3-1
PAM – PULSE AMPLITUDE MODULATION CONCEPTS
PCM – PULSE CODE MODULATION CONCEPTS
3-2
PAM Encoding*
3-3
ANALOG TO DIGITAL CONVERSION
Sampling (pencuplikan)
Quantization (kuantisasi)
Digital signal
Coding (pengkodean)
Analog signal
Discrete-time signal
Quantized signal
3-4
Sampling (pencuplikan)
Sinyal waktu kontinu sinyal waktu diskrit
T = sampling interval
Fs = sampling rate (sampel/detik)
3-5
x a ( t ) A cos(2Ft )
x a (nT ) A cos(2FnT )
2nF
A cos
Fs
F
x (n ) A cos( 2 f n ) f
Fs
1 Fs 1
f max Fmax
2 2 2T
Fs
F ?
2
3-6
x1 ( t ) cos[2(10) t ] F1 10 Hz
x 2 ( t ) cos[2(50) t ] F2 50 Hz
Fs 40 Hz
10
x1 (n ) cos[2 n ] cos( n )
40 2
50 5
x 2 (n ) cos[2 n ] cos( n )
40 2
cos(2 )n cos(2n n ) cos( n ) x1 (n )
2 2 2
x2(n) identik dengan x1(n) F2 (50 Hz) = alias dari F1(10 Hz)
x a ( t ) A cos(2Fk t )
Fk Fo kFs k 1, 2,
x (n ) x a (nT ) A cos(2Fk nT )
Fo kFs
x (n ) A cos 2 n
Fs
x (n ) A cos(2f o n 2k )
x (n ) A cos(2f o n )
Alias dari Fo
3-8
Contoh Soal 1.1
Diketahui sebuah sinyal analog
xa(t) = 3 cos 100t
a) Tentukan Fs minimum
b) Bila Fs = 200 Hz, tentukan x(n)
c) Bila Fs = 75 Hz, tentukan x(n)
d) Berapa 0 < F < Fs/2 yang menghasilkan x(n) sama dengan c)
Jawab:
a) F = 50 Hz Fs minimum = 100 Hz
100
b) x (n ) 3 cos n 3 cos n
200 2
3-9
100 4
c) x (n ) 3 cos n 3 cos n
75 3
2 2
3 cos(2 )n 3 cos( )n
3 3
2 1 1
d) x(n) 3 cos( )n 3 cos(2 )n f
3 3 3
Fo 1
f Fo f Fs (75) 25 Hz
Fs 3
2 Fmaks 2 B FN Frekuensi
Nyquist 3-11
Contoh Soal 1.2
Diketahui sebuah sinyal analog
xa(t) = 3 cos (2000 t) + 5sin(6000 t) + 10 cos (12000 t)
a) Tentukan frekuensi Nyquistnya
b) Bila Fs = 5000 Hz, tentukan x(n)
Jawab:
a) F1 1 kHz F2 3 kHz F3 6 kHz
B Fmaks 6 kHz FN 2 B 12 kHz
3-12
Fs
b) Fs 5 kHz 2,5 kHz
2
2000 6000 12000
x(n) 3 cos n 5 sin n 10 cos n
5000 5000 5000
1 3 6
3 cos(2 )n 5 sin(2 )n 10 cos(2 )n
5 5 5
1 2 1
x(n) 3 cos[2 ( )n] 5 sin[2 (1 )n] 10 cos[2 (1 )n]
5 5 5
1 2 1
x(n) 3 cos[2 ( )n] 5 sin[2 ( )n] 10 cos[2 ( )n]
5 5 5
1 2 1
x(n) 3 cos[2 ( )n] 5 sin[2 ( )n] 10 cos[2 ( )n]
5 5 5
3-13
1 2 1
x(n) 3 cos[2 ( )n] 5 sin[2 ( )n] 10 cos[2 ( )n]
5 5 5
1 2
x(n) 13 cos[2 ( )n] 5 sin[2 ( )n]
5 5
3-14
Contoh Soal 1.3
Diketahui sebuah sinyal analog
xa(t) = 3 cos (50 t) + 10 sin(300 t) - cos (100 t)
a) Tentukan laju pencuplikan minimum yang dibutuhkan untuk menghindari
pengaliasan
b) Bila sinyal tersebut dicuplik dengan laju 100 pencuplikan/sekon, berapa sinyal
waktu diskrit yang diperoleh sesudah pencuplikan
c) Bila sinyal tersebut dicuplik dengan laju 200 pencuplikan/sekon, berapa sinyal
waktu diskrit yang diperoleh sesudah pencuplikan
Jawab:
a) F1 25 Hz F2 150 Hz F3 50 Hz
3-18
0,9t t0
xa (t ) FS 1 Hz T 1 s
0 t0
0,9 n
n0
x ( n)
0 n0
3-19
n x(n) xq(n) xq(n) eq(n)
