SIGNALS & SYSTEMS

CHAPTER 1

1.0 Introduction

1.1 Continuous-Time and Discrete-Time Signals

Signals are represented mathematically as functions of one or more independent variables.

Speech signal: Acoustic pressure as a function of time, t.

Image signal: Brightness as a function of two spatial variables, x and y.

In this course, two basic types of signals will be considered:

a) Continuous-Time Signals b) Discrete-Time Signals

- Independent variable, n, is discrete

- Independent variable, t, is
and is defined only for integer
continuous
values
- x[n] is referred as a discrete-time
sequence

1.1.2 Signal Energy & Power

Remember,

v 2 (t ) 2
Then the instantaneous power: p(t )  v(t )  i (t )   i (t )  R
R

1
 t2 t2 t
2
1 2
The total energy for t 1 ≤ t ≤ t 2 : t  t R  t R  i (t )dt
2
p (t ) dt v (t ) dt
1 1 1

t t t
1 2 1 21 2 1 2
The average power over t 1 ≤ t ≤ t 2 : 
t2  t1 t1
p (t ) dt  
t2  t1 t1 R
v (t ) dt  
t2  t1 t1
R  i 2 (t )dt

Using the similar terminology for power and energy for any CT signal x(t) or any DT signal
x[n]:

t2

 x(t )
2
dt : The total energy for CT signal
t1

n2

 x[n]
2
: The total energy for DT signal
n  n1

t
1 2

2
x (t ) dt : Average power for CT signal
t2  t1 t1

n2
1

2
x[n] : Average power for DT signal
n2  n1  1 nn1

Over an infinite time interval, -∞ < t < ∞ or -∞ < n < ∞ :

T  T
1
E  lim  x(t ) dt   P  lim 
2 2 2
x(t ) dt x(t ) dt
T  T  2T
T  T

N  N
1
E  lim  x[n]   P  lim 
2 2 2
x[n] x[n]
N  N  2 N  1
n  N n  n  N

With these definitions, we can identify three important classes of signals.

1) With finite total energy. That is E∞ < ∞. Such a signal must have zero average power.
E
P  lim 0
T  2T

2) With finite average power, P∞. If P∞ > 0, then E∞ = ∞.

3) Neither P∞ nor E∞ are finite.

1.2 Transformations of The Independent Variable

2
1.2.1 Examples

i. Time Shift

If n0 or t 0 is positive, the signal is delayed.

If n0 or t 0 is negative, the signal is advanced.

ii. Time Reversal

The signal, x(t), is reflected about t = 0 to obtain x(-t).

iii. Time Scaling

Consider transform;

x(t )  x( t   ) where α and β are given numbers.

Such a transformation of independent variable preserves the shape of x(t), except that the
resulting signal may be linearly

3
 - Stretched if |α| < 1 or
- Compressed if |α| > 1
- Reversed in time if α < 0
- Shifted in time if β is nonzero
- If β is negative, the signal is delayed
- If β is positive, the signal is advanced

In general;

A systematic approach to obtain x(αt + β):

- First, delay or advance x(t) in accordance with the value of β,

- Perform time scaling and/or time reversal on the resulting signal in accordance with
the value of α.

Example 1.1, 1.2, 1.3

β = 1 ⇒ advance (shift to left)

|α| = 3/2 ⇒ linearly compress by a factor of 2/3

1.2.2 Periodic Signals

x(t) = x(t + T) ⇒ x(t) is periodic with period T (T being a nonzero constant) for all values of t
in the time domain. Then x(t) is also periodic with period 2T, 3T, … The fundamental period
T0 of x(t) is the value of T.

4
x[n] = x[n + N] for all values of n. x[n] is also periodic with period 2N, 3N, … The
fundamental period N0 is the value of N.

Example: A DT periodic signal with fundamental period N0 = 3.

1.2.3 Even & Odd Signals

Another set of useful properties of signals relates to their symmetry under time reversal.

 x(t) or x[n] is even when x(-t) = x(t) or x[-n] = x[n]

 x(t) or x[n] is odd when x(-t) = - x(t) or x[-n] = - x[n]

An odd signal must necessarily be 0 at t = 0 or n = 0 since x(0) = -x(0) and x[0] = - x[0].

Any signal can be broken into sum of two signals, one of which is even and one of which is
odd.

1
v  x(t )   x(t )  x(t )
2

1
Od  x(t )   x(t )  x(t )
2

1.3 Exponential and Sinusoidal Signals

1.3.1 CT Complex Exponential and Sinusoidal Signals

5
 x(t )  C  eat , C and a are in general complex numbers.

Real Exponential Signals

Here, C and a are real numbers. In this case, x(t) is called a real exponential.

