Sensor Networks and Data Analysis

(SNADA)
Dr Wasiu O. Popoola
w.popoola@ed.ac.uk

Rm 2.2004, JCMB
 Lecture 2
Introduction to noise and bandwidth
Noise: an unwanted signal perturbation that is always
present in any communication system.

Noise is random and non-deterministic. Hence It is

characterised using probability.

The source of the noise could be internal or external

to the system.

Year 2 Signals & Communication Systems

 Internal noise
Noise generated by components within a communication system such as
resistors and solid state active devices.

Internal noise results from the random motion of charge carriers in

electronic components.

Types of internal noise:

Thermal noise: caused by random motion of free electrons in a conductor
or semiconductor excited by thermal agitation.

Shot noise: caused by random arrival of discrete charge carriers in devices

such as thermionic tubes or semiconductor junction devices.

Flicker (1/f) noise: produced in semiconductors and is more severe the

lower the frequency. i.e. it’s strength is inversely proportional to frequency.

Year 2 Signals & Communication Systems

 External noise
Results from sources outside of the communication system such as atmospheric
(thunderstorm), man-made and extra-terrestrial sources.

Atmospheric noise is impulsive. That is, in time domain it is characterised by large

amplitude, short duration bursts.

The strength of atmospheric noise is inversely proportional to frequency. Thus it

affects commercial AM broadcast radio (freq. range: 540 kHz – 1.6MHz) more
than it affects TV and FM which operate above 50MHz.

Man-made noise includes high-voltage power-line corona discharge, commutator

generated noise in electrical motors, automobile and aircrafts ignition noise and
switching-gear noise.

Another man-made noise is radio-frequency transmitter other than the one of

interest. This is often referred to as radio-frequency interference (RFI).

Extra-terrestrial noise sources include our Sun and other hot heavenly bodies
such as stars.

Year 2 Signals & Communication Systems

Noise Calculations

 Noise power from thermal noise is:

Pn = kTDf (Df  B)

 k is Boltzman’s constant (1.38 x 10-23 J/K)

 T is temperature in Kelvin (K) (Kelvin is oC+273 e.g. 20
oC room temperature is 293 K)

 Df is bandwidth of the system being used

Year 2 Signals & Communication Systems

Resistance noise generator.

Year 2 Signals & Communication Systems

Noise Calculations - Resistor

2
(en / 2)
Pn = kTDf = 
R
en = 4kTDf R
 where en is the RMS noise voltage (V).

Year 2 Signals & Communication Systems

Noise effect on a receiver’s amplifier stages.

“Amplifier added”

Minimise noise
by appropriate
use of filtering

Year 2 Signals & Communication Systems

Reminder

 To convert to dB from a power or voltage (amplitude)

ratio:-

P1
=10 log10
P2
V1
= 20 log10
V2
Absolute power of 1 W = 0 dBW or +30 dBm
Year 2 Signals & Communication Systems
Noise Figure for an Amplifier
 Noise ratio NR (or factor, F), Noise Figure, NF (in dB)

( S N ) in
NR =
( S N ) out

NF = 10 log10 ( NR) (dB)

 (S/N)in is the signal power to noise power ratio at the input to

an amplifier (or other device)
 (S/N)out the signal to noise ratio at the output.
 For perfection (i.e. noiseless system) NR=1, NF = 0 dB
Year 2 Signals & Communication Systems
Noise Ratio for cascaded system
 For a cascade of amplifiers 1, 2, etc, overall noise ratio:

NR2 − 1 NRN − 1
NR = NR1 + + .... +
PG1 PG1PG 2 ...PG N −1

 where PGn is the gain of the nth amplifier in the cascade.

 Note the overall NR is dominated by first amplifier, i.e. No. 1!

Year 2 Signals & Communication Systems

Practical Amplifiers

(LNA – Low noise amplifier)

Here f is simply the noise ratio, NR

Year 2 Signals & Communication Systems
 Further Noise considerations

Pn = kTDf (Df  B)

sets the “Noise Floor” in a receiver.

If ‘S’ represents the received signal power level, then S/Pn gives
you the Signal-to-Nose Ratio or SNR.

Year 2 Signals & Communication Systems

Effective Noise Temperature

The effective noise temperate of a system is an alternative method of describing the

noise performance of a system.

The effective noise temperature is a measure of the noisiness of the system referred
to the input.

So, for a system with power gain, 𝑃𝐺 , the effective noise temperature in Kelvin is
given as:

𝑃𝑛
𝑇𝑒 =
𝑃𝐺 𝑘𝐵

Year 2 Signals & Communication Systems

Effective Noise Temperature

The effective noise temp. 𝑇𝑒 of a subsystem (e.g. amplifier) is NEITHER the

temperature of the room NOR the temperature inside its case.

𝑇𝑒 is determined by replacing the noisy network by a noise free one with an

equivalent noise source at its input. The temperature of the equivalent noise source,
𝑇𝑒 , is chosen to make the noise at output of the noise free network equal to the
noise at the output of the noisy network.

That is, if the internal noise, 𝑃𝑛,𝑖𝑛𝑡 , of a subsystem is represented by a noisy resistor
(with noise power 𝑃𝑛,𝑅 ), then 𝑇𝑒 is the temperature of the resistor such that:

𝑃𝑛,𝑅 = 𝑃𝑛,𝑖𝑛𝑡 = 𝑃𝐺 𝑘𝐵𝑇𝑒

Year 2 Signals & Communication Systems

Effective Noise Temperature
Noisy amplifier

Gain,
𝑃𝑛,𝑜𝑢𝑡 Ideal noiseless
𝑃𝐺 amplifier
Noise from
input signal Gain,
𝑃𝑛,𝑜𝑢𝑡
𝑇𝑠 Bandwidth, B;
source noise temp, 𝑇𝑒 𝑇𝑠 𝑃𝐺

𝑇𝑒

Internal noise now referred to the input and the subsystem represented as noiseless. Total
noise at the output is kept the same as:

𝑃𝑛,𝑜𝑢𝑡 = 𝑘𝐵𝑇𝑠 + 𝑘𝐵𝑇𝑒 𝑃𝐺

From source From subsystem

Year 2 Signals & Communication Systems

The relationship between noise ratio, F, and effective noise temperature, 𝑇𝑒 , is:
𝑇𝑒
𝐹 =1+ 𝑇𝑒 = 𝑇0 𝐹 − 1
𝑇0

𝑇0 is the ambient temperature, in Kelvin, of the noisy system/resistor.

For a cascaded system the overall noise ratio is:

F2 − 1 FN − 1
F = F1 + + .... +
PG1 PG1PG 2 ...PG N −1
For a cascaded system the overall effective noise temperature, Te is:

Te 2 Te N
Te = Te1 + + .... +
PG1 PG1PG 2 ...PG N −1
Year 2 Signals & Communication Systems
 Bandwidth

 Bandwidth is the range of frequencies occupied by a signal.

 We always talk about the -3 dB bandwidth

 -3 dB point corresponds to ½ power where the voltage drops

to 0.707 of the peak value.

 This gives the half-power bandwidth

 Why do we use these bandpass filters?

 To select the signal of interest and remove noise and adjacent

channel interference!

