Are you sure?
This action might not be possible to undo. Are you sure you want to continue?
Digital modulation techniques
Knowledge empowers man to use technology wisely.
4.1
INTRODUCTION
The techniques used to modulate digital information so that it can be transmitted via microwave, satellite or down a cable pair are different to that of analogue transmission. However, it must be remembered that when the data is transmitted via satellite or microwave, the radio frequency (RF) stage is an analogue stage and that the data is transmitted as an analogue signal. This means that the techniques used to transmit analogue signals are used to transmit digital signals. The problem is to convert the digital signal to a form that can be treated as an analogue signal. This chapter deals with the different modulation techniques that are used to convert a digital signal into an analogue signal that is then in the appropriate form to either be transmitted down a twisted cable pair or to be applied to the RF stage where it is modulated to a frequency that can be transmitted via microwave or satellite. Although digital transmission is on the increase worldwide, many of the existing circuits are still analogue. This means that the digital signals must be changed into an analogue format to enable them to be transmitted over the existing metallic pairs that form the link. The equipment that implements this conversion for a personal computer (PC) that is used to communicate with other PCs via the public switched telephone network (PSTN) is a modem. The term modem is made up from the words 'modulator' and 'demodulator'. A modem accepts a serial data stream from the PC and converts it into an analogue format that matches the transmission medium. A typical system arrangement is as shown in Fig. 4.1. As can be seen in the figure the data terminal equipment (DTE) is the point at which the data originates and also the end point for received data. The data circuit equipment (DCE) comprises the modem. The connection between the DTE and DCE is via an interface such as the RS232C. The modem is connected to the PSTN by means of a suitable circuit such as a cable pair. There are many different modulation techniques that can be utilised in a modem. Some of these techniques are:
60
Digital modulation techniques
(application ! programs) ~N~ Control [ Line control ]
,ation
UART ~
Interface /
\
Modem
(Originate)
Personal i
Printer computer or terminal
machine Printer
/"
~ Personal ~
Fax machine computer or terminal Telephone
line
/
C,,....
line
,
] .l._[..
.'"
DCE
'
i,
RS232C
Interface
i,

DTE
 o.,ro.er
[
St,~tion
I,'o'A
Modem
UART
(Answer)
t
DCE: Data circuit equipment DTE: Data terminal equipment PSTN: Public switched telephone network
Fig. 4.1
v i station ., . , )M (application t..onlro , .. Vi programs~
~
Lowspeed data link
9 9 9 9 9
Amplitude shift key modulation (ASK). Frequency shift key modulation (FSK). Binaryphase shift key modulation (BPSK). Quadraturephase shift key modulation (QPSK). Quadrature amplitude modulation (QAM).
4.2
AMPLITUDE SHIFT KEY MODULATION
In this method the amplitude of the carrier assumes one of two amplitudes dependent on the logic states of the input bit stream. Unlike an amplitude modulator used for modulating analogue signals, the carrier frequency need not be totally suppressed. The typical output waveform of a ring modulator is shown in Fig. 4.2. The frequency components are the USB and LSB with a small residual carrier frequency. The carrier suppression resistor is omitted from this modulator so that when a logic 0 is sent, a lowamplitude carrier frequency is allowed to be transmitted. This ensures that at the receiver the logic 1 and logic 0 conditions can be recognised uniquely.
Frequency shift key modulation
+2 +1 0
61
Logic 1
Logic 0 Modulating frequency
Logic
1
+3 +2 +1 0 1 2 3 0
I
VVVJV VVV VV
Carrier frequency
+6
I
+1
I
+2
I
+3
I
+4
I
+5
I
+7
I
+8
I
+9
I
+10
I
+11
I
+12
.,,
I
+2+3+1
,4"%Z"%
Output frequency
Fig. 4.2 Amplitude shift key modulation
A typical single sideband modulator is shown in Fig. 4.3(a). The carrier frequency is used to either forward bias or reverse bias the two diodes. When the diodes are forward biased then the digital signals are passed to the output. When the diodes are reverse biased then the digital signal is prevented from reaching the output and only a residual carrier frequency appears at the output. A typical Cowan modulator is shown in Fig. 4.3(b). In this case, when the diodes are forward biased by the carrier frequency the input is short circuited and the input digital signal is prevented from reaching the output. When the carrier frequency reverse biases the diodes then the digital signal is passed to the output. Figure 4.3(c) shows a ring modulator. The phase of the carrier frequency is used to forward bias two of the diodes and reverse bias the other two diodes.
4.3
FREQUENCY SHIFT KEY MODULATION
In this method the frequency of the carrier is changed to two different frequencies depending on the logic state of the input bit stream. The typical output waveform
62
Digital modulation techniques
Data input
Output
(a)
fc
I
Data input
Output
f~
(b)
Data input
~
Output
I11
(c)
Fig. 4.3
f~
(a) Single balanced modulator; (b) Cowan modulator; (c) ring modulator
is shown in Fig. 4.4. The carrier frequency is more conveniently referred to as the centre frequency. Notice that a logic high causes the centre frequency to increase to a maximum and a logic low causes the centre frequency to decrease to a minimum. A typical FSK transmitter and receiver is shown in Fig. 4.5. The circuits shown use phaselocked loop circuits.
The transmitter The input digital signal is applied to a digitaltoanalogue (D/A) converter. The resultant analogue signal is summed with the control voltage produced by the phase comparator. The resultant control voltage is used to change the frequency of the voltagecontrolled oscillator (VCO). The output of the VCO is the FM output which is applied to the input of the next stage. Some of the output signal from the VCO is divided and this frequency is phasecompared with a frequency produced by the reference oscillator. The resultant output of the phase comparator produces a control voltage. The greater the
Frequency shift key modulation 63
+2 +1 0 Modulating frequency +3 +2 +1 0 1 2 3 0
I
Logic l
Logic 0
I
Logic I
[
VVVVVVVVV /V
Carrier frequency +I
I
+2
I
+3
I
+4
I
+5
I
+6
I
+7
!
+8
I
+9
I
+10 +11 +12
, ,
I
I
i
Output frequency Fig. 4.4 Frequency shift key modulation
difference between the two input frequencies the larger the control voltage. The control voltage will be either positive or negative, depending on whether the output frequency of the VCO is higher or lower than the reference frequency at the input to the phase comparator. The divider incorporated into the circuit enables the output frequency of the VCO to be much larger than the reference frequency. If there is no digital input then the output of the VCO is the cartier frequency or centre frequency. Hence by changing the control voltage the centre frequency is either increased to a maximum or decreased to a minimum.
The receiver The FM signal is applied to the receiver input. This input signal is phasecompared with the output of the VCO. The output of the phase comparator produces a d.c. control voltage, which contains the information relating to the original digital signal. The control voltage is used to change the frequency of the VCO. In this way the control voltage is changing in accordance with the original digital signal. The control voltage is then applied to an analoguetodigital (A/D) converter, which produces the original digital signal.
