You are on page 1of 5

ECE301, Homework#3 Solution

August 30, 2007

2.21
(a) The desired convolution is

y[n] = x[n] ∗ h[n]


X∞
= x[k]h[n − k]
k=−∞
n
X
n
= β (α/β)k , for n ≥ 0
k=0
(α/β)n+1 − 1
= βn , for n ≥ 0
α/β − 1
 n+1
− αn+1

β
= u[n], for α 6= β
β−α

(b) Similar to (a), but now replace β with α:

n
X
n
y[n] = α u[n] = (n + 1)αn u[n]
k=0

(c) Case 1, n ≤ 6:

1

X
y[n] = (−1/2)k 4n−k
k=4

X
n
= 4 (−1/8)k
k=4
(−1/8)4
= 4n
1 − (−1/8)
1 n
= 4
83 9
Case 2, n > 6:


X
y[n] = (−1/2)k 4n−k
k=n−2
(−1/8)n−2
= 4n
1 − (−1/8)
= (512/9)(−1/2)n

We have then that:

1 n

83 9
4 , n≤6
y[n] =
(512/9)(−1/2)n , n > 6

2.23
For any value of T , we have
Z ∞
!
∞ X
y(t) = x(t) ∗ h(t) = δ(τ − kT ) h(t − τ )dτ
−∞ k=−∞
∞ Z
X ∞
= δ(τ − kT )h(t − τ )dτ
k=−∞ −∞
X∞
= h(t − kT )
k=−∞

y(t) is plotted below for various values of T in the range t ∈ [−5, 5]


1.5 1.5

1 1

0.5 0.5

0 0

−0.5 −0.5
−5 −4 −3 −2 −1 0 1 2 3 4 5 −5 −4 −3 −2 −1 0 1 2 3 4 5

(a) T = 4 (b) T = 2
1.5 1.5

1 1

0.5 0.5

0 0

−0.5 −0.5
−5 −4 −3 −2 −1 0 1 2 3 4 5 −5 −4 −3 −2 −1 0 1 2 3 4 5

(c) T = 3/2 (d) T = 1

Iterative Convolution If S(t) = u(t + 1) − u(t − 1) is a pulse signal, and


Sn (t) = S ∗ S ∗ ... ∗ S (convolution repeated n times), we will have
that as n → ∞, Sn (t) approaches a zero-mean, zero-variance Gaussian
function(N (0, 0; t)), which is simply a dirac delta function δ(t).

Matlab Portion
Matlab code for the following problem is attached
(a) Plots of S and X are shown below

(b) Proof 1 Linearity


Let x1 [n] and x2 [n] be inputs with corresponding outputs y 1 [n] and
y2 [n]. Then we have
y1 [n] = (x1 [n − 2] + x1 [n − 1] + x1 [n] + x1 [n + 1] + x1 [n + 2])/5
y2 [n] = (x2 [n − 2] + x2 [n − 1] + x2 [n] + x2 [n + 1] + x2 [n + 2])/5
Now let x3 [n] = x1 [n] + x2 [n] have an output y3 [n]. We then have
y3 [n] = (x3 [n − 2] + x3 [n − 1] + x3 [n] + x3 [n + 1] + x3 [n + 2])/5
= (x1 [n − 2] + x2 [n − 2] + x1 [n − 1] + x2 [n − 1] + x1 [n] + x2 [n] +
x1 [n + 1] + x2 [n + 1] + x1 [n + 2] + x2 [n + 2])/5
= y1 [n] + y2 [n]
Thus the system is linear

Proof 2 Time Invariance


If x[n] is an input to the system with output y[n], then we have that
the delayed output y[n − N ] is as follows
y[n − N ] = (x[n − N − 2] + x[n − N − 1] + x[n − N ] + x[n − N + 1] + x[n − N + 2])/5

Now define an input z[n] = x[n − N ], such that its output y 2 [n] is as
follows
y2 [n] = (z[n − 2] + z[n − 1] + z[n] + z[n + 1] + z[n + 2])/5
= (x[n − 2 − N ] + x[n − 1 − N ] + x[n − N ] + x[n + 1 − N ] + x[n + 2 − N ])/5
= y[n − N ]
Thus the system is time invariant.
We find the impulse response of the system simply by letting the input
x[n] = δ[n]. Then
h[n] = (δ[n − 2] + δ[n − 1] + δ[n] + δ[n + 1] + δ[n + 2])/5.

Because the system is LTI, we have that the output, y[n], for any
input, x[n] follows
y[n] = x[n] ∗ h[n]
= x[n] ∗ (δ[n − 2] + δ[n − 1] + δ[n] + δ[n + 1] + δ[n + 2])/5
= (x[n] ∗ δ[n − 2] + x[n] ∗ δ[n − 1] + x[n] ∗ δ[n] + x[n] ∗ δ[n + 1] + x[n] ∗ δ[n + 2])/5
(by linearity of convolution)
= (x[n − 2] + x[n − 1] + x[n] + x[n + 1] + x[n + 2])/5
(by sifting property of convolution).
This matches the definition of y[n], as expected. (c) Plots of H and Y
are shown below. System implements a crude low-pass filter for noise
removal.

Signal System impulse response


1 0.2

0.5 0.15

0 0.1

−0.5 0.05

−1 0
0 10 20 30 40 50 60 70 80 90 100 −2 −1.5 −1 −0.5 0 0.5 1 1.5 2

Signal plus noise System response to noisy signal


2 1.5

1
1
0.5

0 0

−0.5
−1
−1

−2 −1.5
0 10 20 30 40 50 60 70 80 90 100 −20 0 20 40 60 80 100 120

Part (a) Part (c)

You might also like