5.

1 Digital Modulation tradeoffs

There is a fundamental tradeoff in communication

systems. Simple hardware can be used in transmitters and receivers
to communicate information. However, this uses a lot of spectrum
which limits the number of users. Alternatively, more complex
transmitters and receivers can be used to transmit the same
information over less bandwidth. The transition to more and more
spectrally efficient transmission techniques requires more and more
complex hardware. Complex hardware is difficult to design, test, and
build. This tradeoff exists whether communication is over air or wire,
analog or digital.

Fig 5.1 The Fundamental Trade-off

Over the past few years a major transition has occurred from simple
analog Amplitude Modulation (AM) and Frequency/Phase Modulation
(FM/PM) to new digital modulation techniques.

Examples of digital modulation include

• QPSK (Quadrature Phase Shift Keying)

• FSK (Frequency Shift Keying)
• MSK (Minimum Shift Keying)
• QAM (Quadrature Amplitude Modulation)
 Fig 5.2 Trends in the Industry

Complex tradeoffs in frequency, phase, timing, and modulation are

made for interference-free, multiple-user communications systems. It
is necessary to accurately measure parameters in digital RF
communications systems to make the right tradeoffs. Measurements
include analyzing the modulator and demodulator, characterizing the
transmitted signal quality, locating causes of high Bit-Error-Rate and
investigating new modulation types. Measurements on digital RF
communications systems generally fall into four categories: power,
frequency, timing, and modulation accuracy.
5.2 Optimum demodulation of digital signal over band limited channels

All physical channels are band limited, with C(f) = 0 for |f| > W

Nondistorting (ideal) channel: |C(f)| = const. for | f | < W and is

linear

All other channels are nonideal (distort the signal in amplitude,

phase or both)
Channel is assumed to be ideal band-limited

To find pulse shapes that will – Result in zero ISI at the receiver –

Allow us to achieve maximum possible transmission rate with zero

ISI (optimum utilization of the given bandwidth)

The condition for zero ISI:

Consider transmitting data at rate 1/T through a channel with

bandwidth W or through a distorting channel. We would like to find
the optimum (minimum sequence error probability) receiver. Assume
the modulator is filter acting on a infinite sequence of impulses (at
rate 1/T with impulse response f (t). The channel is characterized by
an impulse response of g(t) and the receiver is a filter sampled at rate
1/T with impulse response h(t).

where um is the data symbol transmitted during the m-th signaling

interval assumed to be in the alphabet A and f(t) is the waveform
used for transmission. We assume a transmission of 2N+1 data
symbols (think of N as being very large). The output of the channel
filter is then
5.3 Maximum likelihood sequence detection (Viterbi receiver)

When the signal has no memory, the symbol-by-symbol detector

is used for optimum in the sense of minimizing the probability of a
symbol error. On the other hand, when the transmitted signal has
memory, i.e., the signals transmitted in successive symbol intervals
are interdependent and then the optimum detector is a detector that
bases its decisions on observation of a sequence of received signals
over successive signal intervals.

In this section, we describe a maximum-likelihood sequence

detection algorithm that searches for the minimum Euclidean
distance path through the trellis that characterizes the memory in
the transmitted signal. Another possible approach is a maximum a
posteriori probability algorithm that makes decisions on a symbol-
by-symbol basis, but each symbol decision is based on an
observation of a sequence of received signal vectors.

The Maximum Likelihood Sequence Detector

Modulation systems with memory can be modeled as finite-state
machines which can be represented by a trellis, and the transmitted
signal sequence corresponds to a path through the trellis. Let us
assume that the transmitted signal has a duration of K symbol
intervals. If we consider transmission over K symbol intervals, and
each path of length K through the trellis as a message signal, then
the problem reduces to the optimal detection problem.

The number of messages in this case is equal to the number of paths

through the trellis, and a maximum likelihood sequence detection
(MLSD)algorithm selects the most likely path (sequence)
corresponding to the received signal r(t) over the K signaling interval.
As we have seen before, ML detection corresponds to selecting a path
of K signals through the trellis such that the Euclidean distance
between that path and r(t) is minimized.

Note that since

As an example of the maximum-likelihood sequence detection
algorithm, let us consider the NRZI signal. Its memory is
characterized by the trellis shown in Figure. The signal transmitted
in each signal interval is binary PAM. Hence, there are two possible
transmitted signals corresponding to the signal points

where εb is the energy per bit.

In searching through the trellis for the most likely sequence, it may
appear that we must compute the Euclidean distance for every
possible sequence. For the NRZI example, which employs binary
modulation, the total number of sequences is 2K. However, this is
not the case. We may reduce the number of sequences in the trellis
search by using the Viterbi algorithm to eliminate sequences as new
data are received from the demodulator.
The Viterbi algorithm is a sequential trellis search algorithm for
performing ML sequence detection. It is a decoding algorithm for
convolutional codes. We describe it below in the context of the NRZI
signal detection. We assume that the search process begins initially
at state S0. The corresponding trellis is shown in Figure.

By using the outputs r1 and r2 from the demodulator. The Viterbi

algorithm compares these two metrics and discards the path having
the larger (greater-distance) metric. The other path with the lower
metric is saved and is called the survivor at t = 2T. The elimination of
one of the two paths may be done without compromising the
optimality of the trellis search, because any extension of the path
with the larger distance beyond t = 2T will always have a larger
metric than the survivor that is extended along the same path
beyond t = 2T.
Similarly, for the two paths entering node S1 at t = 2T, we compute
the two Euclidean distance metrics

by using the outputs r1 and r2 from the demodulator. The two metrics
are compared, and the signal path with the larger metric is
eliminated. Thus, at t = 2T, we are left with two survivor paths, one at
node So and the other at node S1 and their corresponding metrics.
The signal paths at nodes So and S1 are then extended along the two
survivor paths.