64
Digital modulation techniques
Phase comparator Reference frequency

i@i
Summer d.c. control voltage ,d
Divider
FM ~
l
I Vco ~J t
Data input
I
output (a)
Phase comparator d.c. control voltage . ~[A/~
VDI
Data output
(b)
Fig. 4.5 (a) FSK transmitter using a VCO; (b) FSK receiver using a VCO
4.3.1 Frequency modulationgeneralities
The modulation index is given by m=
/xf L
When the modulating frequency fa is at a maximum and the frequency deviation A f is at a maximum then the modulation index is referred to as the deviation ratio (D). The deviation ratio is determined by max A f O = ~ max fa A generalised formula to determine the required bandwidth is B = 2(Af +fa) Typical frequency allocation for lowspeed modems is given in Table 4.1. For an FSK modulator the deviation ratio is determined by: D= The bandwidth is given by B = tone separation + bit rate tone separation bit rate
Frequency shift key modulation
Table 4.1 Frequency allocation for a lowspeed modem Bit rate (b/s) Direction Frequencies ( H z ) Logic 0 <300 600 1200 AB BA AB BA AB BA 1180 1850 1700 2100 Logic 1 980 1650 1300 1300 Nominal frequency (Hz) 1080 1710 1500 1700
65
Example 4.1
Calculate the deviation ratio and bandwidth required to transmit a 1200 b/s digital signal. tone separation bit rate Tone separation: from Table 4.1, tone separation = 2100  1300 = 800 Hz
D _~
tone separation bit rate 800 1200
=0.67 The bandwidth is given by B = tone separation + bit rate = 800 + 1200 = 2000 kHz If the bandwidth is limited to the bit rate then the receiver will still be able to recover the original data. The tone difference is 800 Hz which is smaller than the 2 kHz given above. As the deviation ratio is approximately equal to the energy concentrated in the carrier and the firstorder sidebands it is not necessary to transmit all the sidebands. The receiver need only recover the original logic states from the received signal.
66
Digital modulation techniques
4.4
PHASE SHIFT KEY MODULATION
With this method the phase of the carrier changes between different phases determined by the logic states of the input bit stream. There are several different types of phase shift key (PSK) modulators: 9 Twophase (2 PSK). 9 Fourphase (4 PSK). 9 Eightphase (8 PSK). 9 Sixteenphase (16 PSK). 9 Sixteenquadrature amplitude (16 QAM). The 16 QAM is a composite modulator consisting of amplitude modulation and phase modulation. However, it is more closely associated with phase modulation than amplitude modulation. The 2 PSK, 4 PSK, 8 PSK and 16 PSK modulators are generally referred to as binary phase shift key (BPSK) modulators, whereas the QAM modulators are generally referred to as quadrature phase shift key (QPSK) modulators.
4.4.1
Some phase shift key generalities
D symbols/s n
The symbol rate S of a binary PSK modulator can be determined by S =where
D = data rate (b/s) n = number of bits per symbol
The total number of possible symbols at the output is given by M = 2 ~ symbols where M = number of possible symbols at the output n = number of bits per symbol
The phase difference between each symbol is given by P = 360 ~ _ 27r rad M M The maximum input bit rate C can be determined by C = B log2 M where B = bandwidth M = maximum possible number of symbols at the output
4.4.2
Twophase shift key modulation
In this modulator the carder assumes one of two phases. A logic 1 produces no phase change and a logic 0 produces a 180 ~ phase change. The output waveform for this modulator is shown in Fig. 4.6.
Phase shift key modulation
67
+2 +1 Logicl .... I Logic0 ! Logicl l t 0 Modulating frequency +3 +2 +1 0 1 2 V ~ V ~ ~C ~arrierfre~quency ~ ~ ~ ~ V 3
t +1 +2 +3 +4 +5 +6 +7 +8 +9 +10 +11 +12
Output frequency
Fig. 4.6 Phase shift key modulation
A ring modulator (Fig. 4.3(c)) can be used for this purpose. The only difference between the ASK modulator and the 2 PSK modulator is where the carder frequency and the digital modulating signal are applied. The carrier frequency and digital modulating signal are swapped around for the 2 PSK modulator so that the digital signal is used to forward bias and reverse bias the diodes.
Example 4.2 A 2 PSK modulator has an input bit rate of 2400 b/s and works into a commercial speech band circuit. Determine: 1. 2. 3. 4. The The The The number of possible symbols at the output. symbol rate. phase difference between the symbols. maximum bit rate.
68
Digital modulation techniques
Solution
1. Each input bit causes a phase change of either 0 ~ or 180 ~ Hence n = 1 and
M = 2 n
=2
t l
I
/
=2
2.
S =
D
n
symools/s
= 2400 symbols/s i.e. the symbol rate is the same as the bit rate. 3. P= = 360 ~ M 360 ~ 2 = 180 ~
This shows that there are only two output vectors for the output data. 4. The bandwidth is 300 Hz to 3400 Hz (B = 3100 Hz). Hence C = B log2 M = 3100 log2 2 = 3100 b/s This means that the maximum input bit rate that can be applied to the circuit is 3100 b/s. If a higher bit rate was applied to the input then the received data would be errored.
4.4.3
Fourphase shift key modulation
With 4 PSK, 2 bits are processed to produce a single phase change. In this case each symbol consists of 2 bits, which are referred to as a dibit.
Example 4.3
A 4 PSK modulator has an input bit rate of 2400 b/s and works into a commercial speech band circuit. Determine: 1. 2. 3. 4. The The The The number of possible symbols at the output. symbol rate. phase difference between the symbols. maximum bit rate.
Solution
1. Two bits produce a single phase change. Hence n = 2 and M = 2 ~ 22  4
Phase shift key modulation
69
2.
S =
D symbols/s n 2400  ~  1200 symbols/s 2
In this case the output symbol rate is half of the input bit rate. 3. P = 360~ M = 360 ~ = 90 ~ 4
This shows that there are four output phases for the output data. 4. The bandwidth is 300 Hz to 3400 Hz (B  3100 Hz). Hence C = B log2 M = 3100 log2 4 = 3100 logl0 4 logl0 2 = 6200 b/s This means that the maximum input bit rate that can be input into the circuit is 6200 b/s. If a higher bit rate was applied to the input then the received data would be errored. The actual phases that are produced by the 4 PSK modulator are shown in Table 4.2. Because the output bit rate is less than the input bit rate, this results in a smaller bandwidth. A typical 4 PSK circuit is shown in Fig. 4.7(a) and the constellation is shown in Fig. 4.7(b). The input signal is applied to a 2bit splitter or a lto2 demultiplexer. The output bit rate on each of the two outputs is half of that at the input to the splitter. The two outputs are each connected to the digital inputs to the two ring modulators. The carrier frequency is applied directly to the carrier input to the top ring modulator. The carder input to the bottom ring modulator undergoes a 90 ~ phase shift. The phasor output of the top ring modulator is along the Table 4.2 Phases produced by a 4 PSK modulator Dibit 00 01 10 11 Phase change +225~ +135~ +315~ +45~ ~ ~ ~ ~
70
Digital modulation techniques
Logic 0 Voltage2
1 0 +2 V
I+Acsin Oct
O
L o g l
C
! t a g
e
Output Four phase shift key modulation constellation
1 +2
B
(~+A
c cos Oct
02
Ol
11
(a)
O0 B 10
B
9 .................. 0,1 ,, 1 T +2v
+A c cos (Oct
0 1,1
2v
+2v
A I
0 A e sin (Oct
s s s s 9 9 9 9
IA
1 +A c sin mct
0,0"
9
9
A c
cos Oct
_!_
B
""
~'~6
0
2V
(b)
Fig. 4.7
Fourphase shift key modulation" (a) circuit; (b) constellation
horizontal axis and the phasor output from the bottom ring modulator is along the vertical axis. A logic high at the input to each ring modulator produces a voltage of +2 V at the output and a logic low at the input produces an output voltage o f  2 V. The outputs from the two ring modulators are then summed together in a linear summer. The output of the linear summer produces the constellation shown in Fig. 4.7(b). 4.4.4 Eightphase shift key modulation
With this modulator 3 bits are processed to produce a single phase change. This means that each symbol consists of 3 bits.