Year 2 Signals & Communication Systems

Summary

• Sources/types of Noise in signal transmission

• Bandwidth
• Understand the concept of noise
• Understand how noise is measured &
characterised
• Ability to estimate noise level in a system

• Next lecture – Analogue Carrier Modulation

Year 2 Signals & Communication Systems

 Lecture 3

Analogue Carrier Modulation

Year 2 Signals & Communication Systems

 Carrier Modulation
What is modulation?

Impressing the information signal onto a carrier

wave so that the information signal is then
transmitted at the carrier frequency.

The information signal is a baseband signal.

The term baseband is used to designate the band of frequencies of the signal delivered by the
source or the input transducer.

The carrier wave is usually of higher frequency.

Year 2 Signals & Communication Systems

 Carrier Modulation

X(f)

Carrier Modulation
message

message
fcarrier Frequency
Baseband

Year 2 Signals & Communication Systems

 Pass Band Modulation
Why?
▪ By modulating several baseband signals and shifting their spectra to non-
overlapping bands, one can use all the available bandwidth through
frequency division multiplexing (FDM).

▪ To allow multiple transmitters at the same time without interfering with

each other.

▪ To shift the signal spectrum to higher frequencies in order to enable

efficient power radiation using antennas of reasonable dimensions.

▪ To take advantage of the (good) propagation characteristics at the selected

carrier frequency.

▪ To exchange transmission bandwidth for the signal-to-noise ratio (SNR).

▪ To move the information signal to the allocated frequency band!

Year 2 Signals & Communication Systems
Modulation – Frequency Division Multiplexing

X(f)

fc1 fc2 fc3 fc4 Frequency

Multiple signals
Year 2 Signals & Communication Systems
 Modulation – Antenna size

AM/FM receiver
~ 500 kHz - 108 MHz Mobile phone
800MHz – 3.4 GHz

https://spectrum.ieee.org/telecom/wireless/building-
smartphone-antennas-that-play-nice-together

Year 2 Signals & Communication Systems

 Modulation
How?

v(t ) = VP sin(ct +  )
Use the information signal to alter the amplitude, phase
or frequency of the carrier wave.

Year 2 Signals & Communication Systems

 Modulation

The baseband signals can be analogue or digital resulting

in analogue or digital modulation respectively.

Emphasis will be on analogue modulation at first.

Year 2 Signals & Communication Systems

 Analogue Modulation

Amplitude Modulation (AM):-

Sinusoidal carrier amplitude Vp is linearly varied in proportion to the
information signal (base band) signal. Thus, AM is a linear
modulation.

Frequency Modulation (FM):-

Sinusoidal carrier frequency 𝜔𝑐 is varied in proportion to the
information signal (base band) signal.

Phase Modulation (PM):-

Sinusoidal carrier phase Φ is varied in proportion to the information
signal (base band) signal.

FM and PM are similar and are also called Angle Modulation

Year 2 Signals & Communication Systems
 Carrier
Modulation

Analogue Digital
modulation modulation

Amplitude Angle
Amplitude Frequency Phase shift
(linear) (exponential)
modulation modulation shift keying shift keying keying

Frequency Phase
modulation modulation

Year 2 Signals & Communication Systems

 Linear Modulation

Carrier
Modulation

Analogue
Digital modulation
modulation

Amplitude Angle
Amplitude shift Frequency Phase shift
(linear) (exponential)
keying shift keying keying
modulation modulation

Frequency Phase
modulation modulation
 Linear (Amplitude) Modulation

For a sinusoidal carrier, a modulated carrier can be generally represented as:

𝑥𝑐 𝑡 = 𝐴 𝑡 cos 2𝜋𝑓𝑐 𝑡 + 𝜙(𝑡)

When the amplitude 𝐴 𝑡 is linearly related to the modulating signal, the result is
linear modulation.

Year 2 Signals & Communication Systems

Amplitude Modulation (AM)

Conventional AM is an approach that requires the transmission of the carrier

along with the modulated signal

𝑥𝐴𝑀 𝑡 = 𝐴𝑐 cos 2𝜋𝑓𝑐 𝑡 + 𝑚 𝑡 cos 2𝜋𝑓𝑐 𝑡

Amplitude modulated
Carrier
Carrier

Implementation:

Year 2 Signals & Communication Systems

Amplitude Modulation (AM)

So, in conventional AM

• the transmitter needs to transmit much larger power

• making its transmitter rather expensive.

• But the AM envelope detection receiver is simple and cheap.

Year 2 Signals & Communication Systems

Amplitude Modulation (AM)

• The conventional AM is preferred to linear modulation with no carrier

transmission (DSB-SC) in radio broadcast system.

• This is because in radio broadcast, it is more economical to have one

expensive high-power transmitter and simple, less expensive receivers.

• Recall that the AM is represented as:

𝑥𝐴𝑀 𝑡 = 𝐴𝑐 cos 2𝜋𝑓𝑐 𝑡 + 𝑚 𝑡 cos 2𝜋𝑓𝑐 𝑡

= 𝐴𝑐 + 𝑚 𝑡 cos 2𝜋𝑓𝑐 𝑡

Year 2 Signals & Communication Systems

The spectrum of 𝑥𝐷𝑆𝐵−𝐴𝑀 𝑡 becomes:

1 1
𝑋𝐴𝑀 𝑓 = 𝜋𝐴𝑐 𝛿 𝑓 + 𝑓𝑐 + 𝛿(𝑓 − 𝑓𝑐 ) + 𝑀 𝑓 + 𝑓𝑐 + 𝑀(𝑓 − 𝑓𝑐 )
2 2

Note the impulses at ±𝑓𝑐 .

Year 2 Signals & Communication Systems

 An illustration of
m(t) spectrum

Carrier

Lower Upper
side-band side-band

An illustration of
𝑥𝐴𝑀 𝑡 spectrum

Figure: Conventional AM in both the time and frequency domain.

In AM, message signal m(t) can be recovered using ENVELOPE DETECTION.

Year 2 Signals & Communication Systems

Observations

• Note that 𝑀 𝑓 − 𝑓𝑐 is 𝑀(𝑓)shifted to the right by 𝑓𝑐 and 𝑀(𝑓 + 𝑓𝑐 ) is 𝑀(𝑓)

shifted to the left by 𝑓𝑐 .

• The process of modulation shifts the spectrum of the modulating signal to the
left and right by 𝑓𝑐 .

• If the bandwidth of m(t) is W Hz, then the bandwidth of the modulated signal
is 2W Hz.

• The modulated signal spectrum centred at 𝑓𝑐 is composed of two parts:

• (i) Upper side band (USB) – the portion that lies above 𝑓𝑐 .

• (ii) Lower side band (LSB) – the portion that lies below 𝑓𝑐 .

Year 2 Signals & Communication Systems

Observations (contd)

• Similarly the spectrum centred at −𝑓𝑐 also has USB and LSB.

• Hence this scheme has double side bands.

• The modulated signal does contain a discrete component of the carrier

frequency, 𝑓𝑐 .

• Hence it is called double-sideband AM.

• The condition 𝑓𝑐 ≥ W has to be met to avoid overlap of spectra and lost

of the message signal m(t). 𝑓𝑐 = W gives the minimum carrier frequency
for linear modulation

𝑓𝑐
• To avoid any form of distortion the condition ≫ 1 must be satisfied.
𝑊

Year 2 Signals & Communication Systems

To ensure that the envelope of the AM signal contains all the information about the
message, the condition 𝐴𝑐 + 𝑚 𝑡 > 0 must be satisfied for all t.