- If a is positive, then as t increases x(t) is a growing exponential.

- If a is negative, then as t increases x(t) is a decaying exponential.

Also note that for a = 0, x(t) is constant.

Periodic Complex Exponential and Sinusoidal Signals

Here, a is purely imaginary.

Specifically, consider

x(t )  e j0t  This signal is periodic.

A signal closely related to the periodic complex exponential is the sinusoidal signal.

x(t )  A  cos(0t   ) where 0  2 f 0 .

units of t are seconds.

units of ϕ and ω0 are radians and radians per second respectively.

units of f 0 are cycles per second or hertz (Hz).

2
The signal, x(t), is periodic with fundamental period T0 where T0  and ω0 is
| 0 |

fundamental frequency.

Sinusoidal and complex exponential signals are used to describe the characteristics of many
physical processes.

- Using Euler’s relation:

A j j0t A  j  j0t
A  cos(0t   )  e e  e e
2 2

6
 - Periodic signals (in particular complex periodic exponential signal and the sinusoidal
signal) have infinite total energy but finite average power.

Example (The periodic Exponential):

T0 T0
1
E period   e j0t 2
dt   1 dt  T0 and Pperiod  E period  1
0 0
T0

For an infinite # of periods as -∞ < t < ∞, the total energy integrated over all time is infinite.

However, the average power of the signal equals 1 over each period, averaging over multiple
periods always yields an average power of 1.

T
1

2
P  lim e j0t dt  1
T  2T
T

Periodic complex exponentials serve as extremely useful building blocks for many other
signals.

A necessary condition for a complex exponential, ejωt, to be periodic with period T0 is that

e jT0  1 which implies that ωT0 is a multiple of 2π. That is ωT0 = 2πk, k = 0, ±1, ±2, …

General Complex Exponential Signals

x(t) = C·eat

Using C = |C|·ejθ (in polar form) and a = r + jω0 (in rectangular form), then

Ceat | C | e j e( r  j0 )t | C | ert e j (0t  )

Using Euler’s relation:

Ceat | C | ert cos(0t   )  j | C | ert sin(0t   )

For

r=0 ⇒ the real and imaginary parts are sinusoidal

r>0 ⇒ correspond to sinusoidal signals multiplied by a growing exponential

7
 r<0 ⇒ correspond to sinusoidal signals multiplied by a decaying exponential

1.3.2 DT Complex Exponential and Sinusoidal Signals

x[n] = C·αn (C and α are complex numbers in general)

x[n] = C·eβn where α = eβ

Real Exponential Signals (Sequences)

Here, C and α are real.

If |α| > 1, the magnitude of the signal grows exponentially with n.

If |α| < 1, the magnitude of the signal decays exponentially with n.

If α is positive, all the values of C·αn are of the same sign.

If α is negative, then the sign of x[n] alternates.

If α = 1 then x[n] is a constant.

Sinusoidal Signals

β : Purely imaginary (|α| = 1)

x[n]  e j0 n

x[n]  A  cos(0 n   )

Using Euler’s relation:

A j j0 n A  j  j0n
e j0 n  cos 0 n  j sin 0 n  A cos 0 n     e e  e e
2 2

General Complex Exponential Signals

If we write C and α in polar form;

C | C | e j and  |  | e j0

8
Then, C n | C ||  |n cos(0n   )  j | C ||  |n sin(0n   )

1.3.3 Periodicity Properties of DT Complex Exponentials

In terms of some properties, there are differences between CT and DT signals:

e j0t :

1) 0   The rate of oscillation in the signal  . (The signals are distinct for distinct

values of 0 ).

2) e j0t is periodic for any value of 0 .

e j0 n :

1) The signals are not distinct as the signal with frequency 0 is identical to the signals

with frequencies 0 2 , 0 4 , …

e j (0  2 ) n  e j 2 n  e j0 n  e j0 n

Thus, in considering DT complex exponentials, we need consider only a frequency

interval of length 2π. (Usually 0 ≤ ω0 ≤ 2π or –π ≤ ω0 ≤ π)

ω0 : 0→π ⇒ The rate of oscillation in the signal ↑

π → 2π ⇒ The rate of oscillation in the signal ↓

Therefore, the low frequency DT exponentials have values of ω0 near 0, 2π.

the high frequency DT exponentials have values of ω0 near ±π.

2) In order for the signal e j0 n to be periodic with period N > 0,

e j0 ( n N )  e j0n  e j0 N  1

Thus, ω0 N must be a multiple of 2π. That is there must be an integer m such that
0 m 0
0 N  2 m    The signal e j0 n is periodic if is a rational
2 N 2
number. Otherwise, it is not periodic.