Phase shift key modulation
71
Example 4.4
An 8 PSK modulator has an input bit rate of 2400 b/s and works into a commercial speech band circuit. Determine: 1. 2. 3. 4. The The The The number of possible symbols at the output. symbol rate. phase difference between the symbols. maximum bit rate.
So/ut/on 1. Three bits produce a single phase change. Hence n = 3 and M = 2" = 2 3 = 8 2. S =D
n
symbols/s
2400 =   ~ = 800 symbols/s In this case the output symbol rate is a third of the input bit rate. 3. P= 360 ~ M 360 ~ = = 45 ~ 8
This shows that there are eight output vectors for the output data. 4. The bandwidth is 300 Hz to 3400 Hz (B = 3100 Hz). Hence C = B log2 M = 3100 log2 8 = 3100 loglo 8 loglo 2 = 9300 b/s This means that the maximum input bit rate that can be input into the circuit is 9300 b/s. If a higher bit rate was applied to the input then the received data would be errored.
A theoretical 8 PSK modulator Figure 4.8(a) shows a typical circuit for this modulator. With this modulator bit ,4 controls the output polarity of the first digitaltoanalogue converter (DAC 1). Bit B is used to control the output polarity of the second digitaltoanalogue converter (DAC 2) and bit C is used to control the output amplitude of both the digitaltoanalogue converters (DAC 1 and DAC 2).
72
Digital modulation techniques
A
[ DAC ]
C ~, ~ + A
fc
c sin tOct
Output
Input
_7
B [DAC21
(a)
CB
I 900 +AC cos Old ~....// Mixer 2
Quad~
110
OlO ~'~
+Ac cos Oct _._~ Oil ! "ii//i
anti
~ . z . . . . . . . . . . r" +0.5V@/ . . . . 1. .~. . . . . . . . . ~
"11 l
ca
1.21V
f
\i
..O.5V~
/T\
/1 \
~ +0.5V
+I.21V
i/
+Zcs,n c,
1130 . 0 ~ . . . . . . . . . .
CBA
, 0.5V"~''X . . . . . ~. . . . . . . . . . ~
101
xxx
Quad
ant 4

"
o
000
~
~  o
001
A C cos (Oct
CB
(b)
Fig.4.8
Theoretical eightphase shift key modulator: (a) circuit; (b) constellation
The input digital signal is applied to a 3bit splitter or a 1to3 demultiplexer. This means that the output bit rate on each of the three outputs is a third of that at the input to the splitter. The data on lead A and lead B are applied to the inputs of the two digitaltoanalogue converters (DAC 1 and DAC 2). The output on lead C is applied to the second input to DAC 1 directly but is negated before being applied to the second input to DAC 2. The carrier frequency from the oscillator is connected directly to the carrier input on mixer 1 (ring modulator) and it passes through a 90 ~ phase shifter before being applied to the carrier input for mixer 2 (ring modulator). The two mixers are identical circuits. The resultant carrier frequency phases are as shown in the circuit diagram indicated in Fig. 4.8(b). As can be seen the phase of the carrier for quadrants 1 and 2 is positive, i.e. axis CB, and the phase of the carrier for quadrants 4 and 5 is negative, i.e. axis CB. The digitaltoanalogue conversion follows the conditions given in Table 4.3.
m
Phase shift key modulation Table 4.3 Digitaltoanalogue conversion conditions for 8 PSK modulator A 0
1
73
Polarity +
B 0
1
Polarity +
C 0
1
Amplitude 0.5
1.21
C 1
0
Amplitude 1.21 0.5
The conditions shown in Table 4.3 produce the positions as shown in Table 4.4 for all the different input permutations. The constellation diagram can now be drawn according to this table and is shown in Fig. 4.8(b). The operation of each mixer is the same as described in Section 4.5.2. By referring to Table 4.4 it will be seen that in each quadrant there are two phases. The outputs of the two modulators (mixer 1 and mixer 2) are applied to a linear summer. The output of the linear summer then produces the constellation as shown in Fig. 4.8(b).
A practical 8 PSK modulator
In practice the phase of the carrier frequencies is as shown in Fig. 4.9(a). The digitaltoanalogue conversion follows the conditions given in Table 4.3. The circuit is identical to that of the theoretical 8 PSK modulator except for the phasing of the carrier frequency. On the CB axis the polarity must be multiplied by a negative due to the phasing of the carrier frequency. This produces the positions shown in Table 4.5. In this case the phase of the carrier is negative for the CB axis for quadrants 1 and 2 and the phase of the carrier is positive for the CB axis for quadrants 3 and 4. The constellation diagram can now be drawn according to this table and is shown in Fig. 4.9(b). As can be seen, again there are two phases in each quadrant. The operation of the circuit is identical to that of the theoretical 8 PSK circuit.
Table 4.4 Input permutations and positions (theoretical)
..
C
1 1 1 1
C
0 0 0 0
B
0 0 1 1
A
0 1 0 1
C 0 0 0 0
1 1 1 1
.4 0 1 0 1
0 1 0 1
Polarity + +
+ +
. .
Amplitude 0.5 0.5 0.5 0.5
1.21 1.21 1.21 1.21
C' 1 1 1 1
0 0 0 0
B 0 0 1 1
0 0 1 1
Polarity + +
+ +
Amplitude 1.21 1.21 1.21 1.21 0.5
0.5 0.5
Quad
0 O 0 0
1 1 1 1
0 0 1 1
0 1 0 1
0.5
A[DA
Ck Input
+Ac sin tOct Ou!put
7
I
(a)

c
i
I
/,"~A
C c o s 0~ct
+ACcos COct
ooo
a   . j. .........