An AM signal can also be demodulated using coherent detection, but this defeats
the key advantage of AM which is that the very simple envelope detection can be
used to recover m(t)

Year 2 Signals & Communication Systems

Let 𝑚𝑝 be the peak amplitude of m(t), (i.e. 𝑚𝑝 = min[𝑚 𝑡 ] ), hence the condition for
envelope detection becomes:

𝐴𝑐 ≥ 𝑚 𝑝

The minimum carrier amplitude required for the viability of envelope detection is 𝑚𝑝 .

The AM modulation index (or depth), 𝜇 is defined as the ratio of 𝑚𝑝 to 𝐴𝑐 .That is:

𝑚𝑝
𝜇=
𝐴𝑐

It follows therefore that 0 ≤ 𝜇 ≤ 1 is a required condition for the viability of AM envelope

detection.

Year 2 Signals & Communication Systems

 1
Let the normalised message signal 𝑚𝑛 𝑡 = 𝑚(𝑡),
𝑚𝑝

Thus,
𝑥𝐷𝑆𝐵−𝐴𝑀 𝑡 = 𝐴𝑐 1 + 𝜇𝑚𝑛 𝑡 cos 2𝜋𝑓𝑐 𝑡

When 𝐴𝑐 < 𝑚𝑝 , 𝜇 > 1:- this results in over-modulation and envelope detection can not be
used but coherent detection can still be used.

It should be noted that envelope detection is considerably simpler and less expensive than
coherent detection.

Year 2 Signals & Communication Systems

In the case of tone modulation with 𝑚 𝑡 = 𝐴𝑚 cos 2𝜋𝑓𝑚 𝑡

Then the AM signal becomes:

𝑥𝐴𝑀 𝑡 = (𝐴𝑐 +𝐴𝑚 cos 2𝜋𝑓𝑚 𝑡) cos 2𝜋𝑓𝑐 𝑡

𝑥𝐴𝑀 𝑡 = 𝐴𝑐 (1 + μ cos 2𝜋𝑓𝑚 𝑡) cos 2𝜋𝑓𝑐 𝑡

The modulation index/depth is:

𝐴𝑚
𝜇=
𝐴𝑐

𝜇 ≤ 1 is the prerequisite for undistorted envelope.

Practice Q: Using MATLAB sketch 𝑥𝐴𝑀 𝑡 with different values of 𝜇

Year 2 Signals & Communication Systems

 𝑥𝐴𝑀 𝑡 = 𝐴𝑐 (1 + μ cos 2𝜋𝑓𝑚 𝑡) cos 2𝜋𝑓𝑐 𝑡
= 𝐴𝑐 cos 2𝜋𝑓𝑐 𝑡 + μ cos(2𝜋𝑓𝑚 𝑡) × cos(2𝜋𝑓𝑐 𝑡)

Recall that:
1 1
cos(𝐴) × cos 𝐵 = cos 𝐴 − 𝐵 + cos(𝐴 + 𝐵)
2 2

On expansion:

μ μ
𝑥𝐴𝑀 𝑡 = 𝐴𝑐 cos 2𝜋𝑓𝑐 𝑡 + cos 2𝜋(𝑓𝑐 −𝑓𝑚 )𝑡 + cos 2𝜋(𝑓𝑐 +𝑓𝑚 )𝑡
2 2

Carrier LSB USB

Year 2 Signals & Communication Systems

If the modulation signal m(t) is a multi-tone signal, for example:

𝑚 𝑡 = 𝐴1 cos 2𝜋𝑓𝑚1 𝑡 + 𝐴2 cos 2𝜋𝑓𝑚2 𝑡 + 𝐴3 cos 2𝜋𝑓𝑚3 𝑡

Then,
𝑥𝐴𝑀 𝑡 = 𝐴𝑐 1 + 𝑚 𝑡 cos 2𝜋𝑓𝑐 𝑡

𝑥𝐴𝑀 𝑡 = 𝐴𝑐 (1 + μ1 cos 2𝜋𝑓𝑚1 𝑡 + μ2 cos 2𝜋𝑓𝑚2 𝑡 + μ3 cos 2𝜋𝑓𝑚3 𝑡) cos 2𝜋𝑓𝑐 𝑡

The modulation index, μ, becomes:

μ = μ1 +μ2 +μ2

Similarly, for undistorted envelope, μ ≤ 1.

Practice Q: sketch the amplitude spectrum of a multi-tone AM

Year 2 Signals & Communication Systems

 AM waveform with varying intelligence signal (e1).

Year 2 Signals & Communication Systems

 µ = 0.5

µ = 1.0

µ = 1.5

Figure: DSB-AM signal with different value of modulation index, µ

Year 2 Signals & Communication Systems

Modulation by a band of intelligence frequencies - spectrum.

Year 2 Signals & Communication Systems

Example

 A 1.4 MHz carrier is amplitude modulated by a music signal that has

frequency components from 20Hz to 10 kHz.

 Determine the range of frequencies generated by the upper and lower

sidebands.

Year 2 Signals & Communication Systems

Solution

Year 2 Signals & Communication Systems

DSB-AM Transmission Efficiency

In DSB-AM, the transmitted carrier term does not carry any meaningful
information.Thus the carrier power is a waste.

𝑥𝐴𝑀 𝑡 = 𝐴𝑐 cos 2𝜋𝑓𝑐 𝑡 + 𝑚(𝑡) cos 2𝜋𝑓𝑐 𝑡

1 𝑇 2 𝑑𝑡 𝐴𝑐 2
Carrier power, 𝑃𝑐 = ‫׬‬ 𝐴𝑐 cos 2𝜋𝑓𝑐 𝑡 = .
𝑇 0 2

1 𝑇 1
Similarly, the sidebands power, 𝑃𝑠 = ‫׬‬0 𝑚(𝑡) cos 2𝜋𝑓𝑐 𝑡 2 𝑑𝑡 = 𝑚2 (𝑡) .
𝑇 2

Let 𝑃𝑚 = 𝑚2 (𝑡) represents the message signal power.

The sidebands power is useful power, (contains the message power) while the
carrier power is the power wasted for convenience.

Useful power
Power efficiency, 𝜂 =
Total power

Year 2 Signals & Communication Systems

 𝑃𝑠 𝑚2 (𝑡) 𝑃𝑚
𝜂= = × 100% = × 100%
𝑃𝑐 +𝑃𝑠 𝐴𝑐 2 + 𝑚2 (𝑡) 𝐴𝑐 2 + 𝑃𝑚

For tone modulation with 𝑚 𝑡 = 𝜇𝐴𝑐 cos 2𝜋𝑓𝑚 𝑡, then

2
2
𝜇𝐴𝑐
𝑚 (𝑡) =
2
Hence,
𝜇2
𝜂= × 100%
2 + 𝜇2

Note 0 ≤ 𝜇 ≤ 1. 𝜂 increases monotonically with 𝜇 and the maximum efficiency

occurs at 𝜇 = 1.

