UNIT – III

Introduction to systems:

A system is defined as a physical device that performs an operation on an input signal and
produces another signal as output. The concept of system is applicable to the field of electrical,
electronics, mechanical, chemical, economics etc.

Or

A system is a physical device that operates on a signal called excitation or input to produce
another signal called output signal or response.

Eg: motor, amp, turbine, filter etc.

Fig:1-System Block Diagram

Definition: A system is a physical device that generates a response for a given input signal

y(t) = T[x(t)] where T is the Transformation

y(t)= operation on x(t)

Classification of systems:

Continuous time & discrete time Systems.

Classification of continuous time systems:

i) Static & Dynamic

ii) Casual & Non-casual
iii) Linear & Non-linear
iv) Time Variant & Time Invariant
v) Stable & unstable
LTI System:

A continuous time system is linear if it obeys the principle of superposition and it is time
invariant if its input and output relationship does not change with time.When a continuous
time system satisfies the properties of linearity and time invariance then it is called LTI
system(Linear Time Invariant System).

Fig:2- LTI System Block Diagram

Static & dynamic:

A continuous time system is called as static or memory less system if its output at
any instant of time t depends on the present input of the system but not on the past and
future inputs. In any other case the system is called as dynamic system or A system is said
to be dynamic or a memory system if its output depend up on not only the present input
but also depend up on the past and future values.
Ex:
Y(t) = x2(t) static
dx( f )
Y(t) = dynamic
dt
Y(n)=x(n+2) dynamic
Casual & non –casual:
A system is said to be causal if the output of the system at any point of time depend only
on the present input, past inputs and past outputs but does no depend up on the future
inputs and outputs.
If the system at any point of time depends on the future inputs and outputs then the
system is said to be non-causal system.
The causality refers to a system that is realizable in real time. The LTI system is said to be
causal if and only if the impulse response is zero for t<0.That is h(t)=0 for t<0.
ex: y(t) = x(t)+x(t-1) casual
y(t) = t x(t) causal
 y(t)=x(t+3)+x2(t) no-casual

Linear & Non-Linear: A system that satisfies superposition principle is known as Linear.

Super Position Principle: The response to a weighted sum of input signals be equal to a
weighted sum of outputs of individual components

Therefore, a1x1(t) + a2x2(t) = a1T{x1(t)} + a2 T{x2(t)}

Eg: y (t) = t x(t) ---linear system

Y (t) = x2(t)---- non-linear

Time variant & time invariant:

A system is said to be time invariant if its output characteristics does not change with time else
the system is said to be time variant or shift variant.

A system is time invariant or shift invariant if and only if

A system is time variant or shift variant if

Where ‘m’ is the time shift.

Fig:3(a)-original step signal Fig:3(b)-delayed step signal

Eg: y(t ) = 2t x(t) ----time variant

Y(t) = 2 ex(t) ------time invariant

Stable & Unstable: A system is said to be BIBO stable(bounded input-bounded output) system if
and only if every input produces bounded output. The term bounded refers to finite value of t.

Condition for Stability:

Where h(t) is the impulse response

Eg: y(t) = cos (x(t)) -----stable

Impulse response of a system:

The impulse response of a LTI system is the response when the input signal is an unit impulse
function. The system is assumed to be initially relaxed or at zero initial conditions.

x(t) is the input signal and y(t) is the output signal.

Fig:4-LTI system(time domain)

Here the input signal is Impulse function which is denoted by δ(t) and h(t) is the Impulse
response.

y(t) = h(t) x(t)

Therefore, the output is the convolution of Impulse response and input signal.

We know that convolution in time domain is equal to multiplication of their spectrum in

frequency domain
Fourier Transform of h(t) = H(ω) = Transfer Function =

The Transfer Function of a signal is defined as the Transform of the output signal to the
transform of input signal at zero initial conditions.

The output of the system to any arbitrary input is given as convolution of impulse response with
the input.

------- (a)

------- (b)

The equations (a) and (b) are known as convolution integral

Impulse response relates output function to the input function.

Transfer function:

The Transfer Function of a signal is defined as the Transform of the output signal to the
transform of input signal at zero initial conditions.

y(t) = h(t) x(t) <------->

Fig: 5-LTI system(frequency domain)

Say input x (t) = ant the delayed version of the input signal is

We know that

Since, , the above equation becomes

 According to convolution property y(t) = h(t) x(t) <------->

and = Transfer Function = or

and

Therefore

If is a complex function then

Here is the magnitude response which is even symmetric and

is the phase response which is odd symmetric.

Distortion less transmission through a LTI system:

Transmission of a signal through a system is said to be distortion less if the output of the system
is an exact replica of the input signal. However, a constant change in magnitude and a constant
time delay in the output replica is not considered as distortion.

Distortions in linear systems are of two types:

i) Amplitude distortion
ii) Phase distortion

Amplitude distortion:

If the magnitude response of the system is not constant with frequency ,then that results
in amplitude distortion

Phase distortion:

If the phase response is not linear with frequency that results in time delay and is known
as phase distortion.
 Fig:6-condition for distortion less transmission

A transmission is said to be ideal or distortion less if the output waveform is same as input
with some change in amplitude K and a constant time delay of τ sec i.e., the phase should be
linear through origin.

