Digital Transmitter

Noise

Digital Receiver

Fig.-2:-Block diagram of a typical digital communication system.

• The source encoder accepts the digital signal and
encodes it into a shorter digital signal. This is called
source encoding, which reduces the redundancy,
hence the transmission speed. This in turn reduces
the bandwidth requirement of the system.
• The channel encoder accepts the output digital signal
of the source encoder and encodes it into a longer
digital signal. Redundancy is deliberately added into
the coded digital signal so that some of the errors
caused by the noise or interference during
transmission through the channel can be corrected at
the receiver.
• Most often the transmission is in a high-frequency
passband; the modulator thus impresses the encoded
digital symbols onto a carrier.
• For wireless systems an antenna (which is not shown
in Fig.-2) is the final stage of the transmitter.
• The channel, where noise adds to the signal, fading
and attenuation effects appear as a complex
multiplicative factor on the signal. The noise
includes all kinds of random electrical disturbance
from outside or from within the system. The channel
also usually has a limited frequency bandwidth so
that it can be viewed as a filter.
• In the receiver, virtually the reverse signal processing
happens. First the received weak signal is amplified
(and down-converted if needed) and demodulated.
Then the added redundancy is taken away by the
channel decoder and the source decoder recovers the
signal to its original form before being sent to the
user. A D/A converter is needed for analog signals.
 Geometric interpretation of signals
• A message source (as shown in Fig.-3(a)) emits
one symbol, mi per T seconds (say) out of known
M quantization levels denoted by m1, m2, …, mM.
• Assume that M symbols of the message are equal
likely, then probability of mi is emitted =
1
pi  P(mi )  for i  1,2,..., M ...(1)
M
• The transmitter takes the message source mi &
codes it into a distinct signal si(t) suitable for
transmission over the channel.
• The signal
si(t) occupies
the entire
symbol Fig.-3(a):-Block diagram of a generic DC system.
period T
allotted to message, mi.
• Hence, the energy of the
signal, Ei =
T
0
si2 (t )dt ; i  1,2,...M ...(2)

• Assuming the channel is

linear (i.e., having no
distortion) & having only
an AWGN of zero mean Fig.-3 (b):-AWGN model of a channel.
 (as shown in Fig.-3(b)), the received signal, x(t)
xi (t )  si (t )  w(t ) ; 0  t  T for i  1,2,..., M ...(3)
• The receiver has the task of observing the
received signal xi(t) for a duration of T secs &
making a best estimate of the transmitted signal
si(t) for the symbol, mi.
• However, due to presence of channel noise, the
decision making process is statistical in nature &
makes an occasional error.
• Thus we have to design the receiver so as to
minimize the average probability of symbol error,
M
P   p P(m
e i
ˆ  m | m ) ...(4) where, m̂ is the
i i
i 1
 estimate produced by the receiver & P(mˆ  mi | mi )
is the conditional error probability given that ith
symbol was sent. The resulting receiver is said to
be optimum in the minimum probability of error
sense.
• Now, we can geometrically represent any set of M
energy signals, {si(t)} as linear combinations of N
orthonormal basis functions, where N ≤ M, i.e.,
N
 0t T
si (t )   sij j (t ) ;  ...(5)
j 1 i  1,2,...M
where the coefficients of the expansion are,
 T  i  1, 2..., M
sij   si (t ) j (t )dt ;  ...(6)
0
 j  1,2,..., N
• & the real valued basis functions ϕ1(t), ϕ2(t), …,
ϕN(t) are orthonormal, i.e.,
T 1 if i  j
0
i (t ) j (t )dt   ij  
0 if i  j
...(7)

where δij is the Kronecker delta.

