LECT 13&14synchronization

Chapter 5: Carrier & Symbol Synchronization
Graduate Program
Department of Electrical and Computer Engineering
Overview
Signal parameter estimation
Likelihood function
Carrier recovery & symbol synchronization in demodulation
Carrier phase estimation

Symbol timing estimation
Joint estimation of Carrier phase & symbol timing
Performance characteristics of ML estimators
Sem. I, 2012/13
Digital Communications Chapter 5: Carrier and Symbol Synchronization
Signal Parameter Estimation

Propagation delay from the transmitter is generally
unknown at the receiver
How to synchronously sample the output of the
demodulator?
Symbol timing must be derived or extracted from the received
signal
Moreover, frequency offset must be estimated at the receiver
for phase-coherent detection, which results from
Propagation delay
Frequency drift at the local oscillator
What are methods for carrier and symbol synchronization?
Sem. I, 2012/13

Assume the channel delays the transmitted signal and also
adds noise to it
Thus the received signal will be
r(t) = s(t ) + n(t)
s(t) = Re sl (t) e j 2 f ct
where
Where is propagation delay and sl(t) is the equivalent low pass

signal
We can also express r(t) as
r(t) = Re (sl (t ) e j + z(t) e j 2 f ct

Where =-2fc is the phase shift due to delay
Sem. I, 2012/13

Note that is a function of fc and
I.e., we need to estimate both fc and to know
The carrier signal generated at the receiver may in general

not be in synchronous with the transmitter
Over time the two oscillators may be drifting slowly in opposite
directions
Furthermore, the precision with which one may

synchronize in time depends on signal interval T
Estimation error in must be a small fraction of T
Usually 1% of T
But this level of precision may not be adequate in the

estimation of since fc is generally large and small
estimation error results in significant phase error
Sem. I, 2012/13

We have to estimate both and to demodulate and detect
the signal
Express the received signal as
r(t) = s(t; , ) + n(t)
And denote the parameter vector {,} by such that
s(t; , ) = s(t; )
Two criteria widely used in signal parameter estimation
1.
Maximum Likelihood (ML) criterion: is treated as deterministic

but unknown
2.
Maximum a posteriori Probability (MAP) criterion: is modeled as

random & characterized by a priori probability density function
p()
Sem. I, 2012/13
Carrier and Symbol Synchronization

Orthonormal expansion of r(t): Using N orthogonal
functions {fn(t)} we may represent r(t) by vector of
coefficients r [r1, r2, r3,.. rN]
In ML, the estimate of is the value that maximizes p(r)
In MAP the value of that maximizes the a posteriori probability
density function is sought
p( r ) =
p( r ) p( )
p( r )
In the absence of any prior knowledge of the properties of

, we can assume p() is uniform over a range of values of
the parameter
In such a case, the value of that maximizes p(r) also
maximizes p(r), i.e., MAP and ML are identical
Sem. I, 2012/13
Overview
Likelihood function
Carrier recovery and symbol synchronization in signal
Demodulation

Sem. I, 2012/13
The Likelihood Function

In what follows, we view the parameters and unknown
but deterministic
Hence, adopt the ML criterion in estimating them
Also the observation interval T0 T, also called one-shot

observation, is used as a basis for continuously updating the
estimate (tracking)
Since the additive noise n(t) is WG with zero mean
N
[rn sn ( )] 2
1
exp
2
2
2
n =1
p( r ) =
Where
rn =
r(t) f
T0
Sem. I, 2012/13
(t) dt
and
s n ( ) =
s(t; ) f
(t) dt
T0
The Likelihood Function

1
lim
N 2 2
[rn sn (t; )] 2
n =1
1
N0
2
[
]
r(t)
s(t;
)
dt
T0
( Show this?)
The maximization of p(r) with respect to the signal

parameter is equivalent to the maximization of the
likelihood function
2
( ) = exp [r(t) s(t; )] dt
N 0 T0
Sem. I, 2012/13
10
Overview
Likelihood function
Carrier recovery & symbol synchronization in demodulation

