Line 20: Line 20:
 
Image1<br><br>
 
Image1<br><br>
 
Now let's describe this process in the frequency domain.
 
Now let's describe this process in the frequency domain.
 
+
----
 
==Derivation==
 
==Derivation==
 
First we'll take the Discrete Time Fourier Transform of the original signal and the downsampled version of it.<br>
 
First we'll take the Discrete Time Fourier Transform of the original signal and the downsampled version of it.<br>
Line 37: Line 37:
 
<math>\begin{align}
 
<math>\begin{align}
 
\mathcal{X}_2(\omega) &= \sum_{m=-\infty}^\infty s_d[m]x_1[m]e^{-j\omega \frac{m}{D}} \text{ where}\\
 
\mathcal{X}_2(\omega) &= \sum_{m=-\infty}^\infty s_d[m]x_1[m]e^{-j\omega \frac{m}{D}} \text{ where}\\
&s_d[m] = \frac{1}{D}\sum_{k=0}^{D-1} e^{jk2\pi \frac{m}{D}}
+
&s_d[m] = \left\{
 +
\begin{array}{ll}
 +
1, & \text{for }m\text{ multiples of } D\\
 +
0, & \text{else}
 +
\end{array}
 +
\right.\\
 +
&\text{ or }=\frac{1}{D}\sum_{k=0}^{D-1} e^{jk2\pi \frac{m}{D}} \text{ so:}\\
 +
\mathcal{X}_2(\omega) &= \sum_{m=-\infty}^\infty \frac{1}{D}\sum_{k=0}^{D-1} x_1[m]e^{jk2\pi \frac{m}{D}}e^{-j\omega \frac{m}{D}}\\
 +
&= \frac{1}{D}\sum_{k=0}^{D-1}\sum_{m=-\infty}^\infty x_1[m]e^{-jm\frac{\omega -k2\pi}{D}} \\
 
\end{align}</math>
 
\end{align}</math>
 +
 +
<br>Comparing to<br>
 +
<math>\begin{align}
 +
\mathcal{X}(\omega) &= \sum_{n=-\infty}^\infty x[n]e^{-jn\omega}\\
 +
\end{align}</math>
 +
 +
<br> We can rewrite the previous as<br>
 +
<math>\begin{align}
 +
\mathcal{X}_2(\omega) &= \frac{1}{D}\sum_{k=0}^{D-1} \mathcal{X}(\frac{\omega -2\pi k}{D}) \\
 +
\end{align}</math>
 +
 +
<math>\text{Comared to } \mathcal{X}_1(\omega) \text{, } \mathcal{X}_2(\omega) \text{ is } \frac{1}{D} \text{times the magnitude, has its frequencies stretched by }D \text{ and also repeats every }2\pi \text{ (as every DTFT should)}</math>
 +
----
 
==Example==
 
==Example==
 +
Image2<br><br>
 +
It is worth noting that for large enough <math>D</math> it is possible for the ends of repetitions to overlap, leading to aliasing! To prevent this:<br>
 +
 +
<math>\begin{align}
 +
D2\pi T_1 f_{max} < \pi\\
 +
\\
 +
x_1[Dn] \text{ has a sampling period of } T_2 = DT_1\text{, so:}\\
 +
\\
 +
\begin{align}
 +
\frac{T_2}{T_1}2\pi T_1f_{max} < \pi\\
 +
f_{max} < \frac{1}{2T_2}
 +
\end{align}
 +
\end{align}</math>
 +
 +
To make sure this condition is satisfied, we can first pass the original <math>x_1[n]</math> signal through a low-pass filter with <math>f_c = 1/(2T_2)</math> BEFORE downsampling.<br><br>
 +
Since the possible aliasing is caused by the downsampling, trying to low-pass filter after the downsampling will be too late and won't be able to get rid of the aliasing. Thus, the full process of downsampling should look like this:<br>
 +
----
 
==Conclusion==
 
==Conclusion==
  
 
----
 
----

Revision as of 15:00, 9 October 2014


Downsampling in the Frequency Domain

A slecture by ECE student John S.

Partly based on the ECE438 Fall 2014 lecture material of Prof. Mireille Boutin.


Introduction

Remember for time domain, Downsampling is defined as:

Image1

Now let's describe this process in the frequency domain.


Derivation

First we'll take the Discrete Time Fourier Transform of the original signal and the downsampled version of it.

$ \begin{align} \mathcal{X}_2(\omega) &= \mathcal{F }\left \{ x_2[n] \right \} = \mathcal{F }\left \{ x_1[Dn] \right \}\\ &= \sum_{n=-\infty}^\infty x_1[Dn]e^{-j\omega n} \end{align} $
make the substitution of $ n=\frac{m}{D} $

$ \begin{align} \mathcal{X}_2(\omega) &= \sum_{m=-\infty}^\infty x_1[m]e^{-j\omega \frac{m}{D}} \end{align} $


Downsampled signal will only be nonzero for m equal to multiples of D so:
$ \begin{align} \mathcal{X}_2(\omega) &= \sum_{m=-\infty}^\infty s_d[m]x_1[m]e^{-j\omega \frac{m}{D}} \text{ where}\\ &s_d[m] = \left\{ \begin{array}{ll} 1, & \text{for }m\text{ multiples of } D\\ 0, & \text{else} \end{array} \right.\\ &\text{ or }=\frac{1}{D}\sum_{k=0}^{D-1} e^{jk2\pi \frac{m}{D}} \text{ so:}\\ \mathcal{X}_2(\omega) &= \sum_{m=-\infty}^\infty \frac{1}{D}\sum_{k=0}^{D-1} x_1[m]e^{jk2\pi \frac{m}{D}}e^{-j\omega \frac{m}{D}}\\ &= \frac{1}{D}\sum_{k=0}^{D-1}\sum_{m=-\infty}^\infty x_1[m]e^{-jm\frac{\omega -k2\pi}{D}} \\ \end{align} $


Comparing to
$ \begin{align} \mathcal{X}(\omega) &= \sum_{n=-\infty}^\infty x[n]e^{-jn\omega}\\ \end{align} $


We can rewrite the previous as
$ \begin{align} \mathcal{X}_2(\omega) &= \frac{1}{D}\sum_{k=0}^{D-1} \mathcal{X}(\frac{\omega -2\pi k}{D}) \\ \end{align} $

$ \text{Comared to } \mathcal{X}_1(\omega) \text{, } \mathcal{X}_2(\omega) \text{ is } \frac{1}{D} \text{times the magnitude, has its frequencies stretched by }D \text{ and also repeats every }2\pi \text{ (as every DTFT should)} $


Example

Image2

It is worth noting that for large enough $ D $ it is possible for the ends of repetitions to overlap, leading to aliasing! To prevent this:

$ \begin{align} D2\pi T_1 f_{max} < \pi\\ \\ x_1[Dn] \text{ has a sampling period of } T_2 = DT_1\text{, so:}\\ \\ \begin{align} \frac{T_2}{T_1}2\pi T_1f_{max} < \pi\\ f_{max} < \frac{1}{2T_2} \end{align} \end{align} $

To make sure this condition is satisfied, we can first pass the original $ x_1[n] $ signal through a low-pass filter with $ f_c = 1/(2T_2) $ BEFORE downsampling.

Since the possible aliasing is caused by the downsampling, trying to low-pass filter after the downsampling will be too late and won't be able to get rid of the aliasing. Thus, the full process of downsampling should look like this:


Conclusion


Alumni Liaison

Basic linear algebra uncovers and clarifies very important geometry and algebra.

Dr. Paul Garrett