\limits_{-\infty}^{+\infty} x(\tau) h(-\tau+t) \, \fsd \tau +\end{equation} +werden wir in den nächsten Aufgaben benutzen und verstehen. % -\caption{Skizzenfahrplan für Aufgabe \ref{sec:FBE36B0684}, -\textbf{zuerst Zeitverschiebung, dann Zeitskalierung $\theta=-1$}. -Vgl. \texttt{TimeMirrorShift\_FBE36B0684.ipynb}} -\label{fig:FBE36B0684_2} -\end{figure*} +Wir sehen hier, dass genau die hier +diskutierte Zeitverschiebung und negative Zeitskalierung benutzt werden muss, um +die Signaloperationen $x(\tau) \rightarrow x(-\tau+t)$ +oder $h(\tau) \rightarrow h(-\tau+t)$ abzubilden. +\textbf{Wir beachten aber}, dass im Faltungsintegral die temporäre Zeitvariable +$\tau$ (das ist keine RC-Glied Zeitkonstante o.ä.) ist und die Zeit $t$ +im Ergebnissignal $y(t)$ die Verschiebungsvariable in der Faltung entspricht. +\end{Werkzeug} + +% \newpage +% \subsection{Signaloperationen: Zeitverschiebung und Negative Zeitskalierung} +% \label{sec:FBE36B0684} +% \begin{Ziel} +% Es ist wichtig, dass wir uns Zeitverschiebung im Verbund mit negativer Zeitskalierung +% als Konzept der Zeitspiegelung bezüglich einer Achse zu einem speziellen Zeitpunkt +% klarmachen. Diese Signaloperation ist fundamental wichtig in der Faltung. +% \end{Ziel} +% \textbf{Aufgabe} {\tiny FBE36B0684}: Stellen Sie die Signale +% $x_1(t) = \mathrm{e}^{-t-1} \cdot \epsilon(t+1)$ und +% $x_2(t) = \mathrm{e}^{t-2} \cdot \epsilon(-t+2)$ +% grafisch dar und machen Sie deutlich wie aus Signal $x_1(t)$ mittels Zeitverschiebung +% und negativer Zeitskalierung das Signal $x_2(t)$ resultiert. +% \begin{Werkzeug} +% Wir brauchen die Konzepte (a) der Zeitverschiebung, hier speziell Zeitverzögerung, +% also Verschiebung des Signals nach rechts und (b) der Umkehr des Funktionsarguments, +% schlampig ausgedrückt, der Zeitumkehr. Letzteres erreichen wir indem wir +% für den im Werkzeugkasten aus Aufgabe \ref{sec:42DE8C69F7} definierten +% Zeitskalierungsfaktor +% $\theta=-1$ zulassen. Wir machen uns klar, dass wir auch andere negative +% $\theta$ zulassen könnten, dann wird das Signal nicht nur gespiegelt, +% sondern zusätzlich noch gestaucht/gestreckt. Das sollten wir uns mal ganz in Ruhe +% mit Wolfram Alpha als Referenzlösung anschauen. +% \end{Werkzeug} +% \begin{Ansatz} +% Es gibt zwei Varianten um von $x_1(t)$ zu $x_2(t)$ zu gelangen: +% +% Variante I: erst Zeitskalierung $\theta=-1$, dann Zeitverschiebung um $\tau$ +% +% Variante II: erst Zeitverschiebung um $\tau$, dann Zeitskalierung $\theta=-1$ +% +% \end{Ansatz} +% \begin{ExCalc} +% Mit Wolfram Alpha \url{https://www.wolframalpha.com} können wir uns bei +% Unsicherheit Klarheit verschaffen. +% % +% Ansonsten ist das eine exzellente Gelegenheit für Stift und Papier. +% \end{ExCalc} +% \begin{Loesung} +% Für Variante I siehe \fig{fig:FBE36B0684_1}, für Variante II siehe +% \ref{fig:FBE36B0684_2}. +% +% Das Faltungsintegral +% \begin{equation} +% y(t) +% = \int\limits_{-\infty}^{+\infty} x(-\tau+t) h(\tau) \, \fsd \tau +% = \int\limits_{-\infty}^{+\infty} x(\tau) h(-\tau+t) \, \fsd \tau +% \end{equation} +% werden wir in den nächsten Aufgaben benutzen und verstehen. +% % +% Wir sehen hier, dass genau die hier +% diskutierte Zeitverschiebung und negative Zeitskalierung benutzt werden muss, um +% die Signaloperationen $x(\tau) \rightarrow x(-\tau+t)$ +% oder $h(\tau) \rightarrow h(-\tau+t)$ abzubilden. +% \textbf{Wir beachten aber}, dass im Faltungsintegral die temporäre Zeitvariable +% $\tau$ (das ist keine RC-Glied Zeitkonstante o.