• Hence from Eq.-(7) it is clear that the basis
function is normalized to have unit energy & they
are orthogonal to each other over the time interval
0 ≤ t ≤ T.
• Thus si(t) is an N-dimensional vector for each i
=1-to-M.
• Thus, given the N element of the vector si (i.e.,
the components: si1, si2, …, siN) operating as i/p
then we can generate/synthesize the signal si(t)
using the ckt shown in Fig.-4(a).
• Similarly, given the signal si(t) for i=1, 2, …, M
operating as i/p then we can calculate/analyze its
coefficients: si1, si2, …, siN using the ckt shown in
Fig.-4(b).
• Thus each signal si(t) is completely determined by
the vector of its coefficients:  si1 
s 
si   i 2  ; i  1, 2, ..., M ...(8)
 ... 
 
 siN 
 (b):- Analyzer ckt to generate
(a):- Synthesizer ckt to generate si(t). the components of si(t).
Fig.-4
• The vector si is called a signal vector in the N-
dimensional Euclidean space with N mutually
perpendicular axes labeled as ϕ1, ϕ2, …, ϕN.
• It provides the mathematical basis for the
geometric representation of energy signals,
thereby paving the way for the noise analysis of
digital comn systems in a conceptually satisfying
manner.
• Fig.-5 represents a 2-dimensional (i.e., N = 2)
signal space with 3 signals (i.e., M=3).
• In an 2-dimensional Euclidean space, we may
define length (absolute value or norm) of vectors
& angles between vectors.
• The
squared-
length of
any signal
vector si is
the inner/dot
product of si
with itself,
i.e.,
|| si ||  s i si
2 T

N
  sij2
j 1 for
i = 1, 2, …, Fig.-5:-Illustrating the geometric representation of
M …(9) signals for the case when N=2 & M =3.
 where, sij is the jth element/component of si.
• The energy of a signal si(t) of duration T secs is:
 N   N

Ei   s i (t )dt    sij j (t )  sikk (t ) dt
T T
2

 j 1   k 1 
0 0

Interchang ing the order of summation &

int egration & then rearranging terms we got : 
N N N
Ei   sij sik   j (t )k (t )dt   ij i ...(10)
T
s 2
 || s || 2
0
j 1 k 1 j 1

• The inner/dot product of a pair of signals over

the interval [0,T] is: T
 si (t )sk (t )dt  si sk ...(11)
T
0
• The Euclidean distance between two signals si &
sk is = dik= || si - sk|| …(12) &
N
|| si  sk ||   ( sij  skj )   [ si (t )  sk (t )]2 dt ...(13)
T
2 2
0
j 1

• The angle, θik between 2 signal vectors si & sk is:

siT sk
cos  ik  ...(14)
|| si || || sk ||
• The 2 vectors si & sk are orthogonal or
perpendicular to each other if their inner product,
siTsk is zero or θik = 90o.
 Schwarz’s inequality
• Consider any pair of energy signals s1(t) & s2(t).
Then the Schwarz inequality states that:
2
 s (t ) s (t )dt     s 2 (t )dt   s 2 (t )dt  ...(15)


  1 2    1   2 

The equality holds iff s2(t) = cs1(t), where c is any

constant.
• Proof:- In order to prove this important inequality
(Eq.-15), let s1(t) & s2(t) be expressed in terms of
the pair of orthonormal basis functions ϕ1(t) &
ϕ2(t) as follows:-
 s1(t) = s11ϕ1(t) + s12ϕ2(t)
& s2(t) = s21ϕ1(t) + s22ϕ2(t)
• Where ϕ1(t) & ϕ2(t) satisfy the orthonormality
conditions over the entire time interval (-∞,∞),
i.e.,  1 for i  j
 i (t )k (t )dt   ij  0 othewise ...(7)
• Thus,
 s11   s21 
s1    & s2    ...(16)
 s12   s22 
as shown in Fig.-6.
• Since θ is the angle between s1 & s2, & cos2θ ≤ 1,
 T
s s | s1T s2 |2
& cos   1 2
...(14)  2 2
 cos 2
 1
|| s1 || || s2 || || s1 || || s2 ||

Fig.-6:- To explain the Schwarz inequality.

 
|  s1 (t ) s (t )dt | 
*
2
2


  | s (t ) |2 dt   | s (t ) |2 dt   E E ...(17)
  
