University of Kufa 1st-Dec-2023

Pulse Amplitude Modulation

PAM
Haider Hasan Hamood // 3rd Stage -Morning Study-
ECE // University of Kufa

What’s PAM?
Definition
Pulse Amplitude modulation (PAM) is a type of Pulse Modulation technique that uses a series of
regularly occurring pulses (discrete signal) as the carrier signal, in this type of modulation, the
amplitude of the carrier is varied in proportion to the amplitude of the message signal (continuous
signal). In this report, we’ll explain the concepts of 2 types of PAM, which are:
 Natural Sampling PAM

Fig. 1: Modulated Signal form using Natural Sampling method

 Flat-top Sampling PAM

Fig. 2: Modulated Signal form using Flat-top Sampling method

Page | 1 Communication Systems I

University of Kufa 1st-Dec-2023

We’ll also explain the method of implementing them, their differences, and finally, we’ll show
LAB results of both methods.
Before explaining the modulation and demodulation process, we need to know about 2 important
terms, Sampling and Nyquist Rate.
Sampling (or Sampling Theorem): Sampling is the process of converting continuous signal into
a discrete signal or making a continuous signal to occur at a particular interval of time.
Nyquist Rate: It’s the rate at which sampling is done correctly (without errors or loss of
information), or in other words, it’s the rate at which sampling is one so that overlapping of
frequency doesn’t occur. Assume we have a signal 𝑚(𝑡) of bandwidth 𝐵, in frequency domain:

Fig. 3: 𝑀(𝑓) with bandwidth of 𝐵

Ideally, the condition for correct sampling is that the sampling frequency 𝑓𝑠 should be
1 1
𝑓𝑠 = 2𝐵 or for sampling time, 𝑇𝑠 = = , but because that ideal conditions in practical systems
𝑓𝑠 2𝐵
1
are unrealizable, the condition is 𝑓𝑠 ≥ 2𝐵 or for sampling time, 𝑇𝑠 ≥ . If this condition isn’t
2𝐵
satisfied, “Under Sampling” will occur and overlapping of frequency will occur which will result
in loss of information.
𝑆(𝜔)

Fig. 4: sampling cases depending on the sampling frequency

Page | 2 Communication Systems I

University of Kufa 1st-Dec-2023

Modulation and Demodulation Process

In the modulation process, we mainly have 2 signals, the message signal (modulating signal) and
the carrier signal (pulse train).
𝑚(𝑡)

(a)
𝑐(𝑡)

(b)

Fig. 5: message and carrier signal

Using “Natural Sampling Method”, the carrier signal can be expressed in terms of its Fourier
series expansion:
1 2 1 1
𝑐 (𝑡) = + [cos(𝜔𝑐 𝑡) − cos(3𝜔𝑐 𝑡) + cos(5𝜔𝑐 𝑡) − ⋯ ]
2 𝜋 3 5

Fig. 6: Fourier Series of the carrier signal to 𝑛 = 2000 terms

Page | 3 Communication Systems I

University of Kufa 1st-Dec-2023

Multiplying the message signal 𝑚(𝑡) with the equation above resulting in:
1 2 1 1
𝑚(𝑡)𝑐(𝑡) = 𝑚(𝑡) + [𝑚(𝑡) cos(𝜔𝑐 𝑡) − 𝑚(t) cos(3𝜔𝑐 𝑡) + 𝑚(𝑡) cos(5𝜔𝑐 𝑡) − ⋯ ]
2 𝜋 3 5

𝑚(𝑡)𝑐(𝑡)

Fig. 7: Modulated signal (Natural Sampling)

By this, we obtained the modulated signal in the “Natural Sampling” form.