(Truncation) (Rounding) (Rounding)
0 1 1,0 1,0 0,0
1 0.9 0,9 0,9 0,0
2 0.81 0,8 0,8 - 0,01
3 0,729 0,7 0,7 - 0,029
4 0,6561 0,6 0,7 0,0439
5 0,59049 0,5 0,6 0,00951
6 0,5311441 0,5 0,5 - 0,031441
7 0,4782969 0,4 0,5 0,0217071
8 0,43046721 0,4 0,4 - 0,03046721
9 0,387420489 0,3 0,4 0,012579511
3-20
L = level kuantisasi L = 11
= Quantization step = 0,1
xmaks xmin 1 0
0,1 eq (n)
L 1 11 1 2 2
3-21
Kuantisasi sinyal sinusoidal
x(n) A cos( 0t )
3-22
FS 2 B eq (t ) xa (t ) xq (t )
xa(t) dianggap linier diantara level-level kuantisasi
= waktu selama xa(t) berada di dalam level
kuantisasi
1 1 2
Pq e (t )dt 0 eq (t )dt
2
Error power q
(rms) 2
3-23
2
1 2 2
eq (t ) t Pq t dt
2 0 2 2
b = jumlah bit L = 2b + 1
Xmaks-xmin = 2A
2
2A A
b Pq 2b
2 3(2 )
Tp 2
1 A
0 A cos ot dt 2
2
Px
Tp
Px 3 2b
Signal-to-quantization SQNR (2 )
ratio Pq 2 3-24
SQNR (dB) 10 log SQNR 1,76 6,02 b
Word length (jumlah bit) ditambah satu
Level kuantisasi menjadi dua kali lipat
SQNR bertambah 6 dB
Contoh :
Compact disk player
Sampling frequency 44,1 kHz
16-bit sample resolution
SQNR =96 dB
3-25
Coding of Quantized Samples
3-26
Contoh Soal 1.4 :
Diketahui sinyal waktu diskrit : x(n) 6,35 cos( ) n
10
Tentukan jumlah bit yang diperlukan oleh A/D
converter agar resolusinya :
a) = 0,1
b) = 0,02
Jawab:
a) x(n) maksimum pada cos( )n 1 n 0
saat :
10
x(n) minimum pada cos( )n 1 n 10
saat : 10
3-27
xmaks xmin xmaks xmin
L 1
L 1
[6,35(1) 6,35(1)]
0,1 L 1 128
0,1
2b 128 b 7 bit
b)
[6,35(1) 6,35(1)]
0,02 L 1 636
0,02
2b 636 b 10 bit
3-28
Contoh Soal 1.5 :
Diketahui sinyal seismik analog dengan dynamic range
sebesar 1 Volt. Bila sinyal analog ini dicuplik dengan
frekuensi sebesar 20 sample/s menggunakan 8-bit A/D
converter,
Tentukan :
a)Bit rate (bps)
b)Resolusi
c) Frekuensi sinyal maksimum yang ada pada
digital seismic signal
Jawab:
8 bit 20 sample
a) bps 160 bit / s
sample s 3-29
Dynamic range = xmaks - xmin
FS 20
c) Fmaks 10 Hz
2 2
3-30
Contoh Soal 1.5 :
Suatu jaringan komunikasi digital akan digunakan untuk
mentransmisikan sinyal analog :
xa (t ) 3 cos(600t ) 2 cos(1800t )
bps 10000
FS 1000 Hz
b 10
FS
FD 500 Hz
2
b) xa (t ) 3 cos(2 300t ) 2 cos(2 900t )
F1 300 Hz F2 900 Hz
FN 2 Fmaks 2 F2 2(900) 1800 Hz
3-32
300 900
c) x(n) 3 cos(2 n) 2 cos(2 n)
1000 1000
3 cos[2 (0,3)n] 2 cos[2 (0,9)n]
3 cos[2 (0,3)n] 2 cos[2 (1 0,1)n)]
3 cos[2 (0,3)n] 2 cos[2 (0,1)n)]
3-33
3-34
PAM Encoding to PCM Encoding - The D/A process
•When PAM samples are digitized by quantizing and encoding into numbers,
the result in Pulse Code Modulation (PCM).
•Since all possible values can not be encoded digitally, a QUANTIZING
ERROR is always the result of converting PAM to PCM. This error can not be
corrected.