CB
OOl
m~/1 Quad
~'21V i "
t 
" ~ ~ r i
anti
CA,,',
ac sin t~ [Note[ ]CxBA ]
, , , t......~Ak ~ [ [ 1.21V 0.5VI~ / 1 \ ~+0.5V \i ~ ] /T ~ i ~ 110 ~rant~ 3 Quad "01 ACcos O)ct
CB
[
!, 0.5V1i_ ~..... ~ Q u ~ d ,  ,~ ~ 011
'.11 CA +I.21V i/ +Ac sin t~ Ill rant 4
(b)
Fig. 4.9
Practical eightphase shift key modulator: (a) circuit; (b) constellation
Table 4.5 Input permutations and positions C' C B A C A Polarity
1 1 1 1 0 0 0 0 0 0 1 1 0 1 0 1 0 0 0 0 0 1 0 1
+ +
(practical)
Amplitude C B Polarity 0.5
0.5
Amplitude Quad 1.21
1.21
1 0
1 0
(x)=+
( ( x
)=+
0.5
0.5
1
1
1
1
( x +)=
x+)=
1.21
1.21
0 0 0 0
1 0 0 1 0 1 1 1 0 1 1 1
1 0 1 1 + 1 0 1 1 +
1.21 1.21 1.21 1.21
0 0 0 0
0 0 l 1
(x)=+ ( x )=+ ( x+)=( x +)=
0.5 0.5 0.5 0.5
Phase shift key modulation
75
4.4.5
Sixteenphase shift key modulation
With this m o d u l a t o r 4 bits are processed to produce a single phase change. This means that each symbol consists of 4 bits.
Example 4.5
A 16 PSK modulator has an input bit rate of 2400 b/s and works into a commercial speech band circuit. Determine: 1. 2. 3. 4. The The The The number of possible symbols at the output. symbol rate. phase difference between the symbols. m a x i m u m bit rate.
So/m/on 1. Four bits produce a phase change. Hence n = 4 and M = 2 n = 24 = 16 2. S =D
n
symools/s = 600 symbols/s
/ ' 1 /
2400
In this case the output symbol rate is a quarter of the input bit rate.
. P ~ ~
360 ~ M
360 ~ = ~ = 22.5 ~ 16 This shows that there are 16 output vectors for the output data. 4. The bandwidth is 300 Hz to 3400 Hz (B  3100 Hz). Hence C = B log2 M = 3100 log2 16 = 3100 lOgl0 16 lOglo 2 = 12 400 b / s This means that the maximum input bit rate that can be input into the circuit is 12 400 b/s. If a higher bit rate was applied to the input then the received data would be errored.
76
Digital modulation techniques
I I I i I I I .......i.... I
"/" /
O O 0 0
eq
f /
O

9
I I I.
/
I I I I I I t
Fig. 4.10
Constellation for 16 PSK modulation
The constellation of a 16 PSK modulator is shown in Fig. 4.10. Only the constellation for this modulation scheme is shown. A better modulator is the 16 QAM, which is described in more detail below. Notice that in the constellation for the 16 PSK, each phasor is separated from the adjacent phases by 22.5 ~
4.5
SIXTEENQUADRATURE AMPLITUDE MODULATION
With this modulator, 4 bits are processed to produce a single vector. Bits A and C determine the quadrant in which the phasor will lie and bits B and D determine the position within the quadrant that the phasor will assume. Bit A is the MSB and bit D is the LSB. This is a better method than 16 PSK modulation. The complete constellation is shown in Fig. 4.11. A typical circuit is shown in Fig. 4.12. The circuit consists of four identical ring modulators. The carrier to the bottom two modulators undergoes a 90 ~ phase shift. The top two modulator outputs are summed together to produce phases on the horizontal axis through a linear summer. The outputs of the bottom two
Sixteenquadrature amplitude modulation 77
I 1 I I
O
0011 0111
+3V
9
1011
9
1111
+2V
9
0010
 O+lV0110
9
1010
9
1110
3V
2V
IV
+IV
+2V
+3V
I
9
0001
I
9 lv0101
2V
,,, 9
I
9
I
1001
1101
9
0000
9 3v0100
9
1000
9
I100
~ 1
Fig. 4.11
I
L ~ I
[
~
1
~
1
~
1
~
Constellation for 16 QAM showing the 4bit binary codes
modulators are summed together to produce the resultant phases on the vertical axis through a linear summer. The vertical and horizontal vectors are summed together to produce the 16 QAM constellation through a linear summer. The two 6 dB attenuators, or pads, are used to ensure that the output voltage is half of the input voltage. The 6 dB pads are connected to the outputs of the ring modulators for bits B and D. The following mathematical manipulations show that the 6 dB pads cause the output voltage to be half of the input voltage: N  20 loglo ~
Vout
dB
Vout  6 dB  20 logl0 Vin 0.3 = loglo V~
Vin
V~ = 100.3 = 0.5
Via
78
Digital modulation techniques
Logic Voltage
0 1 2 0 +2 01 10 11 1 +I +3
I
Logic 00 Voltage 3
co,ic0
oltage
~ ? +1 B
l I
16 quadrature amplitude modulation constellation
T
!
I
If ~"xl
'
'
6dB Pad
Y
T
"
AB
_....~ 9 0 , ~ t ~ d
CD
P[ I
I ~'LI o+2 ~ [ 1o 1 qt
[ 11g t ; 6dB Pad
i:i
L 11 o 10
ou,ou,
~~c0 g2 ~
c g2,
Fig. 4.12
VIL1 [ o+2 ~ o ! t
!Ol
00
~=]
Modulator circuit for 16 QAM
The circuit produces the phase angles in each quadrant as shown in Fig. 4.13. This figure, together with Fig. 4.14, clearly indicate the three different amplitudes. Figure 4.13 together with Fig. 4.15 indicate the 12 different phases that result. The three different amplitudes and 12 different phases together with the binary codes for each phasor are shown in Fig. 4.16. This figure shows more clearly the 4bit binary codes for each phasor. The modulator works according to the data given in Table 4.6. From this the positions in the constellation can be plotted as shown in Fig. 4.11. When the logic condition in lead A is a logic high, the output of the ring modulator produces a phase having an amplitude of +2 V. When the logic condition on lead A is a logic low then the output of the ring modulator produces a phase having an amplitude o f  2 V. The same applies to the logic conditions on leads B, C and D and the outputs from the other ring modulators. The 6 dB pad that is associated with the output of the ring modulator for lead B produces an output voltage of either + 1 V for an input voltage of +2 V or  1 V for an input voltage o f  2 V. The same applies to the 6 dB pad associated with the ring modulator for lead D.
I
I
I
i
Phase 3
se 2
+3V
9
9
+2V
+IV
I
,,
I
.....
+IV
+2V
+3V
Fig. 4.13
Phase angles in the first quadrant for 16 QAM
I I I I
I
I
i
I
~
!C
3V 2V
O
O
\
IV
+IV
I i
+2V
I
+3V
I
I
9
o,~..
2V 
_~d
"
J
I
I
I
I
I
I
I
I
Fig. 4.14
Constellation for 16 QAM showing three amplitudes
I
I
I
I
I
I
I
!
_

NN~
5
9
~hase4
9 9
~ Phase3
~ e2
9
_

 
m
.
~~
t
Phase6 ~ Phase 1
P
Phase 12
9 9 9 9 
I
I
I
I
I
I
I
I
Fig. 4.15
Constellation for 16 QAM showing 12 phases
i
I
I
I
I
I
i
i
_

Quadrant ~ 2
~
0011
~
~
~
Quadrant!
o ~.
