Difference between revisions of "Aufgaben:Exercise 3.1: Spectrum of the Exponential Pulse"

From LNTwww
 
(One intermediate revision by the same user not shown)
Line 6: Line 6:
  
 
In this task, a causal signal  $x(t)$ is considered
 
In this task, a causal signal  $x(t)$ is considered
*which rises abruptly from zero to  $A$  at time  $t = 0$ , and  
+
*which rises abruptly from zero to  $A$  at time  $t = 0$,  and  
*decreases exponentially with the time constant  $T$  for  $t > 0$ :
+
*decreases exponentially with the time constant  $T$  for  $t > 0$:
  
 
:$$x(t) = A \cdot {\rm e}^{ - t/T} .$$
 
:$$x(t) = A \cdot {\rm e}^{ - t/T} .$$
 
   
 
   
At the saltus at time  $t = 0$ ,   $x(t = 0) = A/2$.
+
At the jumping point at time  $t = 0$,  $x(t = 0) = A/2$.
  
 
Use the following parameters for the numerical calculations:
 
Use the following parameters for the numerical calculations:
Line 17: Line 17:
 
:$$A = 3 \hspace{0.1cm} {\rm V}, \hspace{0.4cm} T = 1 \hspace{0.1cm} {\rm ms} .$$
 
:$$A = 3 \hspace{0.1cm} {\rm V}, \hspace{0.4cm} T = 1 \hspace{0.1cm} {\rm ms} .$$
  
The spectral function  $X(f)$  to be calculated will be complex and therefore can be
+
The spectral function  $X(f)$  to be calculated will be complex and therefore can be represented
 
*by real and imaginary part, but also  
 
*by real and imaginary part, but also  
*by magnitude and phase
+
*by magnitude and phase.
 
 
can be represented. Use the notation
 
 
:$$X( f ) = \left| {X( f )} \right| \cdot {\rm e}^{ - {\rm j} \hspace{0.05cm}\cdot \hspace{0.05cm} \varphi( f )} .$$
 
 
 
  
  
Line 33: Line 28:
  
 
''Hint:''  
 
''Hint:''  
*This exercise belongs to the chapter  [[Signal_Representation/Fourier_Transform_and_Its_Inverse|Fourier Transform and Its Inverse]].
+
*This exercise belongs to the chapter  [[Signal_Representation/Fourier_Transform_and_Its_Inverse|Fourier Transform and its Inverse]].
 +
*Use the notation:
 +
 +
:$$X( f ) = \left| {X( f )} \right| \cdot {\rm e}^{ - {\rm j} \hspace{0.05cm}\cdot \hspace{0.05cm} \varphi( f )} .$$
 
   
 
   
  
Line 41: Line 39:
  
 
<quiz display=simple>
 
<quiz display=simple>
{Calculate the spectral function&nbsp; $X(f)$. What spectral value results at the frequency&nbsp; $f = 0$?
+
{Calculate the spectral function&nbsp; $X(f)$.&nbsp; What spectral value results at the frequency $f = 0$?
 
|type="{}"}
 
|type="{}"}
 
$\text{Re}[X(f=0)] \ = \ $ { 3 3% }  &nbsp;$\rm mV/Hz$
 
$\text{Re}[X(f=0)] \ = \ $ { 3 3% }  &nbsp;$\rm mV/Hz$
 
$\text{Im}[X(f=0)] \ = \  $ { 0. } &nbsp;$\rm mV/Hz$
 
$\text{Im}[X(f=0)] \ = \  $ { 0. } &nbsp;$\rm mV/Hz$
  
{What are the real and imaginary parts of&nbsp; $X(f)$&nbsp; using&nbsp; $f_0 = 1/(2\pi T)$. What are the values when&nbsp; $f = f_0$?
+
{What are the real and imaginary parts of&nbsp; $X(f)$&nbsp; using&nbsp; $f_0 = 1/(2\pi T)$.&nbsp; What are the values when $f = f_0$?
 
|type="{}"}
 
|type="{}"}
 
$\text{Re}[X(f=f_0)] \ = \  $ { 1.5 3% } &nbsp;$\rm mV/Hz$
 
$\text{Re}[X(f=f_0)] \ = \  $ { 1.5 3% } &nbsp;$\rm mV/Hz$
 
$\text{Im}[X(f=f_0)] \ = \  $ { -1.55--1.45 } &nbsp;$\rm mV/Hz$
 
$\text{Im}[X(f=f_0)] \ = \  $ { -1.55--1.45 } &nbsp;$\rm mV/Hz$
  
{Calculate the magnitude function&nbsp; $|X(f)|$. Which values result at the frequency&nbsp; $f = f_0$&nbsp; and at very high frequencies?
+
{Calculate the magnitude function&nbsp; $|X(f)|$.&nbsp; Which values result at the frequency $f = f_0$&nbsp; and at very high frequencies?
 
