Scribd
Upload
15 views11 pages

Lesson 3

This document summarizes key concepts about linear time-invariant (LTI) systems including: 1) The transfer function relates the output and input of an LTI system in the Z-domain and can be used to determine frequency response. 2) The frequency response shows how a system amplifies signals of different frequencies. 3) The impulse response of a system is equal to its transfer function and describes the output over time from a brief input signal. 4) Questions and exercises are provided to calculate transfer functions, difference equations, impulse responses, and frequency responses for example systems.

Uploaded by

Octav Paloaie
Copyright
© © All Rights Reserved
We take content rights seriously. If you suspect this is your content, claim it here.
Available Formats
Download as PDF, TXT or read online on Scribd
15 views11 pages

Lesson 3

This document summarizes key concepts about linear time-invariant (LTI) systems including: 1) The transfer function relates the output and input of an LTI system in the Z-domain and can be used to determine frequency response. 2) The frequency response shows how a system amplifies signals of different frequencies. 3) The impulse response of a system is equal to its transfer function and describes the output over time from a brief input signal. 4) Questions and exercises are provided to calculate transfer functions, difference equations, impulse responses, and frequency responses for example systems.

Uploaded by

Octav Paloaie
Copyright
© © All Rights Reserved
We take content rights seriously. If you suspect this is your content, claim it here.
Available Formats
Download as PDF, TXT or read online on Scribd
You are on page 1/ 11

Lesson 3

 Discrete transfer function

 Frequency behaviour LTI systems

 Relation between transfer function and impuls response

 Questions / exercises

1
Transfer function

Each LTI system can be written as (for FIR systems applies: ai = 0):

𝑴 𝑵

𝒚𝒏 = 𝒃𝒊 ∙ 𝒙 𝒏 − 𝒊 − 𝒂𝒊 ∙ 𝒚 𝒏 − 𝒊
𝒊=𝟎 𝒊=𝟏

In the z-domain applies:

𝑴 𝑵 𝑴 𝑵
𝑿(𝒛) 𝒀 𝒛 𝒃𝒊 𝒂𝒊
𝒀 𝒛 = 𝒃𝒊 ∙ 𝒊 − 𝒂𝒊 ∙ 𝒊 = 𝑿 𝒛 ∙ −𝒀 𝒛 ∙ →
𝒛 𝒛 𝒛𝒊 𝒛𝒊
𝒊=𝟎 𝒊=𝟏 𝒊=𝟎 𝒊=𝟏

𝑵 𝑴 𝑴 𝒃𝒊
𝒂𝒊 𝒃𝒊 𝒀 𝒛 𝒊=𝟎 𝒛𝒊
𝒀 𝒛 ∙ 𝟏+ =𝑿 𝒛 ∙ → =
𝒛𝒊 𝒛𝒊 𝑿 𝒛 𝑵 𝒂𝒊
𝒊=𝟏 𝒊=𝟎 𝟏+ 𝒊=𝟏 𝒛𝒊

2
Transfer function
The transfer function H(z) is the ratio between output Y(z) and input X(z):

𝑴 𝒃𝒊
𝒀 𝒛 𝒊=𝟎 𝒛𝒊
𝑯 𝒛 = =
𝑿 𝒛 𝑵 𝒂𝒊
𝟏+ 𝒊=𝟏 𝒛𝒊

Example: Given the third order LTI system:

y[n] = 2x[n] – x[n-2] + 0.5  y[n-1] - 0.4  y[n-2] + 0.25  y[n-3]

The transfer function of this system is then given by:


𝟏
𝟐− 𝟐 𝟐 ∙ 𝒛𝟑 − 𝒛
𝑯 𝒛 = 𝒛 =
𝟎. 𝟓 𝟎. 𝟒 𝟎. 𝟐𝟓 𝒛𝟑 − 𝟎. 𝟓 ∙ 𝒛𝟐 + 𝟎. 𝟒 ∙ 𝒛 − 𝟎. 𝟐𝟓
𝟏− 𝒛 + 𝟐 − 𝟑
𝒛 𝒛

Remark: it is common to write the transfer function as a quotient


of 2 polynomials (with non-negative integer exponents)
3
Frequency behaviour LTI systems
For LTI systems, the transfer function can be used to determine the
amplification of signals as a function of the frequency.
Consider a sinelike signal: 𝒙 𝒏 = 𝒆𝒋∙𝛀∙𝒏
We define y[n] as: 𝒚 𝒏 = 𝒙 𝒏 − 𝟏 = 𝒆𝒋∙𝛀∙(𝒏−𝟏) = 𝒆−𝒋∙𝛀 ∙ 𝒆𝒋∙𝛀∙𝒏
For the Z-transforms of x[n] and y[n] apply:
∞

𝑿(𝒛) = 𝒆𝒋∙𝛀∙𝒏 ∙ 𝒛−𝒏


𝒏=𝟎

∞ ∞

𝒀 𝒛 = 𝒆−𝒋∙𝛀 ∙ 𝒆𝒋∙𝛀∙𝒏 ∙ 𝒛−𝒏 = 𝒆−𝒋∙𝛀 ∙ 𝒆𝒋∙𝛀∙𝒏 ∙ 𝒛−𝒏 = 𝒆−𝒋∙𝛀 ∙ 𝑿(𝒛)


𝒏=𝟎 𝒏=𝟎

Since y[n] = x[n-1], it also applies: 𝒀 𝒛 = 𝒛−𝟏 ∙ 𝑿(𝒛)

Combining both equations gives: 𝒛 = 𝒆𝒋∙𝛀


4
Example of frequency behaviour

Given: y[n] = 0.156(x[n] - x[n-1] + x[n-2]) + 1.5y[n-1] - 0.8y[n-2]