ä.) ist und die Zeit $t$ +% im Ergebnissignal $y(t)$ die Verschiebungsvariable in der Faltung entspricht. +% \end{Loesung} +% +% \begin{figure*}[h] +% \centering +% \begin{subfigure}{0.45\textwidth} +% \begin{tikzpicture} +% \begin{axis}[ +% width=1\textwidth, +% height=0.45\textwidth, +% domain=-1:4, +% samples=50, +% legend pos=outer north east, +% xlabel = {t}, +% %ylabel = {x(t)}, +% title = {$x_1(t) = \mathrm{e}^{-\left(t+1\right)} \cdot \epsilon(\left(t+1\right))$}, +% xmin=-4, xmax=4, +% ymin=-0.1, ymax=1.1, +% xtick={-4,-3,-2,-1,0,1,2,3,4}, +% ytick={0,0.5,1,1.5,2}, +% ymajorgrids=true, +% xmajorgrids=true +% ] +% \addplot[mark=None, color=C0, ultra thick] +% {exp(-x-1)}; +% \addplot[mark=None, color=C0, ultra thick] +% coordinates {(-4,0)(-1,0)(-1,1)}; +% \end{axis} +% \end{tikzpicture} +% \caption{Ausgangspunkt: abfallende, vorauseilende Exponential-Funktion.} +% %\label{fig:} +% \end{subfigure} +% % +% \begin{subfigure}{0.45\textwidth} +% \begin{tikzpicture} +% \begin{axis}[ +% width=1\textwidth, +% height=0.45\textwidth, +% domain=-4:-1, +% samples=50, +% legend pos=outer north east, +% xlabel = {t}, +% %ylabel = {x(t)}, +% title = {$\mathrm{e}^{+\left(t+1\right)} \cdot \epsilon(-\left(t+1\right))$}, +% xmin=-4, xmax=4, +% ymin=-0.1, ymax=1.1, +% xtick={-4,-3,-2,-1,0,1,2,3,4}, +% ytick={0,0.5,1,1.5,2}, +% ymajorgrids=true, +% xmajorgrids=true +% ] +% \addplot[mark=None, color=C0, ultra thick] +% {exp(x+1)}; +% \addplot[mark=None, color=C0, ultra thick] +% coordinates {(-1,1)(-1,0)(4,0)}; +% \end{axis} +% \end{tikzpicture} +% \caption{Umkehr der Funktionsargumente von (a) ist Spiegelung an der $t=-1$ Achse.} +% %\label{fig:} +% \end{subfigure} +% +% \begin{subfigure}{0.45\textwidth} +% \begin{tikzpicture} +% \begin{axis}[ +% width=1\textwidth, +% height=0.45\textwidth, +% domain=-4:2, +% samples=50, +% legend pos=outer north east, +% xlabel = {t}, +% %ylabel = {x(t)}, +% title = {$x_2(t) = \mathrm{e}^{(+\left(t+1-3\right))} \cdot \epsilon(-\left(t+1-3\right))$}, +% xmin=-4, xmax=4, +% ymin=-0.1, ymax=1.1, +% xtick={-4,-3,-2,-1,0,1,2,3,4}, +% ytick={0,0.5,1,1.5,2}, +% ymajorgrids=true, +% xmajorgrids=true +% ] +% \addplot[mark=None, color=C0, ultra thick] +% {exp(+(x-2))}; +% \addplot[mark=None, color=C0, ultra thick] +% coordinates {(2,1)(2,0)(4,0)}; +% \end{axis} +% \end{tikzpicture} +% \caption{Erst Umkehr der Funktionsargumente von (a) und dann Zeitverzögerung von +% (b) um $3$.} +% %\label{fig:} +% \end{subfigure} +% % +% % +% % +% \caption{Skizzenfahrplan für Aufgabe \ref{sec:FBE36B0684}, +% \textbf{zuerst Zeitskalierung $\theta=-1$, dann Zeitverschiebung}. +% Vgl. \texttt{TimeMirrorShift\_FBE36B0684.ipynb}} +% \label{fig:FBE36B0684_1} +% \end{figure*} +% % +% % +% % +% \begin{figure*}[h] +% \centering +% \begin{subfigure}{0.45\textwidth} +% \begin{tikzpicture} +% \begin{axis}[ +% width=1\textwidth, +% height=0.45\textwidth, +% domain=-1:4, +% samples=50, +% legend pos=outer north east, +% xlabel = {t}, +% %ylabel = {x(t)}, +% title = {$x_1(t) = \mathrm{e}^{-\left(t+1\right)} \cdot \epsilon(\left(t+1\right))$}, +% xmin=-4, xmax=4, +% ymin=-0.1, ymax=1.1, +% xtick={-4,-3,-2,-1,0,1,2,3,4}, +% ytick={0,0.5,1,1.5,2}, +% ymajorgrids=true, +% xmajorgrids=true +% ] +% \addplot[mark=None, color=C0, ultra thick] +% {exp(-(x+1))}; +% \addplot[mark=None, color=C0, ultra thick] +% coordinates {(-4,0)(-1,0)(-1,1)}; +% \end{axis} +% \end{tikzpicture} +% \caption{Ausgangspunkt: abfallende, vorauseilende Exponential-Funktion.