|type="{}"}
 
|type="{}"}
 
$|X(f=f_0)| \hspace{0.25cm} = \  $ { 2.12 3% } &nbsp;$\rm mV/Hz$
 
$|X(f=f_0)| \hspace{0.25cm} = \  $ { 2.12 3% } &nbsp;$\rm mV/Hz$
 
$|X(f\rightarrow \infty)| \ = \  $ { 0. } &nbsp;$\rm mV/Hz$
 
$|X(f\rightarrow \infty)| \ = \  $ { 0. } &nbsp;$\rm mV/Hz$
  
{Calculate the phase function&nbsp; $\varphi(f)$. What values result for this at the frequency&nbsp; $f = f_0$&nbsp; and at very high frequencies?
+
{Calculate the phase function&nbsp; $\varphi(f)$.&nbsp; What values result for this at the frequency $f = f_0$&nbsp; and at very high frequencies?
 
|type="{}"}
 
|type="{}"}
 
$\varphi(f=f_0) \hspace{0.25cm} = \  $ { 0.785 3% } &nbsp;$\rm rad$
 
$\varphi(f=f_0) \hspace{0.25cm} = \  $ { 0.785 3% } &nbsp;$\rm rad$
Line 76: Line 74:
 
X( {f = 0}) = A \cdot T{ = 3 \cdot 10^{ - 3}\; {\rm V/Hz}} \hspace{0.15 cm}\underline{ = 3 \; {\rm mV/Hz}}.$$
 
X( {f = 0}) = A \cdot T{ = 3 \cdot 10^{ - 3}\; {\rm V/Hz}} \hspace{0.15 cm}\underline{ = 3 \; {\rm mV/Hz}}.$$
  
*At the frequency&nbsp; $f = 0$&nbsp;, the spectrum is therefore purely real:
+
*At the frequency $f = 0$&nbsp;, the spectrum is purely real:
 
:$$\text{Re}[X(f=0)] \hspace{0.15 cm}\underline{ = 3 \; {\rm mV/Hz}}&nbsp; \hspace{1.15 cm}\text{Im}[X(f=0)] \hspace{0.15 cm}\underline{ =0}.$$
 
:$$\text{Re}[X(f=0)] \hspace{0.15 cm}\underline{ = 3 \; {\rm mV/Hz}}&nbsp; \hspace{1.15 cm}\text{Im}[X(f=0)] \hspace{0.15 cm}\underline{ =0}.$$
  
Line 83: Line 81:
 
'''(2)'''&nbsp; With the abbreviations&nbsp; $X_0 = A \cdot T$&nbsp; and&nbsp; $f_0 = 1/(2\pi T)$&nbsp; the spectral function is:
 
'''(2)'''&nbsp; With the abbreviations&nbsp; $X_0 = A \cdot T$&nbsp; and&nbsp; $f_0 = 1/(2\pi T)$&nbsp; the spectral function is:
  
 
 
:$$X( f) = \frac{ {X_0 }}{ {1 +{\rm j} \cdot f/f_0 }} = \frac{ {X_0 }}{ {1 + \left( {f/f_0 } \right)^2 }} \cdot \left( {1 - {\rm j} \cdot f/f_0 } \right).$$
 
:$$X( f) = \frac{ {X_0 }}{ {1 +{\rm j} \cdot f/f_0 }} = \frac{ {X_0 }}{ {1 + \left( {f/f_0 } \right)^2 }} \cdot \left( {1 - {\rm j} \cdot f/f_0 } \right).$$
  
 
Divided into real and imaginary parts, this gives:
 
Divided into real and imaginary parts, this gives:
 
   
 
   
$${\mathop{\rm Re}\nolimits} [ {X(f)}] = \frac{ {X_0 }}{{1 + \left( {f/f_0 } \right)^2 }},
+
:$${\mathop{\rm Re}\nolimits} [ {X(f)}] = \frac{ {X_0 }}{{1 + \left( {f/f_0 } \right)^2 }},
 
\hspace{0.5 cm}{\mathop{\rm Im}\nolimits} [ {X(f)}] =  - \frac{ {X_0  \cdot f/f_0 }}{ {1 + \left( {f/f_0 } \right)^2 }}.$$
 
\hspace{0.5 cm}{\mathop{\rm Im}\nolimits} [ {X(f)}] =  - \frac{ {X_0  \cdot f/f_0 }}{ {1 + \left( {f/f_0 } \right)^2 }}.$$
  
Line 97: Line 94:
  
  
 +
'''(3)'''&nbsp; The magnitude of a complex-valued function, which is a quotient, is equal to the quotient of the magnitudes of the numerator and denominator.
 