1
  
2
 1 2

• Hence, square of inner/dot product of 2 signals is

always ≤ product of their energies & it (Eq.-17)
is equals when s2(t) = cs1(t), where c is any
constant. Thus, Eq.-15 (Schwarz inequality) is
proved.
 Amplitude shift keying(ASK)
• When it is required to transmit digital data over a
band-pass channel, it is necessary to modulate the
incoming data onto a sinusoidal carrier wave with
fixed frequency limits imposed by the channel.
• In digital modulation process, either the
amplitude or phase or frequency of the carrier
signal is switched from one value to another (i.e.,
a step change happens) in accordance with the
incoming digital data as shown in Fig.-7.
• Thus, 3 basic signaling schemes are used in
 Waveform for BASK.

Waveform for BPSK

Fig.-7 (c):-Waveform for BFSK.

 passband data transmission (e.g., telephone lines,
microwave radio links, optical fibers, etc.): (1)
ASK (2) PSK (3) FSK.
• In ASK, amplitude of the carrier is switched from
one value to another depending on modulating
digital i/p. Similarly also happens in case of PSK
& FSK systems.
• In practice, PSK & FSK signals are much more
widely used (due to constant envelope) than ASK
(due to variable/nonlinear amplitudes as observed
from Fig.-7).
• Sometimes, a hybrid form of modulation is used.
 For example, changes in both amplitude & phase
of the carrier are combined to produce amplitude-
phase keying (APK).
• Basically, there are 2 types of digital modulation
schemes: (i) Binary schemes & (ii) M-ary
schemes.
• In M-ary signaling schemes, the modulator
produces one of an available set of M = 2n
distinct signals in response to n bits of source
data at a time & binary modulation is a special
case of M-ary modulation with M =2.
• The error performance of M-ary schemes are
poor as compared to the binary schemes.
• To perform demodulation at the receiver, we have
the choice of coherent or noncoherent detection.
• In coherent/synchronous detection, we use to
generate a phase & frequency synchronized
carrier at the receiver to recover the information
signal from the transmitter. Coherent techniques
are complex but yield better performance.
• In coherent detection, the detection is done by
correlating received noisy signal with locally
generated carrier.
• In noncoherent detection, knowledge of the
carrier wave’s phase is not required. The
complexity of the receiver is thereby reduced but
 at the expense of an inferior error performance,
compared to a coherent system.
• Choice of modulation/detection schemes are based
on the following design goals:-
(i) Maximum data rate, (ii) Minimum probability of
symbol error, (iii) Minimum transmitted power, (iv)
Minimum channel BW, (v) Maximum resistance to
interfering signals, (vi) Minimum ckt complexity,
etc.
• The channel BW & transmitted power are 2 primary
comn resources. The BW efficiency is the ratio of
the data rate (bits/s) to the effectively utilized
 channel BW, i.e.,   Rb bits / Hz.
• The BW efficiency BW
depends on (i) multichannel coding & (ii)
spectral shaping.
• Coherent binary ASK (Fig.-7(a)) or on-off keying
(OOK) is the simplest digital modulation
technique. To transmit the BASK waveform,
 2 Ps cos(c t ) for '1'
s (t )   ...(a)
 0 for '0'

• The signal space diagram of ASK is shown in

Fig.-8(a), with ‘0’ at 0 & ‘1’ at PsTb . The distance
between the 2 signal point is: d  PsTb  Eb ...(b)
• Generation of BASK signal:- The BASK modulated
signal can be generated by applying the incoming binary
data (in NRZ unipolar form) & carrier signal as 2 i/ps to a
product /balanced modulator & band-limiting it by a BPF
as shown in Fig.-8(b).

Fig.-8(a):-Signal space diagram of ASK.