} +% %\label{fig:} +% \end{subfigure} +% % +% \begin{subfigure}{0.45\textwidth} +% \begin{tikzpicture} +% \begin{axis}[ +% width=1\textwidth, +% height=0.45\textwidth, +% domain=2:4, +% samples=50, +% legend pos=outer north east, +% xlabel = {t}, +% %ylabel = {x(t)}, +% title = {$\mathrm{e}^{(-\left(t+1-3\right))} \cdot \epsilon(\left(t+1-3\right))$}, +% xmin=-4, xmax=4, +% ymin=-0.1, ymax=1.1, +% xtick={-4,-3,-2,-1,0,1,2,3,4}, +% ytick={0,0.5,1,1.5,2}, +% ymajorgrids=true, +% xmajorgrids=true +% ] +% \addplot[mark=None, color=C0, ultra thick] +% {exp(-(x-2))}; +% \addplot[mark=None, color=C0, ultra thick] +% coordinates {(-4,0)(2,0)(2,1)}; +% \end{axis} +% \end{tikzpicture} +% \caption{Zeitverzögerung von (a) um $3$.} +% %\label{fig:} +% \end{subfigure} +% +% \begin{subfigure}{0.45\textwidth} +% \begin{tikzpicture} +% \begin{axis}[ +% width=1\textwidth, +% height=0.45\textwidth, +% domain=-4:2, +% samples=50, +% legend pos=outer north east, +% xlabel = {t}, +% %ylabel = {x(t)}, +% title = {$x_2(t) = \mathrm{e}^{(+\left(t+1-3\right))} \cdot \epsilon(-\left(t+1-3\right))$}, +% xmin=-4, xmax=4, +% ymin=-0.1, ymax=1.1, +% xtick={-4,-3,-2,-1,0,1,2,3,4}, +% ytick={0,0.5,1,1.5,2}, +% ymajorgrids=true, +% xmajorgrids=true +% ] +% \addplot[mark=None, color=C0, ultra thick] +% {exp(+(x-2))}; +% \addplot[mark=None, color=C0, ultra thick] +% coordinates {(2,1)(2,0)(4,0)}; +% \end{axis} +% \end{tikzpicture} +% \caption{Erst Zeitverzögerung von (a) um $3$ und danach Umkehr der Funktionsargumente von (b). +% Daher Spiegelung des Signals von (b) an der $t=2$ Achse.} +% %\label{fig:} +% \end{subfigure} +% % +% % +% % +% \caption{Skizzenfahrplan für Aufgabe \ref{sec:FBE36B0684}, +% \textbf{zuerst Zeitverschiebung, dann Zeitskalierung $\theta=-1$}. +% Vgl. \texttt{TimeMirrorShift\_FBE36B0684.ipynb}} +% \label{fig:FBE36B0684_2} +% \end{figure*} @@ -1533,7 +1836,7 @@ \subsection{Faltung Rechteckimpuls mit Dreieckimpuls} damit ist (rect: erst zeitlich verspätet um $\frac{5}{4}$, dann zeitlich stauchen wegen $2>1$) \begin{equation} -h(t) = (-2 t + 3) \cdot \mathrm{rect}\left(\left[t-\frac{5}{4}\right] \cdot 2\right) +h(t) = (-2 t + 3) \cdot \mathrm{rect}\left(\left(t-\frac{5}{4}\right) \cdot 2\right) \end{equation} \begin{equation} x(t)= @@ -1545,7 +1848,7 @@ \subsection{Faltung Rechteckimpuls mit Dreieckimpuls} damit ist (rect: erst zeitlich verspätet um 1, dann zeitlich strecken wegen $\frac{1}{2}<1$ und Amplitude strecken um Faktor 2) \begin{equation} -x(t) = 2\cdot\mathrm{rect}\left(\left[t-1\right] \cdot \frac{1}{2}\right) +x(t) = 2\cdot\mathrm{rect}\left(\left(t-1\right) \cdot \frac{1}{2}\right) \end{equation} Fassen wir $h(t)$ hier mal als Impulsantwort eines LTI-Systems und $x(t)$ als Eingangssignal in dieses auf, obwohl es dem Faltungsintegral völlig egal ist,