[[File:P_ID548__Sig_A_3_1_c_neu.png|right|frame|Magnitude spectrum of the exponential pulse]]
 
[[File:P_ID548__Sig_A_3_1_c_neu.png|right|frame|Magnitude spectrum of the exponential pulse]]
'''(3)'''&nbsp; The magnitude of a complex-valued function, which is a quotient, is equal to the quotient of the magnitudes of the numerator and denominator.
+
 
 
*Thus one obtains:
 
*Thus one obtains:
 
   
 
   
Line 111: Line 109:
 
'''(4)'''&nbsp; The general rule for the phase function is:
 
'''(4)'''&nbsp; The general rule for the phase function is:
 
   
 
   
$$\varphi ( f ) = \arctan \left( {\frac{ { - {\mathop{\rm Im}\nolimits}[{X(f)} ]}}{{ {\mathop{\rm Re}\nolimits} [ {X(f)} ]}}} \right) = \arctan \left( {f/f_0 } \right).$$
+
:$$\varphi ( f ) = \arctan \left( {\frac{ { - {\mathop{\rm Im}\nolimits}[{X(f)} ]}}{{ {\mathop{\rm Re}\nolimits} [ {X(f)} ]}}} \right) = \arctan \left( {f/f_0 } \right).$$
  
*For $f = f_0$ this gives&nbsp; $\arctan(1)= \pi /4 \hspace{0.15 cm}\underline{\approx 0.785}$.
+
*For $f = f_0$&nbsp; this gives&nbsp; $\varphi ( f_0 ) =\arctan(1)= \pi /4 \hspace{0.15 cm}\underline{\approx 0.785}$.
*For very large values of&nbsp; $f$&nbsp; the phase function approaches&nbsp; $\arctan(\infty) = \pi /2 \hspace{0.15 cm}\underline{ \approx 1.571}$&nbsp; an.  
+
*For very large values of $f$&nbsp; the phase function approaches&nbsp; $\varphi ( f \to \infty ) =\arctan(\infty) = \pi /2 \hspace{0.15 cm}\underline{ \approx 1.571}$.  
 
*Both specifications are to be understood in radians.
 
*Both specifications are to be understood in radians.
  

Latest revision as of 14:05, 21 April 2021

Exponential pulse

In this task, a causal signal  $x(t)$ is considered

  • which rises abruptly from zero to  $A$  at time  $t = 0$,  and
  • decreases exponentially with the time constant  $T$  for  $t > 0$:
$$x(t) = A \cdot {\rm e}^{ - t/T} .$$

At the jumping point at time  $t = 0$,  $x(t = 0) = A/2$.

Use the following parameters for the numerical calculations:

$$A = 3 \hspace{0.1cm} {\rm V}, \hspace{0.4cm} T = 1 \hspace{0.1cm} {\rm ms} .$$

The spectral function  $X(f)$  to be calculated will be complex and therefore can be represented

  • by real and imaginary part, but also
  • by magnitude and phase.




Hint:

$$X( f ) = \left| {X( f )} \right| \cdot {\rm e}^{ - {\rm j} \hspace{0.05cm}\cdot \hspace{0.05cm} \varphi( f )} .$$



Questions

1

Calculate the spectral function  $X(f)$.  What spectral value results at the frequency $f = 0$?

$\text{Re}[X(f=0)] \ = \ $

 $\rm mV/Hz$
$\text{Im}[X(f=0)] \ = \ $

 $\rm mV/Hz$

2

What are the real and imaginary parts of  $X(f)$  using  $f_0 = 1/(2\pi T)$.  What are the values when $f = f_0$?

$\text{Re}[X(f=f_0)] \ = \ $

 $\rm mV/Hz$
$\text{Im}[X(f=f_0)] \ = \ $

 $\rm mV/Hz$

3

Calculate the magnitude function  $|X(f)|$.  Which values result at the frequency $f = f_0$  and at very high frequencies?