• Modulation causes a shift of the baseband signal
spectrum.
• The PSD of ASK signal is shown in Fig.-8(c) having BW
of 3fb centered at ±fc due to rectangular pulse/data
waveforms. The BW of the ASK signal can be reduced
further by using smoothed versions of the pulse waveform

Product modulator
&

Fig.-8(b):-Generation of BASK waveform.

instead of rectangular pulse waveforms.
• Coherent detection or demodulation of BASK signals:-
The Coherent detector for the BASK signal is shown in
Fig.-8(d).

Fig.-8(c):-PSD of BASK signal.

• It consists of a product modulator which is
followed by an integrator & a decision making
device.
•The integrator operates on the o/p of the product

Fig.-8(d):-Coherent detection of BASK signal.

 modulator for successive bit intervals &
essentially performs a low pass filtering action.
• The o/p of the integrator goes to the i/p of a
decision-making device, which compares the o/p
of the integrator with a preset threshold value to
decide ‘1’ or ‘0’.
• The local carrier is in perfect synchronism with
the carrier, i.e., the frequency & phase of the
locally generated carrier is same as that of the
transmitted carrier.
• Also, bit period timing synchronization is
required for exact decision-making operations as
 per the transmitted binary data.
• The BASK generation & detect circuit is simple
but very sensitive to noise (hence bit rate is low,
limited upto 100 bits/s), hence it finds limited
application in data transmission.
• BER or probability of error, Pe is a very
important parameter to judge the performance of
a digital comn system. It should be as small as
possible.  PsTb  1  Es 
1
• For BASK, Pe  2 erfc  4 N 0   2 erfc  4 N 0  ...(c)
when using the matched filter with PSD for a
white noise i/p = N0/2.
 Binary Phase shift keying(BPSK)
• Generally data rates decreases when channel
condition worsens & there is a movement to a
more noise-robust digital modulation scheme.
• In BPSK the transmitted signal is a sinusoid of
fixed amplitude & it has one fixed phase when
data is at level ‘1’ but the phase difference is by
1800 for level ‘0’ as shown in Fig.-9(a).
• If the sinusoid carrier has amplitude A, then it has
power 1 2
Ps  A  A 2 Ps ...(d )
2
• Thus, the transmitted signal is either
 BPSK modulated wave & modulating digital signal.

Fig.-9(b):-A scheme for BPSK transmission & reception.

 
 2 Ps cos(0t ) for '1'
vBPSK (t )   ...(e)
 2 Ps cos0t      2 Ps cos 0t for '0'


• Thus, assuming polar coding scheme, i.e., b(t) =

1V for symbol ‘1’ & -1V for symbol ‘0’
vBPSK (t )  b(t ) 2Ps cos 0t ...(e' )
• So, Eq.-(e’) resembles a DSB-SC AM signal,
hence BPSK is more power efficient than ASK.
• In practice BPSK signal is generated by applying
b(t) & carrier signal to a balanced modulator as
shown in Fig.-9(b) (left-side).
• The received BPSK signal is:-
vBPSK (t )  b(t ) 2Ps cos(0t   )  b(t ) 2Ps cos 0 (t   / 0 ) ...( f )

• Where θ is the phase shift due to propagation

time delay of θ/ω0 between transmitter &
receiver.
• The original data b(t) is recovered in the coherent
demodulator as shown in Fig.-9(b) (right-side).
• The generated carrier from the squared
synchronizing circuit is multiplied with the
received signal to generate:-
1 1 
b(t ) 2 Ps cos (0t   )  b(t ) 2 Ps   cos 2(0t   ) ...( g )
2

2 2 
• Eq.-(g) is applied to an integrator as shown in Fig.-
9(b) (right-side).
• The bit synchronizer recognizes exactly the starting
& ending of a bit period, Tb. At end of each bit
period it closes switch Sc very briefly to
discharge/dump the integrator capacitor & keeps it
open for the entire bit interval. Hence, the operation
of the bit synchronizer allows us to sense each bit
independently of every other bit.
• Thus, integrator along with bit synchronizer works
as an integrate-&-dump ckt as shown in Fig.-9(b)
(right-side) to collect the binary bit stream.
• Since, the bit synchronizer captures an integral, n
• no.s of cycles of the carrier of frequency f0, hence
the o/p of the BPSK receiver will be:-
kTb 1
v0 (kTb )  b(kTb ) 2 Ps  dt
( k 1)Tb 2

kTb 1
 b(kTb ) 2 Ps  cos 2(0t   )dt
( k 1)Tb 2

Ps
 b(kTb ) Tb ...(h)
2
• As the b(t) is a NRZ polar code which makes
excursions between + P &  P , thens s