1011~
_

o +3"f:~
0111
~
Jill
//J
[
01,10 2V IV
\!
0010 3V
~1
q~
9 +iv
+lV
I/
+2V
+3V
I/
1
ooo,
O
oo
OIV
~ * ~ " 001
~

9
oooo~o,oo
o ~,~:~~
co
qD
~176176 ~
L
9
, , oo



Quadrant 3
I I I
~
I
~
I I
Quadrant4
I I

Fig. 4.16
Constellation for 16 QAM showing amplitudes, phases and codes
Bandwidths
81
Table 4.6 A 0 0 0 0 0 0 0 0
1 1 1 1 1 1 1 1
Input permutations and positions (16 QAM)
.
B 0 0 0 0 1 1 1 1
0 0 0 0 1 1 1 1
C 0 0 1 1 0 0 1 1
0 0 1 1 0 0 1 1
D 0 1 0 1 0 1 0 1
0 1 0 1 0 1 0 1
A 0 0 0 0 0 0 0 0
1 1 1 1 1 1 1 1
B 0 0 0 0 1 1 1 1
0 0 0 0 1 1 1 1
Polarity
Amplitude 3 3 3 3 1 1 1 1
1 1 1 1 3 3 3 3
C 0 0 1 1 0 0 1 1
0 0 1 1 0 0 1 1
Polarity 0 1 0 1 0 1 0 1
0 1 0 1 0 1 0 1
Amplitude 3 1 1 3 3 1 1 3
3 l 1 3 3 1 1 3
Quad
+ + + + + + + +
+ + + +
+ + + +
The resultant constellation consists of three different a m p l i t u d e s distributed in 12 different phases as s h o w n in Figs 4.11 a n d 4.144.16.
4.6
BANDWIDTHS
A n i m p o r t a n t c o n s i d e r a t i o n in any c o m m u n i c a t i o n circuit is b a n d w i d t h . T o d a y the usable frequency s p e c t r u m is b e c o m i n g increasingly crowded, with less and less available. This m e a n s that b a n d w i d t h s m u s t be limited m o r e a n d more.
4.6.1
Nyquist
bandwidth
This is the m i n i m u m b a n d w i d t h required to t r a n s m i t a n d receive a digital signal to ensure errorfree c o m m u n i c a t i o n . T h e N y q u i s t b a n d w i d t h is m u c h smaller t h a n the c o m p l e t e signal b a n d w i d t h . The receiver need only be able to identify the different logic states.
4.6.2
Spectrum
analyser
comparison
C o n s i d e r a m o d u l a t o r using a c a r d e r frequency of 140 M H z when a 143.36 M b / s bit s t r e a m is applied to it. Figure 4.17 shows these c o m p a r i s o n s .
82
Digital modulation techniques
2 PSK modulator (Fig. 4.17(a))
The entire band stretches from 3.36 Hz to 283.36 MHz. This is derived from 140 MHz  143.36 MHz to 140 MHz + 143.36 MHz In practice the band is from 0 H z to 143.35MHz as there are no negative frequencies. The Nyquist bandwidth is 68.32 MHz to 211.68 MHz. The frequency deviation is thus 71.68 MHz.
4 PSK modulator (Fig. 4.17(b))
The entire band stretches from 68.32 MHz to 211.68 MHz. This is derived from 143.36/2 = 71.68 Mb/s 140 MHz  71.68 MHz to 140 MHz + 71.68 MHz The Nyquist bandwidth is 104.16MHz to 175.84MHz. The frequency deviation is thus 35.84 MHz. This is a smaller bandwidth than the 2 PSK.
8 PSK modulator (Fig. 4.17(c))
The entire band stretches from 104.16MHz to 175.84MHz. This is derived from 143.36/4 = 35.84 M b / s 140 MHz  35.84 MHz to 140 MHz + 35.84 MHz The Nyquist bandwidth is 122.08 MHz to 157.92 MHz. The frequency deviation is thus 17.92 MHz. This is a smaller bandwidth than the 4 PSK.
16 PSK modulator (Fig. 4.17(d))
The entire band stretches from 122.08 MHz to 157.92 MHz. This is arrived at by 143.36/8 = 17.92Mb/s 140 MHz  17.92 MHz to 140 MHz + 17.92 MHz The Nyquist bandwidth is 131.04 MHz to 148.96 MHz. The frequency deviation is thus 8.96 MHz. This is a smaller bandwidth than the 8 PSK.
64 QAM
From the above it is obvious that if 64 QAM was implemented then the Nyquist bandwidth would be 146.36 MHz/16 = 8.96 MHz
Differential phase modulation
83
(a)
/ / [o 3.36
i"....71.68MHz ==_ 71.6 " Bandwidth 143.36MSz ~i = ,_ ! 168.321 140 ~
2,80 283.36
(b)
t 0 3.36 32.48 68.32 1104.161 140 [175.841 211.68 247L~~.52283.36
17.92MHz
~~
Bandwidth= 35.84MHz
'/~17.92MHz
(c)
3.36
50.4
86.241122.0811401157,921193.76 229.6 8.9~6~ ~ Bandwidth= 17.92MHz ~  ~ / ~ 8.96MHz
283.36
(d) Fig. 4.17
I3.36}_~ 14i.441177.281 !~13.~21~4011157.921~ I14.561 1131.04 1148,96 i
1238.56] 1283.361
Spectrum analyser comparison: (a) 2 PSK; (b) 4 PSK; (c) 8 PSK; (d) 16 PSK
with a deviation of 4.48 MHz. The advantages of QAM modulation are clearly seen. However, there is a limit, as the higher the modulation scheme becomes the closer the phasors are to one another. When the signal is transmitted and noise is introduced it is possible for phasors to cross over and the result is corrupted data.
4.7
DIFFERENTIAL
PHASE
MODULATION
Differential phase modulation is used on radio and microwave systems which employ BPSK, QPSK or QAM modulators. On these systems the data is transmitted in different phasors and/or amplitudes. If differential modulation techniques are not used then the received signal will be decoded incorrectly as the constellation can and does rotate as it propagates through the air. Differential encoding enables the relative phase and relative amplitude information to be transmitted instead of the actual phase and actual
84
Digital modulation techniques
amplitude information. This ensures the correct detection and restoration of the digital information.
