Signals: Classification and Properties

Dr. J. Aravinth
Continuous-Time Vs. Discrete-Time Signals

A continuous-time signal 𝑥(𝑡) is de ned for all time 𝑡 ,

−∞ < 𝑡 < ∞ .

In contrast, to specify a discrete-time signal, we write

𝑥[𝑛] = 𝑥(𝑛𝑇𝑠 ) , 𝑛 = 0, ±1, ±2, … , where 𝑇𝑠 is the sampling

interval.

Sampling of a continuous-time signal provides a discrete-time

signal; but, some signals are naturally generated in discrete-time.
Continuous-Valued Vs. Discrete-Valued Signals

How to discretize the value of a signal?

Quantization: Converts a continuous-valued signal into a

discrete-valued signal.
Analog Vs. Digital Signals

A continuous-time and continuous-valued signal is called an analog

signal.
Analog Vs. Digital Signals

Sampling of an analog signal provides a discrete-time and

continuous-valued signal.
Analog Vs. Digital Signals

Quantization of an analog signal provides a continuous-time and

discrete-valued signal.

A discrete-time and discrete-valued signal is called a digital signal.

Real Vs. Complex Signals

A real signal takes real number values.

A complex signal takes complex number values.

A complex signal can also be viewed as to be taking two-

dimensional vector values.

There are signals that take multi-dimensional vector values.

Examples??
Deterministic Vs. Random Signals

A deterministic signal is a known function of time.

Example: (i) 𝑓(𝜃) = sin(𝜃) , (ii) 𝑔(𝑥) = log(𝑥) , etc.

A random signal takes random (unpredicted) values and can only

be described statistically.

Example: The random noise in electronic and

communication systems. https://www.youtube.com/watch?
v=CCnCMHNyny8 (https://www.youtube.com/watch?
v=CCnCMHNyny8)
Even Vs. Odd Signals

A signal 𝑥(𝑡) or 𝑥[𝑛] is said to be an even signal if

𝑥(−𝑡) = 𝑥(𝑡) .

𝑥[−𝑛] = 𝑥[𝑛]

Thus, an even signal is symmetric about the vertical axis at the time
origin.
Even Vs. Odd Signals

A signal 𝑥(𝑡) or 𝑥[𝑛] is said to be an odd signal if

𝑥(−𝑡) = −𝑥(𝑡) .

𝑥[−𝑛] = −𝑥[𝑛]

Thus, an odd signal is antisymmetric about the vertical axis at the

time origin.
Even Vs. Odd Signals

It is possible for a signal to be neither even nor odd.

Any signal 𝑥(𝑡) can be decomposed into an even part and an odd
part as

𝑥(𝑡) = 𝑥𝑒 (𝑡) + 𝑥𝑜 (𝑡) ,

where the even part and odd part are given by

𝑥(𝑡) + 𝑥(−𝑡) 𝑥(𝑡) − 𝑥(−𝑡)

𝑥𝑒 (𝑡) = and 𝑥𝑜 (𝑡) = .
2 2

A similar result holds for any discrete-time signal 𝑥[𝑛].

Conjugate Symmetry

For complex-valued signals, we may talk about conjugate

symmetry instead of symmetry or anti-symmetry.

Let 𝑥(𝑡) denote a complex-valued signal, i.e.,

𝑥(𝑡) = 𝑎(𝑡) + 𝑗𝑏(𝑡) .

Let 𝑥∗ (𝑡) denote the complex conjugate of 𝑥(𝑡), i.e.,

∗
𝑥 (𝑡) = 𝑎(𝑡) − 𝑗𝑏(𝑡) .

A complex-valued signal 𝑥(𝑡) is said to be conjugate symmetric if

∗
𝑥(−𝑡) = 𝑥 (𝑡) .

A complex-valued signal is conjugate symmetric if and only if (i)

its real part is even, and (ii) imaginary part is odd.
Periodic Vs. Nonperiodic/Aperiodic Signals (Continuous-Time)

A continuous-time signal 𝑥(𝑡) is said to be periodic with period 𝑇 if

there is a positive nonzero value 𝑇 such that

𝑥(𝑡 + 𝑇 ) = 𝑥(𝑡), for all 𝑡 .

What is the period of the signal shown above?

Periodic Vs. Nonperiodic/Aperiodic Signals (Continuous-Time)

If 𝑥(𝑡) is periodic with a period 𝑇 , then it is also periodic with

periods 2𝑇 , 3𝑇 , 4𝑇 , …

The frequency 𝑓 corresponding to a period 𝑇 is de ned by

𝑓 = 1/𝑇 , measured in hertz (Hz) or cycles per second.

The angular frequency 𝜔 correspnding to a period 𝑇 is de ned by

𝜔 = 2𝜋/𝑇 , measured in radians per second.

The minimum value 𝑇0 of the period for which a signal 𝑥(𝑡) is

periodic is called the fundamental period of that signal, and its
reciprocal 𝑓0 = 1/𝑇0 is called the fundamental frequecy.
Periodic Vs. Nonperiodic/Aperiodic Signals (Continuous-Time)

A continuous-time signal 𝑥(𝑡) is said to be aperiodic or nonperiodic

if it is not periodic, that is, if there is no positive nonzero value of 𝑇
such that

𝑥(𝑡 + 𝑇 ) = 𝑥(𝑡), for all 𝑡.