That is for tone modulation:

𝜂𝑚𝑎𝑥 = 33%

Year 2 Signals & Communication Systems

 Example:
Q1. Find the power efficiency for an AM waveform with μ = 0.25, if the modulating
signal is: (a) a sinusoid (b) the triangular signal shown below.
(Ans: (a) 3.03 % (b) 2.041 %

Q2. Repeat Q1 but with (i) μ = 0. 5 and (ii) μ = 1

Year 2 Signals & Communication Systems

DSB-AM Detection

The detection of DSB-AM (i.e. demodulation) can be:

1. coherent (synchronous) detection or

2. non-coherent (envelope) detection

Year 2 Signals & Communication Systems

AM Coherent Detection:

The principle of coherent detection is to translate the frequency of the sidebands back to
baseband by multiplying the DSB-AM signal by a sinusoid of the same frequency (and
phase) as the carrier.

Since the carrier is actually transmitted with the modulated signal, it can be extracted using
a circuit called phase locked loop (PLL)

If we consider a DSB-AM signal given by:

𝑥𝐴𝑀 𝑡 = 𝐴𝑐 (1 + μ cos 2𝜋𝑓𝑚 𝑡) cos 2𝜋𝑓𝑐 𝑡

To demodulate coherently, multiply 𝑥𝐴𝑀 𝑡 by a scaled version of the carrier, i.e.

Kcos 2𝜋𝑓𝑐 𝑡

Year 2 Signals & Communication Systems

 𝑥𝐷𝑒𝑚_𝐴𝑀 𝑡 = K 𝐴𝑐 (1 + μ cos 2𝜋𝑓𝑚 𝑡) cos 2𝜋𝑓𝑐 𝑡 cos 2𝜋𝑓𝑐 𝑡

𝑥𝐷𝑒𝑚_𝐴𝑀 𝑡 = K 𝐴𝑐 (1 + μ cos 2𝜋𝑓𝑚 𝑡)cos 2 2𝜋𝑓𝑐 𝑡

K𝐴𝑐
𝑥𝐷𝑒𝑚_𝐴𝑀 𝑡 = μ cos 2𝜋𝑓𝑚 𝑡 + 𝑢𝑛𝑤𝑎𝑛𝑡𝑒𝑑 𝑡𝑒𝑟𝑚𝑠
2

Scaled copy of the message/

modulating signal

The unwanted terms will be sidebands of 2𝑓𝑐 and are easily filtered off

Block diagram of DSB-AM coherent detection

DSB-AM
input Carrier LPF
PLL Output
Multiplier

Year 2 Signals & Communication Systems

Non-coherent detection of DSB-AM

Envelope detection:-
The envelope detector reproduces the envelope of the modulated carrier. That is
the detector’s output follows the envelope on the modulated signal.
It is basically a half-wave rectifier.

An envelope detection circuit is shown below.

During each positive cycle the capacitor charges up to the peak voltage of the input
signal and then decays slowly until the next positive cycle.

The detector’s output thus closely follows the envelope of the input.

The capacitor discharge between positive cycles causes a ripple signal (undesirable)
of frequency
Year 2 Signals & 𝜔
𝑐 in theSystems
Communication output.
 Figure: Envelope detection of a conventional AM signal.

Year 2 Signals & Communication Systems

The ripple can be reduced by increasing the time constant RC such that the capacitor
discharge very little between positive peaks.

If the time constant RC is however too large it will be impossible for the capacitor output to
follow the signal envelope.

The following conditions need to be met:

1 1
𝑅𝐶 > and 𝑅𝐶 <
𝑓𝑐 𝑊

where W is the highest frequency in m(t).That is, the bandwidth of the message m(t).

An RC time constant within the above range is easily achievable since for AM, the message
signal bandwidth, 𝑊 = 5 kHz typically and the carrier frequency, 𝑓𝑐 , ranges from 550 kHz to
1600 kHz.

An optimum choice of RC for a good tracking of the AM envelope is the geometric mean of
its lower and upper limits.

1
𝑅𝐶 𝑜𝑝𝑡𝑖𝑚𝑢𝑚 =
𝑊𝑓𝑐
Year 2 Signals & Communication Systems
For a clearly defined envelope 𝜔𝑐 ≫ 2𝜋𝑊.

The envelope detector output is:

𝑉𝑐 𝑡 = 𝐷𝐶 𝑐𝑜𝑚𝑝𝑜𝑛𝑒𝑛𝑡 + 𝑚 𝑡 + 𝑟𝑖𝑝𝑝𝑙𝑒 𝑠𝑖𝑔𝑛𝑎𝑙

The DC component can be blocked out by a capacitor or a simple high-pass filter.

The ripple may be reduced further by another low pass RC filter.

Year 2 Signals & Communication Systems

Envelope detector with diagonal clipping – effect of large RC values on the
performance of the envelope detector

1.5

0.5

-0.5

-1

-1.5

-2
0 0.05 0.1 0.15 0.2 0.25 0.3 0.35

Year 2 Signals & Communication Systems

Envelope detector with excess ripple - effect of small RC values on the
performance of the envelope detector

1.5

0.5

-0.5

-1

-1.5

-2
0 0.05 0.1 0.15 0.2 0.25 0.3 0.35

Year 2 Signals & Communication Systems

 Envelope detector optimised for intelligence signal

1.5

0.5

-0.5

-1

-1.5

-2
0 0.05 0.1 0.15 0.2 0.25 0.3 0.35

Year 2 Signals & Communication Systems

 Other Analogue Linear Modulation Options:

Double Side Band with Suppressed Carrier (DSB-SC): Similar to

the conventional AM but the carrier signal is not transmitted.

Single Side Band (SSB): As name implies, only one side band is
transmitted. It requires only one-half the bandwidth of the DSB signal

• Both DSB-SC and SSB can be seen as methods of saving transmission

power in linear modulation.

Year 2 Signals & Communication Systems

 Voice (Analogue) AM is used for:

 Broadcast radio in the Long Wave and Medium Wave bands.

Strongest station in Edinburgh is Radio Scotland (810 kHz
MW).

 Some emergency service, taxi, airport, military voice

communications, still use AM!

Year 2 Signals & Communication Systems

 Angle (Exponential) Modulation

Modulation

Analogue
Digital modulation
modulation

Amplitude Angle
Amplitude shift Frequency Phase shift
(linear) (exponential)
keying shift keying keying
modulation modulation

Frequency Phase
modulation modulation
Angle Modulation: Frequency modulation (FM) and Phase Modulation (PM)

For a generalised sinusoidal carrier signal given by:

𝑥 𝑡 = 𝐴𝑐 cos 𝜔𝑐 𝑡 + 𝜃0

where 𝜃 𝑡 = 𝜔𝑐 𝑡 + 𝜃0 , the instantaneous angular frequency 𝜔𝑖 at any instant t is:

𝑑𝜃(𝑡)
𝜔𝑖 =
𝑑𝑡

The phase angle then becomes:

𝑡

𝜃 𝑡 = න 𝜔𝑖 (𝛼) 𝑑𝛼
−∞

With angle or exponential modulation, it is therefore possible to transmit information

𝑚 𝑡 by varying the angle 𝜃 𝑡 of the carrier signal.

Year 2 Signals & Communication Systems

The two simple possibilities are: Phase and Frequency modulation (PM & FM).