Using “Flat-top Sampling Method”, the top of the modulated signal remains constant and equal
to the instantaneous value of the modulating signal at the start of the sampling. In other words,
the top of the modulated signal doesn’t follow the contour of the modulating signal unlike what
we saw in “Natural Sampling” in Fig. 7.
In the demodulation process, we can simply use a low pass filter to recover the message signal
from the modulated signal (notice in the above labeled equation, when using a LPF, we obtain
the first term which is the message signal). In the next section, we’ll see the difference in results
when using a 2nd order LPF and when using a 4th order LPF.

Advantages & Disadvantages

 Advantages
 PAM is relatively simpler to implement compared to more complex modulation
techniques.

 Easy to integrate with digital systems because it’s compatible with digital circuits and
systems.

 Demodulation of PAM can be done using straightforward techniques.

 Disadvantages
 It’s difficult to remove the noise, because this will affect the amplitude which carries
information.

Page | 4 Communication Systems I

University of Kufa 1st-Dec-2023

 PAM has high noise interference because the information is carried in the amplitude of the
signal, which makes it more susceptible to noise that causes variations in the amplitude.

 It requires high bandwidth for transmission because it’s related to the pulse duration.

Applications
PAM has several applications, some of them are:
 Used in Ethernet communication
 Used in MCUs to generate control signals
 Used in digital communication systems
 Useful for short-range data transmission applications

Page | 5 Communication Systems I

University of Kufa 1st-Dec-2023

Implementation of PAM in The Lab

We used LAB tools which were U-2920A, U-2920B, U-2920C, and an oscilloscope to produce
the message signals (triangular and sinusoidal) and carrier signal (a CLK with adjustable
frequency) and apply the modulation and demodulation technique we explained in preceding
sections. Images for the results on the oscilloscope and the plots on the graph paper:

Fig. 8(a): Natural Sampling.

blue: message signal
(triangular 𝑓 = 3.2𝑘𝐻𝑧)
yellow: Modulated signal
when using CLK of
𝑓 = 8𝑘𝐻𝑧 as the carrier
signal

Fig. 8(b): Natural Sampling.

blue: message signal
(triangular 𝑓 = 3.2𝑘𝐻𝑧)
yellow: Modulated signal
when using CLK of
𝑓 = 16𝑘𝐻𝑧 as the carrier
signal

Page | 6 Communication Systems I

University of Kufa 1st-Dec-2023

Fig. 8(c): Natural Sampling.

blue: message signal
(triangular 𝑓 = 3.2𝑘𝐻𝑧)
yellow: Modulated signal
when using CLK of
𝑓 = 32𝑘𝐻𝑧 as the carrier
signal

Fig. 8(d): Flat-top Sampling.

blue: message signal
(triangular 𝑓 = 3.2𝑘𝐻𝑧)
yellow: Modulated signal
when using CLK of
𝑓 = 8𝑘𝐻𝑧 as the carrier
signal

Fig. 8(e): Flat-top Sampling.

blue: message signal
(triangular 𝑓 = 3.2𝑘𝐻𝑧)
yellow: Modulated signal
when using CLK of
𝑓 = 16𝑘𝐻𝑧 as the carrier
signal

Page | 7 Communication Systems I

University of Kufa 1st-Dec-2023

Fig. 8(f): Flat-top Sampling.

blue: message signal
(triangular 𝑓 = 3.2𝑘𝐻𝑧)
yellow: Modulated signal
when using CLK of
𝑓 = 32𝑘𝐻𝑧 as the carrier
signal

Fig. 8(g): Demodulation

using 2nd order LPF
blue: message signal (Sine
𝑓 = 2𝑘𝐻𝑧)
yellow: Demodulated signal
when using CLK of
𝑓 = 2𝑘𝐻𝑧 as the carrier
signal

Fig. 8(h): Demodulation

using 2nd order LPF
blue: message signal (Sine
𝑓 = 2𝑘𝐻𝑧)
yellow: Demodulated signal
when using CLK of
𝑓 = 4𝑘𝐻𝑧 as the carrier
signal