9
Determining the Fundamental period and Frequency of DT Complex Exponentials

2
If x[n] is periodic with fundamental period N, its fundamental frequency is .
N

2 0
The fundamental frequency of the periodic signal e j0 n is  .
N m

 2 
The fundamental period is N  m  .
 0 

1.4 Unit Impulse and Unit Step Functions

1.4.1 DT Unit Impulse and Unit Step Functions

The DT Unit Impulse (Sample) Function

0, n  0
 [ n]  
1, n  0

The DT Unit Step Function

0, n  0
u[n]  
1, n  0

The Relationship Between δ[n] and u[n]:

δ[n] = u[n] - u[n-1]

n
u[ n]    [m] or changing
m 
the variable of summation from m to k = n – m

0 
u[n]    [ n  k ] or equivalently u[n]    [ n  k ]
k  k 0

1.4.2 The CT Unit Step and Unit Impulse Functions

The CT Unit Step Function

10
 0, t  0
u (t )  
1, t  0

In Terms of Unit Impulse, δ(t):

t
u (t )    ( )d


du (t )
 (t ) 
dt

However, u(t) is discontinuous at t = 0, thus not differentiable. Therefore, we interpret above

equation by considering an approximation to the unit step uΔ(t).

du (t )
   (t ) 
dt

As Δ → 0, δΔ(t) becomes narrower and higher, maintaining its unit area. Its limiting form,

 (t )  lim   (t )
0

δ(t) has no duration but has unit area.

The area of the impulse A scaled impulse with

not the amplitude! area k.

Changing the variables as in the DT case (τ to σ = t - τ):

11
 t 0
u (t )    ( )d    (t   )  (d )
 

Or equivalently,


u (t )    (t   )  d
0

1.5 CT & DT Systems

A system can be viewed as a process in which input signals are transformed by the system or
cause the system to respond in some way resulting in other signals as outputs.

x(t) → y(t) or x[n] → y[n]

1.5.2 Interconnections of Systems

i. Series (Cascade) Interconnection:

ii. Parallel Interconnection:

iii. Series-Parallel Interconnection:

12
 iv. Feedback Interconnection:

1.6 Basic System Properties

1.6.1 Systems with and without Memory

Memoryless System: If its output for each value of the independent variable at a given time
is dependent on the input at only that same time.

Examples:

 
2
y[n]  2 x[n]  x 2 [n] → Memoryless

Identity System: y[n] = x[n] → Memoryless

n
y[n]   x[k ] or,
k 

n 1
Accumulator: → A DT System with memory
y[n]   x[k ]  x[n] or,
k 

y[n] = y[n-1] + x[n]

Delay: y[n] = x[n-1] → A DT System with memory

A resistor: y(t) = R x(t) → Memoryless

t
1
A capacitor: y (t )   x( )d ( )
C 
→ A CT System with memory

13
1.6.2 Invertibility and Inverse Systems

For an invertible system, distinct inputs lead to distinct outputs.

If a system is invertible, then an inverse system exists that, when cascaded with the original
system, yields an output w[n] equal to the input x[n] to the first system.

Examples:

1) y(t) = 2 x(t) ⇒ The inverse system is w(t) = (1/2) y(t) = x(t) ⇒ Invertible CT system.

n
2) y[n]   x[k ] or y[n] = y[n-1] + x[n] ⇒ The inverse
k 
system is w[n] = y[n] - y[n-1] =

x[n] ⇒ Invertible DT system.

3) y[n] = 0 ⇒ Noninvertible DT system

4) y(t) = x 2 (t) ⇒ Noninvertible CT system

1.6.3 Causality

A system is causal if the output at any time depends on values of the input at only the present
and past times. Also this system is referred as nonanticipative system since the system
output does not anticipate future values of the input.

Examples:

14
 n t
1
1) y[n]  
k 
x[ k ] and y[n] = x[n-1] and y (t )   x( )d ( ) are all causal systems!
C 
2) y[n] = x[n] - x[n+1] ⇒ The system is noncausal!
3) y(t) = x(t+1) ⇒ The system is noncausal!
4) All memoryless systems are causal!
M
1
5) y[n]   x[n  k ] ⇒ The system is noncausal!
2 M  1 k  M
6) The motion of an automobile is causal since it does not anticipate future actions of
the driver.

Example 1.12

i) y[n] = x[-n]

We need to check for all times!

n0 > 0 ⇒ y[n0 ] = x[-n0 ] ⇒ It seems causal. But,

n0 < 0 ⇒ y[n0 ] = x[-n0 ] {For example: y[-4] = x[4]} ⇒ Noncausal!!!

ii) y(t) = x(t) · cos (t+1) ⇒ Causal! Note: Distinguish the effects of the input from
those of any other functions used in the definition of the system.