In PM, the angle 𝜽 𝒕 is varied linearly with 𝒎 𝒕 .That is:

𝜃 𝑡 = 𝜔𝑐 𝑡 + 𝜃0 + 𝑘𝑝 𝑚(𝑡)

Where 𝑘𝑝 is the phase deviation constant measured in radians per unit of 𝑚(𝑡) and 𝜔𝑐 is
the carrier frequency.

If we assume 𝜃0 = 0, then the resulting PM signal becomes:

𝑥𝑃𝑀 𝑡 = 𝐴𝑐 cos 𝜔𝑐 𝑡 + 𝑘𝑝 𝑚(𝑡)

Year 2 Signals & Communication Systems

The phase deviation from the original phase is:

𝜃 𝑡 = 𝜔𝑐 𝑡 + 𝜃0 + 𝑘𝑝 𝑚(𝑡)

∆𝜃 = 𝜃 𝑡 − 𝜔𝑐 𝑡 + 𝜃0 = 𝑘𝑝 𝑚(𝑡)

The peak phase deviation is thus:

∆𝜃𝑚𝑎𝑥 = 𝑘𝑝 𝑚(𝑡)

The instantaneous angular frequency for PM is:

𝑑𝜃(𝑡) 𝑑𝑚(𝑡)
𝜔𝑖 = = 𝜔𝑐 + 𝑘𝑝 ሶ
= 𝜔𝑐 + 𝑘𝑝 𝑚(𝑡)
𝑑𝑡 𝑑𝑡

Hence in PM, the instantaneous frequency varies linearly with the derivative of the
modulation signal.

Year 2 Signals & Communication Systems

In FM, the instantaneous angular frequency 𝝎𝒊 is varied linearly with the
modulating signal such that:

𝜔𝑖 𝑡 = 𝜔𝑐 + 𝑘𝑓 𝑚(𝑡)

Where 𝑘𝑓 is the frequency deviation constant measured in radians per second per unit of
𝑚(𝑡).

The frequency deviation from the carrier is proportional to the message signal, that is:

∆𝜔 = 𝜔𝑖 𝑡 − 𝜔𝑐 = 𝑘𝑓 𝑚(𝑡)
𝑘𝑓
∆𝑓 = 𝑚(𝑡)
2𝜋

Year 2 Signals & Communication Systems

The instantaneous angle is then:

𝜃 𝑡 = න 𝜔𝑐 + 𝑘𝑓 𝑚(𝛼) 𝑑𝛼
−∞
𝑡

= 𝜔𝑐 𝑡 + 𝑘𝑓 න 𝑚(𝛼) 𝑑𝛼 (Constant term removed for simplicity)

−∞

The FM signal is thus:

𝑡

𝑥𝐹𝑀 𝑡 = 𝐴𝑐 cos 𝜔𝑐 𝑡 + 𝑘𝑓 න 𝑚(𝛼) 𝑑𝛼

−∞

Year 2 Signals & Communication Systems

FM representation.

Year 2 Signals & Communication Systems

Generalised concept of angle modulation

If we phase modulate the carrier with the integral of a message, it is

equivalent to the frequency modulation of the carrier with the
original message.

And if we frequency modulate the carrier with the derivative of a

message, the result is equivalent to the phase modulation of the
carrier with the message itself.

Year 2 Signals & Communication Systems

𝑥𝑃𝑀 𝑡 = 𝐴𝑐 cos 𝜔𝑐 𝑡 + 𝑘𝑝 𝑚(𝑡)

𝑥𝐹𝑀 𝑡 = 𝐴𝑐 cos 𝜔𝑐 𝑡 + 𝑘𝑓 න 𝑚(𝛼) 𝑑𝛼

−∞

A comparison of frequency and phase modulators.

Year 2 Signals & Communication Systems

 Figure: Frequency and phase modulations of square and sawtooth waves.

Year 2 Signals & Communication Systems

Power of Angle Modulated wave

Although the instantaneous angle and phase of an angle modulated wave

can vary with time, the amplitude always remain constant.

Hence, regardless of the value of 𝑘𝑓 or 𝑘𝑝 , the power of an angle-

modulated wave (PM/FM) is always:
𝐴2𝑐
P𝐹𝑀/𝑃𝑀 =
2

Year 2 Signals & Communication Systems

Bandwidth of FM wave

The theoretical bandwidth of an FM wave is infinite but the most of the modulated
signal power resides in a finite bandwidth termed the effective bandwidth.

Narrow-band FM (NBFM)

𝑡
Let a t = ‫׬‬−∞ 𝑚(𝛼) 𝑑𝛼 , then

𝑥𝐹𝑀 𝑡 = 𝐴𝑐 cos 𝜔𝑐 𝑡 + 𝑘𝑓 𝑎(𝑡)

If 𝑘𝑓 𝑎(𝑡) ≪ 1, we have:

𝑥𝐹𝑀 𝑡 = 𝐴𝑐 cos 𝜔𝑐 𝑡 − 𝑘𝑓 𝑎(𝑡) sin 𝜔𝑐 𝑡

This represents the expression for NBFM.

Year 2 Signals & Communication Systems

If the message has a bandwidth of B, the NBFM has a bandwidth of 2B.

Similarly, the NBPM is given by:

𝑥𝑃𝑀 𝑡 = 𝐴𝑐 cos 𝜔𝑐 𝑡 − 𝑘𝑝 𝑚(𝑡) sin 𝜔𝑐 𝑡

Year 2 Signals & Communication Systems

Wide-band FM (WBFM)
Describes the situation when the condition 𝑘𝑓 𝑎(𝑡) ≪ 1 is not satisfied.

The maximum frequency deviation in FM is:

𝑘𝑓
∆𝜔𝑚𝑎𝑥 = 𝑘𝑓 𝑚𝑝 (in rad/s) or ∆𝑓 = 2𝜋 𝑚𝑝 (in Hz)

Carson’s rule for effective bandwidth of FM is given by:

𝐵𝐹𝑀 = 2 ∆𝑓 + 𝐵
𝑘𝑓
=2 𝑚 +𝐵
2𝜋 𝑝

∆𝑓
If the deviation ratio, 𝛽 is defined as: 𝛽= 𝐵

Then the Carson’s rule can be expressed as:

𝐵𝐹𝑀 = 2𝐵 𝛽 + 1

𝐵𝐹𝑀 above contains at least 98% of the signal power

Year 2 Signals & Communication Systems

The deviation ratio controls the amount of modulation and consequently plays a role similar
to the modulation index in AM.

For 𝛽 ≪ 1, the result is NBFM with effective bandwidth of 2B.

For the special case of tone-modulated: 𝛽 is called the modulation index.

𝑘 𝑓 𝑚𝑝
Note that for PM, 𝛽𝑝 = 𝑘𝑝 𝑚𝑝 and for FM 𝛽𝑓 = .
2𝜋𝑓𝑚
𝑘𝑓 is in rad/s/volt while 𝑚𝑝 = max 𝑚(𝑡) .