4.7.1
PSK differential modulation
A typical 8 PSK differential modulator circuit is shown in Fig. 4.18(a). Bits C, B and A, together with bits c, b and a, form the address for the EPROM. Bits C, B and A are the input bits, with bit C being the MSB and bit A the LSB. The output of the EPROM produces bits c, b and a, with bit c being the MSB and bit a being the LSB. These bits are then applied to the 8 PSK modulator. By considering the constellation diagram of a practical 8 PSK modulator, as shown in Fig. 4.19, bits B and A determine the quadrant and bit C determines the position within the quadrant. It is important to be able to position the phasors in the quadrant. Figure 4.19(a) shows the axis for a theoretical 8 PSK modulator. As can be seen, the x  x axis contains the binary codes for C A which are 102, 002, 012, 112 from left to right.
oAc
B
17 F.,"
Binary input
EP"OM il
+Ac sin OOct[
c
~Z_~ oultFut +A cos (Oct C
(a) ADC F ~ I ~
I
~
A B
EPROM
Binary output
input
IF ~~] ~ ~ ~ r[ Z
~
+A cos ~ct ~',N C
I
A[DC
oT
I ~_ El .. ~_1 .vI
CLK
(b)
Fig. 4.18
(a) 8 PSK modulator (b) 8 PSK demodulator
Differential phase modulation
v
85
Quadrant 2
BA = I0 C=I C=O
010
BA= I0 C=O ~=1 CB 01 CA _
011
BA= II C=O C=I
Quadrant I BA=II
C=l C=0
110 CA 10
Ill
CA CA
I
I
I
I
J
O0
I
I
CB O0
I I
1
i
I
I
x
01
100
BA = O 0
11 I01
BA =01 C=I C=O
C=I C=O
i C, A ]I XXX
(a)
B,4 =00 C=0 C=I
000
BA = O I
m
C=O C=I 001
Quadrant 3
Quadrant 4
Y
Quadrant 2
BA =O0 C=I _ C=0
000 BA =O0 C=0 C=! CB O0 CA
001
BA =OI C=O C=l
Quadrant 1
BA =01 C=I C=0
100 CA
101
CA CA
10 11o
BA= 10 C=l 6"=0
i
t
O0 CB _
I
I'
I
OI
01
I
il
III BA=II C=I
BA= 10 C=O C=I
BA=il
C=0 C=I
Quadrant 3
010
011
Quadrant 4
(b)
Fig. 4.19 Constellations for (a) theoretical 8 PSK and (b) practical 8 PSK modulation
Likewise the y  y axis contains the binary codes for C"B, which are 102, 0 0 2 , 012, 112 from bottom to top. Figure 4.19(b) shows the axis for the constellation for the practical 8 PSK modulator. In this case the x  x axis contains the binary codes for CA which are 102, 002, 012 and 112. The y  y axis contains the binary codes for CB, which are 112, 012, 002 and 102. As an example, the binary code 0012 produces a phasor in quadrant 4 for the theoretical 8 PSK modulator and quadrant 1 for the practical 8 PSK modulator as CA = 012 and CB = 102. The differential encoding takes place according to the data given in Table 4.7. In this example the logic state of bit C remains unchanged through the differential encoder.
86
Digital modulation techniques
Table 4.7 Bin code Differential encoding data Phase angle
Bib
0 0 1 1
A/a
0 1 0 1 0~ +90 ~ 90 ~ 4180 ~
Example 4.6
A practical 8 PSK differential modulator produces the constellation as shown in Fig. 4.9. The following bit stream is applied to the input of the modulator. The differential encoding takes place according to the data given in Table 4.7. Data=(ll0) 101 000 010 100 011 1012
Determine the codes for the encoded data at the output of the EPROM. Assume the default code to be 1102.
Solution
The encoding is done as shown in Table 4.8. 1. Find the position on the constellation diagram for Find the position on the constellation diagram for Relative to 1102, 1012 is found by moving 1102 by c = b i t C = 1. The data is cba = 1112. 2. Feed 1112 back to the input. Find the position on the constellation diagram for Find the position on the constellation diagram for 1102 ~ 3rd quadrant. 1012 ~ 1st quadrant. 4180~ ~ bits ba = 11, bit
1112 ~ 4th quadrant. 0002 ~ 2nd quadrant.
Table 4.8 Present C B A
Differential encoding for 8 PSK m o d u l a t o r Quad Next C 3rd 4th 4th 3rd 3rd 1st 1 0 0 l 0 l B 0 0 1 0 l 0 A 1 0 0 0 1 1 1st 2nd 3rd 2nd 4th 1st +180 ~ 4180 ~ 90 ~ 90 ~ +90 ~ 0~ Quad Phase change O u t p u t code c b a
I I 0 0 I 0
I I 1 1 I 0
T
0 I 1 0 0 1
1 0 0 1 0 1
1 1 1 1 0 0
I
1 1 0 0 1 0
. . . . . . . . . . . .
Differential phase modulation
87
3.
4.
5.
6.
Relative to 1112, 0002 is found by moving 1112 by + 180 ~ ~ bits ba = bit c = bit C = 0. The data is 0112. Feed 0112 back to the input. Find the position on the constellation diagram for 0112 "'} 4th quadrant. Find the position on the constellation diagram for 0102 ~ 3rd quadrant. Relative to 0112, 0102 is found by moving 0112 by  9 0 ~ ~ bits ba = bit c  bit C  0 . The data is 0102. Feed 0102 back to the input. Find the position on the constellation diagram for 0102 ~ 3rd quadrant. Find the position on the constellation diagram for 1002 ~ 2nd quadrant. Relative to 0102, 1002 is found by moving 0102 by  9 0 ~  , bits ba = 10, c = bit C = 1. The data is 1102. Feed 1102 back to the input. Find the position on the constellation diagram for 1102 } 3rd quadrant. Find the position on the constellation diagram for 0112 * 4th quadrant. Relative to 1102, 0112 is found by moving 1102 by +90 ~ } bits ba = 01, c=bit C=0. The data is 0012. Feed 0012 back to the input. Find the position on the constellation diagram for 0012 * 1st quadrant. Find the position on the constellation diagram for 1012 ~ 1st quadrant. Relative to 0012, 1012 is found by moving 1102 by 0 ~ } bits ba = 00, bit c = C=I. The data is 1002.
11,
10,
bit
bit
bit
This gives the following data streams:
CBA Input data = (110) cba Diff. encoded data = (110) CBA 101 cba 111 CBA 000 cba 011 CBA 010 cba 010 cba 110 CBA 100 cba 001 CBA 011 cba 1002 CBA 1012
The differentially encoded data is eventually transmitted to the distant receiver. The receiver must correctly decode the information to obtain the original data stream. A typical circuit for the differential decoder is shown in Fig. 4.18(b). In this case the intermediate frequency (IF) input is equally split and applied to the two ring demodulators. The IF is demodulated and the outputs are connected to analoguetodigital converters (ADCs). The resultant digital signals on leads b and a are used as part of the address for the EPROM. The ADC also produce the logic condition for lead c which is also used as part of the address for the EPROM. The leads c, b, and a are then connected to the inputs to three D flipflops. The Q
88
Digital modulation techniques
outputs of these flipflops produce the same logic conditions one clock pulse later. These outputs are used as part of the address for the EPROM. The output of the EPROM produce the logic conditions for outputs C, B and A. These three outputs are multiplexed together to produce a serial output bit stream.
Example 4.7
A practical 8 PSK differential demodulator produces the constellation shown in Fig. 4.9. The following bit stream is received from the distant transmitter. The differential decoding takes place according to the data given in Table 4.7. DATA = (110) 111 011 010 110 001 1002
Determine the output data of the decoder. Assume the default code to be 1102.