The input function x(t) is said to be transmitted without distortion if the output y(t) is
defined by .In many applications distortion less transmission is required.

Here K is the change in magnitude and τ is the time delay

The input function is said to be transmitted without distortion if the output is defined
by .

According to the time shifting property,

= FT[K ]= K

Here

Fig:7(a)-Magnitude response Fig:7(b)-Phase response

The bandwidth should be infinite for a distortion less transmission. No system has infinite
bandwidth in practice and hence these conditions are not exactly met. K should be constant
always else that results in distortion. Should be always linear through origin with respect to
frequency else that results in delay or phase distortion.
Filter Characteristics of an LTI system:

Filter:

A filter is a frequency selective network which allows certain band of frequencies and
suppresses the remaining frequencies. The band of frequencies that are allowed to pass is
known as pass band and the suppressed band is known as stop band.

Types:

1) Low pass filter (LPF)

2) High pass filter (HPF)
3) Band pass filter (BPF)

Low pass filter: Allows only low frequencies to pass through it.

High pass filter: Allows only high frequencies to pass through it.

Band pass filter: Allows only particular band of frequencies through it.

Sometimes linear filter may behave like a filter depending up on the characteristics of its
transfer function.

Fig:8(a)-LPF characteristics Fig:8(b)-HPF characteristics

Fig:8(c)-BPF characteristics
 Since the response of the system is related to input function by the relation

, the spectrum of the response is modified depending up on the nature of

If for a band of frequencies = 0 then response will be zero, in that case particular band of
frequencies will not be allowed to pass. Then the shape of the output is modified, if certain
bands of frequencies are not allowed to pass.

Fig:9-Filter characteristics of a LPF

Since the Low pass Filter does not allow higher frequency components to pass, the rise of the
pulse is not instantaneous, rather it is delayed. The sharp rise of the pulse causes the presence
of higher frequency components in the input signal whish is not allowed to pass, hence the
output is modified.

Fig:10-Filter characteristics of a HPF

The tilt produced in the output of the high pass filter is due to the presence of low frequency
components.

Ideal Filters:

An ideal filter is a system that allows distortion less transmission for all frequencies in a
certain band. The amplitude spectrum over the band is constant & the phase spectrum is linear.
Causality and paley-wiener criterion:

Causality:

A LTI system is said to be casual if and only if its impulse response is zero for negative
values of t, i.e., h(t) = 0 for t<0

Consider

A system is said to be casual if the output depends on past and present inputs.

impulse response should be zero (causality condition) for negative values of

for t<0

Convolution integral starts from 0 instead of for a causal system.

Paley-wiener Criterion:

Ideal LPF characteristics – is defined by the Transfer function

Where is the cutoff frequency.

As the LPF Frequency limits from the above equation becomes,

=
 sin ( )

Therefore, is a sinc function which extends from -∞ to ∞, here it is not zero for negative
values of time.

h(t) has the max value at t = τ, where τ is the constant time delay.

Fig:11-spectrum of an idle LPF

The width of the main lobe is . So if then τ 0 that is the output or response
approaches input.

Thus if we give unit impulse as input to ideal LPF at t=0, its output (impulse response) h(t) starts
at of time i.e., even before the input is applied. The system is non-casual and hence cannot
be physically realizable.

For a physically realizable system h (t) must be causal,

i.e., h(t) =0 for t<0

The condition for physically realizability is given by

The above condition is known as paley weiner criterion where is the magnitude.
Signal Bandwidth:

The Band Width occupied by the spectrum of the signal is known as signal bandwidth.
Practical signals have frequency components ranging from ω = -∞ to ∞ and with finite energy.
As a result, as the amplitude becomes significantly small. We neglect the component
that has less energy and select a band of frequency components that contains most of the
energy.

This band of frequencies that contain most of the signal energy is known as Bandwidth and is
denoted by fm.

Eg: sinc pulse (spectrum of rect (t/τ)). Most of the energy (95%) is concentrated in the main lobe
and the remaining (5%) is concentrated in the side lobes .That is considered as the signal
bandwidth of that pulse.

Fig:12-Gate pulse and its spectrum

Fig:13-signal bandwidth

System Bandwidth:

In practice it is impossible to design a distortion less system. But it is possible to design a

system that is distortion less over a certain range of frequencies and that range is known as
system bandwidth.
 Fig:14-system bandwidth

Definition:

The range of frequencies for which the magnitude of the system remains with in of the
maximum limit.

It is also defined as the range over which the response is constant is known as system
bandwidth.

The range over which the response does not fall below 0. 707A or is called as 3db
bandwidth or half power bandwidth.

Rise time (tr):

The time taken by the pulse to rise from 10% of its final value to 90% of its final value.

Relation between Risetime & Bandwidth:

We know that

Since δ(ω) = 1 at ω = 0 ; and

 =

Say K=1

On applying the limits we get

Say on differentiating

Si is the sine integral function.

Properties of Si are

i) Si(x) is an odd function Si(-x) = -Si(x)

ii) Si(0)=0
iii) Si(∞) =

Fig:15-Sketch of Si(y)
If then + =1
If then