$|X(f=f_0)| \hspace{0.25cm} = \ $

 $\rm mV/Hz$
$|X(f\rightarrow \infty)| \ = \ $

 $\rm mV/Hz$

4

Calculate the phase function  $\varphi(f)$.  What values result for this at the frequency $f = f_0$  and at very high frequencies?

$\varphi(f=f_0) \hspace{0.25cm} = \ $

 $\rm rad$
$\varphi(f \rightarrow \infty) \ = \ $

 $\rm rad$



Solution

(1)  With the first Fourier integral we get:

$$X( f ) = \int_0^\infty {A \cdot {\rm e}^{ - t\left( {1/T + {\rm j \hspace{0.05cm} \cdot \hspace{0.05cm}}2\pi f} \right)} } {\rm d}t = \left. {\frac{ { - A}}{ {1/T + {\rm j}2\pi f}} \cdot {\rm e}^{ - t\left( {1/T + {\rm j}2\pi f} \right)} } \right|_0^\infty .$$
  • The upper integral limit  $(t \rightarrow \infty)$  gives zero, the lower limit  $(t = 0)$  gives the value  $1$. Thus:
$$X(f) = \frac{ {A \cdot T}}{ {1 + {\rm j}2\pi fT}}\hspace{0.3 cm}\Rightarrow\hspace{0.3 cm} X( {f = 0}) = A \cdot T{ = 3 \cdot 10^{ - 3}\; {\rm V/Hz}} \hspace{0.15 cm}\underline{ = 3 \; {\rm mV/Hz}}.$$
  • At the frequency $f = 0$ , the spectrum is purely real:
$$\text{Re}[X(f=0)] \hspace{0.15 cm}\underline{ = 3 \; {\rm mV/Hz}}  \hspace{1.15 cm}\text{Im}[X(f=0)] \hspace{0.15 cm}\underline{ =0}.$$


(2)  With the abbreviations  $X_0 = A \cdot T$  and  $f_0 = 1/(2\pi T)$  the spectral function is:

$$X( f) = \frac{ {X_0 }}{ {1 +{\rm j} \cdot f/f_0 }} = \frac{ {X_0 }}{ {1 + \left( {f/f_0 } \right)^2 }} \cdot \left( {1 - {\rm j} \cdot f/f_0 } \right).$$

Divided into real and imaginary parts, this gives:

$${\mathop{\rm Re}\nolimits} [ {X(f)}] = \frac{ {X_0 }}{{1 + \left( {f/f_0 } \right)^2 }}, \hspace{0.5 cm}{\mathop{\rm Im}\nolimits} [ {X(f)}] = - \frac{ {X_0 \cdot f/f_0 }}{ {1 + \left( {f/f_0 } \right)^2 }}.$$

At the frequency  $f_0$ 

  • the real part is equal to  $X_0/2 \hspace{0.15 cm}\underline{ = 1.5 \; {\rm mV/Hz}},$
  • the imaginary part is equal to  $–X_0/2 \hspace{0.15 cm}\underline{ = \hspace{0.1 cm}-1.5 \; {\rm mV/Hz}}.$


(3)  The magnitude of a complex-valued function, which is a quotient, is equal to the quotient of the magnitudes of the numerator and denominator.

Magnitude spectrum of the exponential pulse
  • Thus one obtains:
$$ \left| {X( f)} \right| =\frac{ {X_0 }}{ {\left| 1 +{\rm j} \cdot f/ {f_0 } \right|}} = \frac{ {X_0 }}{{\sqrt {1 + \left( {f/f_0 } \right)^2 } }},$$
$$\Rightarrow \hspace{0.5 cm} \left| {X( {f = f_0} )} \right| = { {X_0 }}/{ {\sqrt 2 }}\hspace{0.15 cm}\underline{ = 2.12 \;{\rm mV/Hz}}.$$
  • At very high frequencies  $(f \rightarrow \infty)$  the magnitude is almost zero (see sketch).


(4)  The general rule for the phase function is:

$$\varphi ( f ) = \arctan \left( {\frac{ { - {\mathop{\rm Im}\nolimits}[{X(f)} ]}}{{ {\mathop{\rm Re}\nolimits} [ {X(f)} ]}}} \right) = \arctan \left( {f/f_0 } \right).$$
  • For $f = f_0$  this gives  $\varphi ( f_0 ) =\arctan(1)= \pi /4 \hspace{0.15 cm}\underline{\approx 0.785}$.
  • For very large values of $f$  the phase function approaches  $\varphi ( f \to \infty ) =\arctan(\infty) = \pi /2 \hspace{0.15 cm}\underline{ \approx 1.571}$.
  • Both specifications are to be understood in radians.