Line 29: Line 29:
 
----
 
----
 
==Question 3==
 
==Question 3==
 +
Consider the discrete-time signal
  
 +
<math>x[n]=2\delta[n]+5 \delta[n-1]+\delta[n-1]- \delta[n-2].</math>
  
 +
a) Determine the DTFT <math>X(\omega)</math> of x[n] and the DTFT of <math>Y(\omega)</math> of y[n]=x[-n].
 +
 +
b) Using your result from part a), compute
 +
 +
<math>x[n]* y[n]</math>.
 +
 +
c) Consider the discrete-time signal
 +
 +
<math>z[n]=\left\{ \begin{array}{ll}x[(-n)\mod 4],& 0\leq n < 3,\\ 0 & \text{else }\end{array} \right. </math>. 
 +
 +
Obtain the 4-point circular convolution of x[n] and z[n].
 +
 +
d) When computing the N-point circular convolution of x[n] and the signal
 +
 +
<math>z[n]=\left\{ \begin{array}{ll}x[(-n)\mod N],& 0\leq n < N-1,\\ 0 & \text{else }\end{array} \right. </math>. 
 +
 +
how should N be chosen to make sure that the result is the same as the usual convolution between x[n] and z[n]? 
 
----
 
----
 
[[2010 Fall ECE 438 Boutin|Back to ECE438, Fall 2010, Prof. Boutin]]
 
[[2010 Fall ECE 438 Boutin|Back to ECE438, Fall 2010, Prof. Boutin]]

Latest revision as of 11:16, 27 October 2010

Homework 8, ECE438, Fall 2010, Prof. Boutin

Due in class, Wednesday November 3, 2010.

The discussion page for this homework is here.


Question 1

Consider two discrete-time signals with the same (finite) duration N. Let $ X_1(z) $ be the z-transform of the first signal, and $ X_2[k] $ be the N-point DFT of the second signal. If we assume that

$ X_2[k]=\left. X_1(z) \right|_{z=\frac{1}{2}e^{-j \frac{2 \pi}{N} k}}, \text{ for }k=0,1,\ldots,N-1, $

then what is the relationship between the two signals?


Question 2

Consider the discrete-time signal

$ x[n]=6\delta[n]+5 \delta[n-1]+4 \delta[n-2]+3 \delta[n-3]+2 \delta[n-4]+\delta[n-5]. $

a) Obtain the six-point DFT X[k] of x[n].

b) Obtain the signal y[n] whose DFT is $ W_6^{-2k} X[k] $.

c) Compute six-point circular convolution between x[n] and the signal

$ h[n]=\delta[n]+\delta[n-1]+\delta[n-2]. $

d) If we convolve x[n] with the given h[n] by N-point convolution, how large should N be to insure that the result is the same as the periodic repetition (with period N) of the usual convolution between x[n] and h[n]?


Question 3

Consider the discrete-time signal

$ x[n]=2\delta[n]+5 \delta[n-1]+\delta[n-1]- \delta[n-2]. $

a) Determine the DTFT $ X(\omega) $ of x[n] and the DTFT of $ Y(\omega) $ of y[n]=x[-n].

b) Using your result from part a), compute

$ x[n]* y[n] $.

c) Consider the discrete-time signal

$ z[n]=\left\{ \begin{array}{ll}x[(-n)\mod 4],& 0\leq n < 3,\\ 0 & \text{else }\end{array} \right. $.

Obtain the 4-point circular convolution of x[n] and z[n].

d) When computing the N-point circular convolution of x[n] and the signal

$ z[n]=\left\{ \begin{array}{ll}x[(-n)\mod N],& 0\leq n < N-1,\\ 0 & \text{else }\end{array} \right. $.

how should N be chosen to make sure that the result is the same as the usual convolution between x[n] and z[n]?


Back to ECE438, Fall 2010, Prof. Boutin

Alumni Liaison

Correspondence Chess Grandmaster and Purdue Alumni

Prof. Dan Fleetwood