Sem. I, 2012/13
11
Carrier Recovery & Symbol Synchronization

Consider the binary PSK (or binary PAM) signal
demodulator and detector block diagram shown below
Carrier phase
estimate for
reference signal
generation for
correlator
Block diagram of a binary PSK receiver
Controls the sampler

and the digital pulse
generator
Sem. I, 2012/13
12

Carrier phase estimate is used in generating the phase
reference signal g (t ) cos(2f c t + ) for the correlator

Symbol synchronizer controls the sampler and the output
of the signal pulse generator
If g(t) is rectangular the signal generator can be omitted
The block diagram of an M-ary PSK demodulator is shown
in the next slide
Two correlators (or matched filters) are used to correlate
the received signal with the two quadrature carrier signals
Phase detector is used (compares the received signal
phases with the possible transmitted signal phases)
Sem. I, 2012/13
13
Block diagram of an M -ary PSK receiver

Sem. I, 2012/13
14

The same arrangement can be used for M-ary PAM by
introducing an automatic gain control at the front end and
making the detector an amplitude detector
Block diagram of an M-ary PAM receiver

Sem. I, 2012/13
15

The block diagram of a QAM demodulator is shown below
Block diagram of a QAM receiver

Sem. I, 2012/13
16
Overview
Maximum-Likelihood carrier phase estimation

Phase-locked loop
Effect of Additive noise on phase estimate
Decision-directed loops

Sem. I, 2012/13
17
Carrier Phase Estimation

There are two methods:
1. Use of pilot signal that allows the receiver to extract the
carrier frequency and phase of the received signal
Pilot signal is unmodulated carrier component that is tracked by a
Phase Locked Loop (PLL) which is designed to be narrow band
2. Derive the carrier phase estimate directly from the

modulated signal
Total transmitter power is used to transmit the information bearing
signal only
This is widely used in practice and in our analysis we assume the
signal is transmitted via suppressed carrier
Sem. I, 2012/13
18

As an illustration of the effect of phase error, consider the
demodulation of DSB/SC AM signal
s(t) = A(t) cos(2 f c t + )
Demodulate the signal using a carrier reference signal
c(t) = cos(2 f c t + )
such that
1
1
c(t) s(t) = A(t) cos( ) + A(t) cos(4 f c t + + )
2
2
The double frequency term is removed by the low pass filter

(integrator) such that the output is
1
y(t) = A(t) cos( )
2
Sem. I, 2012/13
19
Note that the effect of the error ( ) is to reduce the
amplitude by the factor cos( ) and power by the square

of this factor
Note 10o error 0.13dB and 30o 1.25 dB
The effect of phase error is much more severe in QAM and

multiphase PSK which are usually represented by
s(t) = A(t) cos(2 f c t + ) B(t) sin(2 f c t + )

This is demodulated using two quadrature carriers
c c ( t ) = cos( 2f c t + )
c s ( t ) = sin( 2f c t + )
Sem. I, 2012/13
20

Multiplying S(t) by cc(t) followed by low-pass filtering yields
the phase component
1
1
y I (t) = A(t) cos( ) + B(t) sin( )
2
2
And multiplying s(t) by cs(t) and low pass filtering yields the
quadrature component
1
1
y Q (t) = B(t) cos( ) + A(t) sin( )
2
2
Results:
Power reduction by a factor of cos ( )

Cross-talk interference from the in-phase and quadrature
components causing a higher degradation in performance
2
Sem. I, 2012/13
21
Overview

Phase-locked loop
Effect of Additive noise on phase estimate

Sem. I, 2012/13
22
Maximum Likelihood Carrier Phase Estimation

Assume the delay is constant and = 0
The likelihood function will be a function of and not of
1
2
(r(t) s(t, )) dt
( ) = exp
N 0 T0
1
2
1 2
2
r
(t)
dt
r(t)
s(t,
)
dt
s
(t,
)
= C exp
+
N 0 T
N
N
0
0
T
T
0
0
0
1st term is independent of and 3rd term is a constant and

equal to the energy over the observation time T0
r(t)
s(t,
)
dt
Hence, ( ) = C exp
N 0 T
0
Sem. I, 2012/13
23

