Analog Communication Systems (ECE3001)

Dr.Thomas Joseph
SENSE
VIT CHENNAI

May 19, 2021

1 / 37
Contents

1 Module-5
Noise
Noise Figure

2 / 37
 Module-5 Noise

Noise

Noise may be defined, in elecrtical terms, as any unwanted

introduction of energy tending to interfere with the proper reception
and reproduction of transmitted signals
Noise interfere with the signal and modifies it in an unwanted manner
It is ever present and limits the performance of most systems
Noise can limit the range of systems, for a given transmitted power,
can produce or cancel pulses in a pulse communication systems, can
limit the bandwidth, or sensitivity etc.
Very important noise quantities: signal to noise ratio, noise figure,
and noise temperature

3 / 37
 Module-5 Noise

Noise

There are numerous ways of classifying noise

It may be subdivided according to type, source, effect or relation to
the receiver depending on the circumstances
Here we divide noise into two broad groups: noise whose sources are
external to the receiver and noise created within the receiver itself
External noise is difficult to treat quantitatively, and there is often
little that can be done about it.
Radiotelescopes are always located away from the industry to reduce
external noise
Satellite earth stations are located in valleys/remote locations

4 / 37
 Module-5 Noise

External Noise

Atmospheric Noise: Mainly due to spurious radio waves which

induce voltage in the antenna
The majority of these radio waves come from natural sources of
disturbance
They represent atmospheric noise , generally called static
Static is caued by the lightning discharges in thuderstoms and other
natural electric disturbances occuring in the atmosphere
It originate in the form of amplitude-modulated impulses
Atmospheric noise consists of spurious radio signals with components
distributed over a wide range of frequencies
Field strength is inversely proportional to frequency

5 / 37
 Module-5 Noise

External Noise

Atmospheric noise becomes less severe at frequencies above about 30

MHz
First reason: the higher frequencies are limited to line-of-sight
propagation; i.e, less than 80 km
Very little noise (atmospheric) is created in the VHF band by various
sources

6 / 37
 Module-5 Noise

External Noise

Extraterrestial Noise: Solar noise and cosmic noise are the two
sources
Solar Noise: Sun is a contant source of noise, it radiate a wide range
of frequencies including the communication spectrum
It has two staes: one quiet condition
Since sun is a constantly changing star, some peak activity results into
additional noise which is typically coming from a small area of the sun
and its magnitude is many order higher than the quiet time activity
Solar cycle disturbances repeat themselves approximately every 11
years
Cosmic Noise: This is the noise received from other stars which are
at high temperature
Mainly in the RF range and is known as thermal or black-body noise

7 / 37
 Module-5 Noise

External Noise

Indistrial Noise: In the frequency range 1 to 600 MHz, the intensity

of noise made by humans is very high
Automobile, aircraft ignition, electric motors, switching equipments,
leakage from high-voltage sources etc
Received noise increases as the receiver bandwidth is increased

8 / 37
 Module-5 Noise

Internal Noise: Thermal Noise

It is the noise created by any active or passive devices found in

receivers
Noise is randomnly distributed over the entire radio spectrum and
hence the noise power is proportional to the bandwidth over which it
is meaured
Thermal Agitation Noise: The noise generated in a resistance or
resisitive component is random and is referred to as thermal,
agitation, white or Johnson noise
It is due to the rapid and random motion of the molecules inside the
component itself
The noise generated by a resistor is approximately proportional to its
absolute temperature, in addition to the bandwidth over which noise
is to be measured
Pn ∝ T δf = kT δf (1)

9 / 37
 Module-5 Noise

Internal Noise:Thermal Noise

This noise can be meaured or felt by using a sensitive electronic

voltmeter
The average value or the DC value is zero, only the ac meter will
register a reading
At any instsnt of time, there are bound to be more electrons arriving
at one particular end than at the other because their movement is
random
The rate of arrival of electrons at either end of the resistor therefore
varies randomly and so does the potential difference between the two
ends
A random voltage across the resisitor definitely exists and may be
both measured and calculated

10 / 37
 Module-5 Noise

Internal Noise: Thermal Noise

Equivalent circuit of a resistor as a noise generator may be drawn as

follows

Figure: Resistance noise generator

Resistor’s equivalent noise voltage Vn

Assume RL is noiseless and is receiving maximum noise power
generated by R ⇒ R = RL

