EECE 3500

Communication Systems

Topic: Quantization Process

and Pulse Code Modulation

Dr. Hong

1
© Dr. Liang Hong
 Introduction
• Why quantization:
• Continuous signal has a continuous range of
amplitudes. i.e., within the finite amplitude range of the
signal, we find an infinite number of amplitude levels.
• It is not necessary to transmit the exact amplitude of the
samples.
• Definition:
Quantization is the process of transforming the sample
amplitude of a message signal into a discrete amplitude
taking from a finite set of possible amplitudes.
Discrete value signal

Sampling + quantization Digital signal 2

© Dr. Liang Hong
 • Memoryless quantizer:

• Uniform memoryless quantizer:

(a) midtread (b) midrise. 3

© Dr. Liang Hong
 • Example: Textbook page 226, Prob. 5.15 (a)

4
© Dr. Liang Hong
 • Example
Original message
signal

Pulse amplitude
modulated signal

Quantized samples

5
© Dr. Liang Hong
 Quantization Noise
• Definition:
The difference between the input sample m and output
sample v.

6
© Dr. Liang Hong
 Pulse Code Modulation
• Definition:
a process that represent a message signal by a sequence
of coded pulse.

• Basic operations of PCM transmitter:

sampling + quantization + encoding
Analog-to-digital conversion

7
© Dr. Liang Hong
 1. Sampling: f s  2W

2. Quantization:
In telephonic communication, nonuniform quantizer is
used
why
• Protect weak passages
• Large amplitude range occur relatively infrequently
• SNR is relatively insensitive to the input level
8
© Dr. Liang Hong
 • -law: A-law:
 A| m| 1
log(1   | m |)  1  log A 0 | m |
| v | A
log(1   ) | v | 
1  log( A | m |) 1
 | m | 1
 1  log A A

US, Canada, Japan,  = 255 Europe, A=87.6 9

© Dr. Liang Hong
 3. Encoding:
• translate the discrete set of sample values to a more
appropriate form of signal.
• binary code (R bit, 2R quantization level)

Ordinal number of Level number expressed Binary

representation level as sum of powers of 2 code
0 000
1 20 001
2 21 010
3 21  20 011
4 22 100
5 22  20 101
6 22  21 110
7 22  21  20 111 10
© Dr. Liang Hong
 • Line codes:
electrical representations of binary data

(a) Unipolar NRZ

(b) Polar NRZ

(c) Unipolar RZ

(d) Bipolar RZ

(e) Manchester code.

11
© Dr. Liang Hong
 • Example (continued from page 5):

Quantized samples

Tb

Unipolar
A 3-bit PCM
NRZ
signaling t
0 2 3 4 5 6 7 8 T
1
Binary data 010 011 110 111 101 100 001
12
© Dr. Liang Hong
 • Example:
A signal with 5 kHz maximum frequency is transmitted
using PCM with eight levels of quantization and binary
code. What is the minimum bit rate required?

13
© Dr. Liang Hong
 • Signal-to-Quantization-Noise ratio:
• The use of quantization always introduce error (noise)
• Quantization noise = The difference between the input
sample m and output sample v.

14
© Dr. Liang Hong
 Consider a PCM system with R bits per sample in the
construction of the binary code, then the number of
quantization representation levels is
L  2R

Consider an input m of continuous amplitude in the range

(-mmax, mmax). Also assume a uniform quantizer of the
midrise is used, the average quantization noise power is
1 2
  mmax  22 R
2
Q
3
Let P denote the average power of the message signal

P 3P 2 R
( SQNR)   2
 Q2 2
mmax

15
© Dr. Liang Hong
 • Example:
Consider the special case of a full–load sinusoidal
modulating signal of amplitude Am, which utilizes all the
representation levels provided. Find the signal-to-
quantization-noise ratio.

3 2R
( SQNR)  2
2
Expressing in decibels,

10 log10 (SQNR)  1.8  6R

16
© Dr. Liang Hong
 • Example:
Consider the an audio signal comprised of the
sinusoidal term.
s(t )  3 cos 500 t

(a) Find the SQNR when the audio signal is quantized

using 10-bit PCM.
(b) How many bits of quantization are needed to
achieve a SQNR of at least 40 dB?

17
© Dr. Liang Hong
 • Basic operation of PCM receiver:

Regeneration + decoder + reconstruction filter

1. Regeneration circuit:

18
© Dr. Liang Hong
 2. Decoder:
• regroup the clean pulses into code words and map
back into a quantized PAM signal
• Generate a pulse with amplitude that is the linear
sum of all the pulses in the code word, with each
pulse being weighted by its place value (20, 21, …,
2R-1) in the code.

3. Reconstruction filter (low pass filter):

Higher order interpolation Zero order hold

First order hold

19
© Dr. Liang Hong
 • Advantages of PCM:
1.Robustness to channel noise and interference
2.Efficient regeneration of the coded signal along the
transmission path

Digital system

Analog system

20
© Dr. Liang Hong
 3. Efficient exchange of increased channel bandwidth for
improved signal-to-noise ratio
4. A uniform format for the transmission of different kinds
of basedband signal, hence their integration with
other form of digital data in a common network
5. Comparative ease with which message sources may
be dropped or reinserted in a time-division multiplex
system
6. Secure communication through the use of special
modulation schemes or encryption

• Disadvantages:
1.High system complexity
2.Require higher channel bandwidth
21
© Dr. Liang Hong