The transfer function of this system is:

𝟏 𝟏
𝟎. 𝟏𝟓𝟔 ∙ 𝟏 − 𝒛 + 𝟐 𝟎. 𝟏𝟓𝟔 ∙ 𝒛𝟐
−𝒛+𝟏
𝑯 𝒛 = 𝒛 = 𝟐
𝟏. 𝟓 𝟎. 𝟖 𝒛 − 𝟏. 𝟓 ∙ 𝒛 + 𝟎. 𝟖
𝟏− 𝒛 + 𝟐
𝒛
The amplification of the system as a function of  is then:

𝟎. 𝟏𝟓𝟔 ∙ 𝒆𝒋∙𝟐∙𝛀 − 𝒆𝒋∙𝛀 + 𝟏


𝑯(𝛀) = 𝒋∙𝟐∙𝛀
𝒆 − 𝟏. 𝟓 ∙ 𝒆𝒋∙𝛀 + 𝟎. 𝟖

The graphs of the absolute value and angle (phase shift) of H()
are shown on the next page.

5
Example of frequency behaviour

Matlab script (in rough lines):


W = 0 : 0.001 : 2*pi; % discrete frequency
z = exp(1j*W);
H = 0.156*(z.*z-z+1)./(z.*z-1.5*z+0.8); % transfer function
plot (W, abs(H), 'red‘); grid on;
6 figure; plot (W, angle(H), 'blue'); grid on;
Transfer function = impuls response

The following applies:

𝒀 𝒛
𝑯 𝒛 = → 𝒀 𝒛 = 𝑯 𝒛 ∙ 𝑿(𝒛)
𝑿 𝒛

Consider: x[n] = [n]  X(z) = 1  Y(z) = H(z)

So, the transfer function H(z) equals the z-transform of the impuls
respons.

7
Example impulse response
Assume: y[n] = 5 x[n] - 3.6x[n-1] +1.4 y[n-1] - 0.48y[n-2]

The first 10 values of the impuls response can be calculated directly


from the diffence equation:

{ 5, 3.4, 2.36, 1.672, 1.208, 0.8886, 0.6643, 0.5034, 0.3859, 0.2987 }

The transfer function equals:

𝟑. 𝟔
𝟓− 𝒛 𝟓𝒛𝟐 − 𝟑. 𝟔 ∙ 𝒛 𝟓𝒛𝟐 − 𝟑. 𝟔 ∙ 𝒛
𝑯 𝒛 = = = →
𝟏. 𝟒 𝟎. 𝟒𝟖 𝒛𝟐 − 𝟏. 4 ∙ 𝒛 + 𝟎. 𝟒𝟖 𝒛 − 𝟎. 𝟖 ∙ 𝒛 − 𝟎. 𝟔
𝟏− 𝒛 + 𝟐
𝒛

𝟐∙𝒛 𝟑∙𝒛
𝑯 𝒛 = +
𝒛 − 𝟎. 𝟖 𝒛 − 𝟎. 𝟔

The impuls response then equals: 𝐡 𝐧 = 𝟐 ∙ 𝟎. 𝟖𝐧 + 𝟑 ∙ 𝟎. 𝟔𝐧

The first 10 values equal according to this equal: { 5, 3.4, 2.36,


1.672, 1.208, 0.8886, 0.6643, 0.5034, 0.3859, 0.2987 }
8
Questions / Exercises
1. Give the tranfer functions of the difference equations given
below.
a. y[n] = (x[n] + x[n-1] + x[n-2] + x[n-3] + x[n-4] + x[n-5] + x[n-6]) / 7
b. y[n] = x[n] - x[n-2] + 0.9  y[n-2]
c. y[n] = 2x[n] – 1.6x[n-1] + 0.8y[n-1]

2. Give the difference equation of the following transfer functions.


𝟏 𝟐∙𝒛
a. 𝑯 𝒛 = 𝒛−𝟏 b. 𝑯 𝒛 = 𝟏−𝟎.𝟖∙𝒛+𝟎.𝟓∙𝒛𝟐

𝟑 𝟐 𝒛
c. 𝑯 𝒛 = 𝒛+𝟎.𝟓 − 𝒛−𝟎.𝟕 d. 𝑯 𝒛 = 𝟐−𝟎.𝟖∙𝒛+𝟎.𝟓∙𝒛𝟐

3. Calculate the first 10 samples of the impuls response of the


systems described by the transfer functions of question 2.
9
Questions / Exercises
4. Assume an analogue sine x(t) given by: x(t) = Xtop  sin(t+0)

The signal x(t) is sampled at a sample frequency of fsample.

a. Give the relation between the angular frequency of the discrete


signal X[n] and  and fsample.

b. What is the range of the angular frequency of x[n] if the sampling


meets the theorema of Shannon?

5. Reason that the steady state value of a step response equals:

y[n] = H(1)

b. Calculate the steady state value of the step response of the systems
given by question 1 and 2.

10
Questions / Exercises
6. Given a LTI system with difference equation:

y[n] = 0.4x[n] + 0.9y[n-1]

a. Calculate the steady state step response of this system.

b. Assume the input signal of the system is obtained from an


analogue signal x(t) = 2sin(t) that is sampled by a frequency
of 10 [kHz]. Calculate the amplitude and phase shift of the
output y[n] when f=200[Hz].

c. Give in a graph the frequency response plot of this system


(you don’t have to use logarithmic scales). Plot it in the range:

0    . Use matlab to make the plot.

d. What kind of filter is this LTU system?

11

You might also like