C is a constant independent of
Equivalently, we can seek the value of that maximizes
log () such that
ln ( ) = L ( ) =
2
r(t) s(t, ) dt + lnC
N 0 T0
The ML estimate ML is the value of that maximizes ()
2
2
L ( ) =
r(t) s(t, ) dt + lnC
r(t) s(t, ) dt
N 0 T0
N 0 T0
Sem. I, 2012/13
24

Example: Consider the transmission of unmodulated signal
Acos2fct. The received signal is r(t)= Acos(2fct+)+n(t)
Then the log likelihood function will be
2A
L ( ) =
r(t) cos( 2 f c t + ) dt
N 0 T0
Differentiating L() and equating to zero we can find the

value of that maximizes the likelihood function
dL ( )
= r(t) sin(2 f c t + ML ) dt = 0; yields
d
T0
ML
r(t) sin 2 f c t dt
= tan 1 0
r(t) cos 2 f c t dt
Sem. I, 2012/13
25
Note that: r(t) sin(2 f c t + ML ) dt = 0 implies the use of a

T0
loop to extract the estimate as illustrated below

The loop filter is an integrator whose bandwidth is
proportional to the reciprocal of the integration interval To
A PLL for obtaining the ML estimate of the phase of an unmodulated carrier

Sem. I, 2012/13
26
Observe ML
= tan r(t) sin 2 f c t dt

T
0
1
T r(t) cos 2 f ct dt
0
A (one-shot) ML estimate of the phase of an unmodulated carrier

Sem. I, 2012/13
27
Overview
Maximum-likelihood carrier phase estimation

Phase-locked loop
Effect of additive noise on phase estimate

Sem. I, 2012/13
28
Phase-locked Loop
Phase-locked loop (PLL) consists of a multiplier, a loop filter,
and a voltage-controlled oscillator (VCO)
Assume that the input to thePLL is a cos(2fct+) and the
output of the VCO sin(2 f c t + )
Then
e(t) = cos(2 f c t + ) sin(2 f c t + )
Sem. I, 2012/13
1
1
= sin( ) + sin( 4 f c t + + )
2
2
29
Phase-locked Loop
The loop filter is a low-pass filter with transfer function
1 + 2s
G(s) =
1 + 1s
1 and 2 are design parameters (1 >> 2 ) that control the
bandwidth of the loop
Output of the loop filter gives control voltage (t) for VCO
The VCO is basically a sinusoidal signal generator with an
instantaneous phase given by
2 f c t + (t ) = 2 f c t + k v( ) d
where K is a gain constant in rad/V

Sem. I, 2012/13
30
Phase-locked Loop
Neglecting the double-frequency term, the PLL may be
implemented as shown below
It is a non-linear system for sin( )
The linearized PLL is characterized by the closed-loop

transfer function (see pages 342-343 of the text)
1 + 2s
H(s) =
1 + ( 2 + 1 /K) s + ( 1 /K) s 2
Where K is the gain parameter
31
Phase-locked Loop
Frequency response of the closed-loop transfer function
Sem. I, 2012/13
32
Overview

Phase-locked loop

Sem. I, 2012/13
33
Effect of Additive Noise on Phase Estimate

Assume narrowband noise at the input of the PLL and that
the PLL is tracking a sinusoidal signal of the form
s(t) = Ac cos(2 f c t + (t))

That is corrupted by additive narrowband noise
n(t) = x(t) cos 2 f c t y(t) sin 2 f c t

Where x(t) and y(t) are assumed to be statistically
independent, stationary and Gaussian with power spectral
density N0/2 W/Hz
Using trigonometric identities n(t) can be expressed as
n(t) = nc (t) cos(2 f c t + (t)) ns (t) sin(2 f c t + (t))

Sem. I, 2012/13
34

Where
nc (t) = x(t) cos (t) + y(t) sin (t)

ns (t) = x(t) sin (t) + y(t) cos (t)
Note that
nc (t) + jns (t) = ( x(t) + jY(t)) e j (t)

Such that nc(t) and ns(t) have the same statistical
properties as x(t) and y(t)
Sem. I, 2012/13
35

If s(t)+n(t) is multiplied by the output of VCO and the
double frequency terms are ignored, the input to the loop
filter is a noise corrupted signal
e(t) = Ac sin + nc (t) sin ns (t) cos

