Signal and System Fundamentals

Week 2 Notes
Prof. Xiaojing Huang
School of Electrical and Data Engineering
Faculty of Engineering and Information Technology
University of Technology Sydney
 Signals
• A signal is a set of information or data
– Telephone or television signal
– Monthly sales figure of a corporation
– Daily closing prices of a stock market
• They are functions of some independent variables
(time and/or space, …)

| Page 2
 Systems
• A system is an entity that processes a set of signals
(inputs) to yield another set of signals (outputs)
• A system may be made up of physical components
(hardware realisation), or it may be an algorithm that
computes an output from an input signal (software
realisation)
• For example, a filter can be considered as a system. It
can be implemented by either hardware with
electronic components or software via digital signal
processing

| Page 3
 Signal Energy
• Given a signal , the signal energy is defined as
=

for real signal, and

=
for complex valued signal

| Page 4
 Signal Power
• For signal with infinite energy, a more meaningful
measure of the signal size would be the time average
of the energy
• The average signal power is defined as
1
= lim
→

for real signal, and

⁄
= lim ⁄
→
for complex valued signal
• rms (root-mean-squared) value =

| Page 5
 Units of Signal Energy and Power
• The standard units of signal energy and power are the
joule (J) and the watt (W)
• In practice, it is often customary to use logarithmic
scales to describe a signal power in dBw or dBm, i.e.,
– 10log10P (dBw)
– Or 30 + 10log10P (dBm)

| Page 6
 Classification of Signals
1. Continuous time and discrete time signals
2. Analog and digital signals
3. Periodic and aperiodic signals
4. Energy and power signals
5. Deterministic and probabilistic (random) signals

| Page 7
 Continuous Time and Discrete Time Signals
• Continuous time signal: Specified for every value of
time t
• Discrete time signal: Specified only at discrete points
of t = nT

| Page 8
 Analog and Digital Signals
• Analog signal: a signal whose amplitude can have any
value in a continuous range
• Digital signal: a signal whose amplitude can take on
only a finite number of values

| Page 9
Periodic and Aperiodic Signals
• Periodic signal: a signal that satisfies
= +
where the smallest value of is the period of
• Aperiodic signal: a signal that is not periodic

• A periodic signal, by definition, must start from -∞ and conEnue

forever
• may also be considered to be a periodic signal with period
, where is any integer. However, is the period.

| Page 10
Energy and Power Signals
• Energy signal: a signal with finite energy
• Power signal: a signal with finite power
• A signal cannot both be an energy signal and a power
signal
• On the other hand, certain signals are neither energy
nor power signals, for example, the ramp signal below:

| Page 11
Deterministic and Random Signals
• Deterministic signal: a signal whose physical
description is known completely, in either
mathematical or graphical form
• Random signal: a signal is known only in terms of
probabilistic description, such as, mean value, mean
squared value and distribution, rather than its full
mathematical or graphical description
• Most of the noise signals encountered in practice are
random signals
• All message signals are random signals

| Page 12
Some Useful Signal Operations
• For a signal
– Time Shifting
• Delay by T: −
• Advance by T: +
– Time Scaling
• Compressed in time by a factor ( > 1):
"
• Expanded in time by a factor ( > 1):
#
– Time Inversion (or Folding)
• A special case of time scaling with = −1
• To time-invert a signal we replace with − such that the time
inversion of yields − (the mirror image about the vertical
axis)
• Recall that the mirror image of about the horizontal axis is
−

| Page 13
Continuous-Time Convolution
• For a signal $ and a filter ℎ , the convolution is
defined as
& =ℎ ∗$ = $ ( ℎ −( (

= $ −( ℎ ( (

• The signals $ ,& and the filter ℎ can each be

complex

| Page 14
Discrete-Time Convolution
• For a signal $ ) of length * and a filter ℎ ) of length
+, the convolution is defined as
-

& ) =ℎ ) ∗$ ) = , $ )− ℎ
./
0

= , $ ℎ )−
./0 -
• & ) is of length * + + − 1
• Circular convolution vs. linear convolution (see
Assignment 1)
• Fast convolution using Fast Fourier Transform (FFT)
and Inverse FFT (IFFT) (see Assignment 1)
| Page 15
Autocorrelation Function
• For a complex signal , the autocorrelation function
is defined as
1 2 = ∗
+2

• It measures the similarity of the signal with its own

displaced version
• The autocorrelation function provides valuable
spectral information about the signal
• The autocorrelation function for a discrete signal )
of length + can be also defined as
-

1 ) = , ∗ +)
./
| Page 16
Cross-correlation Function of two signals
• For two complex signals and z , the cross-
correlation function is defined as
14 2 = 5 ∗
−2 = 5 +2 ∗