•Samples above become 011 001 110 001 011 110 100 as data to be saved or
transmitted. (Note spaces are shown here only for clarity as real schemes
need to identify “words” of data some other way.)
3-35
What is the quantizing and coding scheme used for music
CDs?
•NUMBER OF CHANNELS 2
•FREQUENCY RANGE 20 Hz to 20,000 Hz
•DYNAMIC RANGE 90 dB
•SIGNAL TO NOISE RATIO 90 dB
•HARMONIC DISTORTION < 0.05%
•SAMPLING FREQUENCY 44,100 samples per second
•QUANTIZING 16 bits/sample
•ENCODING Linear
•CHANNELS 2
•RAW BIT RATE 1.4112 Mbps
•BIT RATE WITH OVERHEAD BITS 4.3218 (includes FEC plus add’l o/h)
•PHYSICAL SIZE 120 mm (4.75 inches) diameter
•ROTATIONAL SPEED 500 – 200 rpm (reads from center out
at a constant LINEAR speed)
•MAXIMUM PLAYING TIME 75 minutes
3-36
COMPANDING
•Consider voice analog transmission, such as in an old fashioned local telephone call.
•In the "local loop" (the connection from the local switching office to your location)
random noise can be picked up by the twisted pair wires.
•This noise--which sounds like a hiss in the phone--is most annoying as you would
imagine when you are listening to a soft spoken person.
•One solution to this problem is to use a technique called "companding" which increases
the soft levels of voice before transmission so that the average signal to noise ratio of
the voice signal is higher.
•This is because the average SIGNAL on local loop is higher, but the noise is of course
the same.
•When the signal is received, the circuit reduces the entire received signal to the correct
average, thus reducing the apparent signal to noise ratio.
•The hiss sounds like it's less.
•Success!
3-37
PROSES PENGKODEAN (ENCODING)
t t
T
Encod
Contoh di atas menunjukkan proses encoding,
1 simbol masukan dikodekan menjadi 8 bit
Jumlah bit untuk mengkodekan tiap simbol ditentukan oleh
perangkat ADC (Analog to Digital Converter)
3-38
Encoding
V 111
110
101
M 2N
100
0
011
010
001
-V 000
11111010001101010010111
t
3-39
Bentuk gelombang/sinyal PCM
+V
1 0 1 1 0 +V
1 0 1 1 0
NRZ-L -V Manchester -V
+V +V
Unipolar-RZ Miller
0 -V
+V +V
Bipolar-RZ 0 Dicode NRZ 0
-V -V
0 T 2T 3T 4T 5T 0 T 2T 3T 4T 5T
3-40
Spectrum sinyal PCM
3-41
BIT RATE KANAL VOICE
3-42
Quantization Error
3-43
Quantization Error
2 1
v v
0
v
v 2 v v
2 2 2
v v
e t f t f Q t e t
2 2
3-44
Signal to Noise Ratio
SignalPowe r
SNR Q
ErrorSigna lPower
1
The average power P lim f 2 t dt
T T
T 2
1
e t
2
e 2
t dt Time average Noise
T T 2
v 2 p v dv
1 Continuous Random Variable
v v
v v time
2 2
3-45
Signal to Noise Ratio[1]
v 2 v 2
1 1 v 3
e t v
2 2
dv
v 2
v v 3 v 2
1 v 3 v 3
3v 8 8
v 2
12
f 2t 12 2 2V
SNR Q 2 f t ; v
e t v
2 M
12
2
M f t
2 2
4V
3M 2 f 2 t 3M 2 where
V2
PeakPower
f 2 t AvgPower
V 2
3-46
Signal to Noise Ratio[2]
3M 2
SNR Q
In dB SNR Q 10 log10 3 20 log10 M 10 log10 (dB)
4.77 20 log10 M 10 log10 (dB)
M 2N
log10 a
log b a N log 2 M No.of level
log10 b
No.of binary digit
per code word
3-47
QUANTISER NON-UNIFORM
tegangan keluaran
(volt)
B
tegangan masukan
A (volt)
Uniform
Compressor
Quantizer
3-48
QUANTISER NON-UNIFORM
- law log(1 μ m )
v where > 0
log(1 μ )
- if = 0 Uniform
Quantizer
Standard
Amerika Utara
3-49
QUANTISER NON-UNIFORM
A - law Am 1
, 0 m
1 log A A
v
1 log( A m ) 1
, m 1
1 log A A