= Ac sin + n1 (t)
Where = - is the phase error
Equivalent PLL model with additive noise
36

If the power of the incoming signal Pc= Ac2 is larger than
the noise power, the PLL may be linearized by making
sin(t) (t)
Linearized PLL model with additive noise

Sem. I, 2012/13
37

The variance of the phase error (t), which is also the
variance of the VCO output phase, is given by
=
2
N o Beq
Ac2
N0 2
= 2 Beq
Ac 2
Beq One side equivalent noise bandwidth of the loop

N0Beq Total noise power within the bandwidth
Ac2 signal power
Note that since
1
Ac2
2
SNR L =
We have =
N 0 Beq
L
(Applies to cases where the SNR is large)

Sem. I, 2012/13
38

For the non-linear case (using sin(t)), the analysis is
mathematically tractable when the loop is 1st order (G(s)=1)
For this case, the probability density function of the phase
error has the form
exp(L cos )
p( ) =
2 I 0 (L )
Where I0(.) is the modified Bessel function of order zero
Exact values of the variance can be obtained from the
above for the 1st order PLL
The figure in the next slide shows this for 1/
The linear approximation is exact for > 3
Sem. I, 2012/13
39

If the SNR drops below a given level deterioration in the
performance results and the PLL looses lock (click effect)
Comparison of VCO
phase variance for exact
and approximate (linear
model) first-order PLL.
Sem. I, 2012/13
40
Overview
Maximum-likelihood carrier phase estimation

Phase-locked loop

Joint estimation of carrier phase & symbol timing
Sem. I, 2012/13
41
Decision-directed Loops
How to maximize () or L() when the signal s(t; )
carries the information sequence {In}
Carrier recovery when the signal is modulated uses
decision-directed loops
In such cases, one can use one of two approaches
1. Assume {In} is known or
2. Treat {In} as a random sequence and average it over its statistics
In decision-directed parameter estimation, we assume the

information sequence {In} over the observation interval has
been estimated
~
In the absence of demodulation error I n = I n
Sem. I, 2012/13
42
In this case s(t; ) is completely known except for the
carrier phase
Consider decision-directed phase estimate for linear
modulation technique for which the received equivalent low
pass may be expressed as
rl (t) = e j I n g(t nT)+ z(t) = sl (t) e j + z(t)
Where sl(t) is a known signal if the sequence {In} is

assumed known
Sem. I, 2012/13
43
The likelihood and log-likelihood functions for the equivalent
low pass signal are
1
j
( ) = Cexp Re
r
(t)
s
(t)
e
dt
l
l
N 0 T0
1
L ( ) = Re
r
(t)
s
l
l (t) dt
N 0 T0
j
e

Substituting for sl(t) and observation interval of T0=KT
j 1 K 1 (n +1)T
I n rl (t) g (t nT) dt
L ( ) = Re e
N 0 n =0
nT
j 1 K 1
= Re e
I n yn
N 0 n =0
Sem. I, 2012/13
44
Where
(n +1}T
yn =
r(t) g (t nT) dt is the output of a matched
nT
filter in the nth interval
Then, L ( ) = Re 1
N
0
I yn cos Im
n =0
N0
K 1
I yn sin
n =0
K 1
Differentiating L() with respect to and equating to zero,

we obtain the phase estimate as
ML
K 1
= tan Im I n yn
n =0
1
K 1
Re I n yn
n =0
This is called decision-directed (or decision feedback) carrier

phase estimate
Sem. I, 2012/13
45
Here E{ ML } = and the estimate is unbiased

Block diagram of double-sideband PAM signal receiver with
decision-directed carrier phase estimation
Sem. I, 2012/13
46
For a DSB PAM signal of A(t)cos(2fct+), where A(t)=Amg(t)
and g(t) is assumed rectangular pulse of duration T
Carrier recovery with a decision-feedback PLL is shown
below
Quadrature
carriers
Sem. I, 2012/13
47
Output of the first multiplier and input to the integrator is
given by
r(t) cos(2 f c t + ) = s (t) =
1
[A(t) + nc (t)]cos
2
1
ns (t) sin + double frequency terms
2
s(t) is used to recover information carried by A(t)=Amg(t)

