S&S - Week 3

Signals & Systems
2021-1 Week 3
School of Electronics Engineering

Hyunyeol Lee, Ph.D.
Chapter 1. Signals and Systems
• 0. Introduction
• 1. Continuous-time and discrete-time signals
• 2. Transformations of the independent variables
• 3. Exponential and sinusoidal signals
• 4. The unit impulse and unit step functions
• 5. Continuous-time and discrete-time systems
• 6. Basic system properties
What is a Signal?
• A signal is a pattern of variation of some form (function or sequence)

that contains or carries information
• Examples of signal:
• Electrical signals: voltages and currents in a circuit
• Acoustic signals: Acoustic pressure (sound)
• Mechanical signals: velocity of a car
• Image/video signals: intensity level of a pixel

One Dimensional Signal
y=f(t)
Two Dimensional Signals
s=f(x,y)
Multi-Dimensional Signals
s=f(x,y,z)
How is a Signal Represented?
• Mathematically, signals are represented as a function (or a sequence)

of one or more independent variables.
• 1D: y=f(t); 2D: s=f(x,y); 3D: s=f(x,y,t), ∙∙∙
• On this course, we shall be exclusively concerned with signals that are

a function (or a sequence) of a single variable: time (t)
Signal Types
• Continuous-Time (CT) Signals
• Most real world signals (e.g., velocity, voltage)
• Both independent and dependent variables are continuous
• Discrete-Time (DT) Signals

• Some real world signals (e.g., stock market, bank balance) and sampled signals
• Independent variable: discrete, dependent variable: continuous
• Digital Signals
• Computer-stored signals (quantized)
• Both independent and dependent variables are discrete
Domain ( 정의역 ):
- Set of real numbers (CT)
- Set of integers (DT)
Range ( 치역 ):
- Set of complex numbers (CT, DT)
Notation:
- CT signals: x(t)
- DT signals: x(n∆t)=x[n]
Signal Energy
• Total energy over a time interval [t1,t2] (CT) or [n1,n2] (DT):
• Total energy over an infinite time interval:

Signal Power
• Instantaneous power: (CT), (DT)
• The time-averaged power over an infinite interval:

Three classes of signals
• 1. Finite total energy (energy signal):

•  zero average power
• ex) y(t)=1 for 0<t<1 and y(t)=0 otherwise
• 2. Finite average power (power signal):

• > 0  / ex) constant signal, periodic exponential signal
• 3. Signals with and ; ex) y(t)=t

Transformations of Independent Variable
• Time-shift: x(t)  x(t-t0), x[n]  x[n-n0]
• Scaling
• CT: x(t)  x(at) (a>0)
• DT: x[n]  x[bn] (b>1)
• Time-reversal
• CT: x(t)  x(-t)
• DT: x[n]  x[-n]
• Combination of time-shift, scaling, and time-reversal with
x(t)  x(at+b)
• |a|<1: signal stretching

• |a|>1: signal compression
• a < 0: signal reflection
• Shifted in time by b
time-shift
compression reflection
time-shift
Two ways of implementing the variable transformation
time-shift by -1 scale by 3/2
scale by 3/2 time-shift by -2/3

Periodic Signals
• A period (T) of a periodic signal satisfies: x(t)=x(t+T) for all t
• If T is a period, mT is also a period for any integer m, because

x(t)=x(t+T)=x(t+T+T)=x(t+T+T+T)=∙∙∙=x(t+mT)
• The smallest period (T0): fundamental period
• Note: if x(t) is a constant signal, a fundamental period is not defined.

But in DT, the fundamental period of a constant signal is 1.
Sum of two periodic signals
• Given two periodic signals with x(t)=x(t+T1) and y(t)=y(t+T2), would
z(t)=x(t)+y(t) be periodic?
Even and Odd Signals
• Even signals satisfy:
x(t) = x(-t)
x[n] = x(-n]
• Odd signals satisfy:

x(t) = -x(-t)
x[n] = -x[-n]
Even-Odd Decomposition
• Any signal can be broken into a sum of even and odd signals
1.3 Exponential Signals
Continuous Time Discrete Time
real
𝒏
𝒙 [ 𝒏 ] =𝑪 𝜶
general periodic
comple comple
x
𝒏 𝒏 𝒋(𝝎 𝟎 𝒏+ 𝜽)
𝒙 [ 𝒏 ] =𝑪 𝜶 =¿𝑪∨¿𝜶∨¿ 𝒆 ¿
x
1.3.1. Continuous-Time Complex
Exponentials and Sinusoidal Signals
Real Exponential
• Continuous time: C and a are real values
Complex Exponential and Sinusoidal Signals
• CT complex exponential signal with frequency ω0 and period T:
• Fundamental period (smallest period):
• Relation btw complex exponential and sinusoidal signals:

