Scribd
Upload
101 views

Signals and Systems: Problem 1. Determine The Z Transform (Including Region of Convergence) For Each of The

This document contains solutions to signal processing homework problems involving Z-transforms. Problem 1 involves determining the Z-transform and region of convergence for various signals. Problem 2 involves determining possible signals given specific Z-transform forms. Problem 3 discusses frequency responses corresponding to different pole/zero configurations. Problem 4 involves determining if signals are eigenfunctions of given systems. Problem 5 uses pole placement to design a filter to discriminate between three input tones.

Uploaded by

prashantgemini
Copyright
© © All Rights Reserved
Available Formats
Download as PDF, TXT or read online on Scribd
101 views

Signals and Systems: Problem 1. Determine The Z Transform (Including Region of Convergence) For Each of The

This document contains solutions to signal processing homework problems involving Z-transforms. Problem 1 involves determining the Z-transform and region of convergence for various signals. Problem 2 involves determining possible signals given specific Z-transform forms. Problem 3 discusses frequency responses corresponding to different pole/zero configurations. Problem 4 involves determining if signals are eigenfunctions of given systems. Problem 5 uses pole placement to design a filter to discriminate between three input tones.

Uploaded by

prashantgemini
Copyright
© © All Rights Reserved
Available Formats
Download as PDF, TXT or read online on Scribd
You are on page 1/ 7

Signals and Systems

Homework Solution #9
Problem 1. Determine the Z Transform (including region of convergence) for each of the
following signals:
a.

The ROC is
|z| > 1/2.

b. We use the Z-transform property for this one:

c. As before, we split the signal up into two regimes, and use the result of the previous part:

d. Use straightforward Z-transform:

Problem 2. Determine and sketch all possible signals with Z Transforms of the following
forms:
a. The system has a pole at z = 1. The possible signals are,

x1[n] = u[n-1],

ROC : |z| > 1,

x1[n] = -u[-n],

ROC : |z| < 1.

b. We write this as

,in the time domain,


x2[n] = [n-1]* -nu[-n-1] = -(n-2)u[-n+1],

ROC: |z| < 1,

x2[n] = [n-2]*nu[n] = (n-2)u[n-2],

ROC: |z| > 1.

Equivalently, we can write,


x2[n] = [n-1]* -nu[-n-1] = -(n-2)u[-n +2],

ROC: |z| < 1,

x2[n] = [n-2]*nu[n] = (n-2)u[n-3],

ROC: |z| > 1.

c. From the text, we know that the signal contains a harmonic term. The poles are complex
2

conjugates on the unit circle.

d. In finding the inverse Z-transform, we normally have 2 options. If the Z-transform is a


rational function, we use Partial Fraction Expansion (PFE) to break up it up into simpler
terms whose Z-transforms are familiar to us. However, when the Z-transform is nonrational, as in this case, we should look into messaging the expression a bit to put it in a
z
form that we know how to deal with. In this case, we have e term in the Z-transform that
z
we dont know its inverse. However, we can use Taylors Series to express e in powers of
z which we can take care of easily. Hence, starting with

We can now replace exponential terms with their Taylor Series expansion (recall that
). In other words,

Now, by linearity and by noticing that z has inverse Z-transform of [n+k], we know that
z
2z
the inverse Z-transform of (1 - 2e + e ) is

Finally, since X4(z) = z times the above Z-transform, its inverse Z-transform is simply v[n]
3

n0

shifted to the right by 2 samples (recall that x[n-n0] has Z-transform of z X(z)). So we have,

It may appear that the signal has a pole at z = 0 and a zero at z = 0, but using Taylors
expansion we can see that as z 0 the numerator goes to zero faster than the denominator
goes to infinity. Hence the signal has no poles, and the ROC is the entire Z-plane. The signal
can be better expressed as:

Problem 3. Diagram 1 is a unit step signal. As we move along the j axis, we move away
from the pole at the origin and the log-magnitude decays linearly. The phase is constant since
the angle between the pole and any point along positive side of the j axis remains constant.
Diagram 4 moves the pole to the left of the origin. Then the signal becomes damped.
As we move along the j axis, we move away from the pole at the origin, and the logmagnitude will eventually decay linearly. Because the pole is not exactly at the origin, this
decay does not start until some later. The phase starts at 0, and eventually moves to -/2. Note
that as we move farther up the j axis, this system behaves like the system of diagram1.
Diagram 3 adds a zero at the origin. When is small, the zero is dominant. As we
move away from = 0, the effect of the zero diminishes and the log-magnitude increases
linearly. For sufficiently large we are far enough that the zero and pole appear to cancel
each other. Hence the magnitude becomes a constant. The zero at the origin adds a constant
term to the derivative of the log-magnitude everywhere. A zero at the origin means taking the
derivative, so we take the impulse response of diagram 4 and take its derivative. A zero at the
origin means that we take the phase response of diagram 4 and add to it.
Diagram 2 contains complex conjugate poles, and has a harmonic term. Since they are
not exactly on the j axis, the impulse response is damped (remember diagram 1 and diagram
4). The magnitude response will eventually decay twice as fast as that of diagram
4. Since there are two poles, there will be a bump at around = 1. At the origin, the angular

contributions of the two poles cancel each other out, hence the angle is zero. As we move up
the jaxis, the angles add up to -, with each pole contributing -/2.
Diagram 6 a zero at the origin, meaning that we take the derivative of the impulse
response of diagram 2. The zero at the origin increases the derivative of the log-magnitude
plot everywhere. It also adds to the phase response everywhere.
Diagram 5 complex conjugate poles and zeros at the same frequency . This means
that the effect on magnitude cancel each other out, and we get a constant log-magnitude plot.
The phase response at = 0 is zero, as the contributions cancel each other out. As we move
past = 1 where the conjugates are located, the phase moves in the negative direction faster,
but eventually settles back at 0 as we move farther and the contributions again cancel each
other out. The zeros correspond to differentiation, hence we can derive the impulse response
from the previous case.

Problem 4.
a. Since the signal is two-sided, only B could represent the given signal. We write the signal
as
|n|

x[n] = a

0 < a < 1.

Then the Z-transform is given by:

The Z-transform then has poles at z = a, z = 1/a, and a zero at z = 0. Therefore the
answer is none of the above.
n

b. The signal x[n] = 1 can be written as x[n] = (1) . Therefore z = 1 must be in the ROC on the
z-plane. This is possible for A,B,C depending on the choice of ROC but not possible for D.
Therefore the answer is that x[n] = 1 for all n cannot be an eigenfunction of all 4 systems.
c. For stability, the ROC has to include the unit circle. A causal system must also be rightsided. A right-sided system has its ROC to be of form |z| > a, therefore only A,C can be such
a system. A causal system has h[n] = 0, n < 0, therefore its Z-transform has only negative
powers of z, meaning that it has more poles than zeros.

Problem 5. For this problem consider the following input signal:

This signal consists of eigenfunctions of an LTI system, which are exp(j1) and exp(-j1).
Both of these are points on the unit circle, with angles given by 1. Given a system with
two poles, suppose the two poles are at zA = r exp(j0) and zB = r exp(j0). Then we can
write

or in other words, as a difference equation:


2

z Y(z) - (zA + zB)zY(z) + zAzBY(z) = X(z),


y[n +2] - (zA + zB)y[n - 1] + zAzBy[n] = x[n].
a. Consider a system given by two poles at z = r exp(j0). The frequency response of this
system will have its magnitude peak at frequency 0. Hence the signal x[n] given above will
be amplified more when 0 1. We use this to discriminate the three possible tones, by
setting 0 = 20.1209.
r = 0.99;
w_0 = 2*pi*0.1209;
zA = r*exp(j*w_0);
zB = r*exp(-j*w_0);
y = zeros(1,3000);
for n = 1:(length(y)-2)
y(n+2) = (zA+zB)*y(n+1) - (zA*zB)*y(n) + x(n);
end

b. The closer 0 is to 20.1209, the larger the magnitude at the output corresponding to that
tone is. If 0 approaches one of the other tones, then the magnitude at the output
corresponding to those respective tones are larger.
c. We wish to choose r to be close to 1 but without making the system unstable. Choosing r
to be between 0.9 and 0.99 gives good discrimination of the tones without incurring
instability. This is shown in the following figure. When r is chosen to be too close to 1, or
greater than 1, the system becomes unstable.

You might also like