Detector makes decision on received symbols every T sec.
In the absence of error it reconstructs A(t) free of any noise
Reconstructed signal multiplied by 2nd quadrature carrier

1
A(t){[A(t) + nc (t)] sin ns (t) cos }+ Double frequency terms
2
1
1
= A2 (t) sin + A(t)[nc (t) sin ns (t) cos ] + ...
2
2
e(t) =
Then:
Sem. I, 2012/13
48
The desired component
A2(t)sin , which
contains the phase
error, drives the loop
filter
Carrier recovery system
for M-ary PSK using
decision feedback PLL
is shown in the figure
Sem. I, 2012/13
49
For the case of M-ary PSK (DF PLL), the received signal is
demodulated to yield the phase estimate
2
m =
(m 1) m = 1,2,...... M
M
Which, in the absence of decision error is phase of the

transmitted signal m
The two outputs of quadrature multipliers are delayed by
symbol duration T & multiplied by cosm and sinm to yield
1.
1
r(t) cos(2 f c t + ) sinm = [Acosm + nc (t)] sinm cos( )
2
1
[Asinm + ns (t)] sinm sin( ) + double freq. terms
2
Sem. I, 2012/13
50
2.
1
r(t) sin(2 f c t + ) cosm = [Acosm + nc (t)]cosm sin( )
2
1
[Asinm + ns (t)]cosm cos( ) + double freq. terms
2
The error signal is the sum of these two which reduces to
1
1
e(t) = Asin( ) + nc (t) sin( m )
2
2
1
+ ns (t) cos( m ) + Double frequency terms
2
The error signal drives the loop filter that provides the control
signal to the VCO
Note that the two quadrature noise components are additive
Sem. I, 2012/13
51
There are no product of the noise terms and thus no
additional power loss associated with decision-feedback PLL
The ML estimate of given by the earlier equation is also
appropriate QAM
(Readings on non-decision directed loops such as the
squaring loop, Costas loop etc pages 352-358 of text)
Sem. I, 2012/13
52
Overview
ML timing estimation

Sem. I, 2012/13
53
Symbol Timing Estimation

Modulator output must be sampled periodically at the
symbol rate, i.e., at precise sampling time instants
tm = mT+
Where T is the symbol interval and is the nominal delay
Periodic sampling requires clock signal at the receiver

Extraction of clock signal called symbol synchronization or
timing recovery
Is critical for a synchronous digital communication system
The receiver must know:

The frequency 1/ T and
Where to take the samples within each symbol interval
The choice of sampling instant within the symbol interval of

duration T is called the timing phase
Sem. I, 2012/13
54
Symbol Timing Estimation

Symbol synchronization can be accomplished in one of the
following ways:
1. Tx and Rx clocks are synchronized to a master clock
which provides precise timing signal (VLF < 30 KHz)
2. Tx transmits the clock frequency (1/T) or (n/T) along with
the information symbol
Rx employs narrowband filter to extract the clock signal for

sampling (simple but power inefficient)
3. Clock signal is extracted from the received data symbol (our

focus is on this)
Will cover a decision-directed method next
Sem. I, 2012/13
55
Symbol Timing Estimation - ML Timing Estimation

Consider the problem of estimating the time delay for a
baseband PAM waveform
r(t; ) = s(t; ) + n(t)

where
s(t; ) = I n g(t nT )
Assume the information symbols from the output of the demodulator are known transmitted sequences
Then the log-likelihood function has the form
L ( ) = C L
r(t) s(t; ) dt
T0
= C L I n r(t) g(t nT ) dt = C L I n yn ( )
n
Sem. I, 2012/13
T0
56

Where yn(t) is defined as
yn ( ) = r(t) g(t nT ) dt
T0
A necessary and sufficient condition for to be the ML

estimate of is that
dL ( )
d
r(t) g(t nT ) dt
= In
d
d T0
n
d
= I n [ yn ( )] = 0
d
n
The above result suggests the implementation of the
tracking loop shown in next slide
Sem. I, 2012/13
57