For a simple FM where the modulating signal is a sinusoid of frequency fm, (𝑚 𝑡 =

𝐴𝑚 cos 2𝜋𝑓𝑚 𝑡 ), the FM wave is given as:

𝑥𝐹𝑀 𝑡 = 𝐴𝑐 cos 2𝜋𝑓𝑐 𝑡 + 𝛽 sin 2𝜋𝑓𝑚 𝑡

𝑥𝐹𝑀 𝑡 above is a complex expression that can be simplified as:

∞

𝑥𝐹𝑀 𝑡 = ෍ 𝐴𝑐 𝐽𝑛 (𝛽) cos 2𝜋 𝑓𝑐 + 𝑛𝑓𝑚 𝑡

𝑛=−∞

where 𝐽𝑛 (. ) is the Bessel function of the first kind order n.

Year 2 Signals & Communication Systems
 ∞

𝑥𝐹𝑀 𝑡 = ෍ 𝐴𝑐 𝐽𝑛 (𝛽) cos 2𝜋 𝑓𝑐 + 𝑛𝑓𝑚 𝑡

𝑛=−∞

The FM representation above shows that for the simple case, the angle-modulated signal
contains all frequencies of the form:

𝑓𝑐 + 𝑛𝑓𝑚 for 𝑛 = 0, ±1, ±2, …

Therefore, the actual bandwidth of the angle modulated signal is infinite.

Hence, the FM has higher bandwidth occupancy and a higher implementation complexity
compared with AM.

However, FM sacrifices bandwidth for high-noise immunity. For this primary reason, the FM
system is widely used in high-fidelity music/audio broadcasting and in point-to-point
communication systems particularly where the transmitter power is quite limited.

The constant envelope of angle-modulated signal is also advantageous when the signal is
amplified by non-linear amplifiers.
Year 2 Signals & Communication Systems
The Bessel functions of the first kind 𝐽𝑛 (β) are defined as the solutions to
the Bessel differential equation:

𝑑2𝑦
2
𝑑𝑦
β 2
+β + β2 − 𝑛2 𝑦 = 0
𝑑β 𝑑β

We don’t need to solve the Bessel differential equation!

The values of 𝐽𝑛 (β) are widely available in graphical and tabular forms.

Note:

𝐽−𝑛 β = −1 𝑛 𝐽𝑛 (β)

Year 2 Signals & Communication Systems

FM spectrum (constant modulating frequency, variable modulation
index and deviation).
(β = mf)

Note how increasing mf increases the occupied bandwidth!

Year 2 Signals & Communication Systems
Commercial FM frequency allocation, adjacent stations.

Year 2 Signals & Communication Systems

AM vs FM
 AM
 Simplest to implement
 Low bandwidth
 No information in carrier transmission
 Low efficiency (<33% for tone message signal)

 FM
 More complex design
 Much higher efficiency
 Wider occupied bandwidth
 Superior noise performance – hence FM radio
 We expect FM broadcast to be completely replaced by DAB!

 We will investigate digital phase modulation later

Year 2 Signals & Communication Systems

 Angle Modulation example - 1
1.(a) In the UK, commercial FM broadcasting stations restricts the maximum
deviation ∆𝑓 to ±75kHz. A single modulating tone of frequency 15kHz producing a
peak deviation of the carrier would yield a modulation index, 𝛽 = 5.

∆𝑓
(Use 𝛽 = )
𝐵

(b) The number of sideband terms with an amplitude of 0.01 or greater is obtained
from the Bessel function as 8.

Thus without using Carson rule, the bandwidth is:

𝐵𝐹𝑀 = 2 × 8 × 𝐵 = 15kHz = 240kHz

If the peak deviation is produced by a 5kHz modulating tone. Find the modulation index, the
number of significant sideband and the bandwidth without using Carson rule. (Ans: 15,19 and
190kHz)

Year 2 Signals & Communication Systems

 Angle Modulation example - 2

2. Let the carrier be 𝐶 𝑡 = 10 cos 2𝜋𝑓𝑐 𝑡 and let the message be 𝑚(𝑡) =
cos 20𝜋𝑡 . Further assume that the message is used to frequency modulate the
carrier with 𝑘𝑓 = 50 Hz/V. Find the expression for the modulated signal and
determine how many harmonic should be selected to contain 99% of the
modulated signal power.

(Ans: 𝑈 𝑡 = 10 cos 2𝜋𝑓𝑐 𝑡 + 5 sin 20𝜋𝑡 ; mod index = 5; 7 harmonics, BW =

140Hz ; not from Carson rule).

Year 2 Signals & Communication Systems

 Angle Modulation example - 3

3) In the UK, commercial FM broadcasting stations restricts the maximum deviation ∆𝑓

to ±75kHz. A single modulating tone of frequency 75kHz producing a peak deviation of
the carrier would yield a modulation index, 𝛽 = 1.

∆𝑓
(Use 𝛽 = )
𝐵

(b) Find the number of sideband terms with a power level of at least 1% of the
unmodulated carrier power.

(c) Find the amplitude as a percentage of the unmodulated carrier signal amplitude and
frequency of all sidebands in (b) if fc = 10MHz.

(d) Find the bandwidth using the result of (b) and using the Carson’s rule

(Ans: b: n = 0,1,2, c: Ampl.: 77%, 44%,11%; freq: fc+nfm and fc- nfm d: 2xnxfm, Carson’s rule,
2fm(β +1))

Year 2 Signals & Communication Systems

 Multiplexing

Year 2 Signals & Communication Systems

Now that you are comfortable with time & frequency

 Lets think about how we combine signals from different

sources.

 We call this multiplexing.

Year 2 Signals & Communication Systems

Multiplexing

Multiplexing:- Combining separate message signals into a composite

signal for transmission over a common channel.

Separate
message . Composite
Multiplexer
signals . (mux) signal
.

The two common methods are:

➢ Frequency division multiplexing (FDM) and
➢ Time division multiplexing (TDM).

TDM is usually used to transmit digital information.

Year 2 Signals & Communication Systems
1. Frequency Division Multiplexing (FDM)
• In FDM the message signals are separated in frequency.

• The filters at the transmitter ensure that the bandwidth of the message signals
is limited to avoid overlap with adjacent message.

Year 2 Signals & Communication Systems

1. Frequency Division Multiplexing (FDM)

• At the receiver, the message signals are separated with a parallel bank of
bandpass filters.

• Each filter is tuned to one of the carrier frequencies and has a bandwidth
that is wide enough to pass the desired signal.

• The output of each bandpass filter is demodulated and passed through a

lowpass filter to extract the baseband message.

• FDM widely used in radio communication and telephony.

Year 2 Signals & Communication Systems

FDM Output Spectra for Speech Signals

Speech signals multiplexed every 4 kHz.

Fibre optic cables use Wavelength Division Multiplex

Year 2 Signals & Communication Systems

FDM Example

 The best example is broadcast radio………

 In AM, 10-20 kHz channel spacing

 In FM, 200 kHz channel spacing

 Must tune radio to select desired station

Year 2 Signals & Communication Systems

2. Channel baseband Time Division Multiplexer (TDM)

• In TDM a given time interval, Tf is selected as a frame.

• Each frame is subdivided into N subintervals of duration Tf/N. N is the number

of users or signals to be multiplexed.

Year 2 Signals & Communication Systems

2. Channel baseband Time Division Multiplexer (TDM)

• In digital speech transmission over telephone lines via PCM, there is a

standard TDM hierarchy.