Page | 8 Communication Systems I

University of Kufa 1st-Dec-2023

Fig. 8(i): Demodulation

using 2nd order LPF
blue: message signal (Sine
𝑓 = 2𝑘𝐻𝑧)
yellow: Demodulated signal
when using CLK of
𝑓 = 8𝑘𝐻𝑧 as the carrier
signal

Fig. 8(j): Demodulation

using 4th order LPF
blue: message signal (Sine
𝑓 = 2𝑘𝐻𝑧)
yellow: Demodulated signal
when using CLK of
𝑓 = 2𝑘𝐻𝑧 as the carrier
signal

Fig. 8(k): Demodulation

using 4th order LPF
blue: message signal (Sine
𝑓 = 2𝑘𝐻𝑧)
yellow: Demodulated signal
when using CLK of
𝑓 = 4𝑘𝐻𝑧 as the carrier
signal

Page | 9 Communication Systems I

University of Kufa 1st-Dec-2023

Fig. 8(l): Demodulation

using 4th order LPF
blue: message signal (Sine
𝑓 = 2𝑘𝐻𝑧)
yellow: Demodulated signal
when using CLK of
𝑓 = 8𝑘𝐻𝑧 as the carrier
signal

Fig. 8: Oscilloscope Results for the Experiment

Discussion
1- What is the minimum rate at which a speech signal can be sampled for the
purpose of PAM?
Ans: Speech signals frequency is in the range of (300 Hz to 3.1kHz). Therefore, the minimum
sampling frequency is double the highest frequency which equals 6.2kHz.

2- Which is better, natural sampling or flat topped sampling and why?

Ans: This depends on the situation in which we’re using PAM in it. For example, if the message
signal is periodic, then “Flat-top Sampling” is better suited for it, but if it’s aperiodic, “Natural
Sampling” will be a better choice.
Flat-top sampling isn’t sensitive to frequency variations, while Natural sampling is sensitive.
Flat-top sampling is used when the message signal is known beforehand, while Natural sampling
is used when the message signal isn’t known beforehand. Concluding that each sampling method
has its applications, advantages, and disadvantages.

3- Is PAM & Demodulation is sensitive to Noise?

Ans: Yes, of course they can be sensitive to noise, PAM is a type of amplitude modulation.
Hence, it also has the major disadvantage of AM, which is noise. Noise can cause variations in
the amplitude of the pulses, which can lead to errors in the demodulation process.

Page | 10 Communication Systems I

University of Kufa 1st-Dec-2023

4- Give an application for PAM.

Ans: PAM can be used in “Ethernet Communication”, Ethernet is a widely used local area
network (LAN) technology that uses physical wired connections to link devices. In Ethernet
communication, data is transmitted over the network in the form of frames.
Each frame contains a payload of data along with source and destination addresses, and other
control information. PAM is used to modulate the data onto the carrier signal for transmission. In
PAM, the message information is encoded in the amplitude of a series of signal pulses.
This allows the data to be transmitted over the network in a format that can be easily decoded by
the receiving device. One of the key advantages of using PAM in Ethernet communication is that
it allows for high-speed data transmission. This is because PAM can effectively use the available
bandwidth of the network, enabling the transmission of large amounts of data at high speeds.

References
[1] Modern Digital and Analog Communication Systems 5th Edition. B. P. Lathi, Zhi Ding
[2] Sampling Theorem : Statement, Waveforms, Proof and Applications (elprocus.com)
[3] Signals Sampling Theorem (tutorialspoint.com)
[4] Analog Communication - Pulse Modulation (tutorialspoint.com)
[5] Difference between Natural Sampling and Flat-Top Sampling (aspiringyouths.com)
[6] https://cnj.atu.edu.iq/wp-content/uploads/2022/11/LECTURE16.pdf
[7] https://www.scesasaram.in/wp-content/uploads/2020/04/EEE_6_ICS_6.1__Analog-Pulse-Modulation.pdf
[8] PAM & Ethernet: A Perfect Match - EDN

Page | 11 Communication Systems I