The summation in the loop serves as the loop filter whose
bandwidth is controlled by the length of the sliding window
in the summation
Output of the loop filter drives the voltage controlled clock
(VCC) or VCO which in turn controls the sampling times for
the input to the loop
Decision-directed ML estimation of timing for baseband PAM

Sem. I, 2012/13
58

Note that since the information sequence {In} is assumed
known and used in the estimation of , the estimate is
decision-directed
This method can be extended to signal formats such as
QAM and PSK by using the equivalent low pass form of
these signals
Sem. I, 2012/13
59
Overview

Joint estimation of carrier phase and symbol timing
Sem. I, 2012/13
60
Joint Estimation of Carrier Phase and Symbol Timing

Joint ML estimation of two or more signal parameters
yields estimates that are good and usually better than
estimates from separate maximization of the likelihood
functions; i.e.,
2
2joint sep
The log-likelihood function for the phase and symbol timing

in equivalent low-pass signal is
1
*
L ( , ) = Re r(t) sl (t; , ) dt
N0 T
j
sl (t; , ) = e I n g(t nT ) + j J n w(t nT )
n
n
Sem. I, 2012/13
61
Joint Estimation of Carrier Phase and Symbol
Where {In} and {Jn} are the two information sequences

For PAM Jn =0, n and In is real
For QAM and PSK, we set Jn = 0, n and In is complex
For decision-directed ML estimation, the log-likelihood
function becomes
e j
L ( , ) = Re
N0
*
+
I
y
(
)
jJ
x
(
)
n n n
n n
Where
yn ( ) = r(t) g * (t nT ) dt
T
xn ( ) = r(t) w* (t nT ) dt
T
Sem. I, 2012/13
62

The ML estimate of and must simultaneously satisfy the
following two conditions
L ( , )
L ( , )
= 0 and
=0
Define
A( ) + jB( ) =
Such that
1
N0
(I
yn ( ) + jJ n* xn ( )
L ( , ) = A( ) cos B( ) sin
The condition for the joint ML becomes
L ( , )
= A( ) sin B( ) cos = 0
L ( , ) A( )
B( )
cos
sin = 0
=
Sem. I, 2012/13
63

The solution to the above equation results in
ML
And
B(
)
ML
= tan 1
A( ML )
A( )
B( )
+ B( )
=0
A( )
= ML
The decision directed tracking loop for QAM (or PSK)

obtained from the above equations is illustrated in the
figure in next slide
Note also that one can also combine the estimation
problem with demodulation of the information sequence {In}
Thus, one needs to consider the ML estimation of {In},
carrier phase , and symbol timing parameter
Sem. I, 2012/13
64
Decision-directed joint tracking loop for carrier phase and

symbol timing in QAM and PSK
Sem. I, 2012/13
65
Overview

Joint estimation of carrier phase & symbol timing
Sem. I, 2012/13
66
Performance Characteristics of ML Estimators

Measures for quality of parameter estimate: bias and variance
Consider the sequence of observations x =[x1, x2, x3,.xn]
with PDF p(x|) from which we extract
The bias of an estimate of is defined as
bias = E{ }
Where is the true value. If the bias is zero, then the
estimate is unbiased and the variance of the estimate is
= E{[ ] } = {E[ ]}2 ( Difficult to compute)

2
Usually a bound is sought for the mean square error

One possible lower bound is given by Cramer-Roa bound as follows
Sem. I, 2012/13
67
E{[ ] }
E{[ ] 2 }

E [
ln p(x )] 2
When the estimate is unbiased the numerator is unity and the

lower bound on the variance of the estimate becomes
E{[ ] 2 }
1

2
E [ lnp( x )]
And since ln p(x/) differs from L() by only a constant

factor
2
2
2
E [ ln p(x )] = E [ ln ( )] = E 2 ln ( )
Sem. I, 2012/13
68

Thus for unbiased estimators, the lower bound on variance is
1
2
E
2 ln (
The lower bound is very useful since it provides a benchmark for comparing any practical estimates
Any unbiased estimate whose variance achieves the lower
bound is called an efficient estimate
Such efficient estimates are rare
Sem. I, 2012/13
69