- A = 1 Uniform Quantizer
- Practical value of A A 100
- Reciprocal slope
Standard Eropa
(digunakan di Indonesia)
3-50
NON-LINEAR ENCODING ALLOWS FOR MORE LEVELS TO
BE ENCODED FOR SMALLER SIGNALS, THUS IMPROVING
THE S/N RATIO
3-51
QUANTISER NON-UNIFORM
- law log(1 μ m )
v where > 0
log(1 μ )
- if = 0 Uniform
Quantizer
Standard
Amerika Utara
52
3-52
QUANTISER NON-UNIFORM
A - law Am 1
, 0 m
1 log A A
v
1 log( A m ) 1
, m 1
1 log A A
- A = 1 Uniform Quantizer
- Practical value of A A 100
- Reciprocal slope
Standard Eropa
(digunakan di Indonesia)
53
3-53
Multiplexing TDM
Multiplexing merupakan proses penggabungan beberapa kanal
sinyal informasi kedalam satu kanal informasi dengan tujuan agar sinyal
informasi dapat dikirimkan secara simultan dalam satu kanal
Time Division Multiplexing merupakan proses multiplexing
dengan cara membagi waktu menjadi slot-slot waktu yang
menyatakan informasi dari tiap kanal
TDM – PCM (Time Division Multiplexing – Pulse Code Modulation)
merupakan proses multiplexing sinyal yang menggunakan teknik
pengkodean PCM
54 3-54
Multiplexing TDM ( cont )
8 bit d1(t)
Analog
Information-1 ADC
proses
d2(t)
Analog 8 bit
Information-2 ADC
Multi-
plexing
Analog 8 bit dN(t) X(t) : Sinyal multiplex
Information-m ADC TDM
Snk(t)
Sinyal
Sinkronisasi
55
3-55
PCM-30 (E-1, Standar Eropa)
1 - 15 dan 17 - 30 adalah sinyal
telephon yang dikodekan/ data digital
0 1 2 15 16 17 29 30 31
8
bit
Informasi signaling
Frame alignment signal x = bit reserved for international use
pada frame 1, 3, 5, dst Y = bit reserved for national use
x 0 0 1 1 0 1 1
A = 0 tidak ada alarm
A = 1 ada alarm
bergantian
1 TS = 8 bit
Terdiri dari 32 TS = 30 kanal suara + 1 sinkronisasi + 1 signaling
Sinkronisasi : TS 0
Signaling : TS 16
Voice : TS 1 – 15 + TS 17 – 31
Dalam 1 detik tdp 8000 sample, sehingga :
Bit rate = (8 x 8000 ) x 32 = 2048 kbps
56
3-56
Multiframe PCM-30
ms
s
0 1 16 31 0 1 16 31 0 1 16 31
0 0 0 0 1 A U 1 a b c d a b c d a b c d a b c d
1 MF = 16 frame
Signaling lengkap untuk 30 kanal voice (1 TS 16 untuk signaling 2
kanal voice)
TS-16 untuk frame ke-0 digunakan untuk alignment / sinkronisasi
multiframe
57
3-57
PCM-24 (T-1, Standar Amerika)
T1 (DS-0) System
24 voice channels are time-division multiplexed
Each voice signal is sampled at a rate of 8000
samples/sec. (sample duration = 125 sec)
Each sample is quantized in amplitude into one of 256
levels (8 bits are used to represent each level)
T1 rate = (24*8 + 1)/125 sec = 1.544 Mbps
58
3-58
PCM-24 (T-1, Standar Amerika)
Multi 1 2 3 4 5 6 7 8 9 10 11 12
Frame
1 TS = 8 bit
Terdiri dari 24 TS = 24 kanal suara Dalam 1 detik tdp 8000 sample
Sinkronisasi menggunakan 1 bit tambahan (=BF)
Signaling diambil pada bit ke-8 tiap TS pada frame ke-6 dan kelipatannya
Bit Rate = ((24 x 8) + 1) x 8000 = 193 x 8000 = 1544 kbps
1 MF = 12 frame
59
3-59
Perbandingan 3 standar
(Amerika, Eropa, Jepang)
Eropa
Amerika Jepang
Level Utara
Bit Rate (Mbps)
1 2.048 1.544 1544
1C - 3.152 -
2 8.448 6.312 6.312
3 34.368 44.736 32.064
4 139.264 274.176 97.728
5 564.992 400.352
60
3-60
Encoder (Konverter) Line coding
-output ADC
- sinyal TDM Encoder Saluran
Line Coding
-Sinyal data text Kabel
-Output scrambler
-Output FEC Sinyal base-band
dijital
61
3-61
Delta (Differential) Modulation