• 1st level TDM hierarchy (DS-1): 24 digital subscriber signal are TDMed into a
single stream of 1.544Mbps (24 x 64kbps plus a few extra bits for control).

• 2nd level (DS-2): four DS-1 channels are multiplexed into a single stream at
6.312Mbps. And so on.

• The above North American hierarchy is also called the T-hierarchy or T-carrier.
There is an equivalent E-carrier standard in Europe.

• Note that in mobile cellular radio speech transmission, the available bit rate
per user is small and cannot support high bit rates required by PCM. So
speech compression is used.

Year 2 Signals & Communication Systems

 The TDM hierarchy for the North American telephone system

Year 2 Signals & Communication Systems

 You should now be able to:

Explain the term ‘multiplexing’ in one paragraph

Highlight the differences between TDM and FDM

Year 2 Signals & Communication Systems

 FDM vs TDM
• In FDM, the different signals are separated in frequency

• In FDM, each of the different signals occupy a limited portion of the entire
available channel spectrum (bandwidth)

• The different signals in FDM occupy their allocated frequency (or range of
frequencies) for the entire duration of the transmission period

• In FDM, the different signals are band-limited to avoid any overlap/interference

with any adjacent signals

• In TDM, the different signals are separated in time

• In TDM, each of the different signal sources is given access to the entire channel
spectrum/bandwidth during its allocated time slot

• A fraction of the total transmission time is allocated to each signal source

• TDM is often used to multiplex digital signals

Year 2 Signals & Communication Systems
Digital Transmission
Lets move to Digital Transmission

Why digital?
 Only 2 states (0/1)
 Transmission amplitude is fixed
 Can operate at lower received power level
 Typically 20 dB rather than 40+ dB

 Disadvantage is sample rate must increase!

Discuss Five advantages of Digital Modulation over Analogue Modulation

Year 2 Signals & Communication Systems

Advantages of digital modulation over analogue modulation
techniques:

▪ Digital communication can use channel coding (forward error control) to

improve performance.
▪ Digital communication systems are more robust to noise and distortion.
▪ Digital communication can use regenerative repeaters to extend the
reach of a communication system.
▪ Flexible implementation of digital communication systems by taking
advantage of the advances in digital signal processing.
▪ Easier and more efficient to multiplex several digital signals.
▪ Digital communication benefits from the availability and low cost of
digital signal processing hardware.
▪ Digital signals are easier to encrypt/compress. This can maintain secrecy
of information.
▪ Digital data transmission can be easily implemented as part of digital
integrated circuits.

Year 2 Signals & Communication Systems

Baseband Transmission

Shannon Limit

C = B log 2 (1 + S N ) bit/s
This links data rate, C, to bandwidth, B, and SNR

Year 2 Signals & Communication Systems

Implication of Shannon limit!

 For 3.5 kHz speech channel at 20 dB SNR (S/N = 100), then

C = 25 kbit/s.

 For 3.5 kHz speech channel at 40 dB SNR (S/N=10,000), C =

50 kbit/s.

 On a 2 MHz line at 40 dB SNR, C = 28 Mbit/s, as used on

your home phone line into the exchange for DSL or
broadband access!

Year 2 Signals & Communication Systems

 Digital Baseband Transmission

Year 2 Signals & Communication Systems

 Line Coding - Introduction

• We assume that the information is digital (the process of converting an analogue signal to digital has
been covered with Dr Hopgood)

• Only Binary data transmission for now (data consists of only binary digits (bits) 1s and 0s).

• How do we actually transmit bits over a wire or fiber? – By ‘Line Coding’

• Line Coding - A distinct waveform (or pulse) is assigned to each of these symbols and
the resulting sequence of pulses is sent over a communication channel.

• The receiver detects these pulses and converts back to a sequence of 1s and 0s.

• We will discuss different waveforms/pulses that can be used to transmit a digital data.

• The features of these waveforms shall be considered but not effect of random noise.

Year 2 Signals & Communication Systems

Block Diagram of a typical Baseband Digital Transmission

• Line coding:- chooses the most appropriate waveform/pulse to represent a

digital data

• Pulse shaping:- shapes the spectrum (frequency properties) of the pulse (or
transmitted signal) to best suit the transmission channel.
The pulse shaping filter has to be carefully selected to avoid
transmitted pulses interfering with one another. This type of
degradation is termed intersymbol interference (ISI).

• Synchronisation:- ensures that the sampling times coincide with the

received pulse/symbol time duration.

Year 2 Signals & Communication Systems

 Factors that determine the choice of line codes:

• Small transmission bandwidth

• Power efficiency: as small as possible for required data rate and error probability

• Error detection/correction: It should be possible to detect, and preferably correct any

detected error

• Suitable power spectral density: e.g. little low frequency content

• Timing information: clock must be extracted from data

• Transparency: a line code is transparent if the timing information can be easily extracted
form it. So, all possible binary sequences can be transmitted.

Year 2 Signals & Communication Systems

Rectangular Pulse

Pulse amplitude, A
Non-Return-to-Zero (NRZ)
(100% pulse)

Pulse width/duration, T

Return-to-Zero (RZ)
(<100% pulse)

Year 2 Signals & Communication Systems

 Line Code Examples

Year 2 Signals & Communication Systems

On-Off: Simplest line code. Bit 1 represented by a pulse p(t) and symbol 0 by no
pulse.

Polar: Symbol 1 is transmitted by a pulse p(t) and 0 by a pulse −p(t). It is a power

efficient code.

Altenate Mark Inversion (AMI): Also called bipolar or pseudoternary line code. Bit 0
is encoded by no pulse while 1 is represented by a pulse p(t) or −p(t),
depending on whether the previous 1 is encoded by −p(t) or p(t). Pulses
representing consecutive 1s alternate in sign. Can be detected the occurrence of
one single error in a sequence.

Year 2 Signals & Communication Systems

 Timing Signal in Line Codes (For synchronisation)

• It highly desirable to be able to easily extract the timing information (clock signal) from a
line code.

• The timing information (clock signal) is needed for synchronisation

• Example: consider the polar RZ r(t) shown below. If we take the absolute value (or rectify
r(t)) we get a timing signal, c(t).

• A line code where it is easy to extract the timing signal is called a transparent code. So,
the polar line code is transparent.

Year 2 Signals & Communication Systems

Timing Signal in Line Codes (For synchronisation)

NRZ codes can be more problematic. Long strings of 1’s or 0’s can cause loss of
synchronization.

Difficult to extract
clock signal from long
strings of 1 or 0

Many codes limit the number of consecutive runs of 1’s or 0’s, and force bit changes
after a given number of bits.

Year 2 Signals & Communication Systems

 Power Spectral Density (PSD) of Line Codes

• The PSD gives us the signal power distribution per frequency. Its unit is W/Hz

• The bandwidth of the line code is obtainable from the PSD.

• If available channel bandwidth is less than the range of frequencies occupied by the line
code, the pulses become distorted making detection error likely.

• To avoid distortion, the line code bandwidth/spectrum should be less than the channel
bandwidth.