Example: ML estimate of the phase of unmodulated carrier is
satisfies
r(t) sin(2 f
+ ML ) dt = 0
T0
Where
r(t) = s(t; ) + n(t)

= Acos(2 f c + ) + n(t)
After some manipulation this reduces to

2
N0
N 0 2 T0
1
=
=
;
2
2
A T0 1 / 2 A
L
A2
where L =
is the loop SNR
2 N o Beq
Sem. I, 2012/13
70

Recall this is also the same variance obtained for the phase
estimate from the decision-directed PLL estimation
In addition to the dependence on SNR, the quality of symbol
timing estimate is also a function of the pulse shape
The most commonly used pulse shape is the raised cosine
spectrum. (See performance graph on page 370 of text)
Sem. I, 2012/13
71

LECT 13&14synchronization

Загружено:

Сведения о документе

Оригинальное название

Авторское право

Доступные форматы

Поделиться этим документом

Поделиться или встроить документ

Параметры публикации

Этот документ был вам полезен?

Это неприемлемый материал?

Авторское право:

Доступные форматы

LECT 13&14synchronization

Загружено:

Авторское право:

Доступные форматы

Chapter 5: Carrier & Symbol Synchronization

Carrier phase estimation

Digital Communications Chapter 5: Carrier and Symbol Synchronization

Signal Parameter Estimation

What are methods for carrier and symbol synchronization?

Digital Communications Chapter 5: Carrier and Symbol Synchronization

Signal Parameter Estimation

r(t) = s(t ) + n(t)

Where is propagation delay and sl(t) is the equivalent low pass

We can also express r(t) as

r(t) = Re (sl (t ) e j + z(t) e j 2 f ct

Digital Communications Chapter 5: Carrier and Symbol Synchronization

Signal Parameter Estimation

The carrier signal generated at the receiver may in general

Furthermore, the precision with which one may

But this level of precision may not be adequate in the

Digital Communications Chapter 5: Carrier and Symbol Synchronization

Signal Parameter Estimation

Maximum Likelihood (ML) criterion: is treated as deterministic

Maximum a posteriori Probability (MAP) criterion: is modeled as

Digital Communications Chapter 5: Carrier and Symbol Synchronization

Carrier and Symbol Synchronization

In the absence of any prior knowledge of the properties of

Digital Communications Chapter 5: Carrier and Symbol Synchronization

Carrier phase estimation

Digital Communications Chapter 5: Carrier and Symbol Synchronization

The Likelihood Function

Also the observation interval T0 T, also called one-shot

Digital Communications Chapter 5: Carrier and Symbol Synchronization

The Likelihood Function

The maximization of p(r) with respect to the signal

Digital Communications Chapter 5: Carrier and Symbol Synchronization

Carrier phase estimation

Digital Communications Chapter 5: Carrier and Symbol Synchronization

Carrier Recovery & Symbol Synchronization

Block diagram of a binary PSK receiver

Controls the sampler

Digital Communications Chapter 5: Carrier and Symbol Synchronization

Carrier and Symbol Synchronization

reference signal g (t ) cos(2f c t + ) for the correlator

Digital Communications Chapter 5: Carrier and Symbol Synchronization

Carrier and Symbol Synchronization

Block diagram of an M -ary PSK receiver

Digital Communications Chapter 5: Carrier and Symbol Synchronization

Carrier and Symbol Synchronization

Block diagram of an M-ary PAM receiver

Digital Communications Chapter 5: Carrier and Symbol Synchronization

Carrier and Symbol Synchronization

Block diagram of a QAM receiver

Digital Communications Chapter 5: Carrier and Symbol Synchronization

Maximum-Likelihood carrier phase estimation

Symbol timing estimation

Digital Communications Chapter 5: Carrier and Symbol Synchronization

Carrier Phase Estimation

2. Derive the carrier phase estimate directly from the

Digital Communications Chapter 5: Carrier and Symbol Synchronization

Carrier Phase Estimation

Demodulate the signal using a carrier reference signal

The double frequency term is removed by the low pass filter

Digital Communications Chapter 5: Carrier and Symbol Synchronization

Carrier Phase Estimation

Note that the effect of the error ( ) is to reduce the

amplitude by the factor cos( ) and power by the square