1.6.4 Stability

Informally, a stable system is one in which small inputs lead to responses that do not
diverge.

Formally, if the input to stable system is bounded (i.e. if its magnitude does not grow
without bound) then the output must also be bounded and therefore cannot diverge.

|x[n]| ≤ Bx < ∞ ⇒ |y[n]| ≤ By < ∞ Then the system is stable!

Examples:

M
1
i) y[n]   x[n  k ] → If |x[n]| is bounded then |y[n]| is bounded. Then the
2 M  1 k  M
system is stable.

15
 n
ii) y[n]   x[k ] → This
k 
system sums all of the past values of input. Even |x[n]| is

bounded, the output, y[n], will be unbounded. Therefore, this system is unstable!

n
For example, x[n] = u[n] ⇒ y[n]   u[k ]  (n  1)u[n] . That is y[0] = 1, y[1] = 2,
k 

y[2] = 3 and so on, and y[n] grows without bound.

iii) y(t) = t · x(t) → To check stability:

Assume, x(t) = 1. This input yields y(t) = t, which is unbounded. The system is
unstable!

iv) y(t) = e x(t) →

| x(t)| < β {or –β < x(t) < β} ⇒ e-β < | y(t)| < eβ ⇒ Stable!

1.6.5 Time Invariance

A system is time-invariant if the behavior and characteristics of the system are fixed over
time.

Specifically, a system is time-invariant if a time shift in the input signal results in an

identical time shift in the output signal. That is

Examples:

1) y(t) = sin[x(t)]

Check TI: Let x 1 (t) be an arbitrary input. Then,

y1 (t) = sin[x 1 (t)]

Another input x 2 (t) = x 1 (t-t 0). Then,

16
 y2 (t) = sin[x 2 (t)] = sin[x 1 (t- t 0 )]

Similarly,

y1 (t- t 0 ) = sin[x 1 (t- t 0 )]

y2 (t) = y1 (t- t 0 ) ⇒ This system is TI.

2) y[n] = n x[n] ⇒ Time varying system.

3) y(t) = x(2t) ⇒ Time varying system.

1.6.6 Linearity

A linear system is a system that possesses the property of superposition.

The system is linear if

1) The response to x 1 (t) + x 2 (t) is y1 (t) + y2 (t) (Additivity property)

2) The response to a·x 1 (t) is a·y1 (t) where a is any complex constant. (Scaling or
homogeneity property)

Together: a·x 1 (t) + b·x 2 (t) → a·y1 (t) + b·y2 (t)

Examples:

1) y(t) = t · x(t)

Check:

x 1 (t) → y1 (t) = t · x 1 (t)

x 2 (t) → y2 (t) = t · x 2 (t)

x 3 (t) → a · x 1 (t) + b · x 2 (t) → ?

17
y3 (t) = t · x 3 (t) = t {a · x 1 (t) + b · x 2 (t)} = a · t · x 1 (t) + b · t · x 2 (t) = a · y1 (t) + b · y2 (t) ⇒
The system is linear!

2) y(t) = x 2 (t)

Check:

x 1 (t) → y1 (t) = x 1 2 (t)

x 2 (t) → y2 (t) = x 2 2 (t)

x 3 (t) → a · x 1 (t) + b · x 2 (t) → ?

y3 (t) = x 3 2 (t) = a2 · x 1 2 (t) + b2 · x 2 2 (t) + 2 a b· x 1 (t) · x 2 (t) = a2 · y1 (t) + b2 · y2 (t) + 2 a b·

x 1 (t)· x 2 (t) ≠ a · y1 (t) + b · y2 (t) ⇒ The system is not linear!

3) y[n] = Re{x[n]}

Check:

x 1 [n] = r[n] + j s[n] → y1 [n] = r[n] Additivity is OK!

x 2 [n] = j x 1 [n] = j (r[n] + j s[n]) = - s[n] + j r[n]

x 2 [n] → y2 [n] = Re{x 2 [n]} = - s[n] ≠ j r[n] ⇒ Not linear!

4) y[n] = 2 x[n] + 3

Check:

Assume; x 1 [n] = 2 and x 2 [n] = 3 then

x 1 [n] → y1 [n] = 2 x 1 [n] + 3 = 7

x 2 [n] → y2 [n] = 2 x 2 [n] + 3 = 9

x 3 [n] = x 1 [n] + x 2 [n] →

y3 [n] = 2[x 1 [n] + x 2 [n]] + 3 =13 ≠ y1 [n] + y2 [n] = 16 ⇒ Not linear!

18