• Line code PSD depends on

− pulse duration or rate - spectrum widens with pulse rate(1/pulse duration)
− pulse shape (smoother pulses have narrower PSD)
− pulse distribution

Year 2 Signals & Communication Systems

 Power Spectral Density (PSD) of Line Codes

• Distortion can result in smeared pulses at the channel output leading to a received
pulse being affected by previous input symbols. This is called Inter-symbol
interference (ISI)

Channel input Channel output

ISI due to pulse
spreading

Year 2 Signals & Communication Systems

 Power Spectral Density (PSD) of Line Codes

The PSD, 𝑆𝑥 𝑓 , is given as the Fourier transform of the autocorrelation function 𝑅𝑥 𝜏 , of

the pulse sequence, 𝑥(𝑡).

∞
𝑃 𝑓 2
𝑆𝑥 𝑓 = ෍ 𝑅𝑚 𝑒 −𝑗2𝜋𝑚𝑇𝑓 PSD equation to find line
𝑇 codes spectrum
𝑚=−∞

Year 2 Signals & Communication Systems

Note for a Rectangular Pulse

Non-Return-to-Zero (NRZ)
(100% pulse)

Pulse amplitude, A
𝑡
𝑝 𝑡 = 𝐴π
𝑇
Pulse width/duration, T 𝑃 𝑓 = 𝑇sinc 𝜋𝑇𝑓
2
𝑃 𝑓 = 𝑇 2 sinc 2 𝜋𝑇𝑓

Return-to-Zero (RZ)
(<100% pulse)

𝑡
𝜏𝑇 𝑝 𝑡 = 𝐴π 𝜏<1
𝜏𝑇
𝑃 𝑓 = 𝜏𝑇sinc 𝜋𝜏𝑇𝑓
2
𝑃 𝑓 = 𝜏𝑇 2 sinc 2 𝜋𝜏𝑇𝑓

Year 2 Signals & Communication Systems

PSD of Polar Signalling: Half-Width Pulse, 𝜏 = 1/2

• Significant signal power at DC (i.e. f

= 0)
• Bandwidth taken at 1st zero crossing
= 2Rb
1
• 𝑅𝑏 = 𝑇 = bit rate in bits per second
(bps or b/s)

Year 2 Signals & Communication Systems

 PSD of On-Off Keying (RZ): Half-Width Pulse, 𝜏 = 1/2

• The DC term results in impulses in the PSD. These are undesirable.

• This corresponds to extra frequency components that don’t carry
information and DC current that just heats up the wires

PSD of Bipolar/AMI: Half-Width Pulse, 𝜏 = 1/2

• The PSD has a null at DC, which aids in transformer coupling.

Year 2 Signals & Communication Systems

 Power Spectral Density (PSD) of Line Codes

Year 2 Signals & Communication Systems

 Digital Carrier Modulation

Year 2 Signals & Communication Systems

 Why Digital Carrier Modulation

▪ Baseband digital signalling is limited in rate.

▪ Carrier modulation moves the information to a suitable

transmission frequency.

▪ Enables us to select appropriate frequency with good

propagation and minimal interference.

▪ Can only use allocated or licensed spectrum!

▪ Enables multiplexing (FDM) - sending multiple signals

simultaneously through a common channel by sharing the
available channel bandwidth.
Year 2 Signals & Communication Systems
Digital (binary) modulation schemes

There are three basic types:

 on-off keying (OOK or ASK),

 frequency shift keying (FSK) and

 phase shift keying (PSK)

Year 2 Signals & Communication Systems

OOK modulation

on-off keying (OOK)

 Simplification of AM
 Only binary information (1/0 or 1/-1)
 This reduces ASK to OOK
 Signal amplitude is fixed
 Switch the carrier on and off

Year 2 Signals & Communication Systems

 Binary Amplitude (On-Off) Keying

In binary amplitude shift keyed (BASK) systems the two digital

symbols 0 and 1 are represented by pulses of a sinusoidal carrier
(frequency, fc) with amplitudes A0 and A1.

In practice, one of the amplitudes, A0, is invariably chosen to be zero

resulting in on-off keyed (OOK) IF modulation:

𝐴1 Π 𝑡/𝑇𝑜 cos 2𝜋𝑓𝑐 𝑡 , for a digital 1

𝑥𝐴𝑆𝐾 𝑡 = ቊ
0, for a digital 0

where 𝑇𝑜 is the symbol duration.

Π(𝑡/𝑇𝑜 ) denotes a rectangular pulse of width/duration 𝑇𝑜 .
Year 2 Signals & Communication Systems
 Carrier waveforms, On-Off Keying (OOK).

Year 2 Signals & Communication Systems

OOK modulation and spectra

Year 2 Signals & Communication Systems

 Frequency Shift Keying

Year 2 Signals & Communication Systems

Frequency Shift Keying

Year 2 Signals & Communication Systems

FSK waveforms and associated spectra

Year 2 Signals & Communication Systems

Binary Frequency Shift Keying

Here the symbol period T0 = 1/f0

Year 2 Signals & Communication Systems
Simplified FSK transmitter.

A Mark or ‘1’ is transmitted at one frequency

and a Space or ‘0’ at a different frequency.
FSK summary

 Transmissions on 2 separate frequencies

 These are thus 2 (OOK) spectral bands
 Overall this must occupy a wider spectral band than for
example OOK
 Continuous transmissions

Year 2 Signals & Communication Systems

PSK modulation

Phase shift keying (PSK)

 Extension of OOK
 Consider binary information as 1/-1 rather than 1/0
 The carrier is always on
 Switch the phase 0 or pi (180 degrees)
 Benefit is continuous transmission
 More efficient

Year 2 Signals & Communication Systems

 Binary Phase Shift Keying (BPSK)

BPSK codes baseband information onto a carrier by changing the

carrier’s phase in sympathy with the baseband digital data:

𝐴Π 𝑡/𝑇𝑜 cos 2𝜋𝑓𝑐 𝑡 , for digital 1

𝑓 𝑡 = ቊ
𝐴Π 𝑡/𝑇𝑜 cos(2𝜋𝑓𝑐 𝑡 + ∅) , for digital 0

Usually antipodal states are chosen, i.e.∅ = 180°.

Year 2 Signals & Communication Systems
 BPSK modulation illustration for a 101101 input

Binary Data Input

1
Amplitude (V)

0.5

0
0 1 2 3 4 5 Time (ms)6

1 Carrier Signal
Amplitude (V)

-1
0 1 2 3 4 5 Time (ms) 6
Phase Shift Keying (PSK) Modulated Signal
1
Amplitude (V)

sin w t -sin wct sin w t sin w t -sin w t sin w ct

c c c c
-1
0 1 2 3 4 5 Time (ms)6

Year 2 Signals & Communication Systems

PSK generation and spectra

Year 2 Signals & Communication Systems

PSK issue

 Information is in the phase

 Constant envelope signal
 Receiver needs access to a recovered synchronised
carrier to ascertain if received phase is 0 or 180 degrees!
 Thus need to get this from the received signal itself.

Year 2 Signals & Communication Systems

A BPSK receiver using coherent carrier recovery
and a mixer circuit.

Year 2 Signals & Communication Systems

fig_08_13
Digital Bit Error Ratio Curves (ASK = OOK)

Year 2 Signals & Communication Systems

Summary

 OOK wastes available power

 FSK occupies excessive bandwidth

 PSK is most efficient but more complex

Year 2 Signals & Communication Systems