Complex Exponential and Sinusoidal Signals
• Total energy and average power in one period
• The total energy integrated over all time is infinite:

• The average power is always 1.
Harmonically Related Complex Exponentials
• Sets of periodic exponentials, all of which are periodic with a common
period T0.
• 1  1 for any integer k
• Let ω=kω0 
φk: harmonically related complex exponentials
General Complex Exponential Signals
1.3.2. Discrete-Time Complex Exponentials
and Sinusoidal Signals
Real Exponential
• Discrete time: 𝒙 [ 𝒏 ] =𝑪 𝜶 𝒏 C and α are real values with
Sinusoidal Signals
• The math in continuous time signals is identical in discrete time
signals with t  n and x(t)  x[n]
• Infinite total energy and finite average power ( )

General Complex Exponential Signals
|α| > 1
(C,α: complex)
|α| < 1
Periodicity Properties of DT Complex
Exponentials
• For the continuous-time complex exponentials:
• larger ω0  more rapid oscillations
• is periodic for any ω0
• Not the case in the discrete domain

Exponentials
• The exponential at frequency ω0 +2πm (m: integer) is identical to that
at frequency ω0.
• We could only consider a frequency interval of length 2π

 typically in the range - π ≤ ω0 < π or 0 ≤ ω0 < 2π
• As opposed to the CT case, in DT the highest possible frequency

exists, which is π
Exponentials
• A periodic complex exponential with a period N meets:
• Then, the following condition must be satisfied for ω0N
m: integer
• m/N has to be a rational number for a discrete-time complex

exponential signal to be periodic.
Exponentials
• Fundamental frequency of the periodic signal
• Fundamental period N, m having no common factors
ω0= 2π/12  N=12, m=1 ω0= 8π/31  N=31, m=4

ω0= 1/6  cannot find integer m and N that meets m/M being rational
 not a periodic signal (although it looks like)

Harmonically Related Periodic Exponentials
• Periodic exponentials with a common period N
• There are only N distinct periodic exponentials in DT
• Note: there are infinite number of distinct periodic exponentials in CT

S&S - Week 3

Uploaded by

Copyright:

Available Formats

S&S - Week 3

Uploaded by

Document Information

Original Description:

Copyright

Available Formats

Share this document

Share or Embed Document

Sharing Options

Did you find this document useful?

Is this content inappropriate?

Copyright:

Available Formats

S&S - Week 3

Uploaded by

Copyright:

Available Formats

Signals & Systems

School of Electronics Engineering

• A signal is a pattern of variation of some form (function or sequence)

• Acoustic signals: Acoustic pressure (sound)

• Mechanical signals: velocity of a car

• Image/video signals: intensity level of a pixel

• Mathematically, signals are represented as a function (or a sequence)

• 1D: y=f(t); 2D: s=f(x,y); 3D: s=f(x,y,t), ∙∙∙

• On this course, we shall be exclusively concerned with signals that are

• Discrete-Time (DT) Signals

• Total energy over an infinite time interval:

• Instantaneous power: (CT), (DT)

• The time-averaged power over an infinite interval:

• 1. Finite total energy (energy signal):

• ex) y(t)=1 for 0<t<1 and y(t)=0 otherwise

• 2. Finite average power (power signal):

• 3. Signals with and ; ex) y(t)=t

• |a|<1: signal stretching

time-shift by -1 scale by 3/2

scale by 3/2 time-shift by -2/3

• If T is a period, mT is also a period for any integer m, because

• The smallest period (T0): fundamental period

• Note: if x(t) is a constant signal, a fundamental period is not defined.

• Odd signals satisfy:

• Fundamental period (smallest period):

• Relation btw complex exponential and sinusoidal signals:

• The total energy integrated over all time is infinite:

• 1  1 for any integer k

• Infinite total energy and finite average power ( )

• larger ω0  more rapid oscillations

• is periodic for any ω0

• Not the case in the discrete domain

• We could only consider a frequency interval of length 2π

• As opposed to the CT case, in DT the highest possible frequency

• Then, the following condition must be satisfied for ω0N

• m/N has to be a rational number for a discrete-time complex

• Fundamental period N, m having no common factors

ω0= 2π/12  N=12, m=1 ω0= 8π/31  N=31, m=4

 not a periodic signal (although it looks like)

• There are only N distinct periodic exponentials in DT

• Note: there are infinite number of distinct periodic exponentials in CT

You might also like