V2 V2 (Vn /2)2 V2
Pn = = = = n
RL R R 4R (2)
2
√
Vn = 4RPn = 4RkT δf ⇒ Vn = 4RkT δf

11 / 37
 Module-5 Noise

Internal Noise: Thermal Noise

Square of the rms noise voltage associated with a resistor is

proportional to the absolute temperature, the value of resistance, and
the bandwidth over which the noise is meaured
Noise voltage is quite independent of the frequency at which it is
measured ⇒ noise is random and is evenly distributed over the
frequency spectrum

12 / 37
 Module-5 Noise

Internal Noise: Shot Noise

Shot noise is another important noise apart from Thermal noise

This is virtually present in all active devices
It is caused by the random variations in the arrival of electrons or
holes at the output electrode of an amplifying device and appears as a
randomly varying noise current superimposed on the output
When amplifying, it is supposed to sound as though a shower of lead
shot were falling on a metal sheet and hence the name shot noise
Shot noise behaves in a similar manner to thermal agitation noise
apart from the fact that it has a different source
Eg. the average collector current is constant, however, there is minute
variations always present which is the cause of shot noise

13 / 37
 Module-5 Noise

Internal Noise: Shot Noise

Many variables are involved in the generation of shot noise and is

customary to use approximate equations for it
Shot-noise current is a little difficult to add to thermal-noise voltage
in calculations, so shot noise formulas used are generally simplified
For a diode the formula is exact and is given by
p
in = 2eip δf (3)

where in = rms shot noise current

e = charge of an electron
ip = direct diode current
δf =bandwidth of system

14 / 37
 Module-5 Noise

Internal Noise

The most convenient method of dealing with shot noise is to find the
value or formula for an equivalent input noise resistor
This precedes the device, which is now assumed to be noiseless and
has a value such that the same amount of noise is present at the
output of the equivalent system as in the practical amplifier
The noise current has been replaced by a resistance so that it is now
easier to add shot noise to thermal noise
The value of the equivalent shot-noise resistnace Req of a device is
generally quoted in the manufacturer’s specifications

15 / 37
 Module-5 Noise

Internal Noise: Transit Time Noise

If the time taken by an electron to travel from the emitter to the

collector of a transistor becomes significant to the period of the signal
being amplified, i.e., at frequencies in the upper VHF range and
beyond, the so called transit-time effect takes place and the noise
input admittance of the transistor increases
The minute currents induced in the input of the device by random
fluctuations in the output current becomes of great importance at
such high frequiencies and create random noise
Once this high frequency noise makes its presence felt, it goes on
increasing with frequency at a rate that soon approaches 6 dB per
octave and this random noise quickly predominates over the other
forms
It is preferable to measure noise at high frequiencies instead of trying
to calculate an input equivalent resistance for it.
A noise figure of as low as 1 dB is possible with transistor amplifiers
well into the UHF band
16 / 37
 Module-5 Noise

Miscellaneous Noise

Flicker: At low frequencies, a poorly understood form of noise called

flicker or modulation noise is found in transistors.
It is proportional to emitter current and junction temperature and is
inversely proportional to frequency
It can be ingnored aboe 500 Hz
It is also known as 1/f noise or pink noise
Resistance: Thermal noise, sometimes called resistance noise is
present in transistors due to base, emitter, collector internal
resistances and among these, base resistance makes significant
contribution
Noise in Mixers: Mixers are much noisier than amplifiers using
identical devices
Conversion transconductance of mixers is much lower than the
transconductance of amplifiers
If image frequency rejection is inadequate, noise associated with the
image frequency will also be accepted
17 / 37
 Module-5 Noise

Addition of Noise due to Several Sources

Lets assume that

√ two sources of thermal
√ agitation noise generators in
series: Vn1 = 4kTR1 δf and Vn2 = 4kTR2 δf
The sum of two such rms voltages in series is given by the square root
of the sum of their squares
q
2 + V2 =
√
Vn,tot = Vn1 n2 4kTR1 δf + 4kTR2 δf
p √
= 4kT δf (R1 + R2 ) = 4kT δfRtot
Rtot = R1 + R2 + ....
Similarly if resistances are connected in parallel find the equivalent
reistance and use it fro noise voltage calculations
In all these cases the noise power remains the same since it is
independent of resistance value