2
 sin fTb 
Gb ( f )  PsTb   ...(i )
 fTb 
 & BPSK waveform will be:-
PsTb  sin  ( f  f 0 )Tb   sin  ( f  f 0 )Tb  
 2 2

GBPSK ( f )        ...( j )

2   ( f  f 0 )Tb    ( f  f 0 )Tb  
 
which is shown in Fig.-10(a) & (b).
• Signal space representation of BPSK signals:-A
BPSK signal can be represented in terms of one
orthonormal signal u1(t) = (2 / T ) cos  t (as shown in
b 0

Fig.-10©), since v (t )  b(t ) 2 P cos  t

BPSK s 0
 Ps
2
 [ PsTb b(t )] cos 0t
Tb
 [ PsTb b(t )]u1 (t ) ...(e' )
 (a) Plot of Gb(f).

Fig.-10(b):-PSD of BPSK signal.

• The distance, d between 2 antipodal signals is:-
d  2 PsTb  2 Eb ...(k )

• The distance, d is inversely proportional to the probability

that we may make an error, in the presence of noise.

Fig.-10(c):-Geometrical representation of BPSK signals.

 Differential PSK(DPSK)
• DPSK & differential encoded PSK (DEPSK) are
modifications of BPSK to eliminate the
ambiguity about whether the demodulated data is
or is not inverted.
• Also, DPSK avoids the need to provide the
synchronous carrier required at the demodulator
for detecting a BPSK signal.
• A means for generating a DPSK signal is shown
in Fig.-11(a). The data stream, d(t) to be
transmitted, is applied to one i/p of an exclusive-
OR logic gate. The one bit period delayed version
Fig.-11(a):-Generation of a DPSK signal.
 of b(t), i.e., b(t-Tb) is applied as the 2nd i/p of the
gate with b(t) as the o/p of the Ex-OR logic gate,
i.e., b(t) = d(t)  b(t-Tb).
• Fig.-11(b) illustrates the response b(t) to an i/p
d(t) assuming that b(0) = 0. We observe that b(t)
changes level at the beginning of each interval in
which d(t) =1 & b(t) does not change level when
d(t) = 0. This behavior is to be anticipated from
the truth table of Ex-OR logic gate.
• Thus, When d(t) = 0, b(t) = b(t-Tb) & when d(t) =
1, b(t) = b(t  Tb ). Thus, Level of b(t) changes only
whenever d(t) = 1 & no change in b(t) when d(t)
= 0.
• Now b(t) is applied to the balanced modulator along with

Fig.-11(b):-Logic waveforms to show the response b(t) to an i/p d(t).

 carrier signal as shown in Fig.-11(a). The
modulator o/p is transmitted as:-
vDPSK (t )  b(t ) 2 Ps cos 0t
  2 Ps cos 0t ...(l )

• A method of recovering data bit stream from

DPSK signal is shown in Fig.-11(c). The
multiplier o/p is:-
b(t )b(t  Tb )2 Ps cos(0t   ) cos[0 (t  Tb )   ]
   Tb  
 b(t )b(t  Tb ) Ps cos 0Tb  cos 20  t    2   ...(m)
   2 
• Then the multiplier o/p is applied to an integrate-&-dump
ckt as shown in Fig.-9(b) (right-side) to collect the binary
bit stream.