• The cross-correlation function is an indication of

similarity (correlation) of two signals
• The cross-correlation function for discrete signals can
be defined similarly

| Page 17
Unit Impulse Signal (1)
• The unit impulse function 6 was first defined by
Paul Adrien Maurice Dirac (hence often known as the
“Dirac delta”) as
6 =0 for ≠ 0 and 6 =1
• The unit impulse can be regarded as rectangular pulse
with a width that has become infinitesimally small, a
height that has become infinitely large and an overall
area that remains constant at unit

| Page 18
Unit Impulse Signal (2)
• Multiplication of a Function by an Impulse
9 6 =9 0 6
9 6 − =9 6 −

• Sampling Property of Unit Impulse Function

9 6 − =9 6 − =9

– The area under the product of a function with an impulse

6 is equal to the value of that function at the instant
where the unit impulse is located

| Page 19
Unit Step Function
• The unit step function is defined as
1 ≥0
( =:
0 <0
• If we want a signal to start at = 0, we need only
multiply the signal by (
– For example, = #" represents an exponential that starts at
= −∞
– If we want this signal to start at = 0 (the causal form), it
can be described as = #" (
• The relationship between ( and 6
" 0 <0 ?@ "
6 2 2=: =( =6
1 ≥0
or ?"

| Page 20
Phasor (or Phase Vector) Representation
• Euler Theorem
= ±BC = DE$F ± G$H)F
• Let F = I + 9. A sinusoid can be represented as the
sum of two complex-valued functions
= B LM "NO + = B LM "NO
KDE$ I + 9 = K
2
Or, as the real part of one of the functions
KDE$ I + 9 = Q= K= B LM "NO =Q= K= BO = BLM "
Im

A
I +9
Re

| Page 21
Continuous-Time Fourier Transform

R S = =TU −G2VS

R S

= R S =TU +G2VS S

| Page 22
Discrete Fourier Transform (DFT)
Y
2VW)
R W =, ) =TU −G
X
0/

) R W
) = 0, ⋯ , X − 1 W = 0, ⋯ , X − 1
Y
1 2VW)
) = , R W =TU +G
X X
\/

For any integer +:

) + +X = )
R W + +X = R W

| Page 23
Fourier Transform Properties

| Page 24
Fourier Transform Pairs

| Page 25
Linear Systems
• Linear time invariant system

T ℎ & =ℎ ∗T
] S _ S ^ S =_ S ] S
• ℎ is the impulse response and _ S is the
frequency response (or transfer function)
• For discrete time system

T ) ℎ ) & ) =ℎ ) ∗T )
] = BL _ = BL ^ = BL = _ = BL ] = BL

• ℎ ) is the impulse response and _ = BL is the

frequency response (or transfer function)
• _ = BL can be evaluated via z-transform of ℎ )
| Page 26
Linear Frequency-Modulated (FM) Signals
• Time Domain Representation
= `=D =TU GV*
where * is the linear FM rate

• The phase of the pulse is φ = V* and the

?c "
instantaneous frequency is S = =*
b ?"

• The bandwidth of the pulse is de = |*|

• The time bandwidth product (TBP) is d = |*|

| Page 27
Spectrum of Linear FM Pulse
• The exact derivation is not straightforward, and a
convenient approximate expression can be obtained as
S S
G S = `=D =TU −GV
* *
• Some important properties
– The spectrum has a similar linear FM structure as that of the
time domain signal
– The envelope is approximately the same as the rectangular
envelope of the time domain signal
– The phase is approximately quadratic in the frequency
domain as it is in the time domain

| Page 28
Matlab Exercises
• Make a Matlab function to implement a unit sample
sequence, the discrete version of unit impulse 6 ,
6 ) − ) over the ) ≤) ≤) interval
• Make a Matlab function to implement a unit step
sequence, the discrete version of unit step function
( , ( ) − ) over the ) ≤) ≤) interval
• Draw the waveform, the phase and frequency of a
linear FM signal
• Draw the spectrum, the magnitude and phase of the
spectrum for a linear FM signal using DFT
• Compare spectra obtained by DFT with those by
approximate expression under different TBPs
| Page 29
Summary
• Signal energy, power and classification (slides 4-17)
– Definitions and examples, signal operations
• Special signals (slides 18-20)
– 6 and u
• Signal and system representations (slides 21-26)
– Phasor representation and time and frequency domain descriptions,
Fourier transform
– Linear systems
• Linear FM signal (slides 27-28)

| Page 30
 Topics for Week 3
• Orbital parameters
– geometry of an ellipse, eccentricity
• Kepler’s laws of planetary motion
• Newton’s laws of motion
• Orbital elements
• Orbital properties
• Useful orbits

| Page 31