18 / 37
 Module-5 Noise

Addition of Noise due to Several Amplifiers in Cascade

19 / 37
 Module-5 Noise Figure

Signal to Noise Ratio

We measure the equivalent noise resistance of an amplifier for two

purposes
First one is for comparing two devices for evaluating their performance
Second is the comparison of noise and signal at the same point to
ensure that the noise is not excessive
When equivalent noise resistance is difficult to obtain we generally use
signal-to-noise ratio (S/N) for comparison

S Xs V 2 /R  V 2
s
= = s2 = (4)
N Xn Vn /R Vn

20 / 37
 Module-5 Noise Figure

Noise Figure

If two devices are operating at two different impedance levels then the
comparison with equivalent resistnace is difficult
So we use Noise Figure or Noise Factor
The noise figure F is defined as the ratio of the signal-to-noise power
supplied to the input terminals of a receiver or amplifier to the
signal-to-noise power supplied to the output or load resistor

input S/N
F = (5)
output S/N

A practical receiver generate noise and hence the output S/N will be
lower than the input S/N
So minimum value of noise figure is 1
F can be defined in terms of ideal system also, F is the ratio of S/N
of ideal system to that of the system under consideration
21 / 37
 Module-5 Noise Figure

Noise Temprature

Noise Figure is not very convenient measure of noise in dealing with

UHF and microwave low-noise antennas, receivers, or devices
Noise temperature is extensively used for antennas and low -noise
microwave amplifiers
Pt = kT δf
= P1 + P2 = kT1 δf + kT2 δf
(6)
kTt δf = kT1 δf + kT2 δf
Tt = T1 + T2
Teq is the equivalent noise temperature of the amplifier or receiver
whose noise figure is F
Teq Req
F =1+ =1+
T0 R (7)
Teq = T0 (F − 1)
where T0 = 290 K or 17◦ C
22 / 37
 Module-5 Noise Figure

White Noise

A random noise with constant power spectral density contains all the
frequency components in equal proportions, which is called white
noise in comparison with white light
Suppose the white noise is the input to a bandpass filter with lower
cut-off frequency fL and upper cut-off frequency fH the output power
is given by
Z ∞
Pout = GX (f )|H(f )|2 df
−∞
Z fH
=2 GX (f )df
fL
(8)
Z fH
N0
=2 df
fL 2
N0
= 2 (fH − fL ) = N0 (fH − fL )
2
23 / 37
 Module-5 Noise Figure

White Noise

The total power in a band of frequiencies is the product of N0 with

the bandwidth of the BPF
The autocorrelation and power spectral density forms a Fourier
Transform pairs
N0
RXX (τ ) = δ(τ ) (9)
2
Average power of the white noise is RXX (0) which is infinity
White noise is not existing in real life
However, many noises encountered in real life can be assumed to be
or approximated to be white
The autocorrelation of white noise is a delta function ⇒ two time
samples of white noise is completely uncorrelated even if the sample
points are very close to each other. So removing white noise from
wanted signal is very difficult
24 / 37
 Module-5 Noise Figure

Thermal Noise

Thermal noise is produced by the random motion of electrons in a

medium and its intensity increases with temperature
Thermal noise is zero at absolute zero temperature
If the voltage across a resistor is measured using a sensitive
oscilloscope a voltage will be displayed on the screen and its power
spectral density is of the form shown below
A|f |
G (f ) = (10)
e B|f |−1
where A and B are constants depending on temperature and other
physical constants
For frequiencies below the knee of the curve, G(f) is almost constant
If we operate in this range the thermal noise can be treated as white
Thermal noise appers to be approximately white for extremely high
frequiencies on the order of 1013 Hz
25 / 37
 Module-5 Noise Figure

Thermal Noise

For frequiencies in this range the mean square value of the voltage
[R(0)] across the resistor has been shown to be
v¯2 = 4kTBR (11)
where k= Boltzmann’s constant, T= absolute temperature, B =
Bandwidth, R = resistance

Figure: Power density spectrum of thermal noise

The actual noise power generated by the resistor when it is connected

to an external circuit is the primary concern
26 / 37
 Module-5 Noise Figure

Thermal Noise - Noise Temperature

Maximum power transferred to the external circuit when there is

impedance matching and the maximu power transferred is given by

v2
N= = kTB (12)
4R
N
The power spectral density of available power is 2B = kT/2
If we have a system with many number of noise sources then we use a
new term known as system noise temperature, Te (in Kelvin), which
is the temperature of a single source that would produces the same
total noise power at the output
If the input to a system contains noise, system will add its own noise
to the output, so that total noise output increases

27 / 37
 Module-5 Noise Figure

Thermal Noise - Noise Figure

The ratio of systems noise power at the output to the input noise
power is known as noise figure
Noise figure is usually expressed in dB
3dB noise figure indicates that the system is adding an amount of
noise equal to that appear at the input, so that the output noise
power is double that of input noise power.

