EE2004: Digital Signal Processing
Tutorial 3
(February 12, 2026)
1. Find the z-transform of the following sequences and state clearly the RoC in each case:
(a) (0.8)n u[n] + 2n u[−n − 1], (b) 2n u[n] − 3n u[−n − 1], (c) (0.8)n u[n] + (0.9)n u[−n −
1], (d) [(0.8)n + 3 (0.4)n ] u[−n − 1], (e) [(0.8)n + 3 (0.4)n ] u[n], and (f) (0.5)|n| .
2. Let x [n] = e jω0 n u[n]. Find X (z) and its corresponding RoC.
3. Let x [n] = cos(ω0 n) u[n]. Find X (z) and its corresponding RoC. Repeat for x [n] =
sin(ω0 n) u[n]. In each case, plot the pole-zero plot of X (z).
( e −2 − 2) z
4. Let H (z) = . List all possible RoCs corresponding to the given H (z).
(z − e−2 )(z − 2)
Also state the nature of the corresponding time-domain sequence for each case, i.e.,
left-sided, right-sided, or two-sided.
5. Find X (z) when x [n] equals (a) δ[n], (b) δ[n − n0 ] (n0 > 0), and (c) δ[n + n0 ] (n0 > 0).
What is the RoC in each case?
6. Find X (z) when x [n] = u[n] (the unit step sequence). What is its RoC?
7. Pole−Zero plot
1
j 2π
20 k
j 2π
20 2
Consider the pole-zero plot shown on
0.5 the left. Find the corresponding H (z)
j 2π
20 whose pole-zero plot is consistent with
21 0.8 the given information.
Img
0
−1 1
j 2π
20 19
−0.5
−1
−1 −0.5 0 0.5 1
Real
sin[(n + 1)θ ]
8. Let x [n] = r n u[n]. Find X (z) and indicate the RoC. You should not as-
sin θ
sume the z-transform of either cos(ω0 n) u[n] or sin(ω0 n) u[n] while deriving the final
expression.
nπ
9. Computer assignment Create the signal x [n] = cos for 0 ≤ n ≤ 19. Delay
5
this signal by 2.5 samples as follows: (i) multiply the signal’s DTFT (implemented us-
ing Matlab’s fft command) by exp(− j 2.5ω ) (after discretizing the frequency variable)
over the the interval [0, π ]; (ii) for the samples in the range (π, 2π ), fill them using
the time-reversed, complex-conjugated samples taken from the interval [0, π ]; (iii) find
EE2004 DSP Page 1 of 2 APK/CSR
the inverse transform to obtain a sequence that is 20 samples in length (it must be
real-valued, and any imaginary part present
nπ must be very small and should be zeroed
π
out). Plot this alongside of y[n] = cos − . Also compare this with delayseq(x,
5 2
2.5).
10. Computer assignment Load the speech waveform contained in the file [Link]
(look up Matlab’s audioread command). Add a delayed version of the speech to itself.
Consider delays in the range 50–100 ms. Note the difference in sound quality. Increase
the delay and observe what happens for values much greater than 100 ms.
11. Computer assignment Load the speech file [Link]. Pass it through a block that
introduces a time-varying delay. That is, this block produces the output is x [n − dn ].
Choose d = 20 + 20*sin(2*pi*5*n/Fs). Implement this block using Matlab’s delayseq
command, which has the capability to introduce fractional delays. A vibrato effect is
imparted to the output. Experiment with variable delay parameters to see how they
influence the vibrato effect.
12. Computer assignment Load the speech file [Link]. Consider the following two
different variable delay values, i.e., d1 = 200 + 32*sin(2*pi*2*n/Fs) and d2 = 200
+ 50*sin(2*pi*3*n/Fs). Pass the original signal through each of these two variable
delay blocks. Listen to each output and compare it with the original. In particular,
are you able to note the changes in the words “along” and “one mile lane”?; are there
any other noticeable differences? Next, add the two outputs to the original and listen
to combined signal. This produces a chorus effect. Vary the parameters of the delay
blocks and observe how they change the sound quality.
EE2004 DSP Page 2 of 2 12.02.2026