Fig.-11(c):- Method of recovering d(t) from DPSK signal

• The integrator will suppress the double frequency
term as we have selected ω0Tb = 2nπ, with n an
integer & cosω0Tb = 1.
• The transmitted data bit d(t) can readily be
determined from the product b(t)b(t-Tb). If d(t) = 0,
then there is no phase change & b(t) = b(t-Tb), i.e.,
b(t)b(t-Tb) = 1. If d(t) = 1, then there is a phase
change & b(t) = -b(t-Tb), i.e., b(t)b(t-Tb) = -1.
• In DPSK, a bit is determined on the basis of the
signal received in 2 successive bit intervals. Hence
noise in one bit interval may cause bit errors in pairs
in a noisy system. Thus, the error rate in DPSK is
greater than in BPSK.
 Quadrature Phase Shift Key (QPSK)
• The provision of reliable performance is exemplified
by a very low probability of error.
• Another goal of DCS is the efficient utilization of
channel BW.
• QPSK is an example of quadrature-carrier
multiplexing system to conserve BW & produces a
modulated wave described as follows:- s(t) =
sI(t)cos(2πfct) - sQ(t)sin(2πfct) …(n)
Where sI(t) is the in-phase component/bit of the
modulated wave & sQ(t) is the quadrature
component/bit.
• Both sine & cosine of carrier waves are phase-
 quadrature with each other & both sI(t) & sQ(t) are
related to the i/p data stream.
• In QPSK, the phase of the carrier takes on one of
4 equally spaced values, such as π/4, 3π/4, 5π/4,
& 7π/4, i.e.,

 2E  
 cos 2f c t  (2i  1)  ; 0  t  T
si (t )   T  4 ; i  1,2,3 & 4. ...(o)

 0 ; elsewhere

Here, E is the transmitted signal energy per

symbol, T (= nTc= n/fc) is the symbol duration for
some fixed integer n.
• Each possible value of the phase corresponds to a
unique pair of bits called a dibit. Here, we will
represent the Gray-encoded set of dibits: 10, 00,
01, &11.
 2E  
 cos (2i  1)  cos(2f ct )
 T  4
 si (t )   2E   ...(o)
  sin (2i  1)  sin(2f ct ) ; 0t T
 T  4
 0 ; elsewhere

• Here, 2 orthonormal basis functions, ф1 & ф2 are

required to represent the si where,
2 2
1 (t )  cos(ct ) & 2 (t )  sin(ct ) ; 0  t  T ...( p)
T T
    
 si (t )  E cos (2i  1)  1 (t )  E sin (2i  1) 2 (t )
 4  4
 si11 (t )  si 22 (t ) for i  1, 2, 3, & 4 ...(q)
• The si1 & si2 values are shown in Table-1.
• Thus, QPSK is characterized by a 2-dimensional
signal constellation having 4 message points (i.e.,
M = 4) as shown in Fig.-12(a).
• To detect the transmitted message points signal
space is divided into 4 regions as shown in Fig.-
12(a). Any point (si) falling to any region belongs
to it.
• Fig.-12(b) shows the QPSK waveforms for an i/p
binary sequence 01101000. The in-phase &
quadrature components of QPSK are 2 BPSK
signals, shown in Fig.-12(b) (b) & © respectively.
• We got the QPSK signal by adding both the in-
phase & quadrature components together.
 d = 2 Eb
s4
s3

s1
s2

Fig.-12(a):-Signal space diagram for Coherent QPSK system.

 I/p binary sequence.

Odd-numbered i/p bits & associated binary PSK wave.

Even-numbered i/p bits & associated binary PSK wave.

QPSK waveform.
Fig.-12(b):- An Example on QPSK waveforms.
• The average probability of symbol error, Pe for
coherent QPSK is:-
 Es   Eb 
Pe  1  Pc  erfc 

  erfc 
 

 ...(i )
 2N0   N0 
where Es = 2Eb with 2-sided PSD for a AWGN i/p
of N0/2.
• Fig.-12© shows the block diagram for the QPSK
transmitter & receiver. Considering polar coding
scheme for binary bits ‘0’ (by - Eb ) & ‘1’ (by
 Eb ), the i/p bit stream is divided into 2 separate
waves by a demultiplexer as odd & even bit wave.
 QPSK Transmitter

Fig.-12(c) :- b) QPSK receiver.