28 / 37
 Module-5 Noise Figure

Shot Noise or Quantum Noise

Shot noise is the variation of current around the average value

Current is not a continuous phenomenon rather it is discrete.
Current occurs in dsicrete pulses when an electron crosses an
observation point

Figure: Shot Noise

The power spectral density of shot noise is approximately flat in the

frequency band of interest

29 / 37
 Module-5 Noise Figure

Narrowband Noise

In communication engineering we generally deals with bandpass filters

which convert the white noise into bandpass noise
If the bandwidth of the noise is small compared to the center
frequency of the noise, then it is known as narrowband noise
We can find the power spectral density and autocorrelation function
of the bandpass noise by white noise approach
However, we often use multipliers and mixers, which are nonlinear
devises in that case expressing the noise in trignometric form is much
useful
Trignometric expression of narrowband noise is given by

n(t) = x(t) cos(2πf0 t) − y (t) sin(2πf0 t) (13)

30 / 37
 Module-5 Noise Figure

Signal to Noise Ratio

In many communication systems we are interested in the average

power
However, the power can be increased by passing the received signal
through an amplifier, which amplifies the noise also
Now the interested parameter becomes the ratio of signal power to
the unwanted noise power and is known as signal-to-noise ratio S/N
or SNR

31 / 37
 Module-5 Noise Figure

Noise: Thermal Noise

Noise is mainly of concern in receiving systems, where it sets a lower

limit on the size of signal that can be usefully received.
Even when precautions are taken to eliminate noise from faulty
connections or that arising from external sources, it is found that
certain fundamental sources of noise are present within electronic
equipment that limit the receiver sensitivity
Adding amplifiers to a receiving system also adds noise, and the
signal-to-noise ratio, which is the significant quantity, may be
degraded by the addition of the amplifiers
The electrons are in motion, and this motion in turn is randomized
through collisions with imperfections in the structure of the conductor.
As a result, the electron density throughout the conductor varies
randomly, giving rise to a randomly varying voltage across the ends of
the conductor
Since the noise arises from thermal causes, it is referred to as thermal
noise (and also as Johnson noise, after its discoverer)
32 / 37
 Module-5 Noise Figure

Thermal Noise

The average or mean noise voltage across the conductor is zero, but
the root-mean-square value is finite and can be measured
It is found that the mean-square value of the noise voltage is
proportional to the resistance of the conductor, to its absolute
temperature, and to the frequency bandwidth of the device measuring
(or responding to) the noise
Consider a conductor that has resistance R, across which a true rms
measuring voltmeter is connected, and let the voltmeter have an ideal
band-pass frequency response of bandwidth Bn
The mean-square voltage measured on the meter is found to be

En2 = 4kRTBn (14)

33 / 37
 Module-5 Noise Figure

Thermal Noise

En = root mean square noise voltage

R = resistance of the conductor in ω
T = conductor temperature in Kelvins
Bn = Noise bandwidth in Hertz
k = Boltzmann constant
The rms noise voltage is given by
p
En = 4kRTBn (15)

Can we use thermal noise as a source of energy?

To abstract the noise power, the resistance R would have to be
connected to a resistive load, and in thermal equilibrium the load
would supply as much energy to R as it receives

34 / 37
 Module-5 Noise Figure

Thermal Noise
The fact that the noise power cannot be utilized as a free source of
energy does not prevent the power being calculated.
In analogy with any electrical source, the available average power is
defined as the maximum average power the source can deliver
For a generator of emf E volts (rms) and internal resistance R, the
available power is E 2 /4R.
P = kTBn (16)

Figure: Thermal noise voltage

35 / 37
 Module-5 Noise Figure

Thermal Noise

Measurement of Thermal noise is shoiwn below

Figure: Measurement of Thermal noise

36 / 37
 Module-5 Noise Figure

Thermal Noise

Thevenin equivqlent and Nortan equivalent representation of Thermal

noise is shown below

Figure: Equivalent sources for Thermal noise (a) voltage source (b current source)

Reactances do not generate thermal noise

This follows from the fact that reactance cannot dissipate power.

37 / 37