So/ut/on
To enable the decoding to be done refer to the constellation diagram for the practical 8 PSK modulator shown in Fig. 4.9(b). The decoding is done as shown in Table 4.9. 1. Find 1102 o n the constellation , 3rd quadrant. Determine where 1102 is moved to. Next input data is 1112 ~ C ~ 1, b a  11. 112 = 4180~ 1102 is moved to the 1st quadrant ~ B A = 01. D a t a  1012. 2. Feed 1112 back to the input. Find 1112 on the constellation , 4th quadrant. Determine where 1112 is moved to. Next input data is O112 ~ C = O, b a = 11. 112 = t180~
Table 4.9 Present
Differential decoding for 8 P S K d e m o d u l a t o r Quad Next Phase change O u t p u t code
cba
1 1 0 0 1 0 1 1 1 1 1 0 0 1 1 0 0 1 3rd 4th 4th 3rd 3rd 1st
cba
1 0 0 1 0 1 1 1 1 1 0 0 1 1 0 0 1 0
+180 ~ 4180 ~ _90 ~ 90 ~ +90 ~ 0o
CBA
1 0 000 0 1 1 0 0 1 1 0 1
0 0 1 1
Differential phase modulation
89
3.
4.
5.
6.
1112 is moved to the 2nd quadrant , B A = 00. Data = 0002. Feed 0112 back to the input. Find 0112 on the constellation , 4th quadrant. Determine where 0112 is moved to. Next input data is 0102 ~ C = 0, ba = 10. 102 = 90 ~ 0112 is moved to the 3rd quadrant ~ B A = 10. Data = 0102. Feed 0102 back to the input. Find 0102 o n the constellation ~ 3rd quadrant. Determine where 0102 is moved to. Next input data is 1102 ~ C = 1, ba = 10. 102  90 ~ 0102 is moved to the 2nd quadrant  , B A = 00. Data = 1002 Feed 1102 back to the input. Find 1102 o n the constellation ~ 3rd quadrant. Determine where 1102 is moved to. Next input data is 0012 * C  0, ba = O1. 012 = +90 ~ 1102 is moved to the 4th quadrant ~ B A  11. Data = 0112. Feed 0012 back to the input. Find 0012 on the constellation , 1st quadrant. Determine where 0012 is moved to. Next input data is 1002 ~ C = 1, ba = 00.
002  0 ~
0012 remains in the 1st quadrant  , B A  0 I. Data  1012. This gives the following data stream:
CBA CBA CBA CBA CBA CBA
Output data =
101
000
010
100
011
1012
This is the same as the original data at the transmitter prior to the differential encoding.
4.7.2
Sixteenquadrature amplitude differential modulation
A typical circuit for a 16 QAM differential modulator is shown in Fig. 4.20. Bits A, B, C and D together with bits a, b, c and d form the address for the EPROM. Bits A, B, C and D are the input bits with bit A being the MSB and bit D the LSB. The output of the EPROM produces bits a (MSB), b, c and d (LSB). These bits are then applied to the 16 QAM modulator.
90
Digital modulation techniques
Binary input
16 quadrature amplitude modulation constellation
~o
9
0~0/.,0
T ~
9
~t~0
9
6dB Pad EPROM
; ~ r : AB ~ , do, ~ ,,~o, ,,o,
000@ 0100 ~lF I(l~t
T
I1~]
CD
output
6dB Pad
C
Fig. 4.20
A 16 QAM differential modulator
The differential modulation method given in the solution to Example 4.8 relies on being able to position the phasors in the correct quadrant in the constellation. By referring to Fig. 4.21 it can be seen that the x  x axis contains the binary codes for AB, which are 002, 012, 102, 112 from left to right. Likewise the y  y axis contains the binary codes for CD which are 002, 012, 102, 112 from bottom to top. Using this information it is a fairly simple matter to position each phasor in the correct quadrant. As shown, A C = 112 for quadrant 1, AC = 012 for quadrant 2, A C = 002 for quadrant 3 and A C = 102 for quadrant 4. Hence the logic conditions for A C determine the quadrant in which the phasor lies. The logic conditions for BD determine the position within the quadrant as in each quadrant the logic states of BD assume values of 002, 012, 102 or 112. As an example the phasor having the code 00102 means that A C = 012 and BD = 002. The code 10002 causes the phasor to lie in quadrant 4 as A C = l0 and BD = 002. The differential encoding takes place according to the data given in Table 4.10.
Differential phase modulation
I I I
91
'
I
Y
I
I
I
I
Quadrant 2 0011
0111 1011
Quadrant I
1111
O
AC=01 BD=01
AC=OI BD = O0
9
CO__
9
9
AC=01 B D = 11
AC=01 B D = 10
A C = 11
BD=01 A C = 11
BD = O0
A C = 11 B D = 11
0
0010
AB
~X
9
c~__
9
A C = 11 BD = 10
9

0110
AB
1010
AB
1110
AB
I
I
00 0001
01 0101
I
I
co_
10 1001
I
I
II 1101
Ix
0
AC=O0
9
9
9
BD=01
AC=O0 BD = O0
AC =O0 B D = 11
AC=O0 B D = IOco ~ !
AC = 10 BD=01
A C = 10 BD = O0
A C = 10 B D = 11 AC = 10 BD = 10
0
0000
9

9
1000
9
1100
0100
Quadrant 3
Y
I
Quadrant 4
I I showing
i Fig. 4.21 Constellation
....
I
I
positioning
I of phasors
I
I
I
for 16 QAM
Example 4.8 A 16 QAM differential modulator produces the constellation as shown in Fig. 4.11. The following bit stream is applied to the input of the modulator. The differential encoding takes place according to Table 4.10. Data = (0110) 1101 1000 0001 1100 10102
Determine the codes for the encoded data at the output of the EPROM. Assume the default code to be 01102. Table 4.10 Differential encoding data Bin code
A/a C/c
Phase angle
Bin code
Bib D/d
Amplitude
b/B diD
0 0 1 1
0 1 0 1
0~ 90 ~ +90 ~ +180 ~
0 0 1 l
0 1 0 l
1 1 0 0
1 0 1 0
92
Digital modulation techniques
Solution
The differential coding is done according to the data given in Table 4.11. 1. Find the position on the constellation diagram for 01102 o 2nd quadrant. Find the position on the constellation diagram for 11012 0 4th quadrant. Relative to 01102, 11012 is found by moving 01102 by + 180 ~ 0 bits a c = 11. Bits B D = 11, therefore bits b d = 0 0 . The data is 10102. 2. Feed 10102 back to the input. Find the position on the constellation diagram for 10102 0 1st quadrant. Find the position on the constellation diagram for 10002 0 4th quadrant. Relative to 10102, 10002 is found by moving 10102 by  9 0 ~ ~ bits a c = 0 1 . Bits B D  0 0 , therefore bits b d = 1 I.
T h e d a t a is 011123. F e e d 01112 b a c k to the input. F i n d the position on the constellation d i a g r a m for 01112 ~ 2nd q u a d r a n t . F i n d the position on the constellation d i a g r a m for 00012 .o 3rd q u a d r a n t . Relative to 01112,00012 is f o u n d by m o v i n g 01112 by + 9 0 ~ 0 bits a c = I O. Bits B D = 01, therefore bits b d = 10. The d a t a is 11002. 4. F e e d 11002 b a c k to the input. F i n d the position on the constellation d i a g r a m for 11002 ~ 4th q u a d r a n t . F i n d the position on the constellation d i a g r a m for 11002 ~ 4th q u a d r a n t .