The foregoing de nition of periodicity is unde ned for a constant

or DC signal. We will study about the frequency of a DC signal in a
later lecture.
Periodic Vs. Nonperiodic/Aperiodic Signals (Discrete-Time)

A discrete-time signal 𝑥[𝑛] is said to be periodic with period 𝑁 if

there is a positive integer 𝑁 such that

𝑥[𝑛 + 𝑁] = 𝑥[𝑛], for all 𝑛 .

Periodic Vs. Nonperiodic/Aperiodic Signals (Discrete-Time)

If 𝑥[𝑛] is periodic with a period 𝑁 , then it is also periodic with

𝑚𝑁 where 𝑚 is any positive integer.

The discrete (angular) frequency Ω corresponding to the period 𝑁

is de ned by Ω = 2𝜋/𝑁 , measured in radians per sample, or
simply, in radians.

The smallest period 𝑁0 of a periodic signal 𝑥[𝑛] is called its

fundamental period, and the fundamental (angular) frequency is
given by Ω0 = 2𝜋/𝑁0 .
Energy Vs. Power Signals

In electrical/electronic systems, a signal 𝑥(𝑡) often represents a

voltage 𝑣(𝑡) or current 𝑖(𝑡).

When a voltage 𝑣(𝑡) or current 𝑖(𝑡) is applied through a resistor of

𝑅 ohm, the instantaneous power 𝑃 (𝑡) dissipated in the resistor is

𝑣 (𝑡)/𝑅 or 𝑖 (𝑡)𝑅.
2 2

In both the cases, the instantaneous power 𝑃 (𝑡) is

proportional to the square of the signal.

In signal analysis, we take 𝑅 = 1 ohm to eliminate the

dependence on the resistance.

Then, the instantaneous power 𝑃𝑥 (𝑡) corresponding to a

signal 𝑥(𝑡) is equal to the square of the signal, that is, we
write

2
𝑃𝑥 (𝑡) = 𝑥 (𝑡) .
Energy Vs. Power Signals

The total energy of the signal 𝑥(𝑡) is then given by

∞ ∞ 𝑇 /2
2 2
𝐸𝑥 = 𝑃𝑥 (𝑡)𝑑𝑡 = 𝑥 (𝑡)𝑑𝑡 = lim 𝑥 (𝑡)𝑑𝑡 .
∫ ∫ 𝑇 →∞ ∫
−∞ −∞ −𝑇 /2

The time-averaged power or simply average power of the signal

𝑥(𝑡) is given by

𝑇 /2 𝑇 /2
1 1 2
𝑃𝑥 = lim 𝑃𝑥 (𝑡)𝑑𝑡 = lim 𝑥 (𝑡)𝑑𝑡 .
𝑇 →∞ 𝑇 ∫−𝑇 /2 𝑇 →∞ 𝑇 ∫−𝑇 /2

The square root of the average power 𝑃𝑥 is called the root mean
square (rms) value of the signal 𝑥(𝑡).
Energy Vs. Power Signals

For discrete-time signals, we replace the integrals with summations.

The total energy of the signal 𝑥[𝑛] is de ned by

2
𝐸𝑥 = 𝑥 [𝑛] .
∑
𝑛=−∞

The average power of the signal 𝑥[𝑛] is de ned by

𝑁
1 2
𝑃𝑥 = lim 𝑥 [𝑛] .
𝑁→∞ 2𝑁 + 1 ∑
𝑛=−𝑁
Energy Vs. Power Signals

For periodic signals, the calculation of average power simpli es as

follows:

For a periodic continuous-time signal 𝑥(𝑡) with

fundamental period 𝑇0 , we have

𝑇0 /2
1 2
𝑃𝑥 = 𝑥 (𝑡)𝑑𝑡 .
𝑇0 ∫−𝑇 /2
0

For a periodic discrete-time signal 𝑥[𝑛] with fundamental

period 𝑁0 , we have

𝑁 0 −1
1 2
𝑃𝑥 = 𝑥 [𝑛] .
𝑁0 ∑
𝑛=0
Energy Vs. Power Signals

A signal 𝑥(𝑡) or 𝑥[𝑛] is called an energy signal if its total energy

𝐸𝑥 satis es

0 < 𝐸𝑥 < ∞ .

For an energy signal, the average power 𝑃𝑥 = 0 .

A signal 𝑥(𝑡) or 𝑥[𝑛] is called a power signal if its average power

𝑃𝑥 satis es

0 < 𝑃𝑥 < ∞ .

For a power signal, the total energy 𝐸𝑥 = ∞ .

Energy Vs. Power Signals

The energy and power classi cation of signals is multually

exclusive, that is, a signal cannot be both an energy signal as well as
a power signal.

Periodic signals and random signals are usually viewed as power

signals.

A signal that is both deterministic and aperiodic is an energy signal.

For complex valued signals, we must replace 𝑥2 (𝑡) with |𝑥(𝑡)|2

and 𝑥2 [𝑛] with |𝑥[𝑛]|2 , respectively, where | ⋅ | denotes the
modulus of the complex number.
References:
[1] Simon Haykin and Barry Van Veen, Signals and Systems, Second
Edition, John Wiley and Sons, 2003.

[2] Lecture Notes by Richard Baraniuk.

https://www.di.univr.it/documenti/OccorrenzaIns/matdid/matdid018094.pdf
(https://www.di.univr.it/documenti/OccorrenzaIns/matdid/matdid018094.pdf)