• Then the 2 binary waves b1(t) & b2(t) are used to
modulate a pair of quadrature carriers or
orthonormal basis functions ф1(t) & ф2(t)
respectively to produce the 2 BPSK signals,
which are added to produce the desired QPSK
wave as shown in Fig.-12© (a).
• For a given transmission BW, a QPSK wave
carries twice as many bits of information as the
corresponding BPSK wave.
• The QPSK receiver (as shown in Fig.-12© (b))
consists of a pair of correlators with a common
i/p & supplied with a locally generated pair of
coherent reference signals ф1(t) & ф2(t).
• The correlator o/ps x1 & x2, are each compared
with a threshold of 0 volts. If x1or 2 > 0, a decision
is made in favor of bit ‘1’ otherwise bit ‘0’.
Finally these binary bits combined in a
multiplexer to produce the original binary
sequence at the receiver with the minimum
probability of symbol error.
 M-ary PSK
• In BPSK we transmitted 2 symbols using the
sinusoids differing in phase by amount 2π/2 =
1800 whereas in QPSK we transmitted 4 symbols
each of 2-bit information using the sinusoids
differing in phase by amount 2π/4 = 900.
• Similarly we can extend the scheme to lump N-
bits together over the time NTb for 2N = M
possible symbols. Each symbol is represented by
sinusoids of duration NTb = Ts which differ from
one another by the phase = 2π/M, known as M-
ary PSK system.
• Thus, in M-ary PSK the waveforms, vm(t) used to
identify the symbols are:-
vm (t )  2 Ps cos(0t  m ) ; m  0, 1, 2, ..., M  1
...(r )

with phase angle:- m  (2m  1) ...( s)
M
• The voltage/signal waveforms are shown by the
dots over the circumference of the circle in Fig.-
13(a) using 2 orthonormal coordinates:-
u1 (t )  (2 / Ts ) cos 0t & u2 (t )  (2 / Ts ) sin 0t.
• The distance of each dot from the origin is
Es  PsTs .
Fig.-13(a):-Geometrical representation of M-ary PSK signals.
 vm (t )  2 Ps cos m cos 0t  2 Ps sin m sin 0t
 vm (t )  Pe cos 0t  Po sin 0t ...(r )
where, Pe  2 Ps cos m & Po  2 Ps sin m .
• Both Pe & Po can change every Ts = NTb
depending on any possible M value, hence Pe, Po
& ϕm are random processes. The PSDs of Pe & Po
are given by Eq.-(t).
 sin fTs 
Ge ( f )  Go ( f )  Es   ...(t )
 fTs 
• The modulated spectrum is
centered around carrier frequency, f0 with a BW,
 B = 2 fs = 2 fb/N ...(u)
• Thus we observed that by increasing the no. of
bits N per symbol the BW becomes progressively
smaller, but the distance, d between symbol
points becomes smaller as M increases
(observing Fig.-13(a)) where,
d 2 Es sin 2 ( / M )  2 NEb sin 2 ( / 2 N ) ...(v)
with Es = PsTs = NEb =NPsTb.
• Thus, as N increases probability of error becomes
higher & higher.
• M-ary PSK transmitter is shown in Fig.-13(b).
• The bit stream b(t) is applied to a N-bit serial to parallel
converter to convert digital-to-analog voltage v(sm) whose
phase is assigned by ϕm depending on value of m = 0,1, 2,
…, M-1.
• In M-ary PSK, the o/p is a fixed amplitude of sinusoid,
but phase is one-to-one correspondence as per N-bit
symbol. The
phase can
change
once
per
symbol
time.

Fig.-13(b):-M-ary PSK transmitter.

• The coherent M-ary PSK receiver is shown in Fig.-13©.

Fig.-13(c):-Coherent M-ary PSK receiver.