PULSE CODE MODULATION

Content:
 What is PCM?
 Advantages of PCM
 Block diagram of PCM
 Quantization and Encoding Technique
 What is PCM ?
In modern communication technology the PAM signals are
digitized before transmission. The digital signal may then be
encoded into any equivalent form desired. Systems embodying
the transmission of digitized and coded signals are commonly
called PCM system.

Advantages of PCM:
 The signal can be regularly regenerated or reshaped
during transmission
 All-digital circuitry can be used throughout the
system
 Signals can be digitally processed as desired
 Noise and interference can be minimized by
appropriate coding of the signals
 Block diagram of PCM
System
x(t) LPF Sampler P/S PCM
Quantizer Encoder
Antialiasing S/H Converter

PCM Transmitter

PCM+
Noise S/P Sampler
x(t)
Regenerator Decoder LPF
Converter S/H

PCM Receiver
Components of PCM encoder
A PCM encoder has three processes
 The analog signal is sampled.
 The sampled signal is quantized.
 The quantized values are encoded as streams of bits.
 PCM System
Code No. x(t) [V]
8 4
7 3
6 2
5 1
4 0
3 -1
2 -2
1 -3
0 -4

Natural value 2.5 3.7 3.0 1.2 -0.6 -1.6

Quantized value 3.0 4.0 3.0 1.0 -1.0 -2.0
Code No. 7 8 7 5 3 2

PCM sequence 0111 1000 0111 0101 0011 0010

 PCM Coding
Digit Binary code Gray code
Binary to gray conversion
b1 b2 b b4 g g g g
3 1 2 3 4 g1 b1
0 0 0 0 0 0 0 0 0
1 0 0 0 1 0 0 0 1
g k bk  bk  1 , k 2
2 0 0 1 0 0 0 1 1
3 0 0 1 1 0 0 1 0
. . . . . . . . .
. . . . . . . . .
7 0 1 1 1 0 1 0 0
8 1 0 0 0 1 1 0 0
9 1 0 0 1 1 1 0 1 b1  g1
. . . . . . . . .
. . . . . . . . .
bk  g k  bk  1 , k 2
15 1 1 1 1 1 0 0 0
Components of PCM decoder
 Bandwidth requirement for PCM

f1(t) 8000 rotation/sec

f2(t)
8-level
(1) Binary
(2)
f3(t) PCM
quantizer encoding

f10(t)
5 6 4
3

(1)
12.5µs
125µs 125µs
101 010 110 100
Bandwidth 1
(2)
(1) 80 KHz B
12.5µs (2) 240 KHz 
4.2µs
 Sources of corruption
The analog signal recovered from the sampled, quantized
and transmitted pulses will contain corruption from several
sources.
Sampling and Quantization effects
Quantization Noise
Quantization Saturation
Timing Jitter

Quantization Noise Quantization Saturation

 Sources of corruption
Channel Effects

Channel noise Intersymbol

interference

Intersymbol
Thermal Interference from Interference interference
noise other users from switching
process

U1 f1

U2 f2
 Quantization Noise
M: Quantization level
a: Quantization spacing = P/M
P: Maximum signal excursion
Aj  a 2
Aj a Jth Aj
level
3a/2 Aj  a 2
a/2 A P
-a/2
-3a/2
The instantaneous
voltage of the signal
will be A j   with
 a 2  a 2
 : Error voltage
 Quantization Noise
The mean squared value of  will be
a 2 a 2 2
1 2 a
  p( ) d    d 
2 2

a 2 a 2
a 12

p ( ) 1 a : is the probability density function of the

quantization error
The rms noise voltage  a 12 a 2 3
The ratio of peak signal voltage to rms noise will be

S ov N ov P (a 2 3 ) P aM
2
S
The corresponding power ratio o N o 12M
 Quantization Noise
So N o  12M 2 12n 2 m M n m

m : Number of pulses in the code group

n : Number of code levels

So N o dB 10.8  20m log10 (n)