Relative to 11002, 11002 is found by moving 11002 by 0 ~ o bits a c = 0 0 . Bits B D = 10, therefore bits b d = 0 1 . The data is 00012. 5. Feed 00012 back to the input. Find the position on the constellation diagram for 00012 o 3rd quadrant. Find the position on the constellation diagram for 10102 ~ 1st quadrant. Relative to 00012, 10102 is found by moving 00012 by + 180 ~ ~ bits a c = 11. Bits B D = 00, therefore bits b d = l l. The data is 11112.
Table 4.11 Present
AB CD
Differential encoding for 16 QAM modulator Quad Next
AB CD
Quad
Phase 2hange l a c 1 1
0
Amplitude B D b
1 0
Output data d
0
a
1
b
0
c
1
d
0
0
1
1
0
1
1
0
0
0 1 0
1 1 0
1 0 0
1 0 1
2nd 1st 2nd 4th 3rd
1 1 1 0 0 0 1 1
1 0
0 0 0 0
1
1 0 1 0
0
4th 4th 3rd 4th 1st
180~ 90 ~
~90o
]1 ]0
1
0
1
1
1
1
0
0
1
1
I
1
0
1
0
0~ 180 ~
0 0 1 1
0 0
0 1
1 1
0
1
0
1
0
1
1
1
!
Differentia/phase modulation This gives the following data streams: Input data =
ABCD 1101 ABCD 1000 ABCD 0001 ABCD 1100 ABCD 1010 2 abcd 11112
93
abcd Diff. encoded data  (0110)
abcd abcd abcd abed 1 0 1 0 0 1 1 1 1 1 0 0 0001
Once the data has been differentially encoded it will eventually be transmitted to the distant receiver. The receiver must be able to correctly decode the data to obtain the original data. A typical circuit is shown in Fig. 4.22. The IF input is equally split to the four mixers where each input is demodulated with the same carrier frequency. The carrier oscillator is connected directly to the top two mixers and the carrier frequency is shifted by 90 ~ before being applied to the bottom two mixers. The top two mixers produce bits a and c while the bottom two mixers produce bits b and d. These four bit streams are then applied to the EPROM as part of the address. They are also applied to the D inputs of four flipflops and one
a v
c A r
6dB Pad IF input } 16 quadrature amplitude modulation constellation ~ o?,,1 ~,, ,~
0010 0110
T
~ EPROM
Binary
d
v
,ah
1010
III0
"=;;;,;AB
0000
9
0100
Pad b
9
/
~
I000
9
I100
9
CD
r
A IIW
I
Fig. 4.22
A 16 QAM differential demodulator
94
Digital modulation techniques
clock cycle later the same logic conditions appear on the four Q outputs and form the other part of the EPROM address. The output of the EPROM produces the bit streams on leads A, B, C and D. These four bit streams are then multiplexed together to form a serial output bit stream.
Example 4.9
A 16 QAM differential demodulator produces the constellation as shown in Fig. 4.11. The following bit stream is received from the distant transmitter. The differential decoding takes place according to the data given in Table 4.10. D a t a  (0110) 1010 0111 1100 0001 11112
Determine the output data of the decoder. Assume the default code to be 0110 2.
Solution
The decoding is done as shown in Table 4.12. 1. Find 01102 on the constellation ~ 2nd quadrant. Determine where 01102 is moved to. Next input data is 10102 o b d = 00 and B D = 11 ~ ac = 11. 112  =t=180~ 01102 is moved to the 4th quadrant. A C = 10. Data = 11012. 2. Feed 10102 back to the input. Find 10102 on the constellation o 1st quadrant. Determine where 1010 2 is moved to. Next input data is 01112 ~ b d = 11 and B D = O0 ~ a c = 01. 012 =  9 0 ~ 10102 is moved to the 4th quadrant. A C = 10. Data = 10002.
Table 4.12 Present
Differential decoding for 16 Q A M d e m o d u l a t o r Q u a d Next Phase Amplitude Change 180 ~ 90 ~ +90 ~ 0~ 180 ~ Quad A C 4th 4th 3rd 4th 1st 1 1 0 1 1 0 0 0 0 1 O u t p u t data
a b c d
0 1 1 0 1 0 1 0 0 1 1 1 1 1 0 0 0 0 0 1
X
abc
2nd 1st 2nd 4th 3rd
d
a
c
b d B D
0 0 1 1 1 1 0 0 1 1 1 1 0 0 0 1 1 0 0 0
ABCD
1 1 0 1 1 0 0 0 0 0 0 1 1 1 0 0 1 0 1 0
1 0 1 0 0 1 1 1 1 0 0 0 0 1 1 1 1
I
1 0
1
1 1 0 1 1 0 0 0 1 1
Review questions 95
3. Feed 01112 back to the input. Find 01112 on the constellation  , 2nd quadrant. Determine where 01112 is moved to. Next input data is 11002 , b d = l 0 and B D = 0 1  , ac 102 = +90 ~ O1112 is moved to the 3rd quadrant.
A C = 00.
=
lO.
Data = 00012. 4. Feed 11002 back to the input. Find 11002 on the constellation ~ 4th quadrant. Determine where 11002 is moved to. Next input data is 00012 0 b d = 01 and B D = 10 ~ ac = 00. 002 = 0 ~ 11002 remains in the 4th quadrant. A C = 10. Data = 11002. 5. Feed 00012 back to the input. Find 00012 on the constellation ~ 3rd quadrant. Determine where 00012 is moved to. Next input data is 11112 ~ b d = 11 and B D = O0 ~ ac = 11. 112 = +180 ~ 00012 is moved to the 1st quadrant. A C = 11. Data = 10102. This gives the following data streams:
ABCD ABCD ABCD ABCD ABCD
Output data =
1101
1000
0001
1100
1010 2
This is the same as the original data that was input into the transmitter.
4.8
REVIEW QUESTIONS
4.1 Describe the operation of the ASK modulators shown in Fig. 4.3. 4.2 Explain with the aid of a diagram how a VCO can be used as an FSK modulator. 4.3 Explain with the aid of a diagram how a VCO can be used as an FSK demodulator. 4.4 Draw a typical circuit diagram of a PSK modulator. 4.5 Draw the constellation of a typical theoretical 8 PSK modulator. 4.6 Show the positions on the constellation of a practical 8 PSK modulator for the following data stream. Data = 011 101 110 0012
96
Digital modulation techniques
4.7 Determine the output data for a 16 QAM differential modulator given that the default code is 10012. The modulator codes the data according to the information given in Table 4.10. The input data stream is as follows"
D a t a = 0111 0011 0000 0110 11112
4.8 Decode the bit stream obtained in Question 4.7 using a 16 QAM demodulator which uses a default code of 10012. The demodulator decodes the data according to the information given in Table 4.10.
This action might not be possible to undo. Are you sure you want to continue?