For a binary code n=2 So N o dB 10.8  6m

Since the bandwidth is proportional to m, the number of

pulses in the code group, the output SNR increases
exponentially with bandwidth
Uniform and Nonuniform
Quantization
So N o dB 10.8  6m
 a 12 a 2 3
 Companding
The input signal is first compressed by using a nonlinear functional device
and then a linear quantizer is used. At the receiver end, the quantized
signal is expanded by a nonuniform device having an inverse characteristic
of the compression at the sending end. The process of first compressing
and then expanding is referred to as companding.
Input Output
signal signal
A/D D/A

Compression Expansion
 Companding Characteristics

µ-law compander

Used in North America µ= 0 for uniform quantizer

and Japan µ= 255 standard value
Companding Characteristics

A-law compander

Used in Europe A= 0 for uniform quantizer

µ= 100 standard value
 Differential PCM
VH
S (VH  VL ) 28
m(k-1) m(k) m(k+1) VH  VL 256 S
VL
If m(k) –m(k-1) extends over the range of  2S, then quantized levels are
needed are at 0.5S and at 1.5S. So we need only four levels and only 2
bits per sample difference are sufficient.

 In analog system the difference can transmit exactly, but in digital

(quantized) system differences are not transmitted exactly due to
quantization error.
 If the difference is larger there might be a large discrepancy between the
original signal and the quantized signal, because of the restricted
number of encoding bits provided for encoding the difference signal.
 Differential PCM
Difference Amplifier
(k )
m(t )  Q (k )
S/H Q S 0 (t )
~ (k )
m

Accumulator Predictor

PCM Transmitter
~ (k )
m
S 0 (t ) Predictor Accumulator Filter m(t )

PCM Receiver
 Delta Modulation (DM)
Delta modulation is a DPCM scheme in which the difference
signal is encoded into just a single bit. The single bit, providing
for just two possibilities, is used to increase or decrease for
estimating m ~ (k )

DM

Linear delta Adaptive Delta

Modulator Modulator
Linear Delta Modulation (LDM)
Sampler

m(t)
a>b DM
a
Comparator
+Step Staircase
a<b
Generator ~ b
m(t )
-Step
m(t) ~ (t )
m
Step size and
step height are
constant in
LDM

Ts

1 1 1 0 1 1 0 0 0
 Problem in LDM
Due to
background m(t) ~ (t )
m
level of the
signal

Initial stage
error
~ (t )
m
m(t) Due to steep
change in
slope of the
signal
Slope
Overload error
 Adaptive Delta Modulation(ADM)
A modification of LDM is called ADM in which the step size is
not kept fixed. Rather, when slope overload occurs the step
height becomes progressively larger, thereby allowing m~ (t )
to catch up with m(t) more rapidly.

Comparator
m(t ) S 0 (t ) e(k )
S/H
~ (k )
m

D/A Digital
Converter Processor

Clock
 Adaptive Delta Modulation
~ (t )
m At each active edge of the
clock waveform, generates
m(t) a step size S, which
augments or diminishes the
accumulator. The step size
is not fixed but a multiple of
kth
a basic step . S0
S0
~
e(k) = +1 if m(t) > m(t ) immediately before the kth edge
~
e(k) = -1 if m(t) < m(t ) immediately before the kth edge
At any sampling time k, the step size S(k) is to be given by
S (k )  S (k  1) e(k )  S 0 e(k  1)
 Assignment
 A sampler is used to sample signals from four input channels
having frequencies 1KHz, 2KHz, 3KHz, and 4KHz and then the
signals are time multiplexed in PAM form. Calculate the minimum
Nyquist sampling rate. Also calculate the total number of samples
in cycle for each channel for the same sampling rate.

 Calculate the number of bits in a PCM frame, frame rate, and bit
rate when the PAM signal is encoded by a 256-encoder. In this
case use minimum sampling frequency 10KHz. For roll no.
1503001, the sampling frequency will be 10+1=11KHz and so on.

 Also graphically show the TDM form of PAM and PCM signal. Use
colors to identify signal from each channel.