Upon receipt of r3 at t = 3T, we compute the metrics of the two paths

entering state S0. Suppose the survivors at t = 2T are the paths (0, 0)
at So and (0, 1) at s1. Then the two metrics for the paths entering So
at t = 3 T are

These two metrics are compared, and the path with the larger
(greater-distance) metric is eliminated. This process is continued as
each new signal sample is received from the demodulator. Thus, the
Viterbi algorithm computes two metrics for the two signal paths
entering a node at each stage of the trellis search and eliminates one
of the two paths at each node. The two survivor paths are then
extended forward to the next state.

Therefore, the number of paths searched in the trellis is reduced by a

factor of 2 at each stage. It is relatively easy to generalize the trellis
search performed by the Viterbi algorithm for M-ary modulation. For
example, consider a system that employs M = 4 signals and is
characterized by the four-state trellis shown in Figure. We observe
that each state has two signal paths entering and two signal paths
leaving each node. The memory of the signal is L = 1. Hence, the
Viterbi algorithm will have four survivors at each stage and their
corresponding metrics. Two metrics corresponding to the two
entering paths are computed at each node, and one of the two signal
paths entering the node is eliminated at each state of the trellis.
Thus, the Viterbi algorithm minimizes the number of trellis paths
searched in performing ML sequence detection.
From the description of the Viterbi algorithm given above, it is
unclear how decisions are made on the individual detected
information symbols given the surviving sequences. If we have
advanced to some stage, say K, where K>>L in the trellis, and we
compare the surviving sequences, we shall find that with high
probability all surviving sequences will be identical in bit (or symbol)
positions K- 5L and less. In a practical implementation of the Viterbi
algorithm, decisions on each information bit (or symbol) are forced
after a delay of 5L bits (or symbols), and hence the surviving
sequences are truncated to the 5L most recent bits (or symbols).
Thus, a variable delay in bit or symbol detection is avoided. The loss
in performance resulting from the suboptimum detection procedure
is negligible if the delay is at least 5L. This approach to
implementation of Viterbi algorithm is called path memory truncation.
 5.4 Equalization techniques

Equalization is a technique used to combat intersymbol

interference (ISI). An Equalizer within a receiver compensates for the
average range of expected channel amplitude and delay
characteristics. ‰

Equalizers must be adaptive as the channel is generally unknown

and time varying. ‰ISI has been recognized as the major obstacle to
high speed data transmission over mobile radio channels.
The equalizer is a device that attempts to reverse the distortion
incurred by a signal transmitted through a channel. In digital
communication its purpose is to reduce inter symbol interference to
allow recovery of the transmit symbols.

The goal of equalizers is to eliminate intersymbol interference (ISI)

and the additive noise as much as possible.
Intersymbol interference arises because of the spreading of a
transmitted pulse due to the dispersive nature of the channel, which
results in overlap of adjacent pulses.

In Figure, there is a four‐level pulse amplitude modulated signal

(PAM), x(t). This signal is transmitted through the channel with
impulse response h(t). Then noise n(t) is added. The received signal
r(t) is a distorted signal.

Equalizers are used to overcome the negative effects of the channel.

In general, equalization is partitioned into two broad categories;

1. Maximum likelihood sequence estimation (MLSE) which entails

making measurement of channel impulse response and then
providing a means for adjusting the receiver to the transmission
environment. (Example: Viterbi equalization)
2. Equalization with filters uses filters to compensate the distorted
pulses.

It can be a simple linear filter or a complex algorithm. The types of

commonly used equalizers in digital communications are

Linear Equalizer:
It processes the incoming signal with a linear filter.

MSME equalizer:
It designs the filter to minimize E[|e|2], where e is the error
signal that is the filter output minus the transmitted signal.

Zero forcing Equalizer:

It approximates the inverse of the channel with a linear filter.

Decision feedback equalizer:

It augments a linear equalizer by adding a filtered version of
previous symbol estimates to the Original filter output filter.

Blind Equalizer:
It estimates that the transmitted signal without knowledge of
the channel statistics and uses only knowledge of the transmitted
signal‘s statistics.

Adaptive Equalizer:
It is typically a linear equalizer or a DFE, which updates the
equalizer parameters (such as the filter coefficients) as it is processes
the data. It uses the MSE cost function and it assumes that it makes
the correct symbol decisions and uses its estimate of the symbols to
compute e which is defined above.

Viterbi Equalizer:
It finds the optimal solution to the equalization problem. It is
having a goal to minimize the probability of making an error over the
entire sequence.

BCJR Equalizer:
It uses the BCJR algorithm whose goal is to minimize the
probability that a given bit was incorrectly estimated.
Turbo Equalizer:
It applies turbo decoding while treating the channel as a
convolutional code.

Depending on the time nature, These type of equalizers can be

grouped as preset or adaptive equalizers.

Preset equalizers assume that the channel is time invariant and try
to find H(f) and design equalizer depending on H(f). The examples of
these ADAPTIVE EQUALIZERS are zero forcing equalizer, minimum
mean square error equalizer, and decision feedback equalizer.

Adaptive equalizers assume channel is time varying channel and try

to design equalizer filter whose filter coefficients are varying in time
according to the change of channel, and try to eliminate ISI and
additive noise at each time. The implicit assumption of adaptive
equalizers is that the channel is varying slowly.

As the mobile fading channels are random and time varying,

equalizers must track the time varying characteristics of the mobile
channel, and thus are called adaptive equalizers.

The working principles of adaptive equalizers are:

The received signal is applied to receive filter. In here, receive filter is

not matched filter. Because we do not know the channel impulse
response. The receive filter in here is just a low‐pass filter that rejects
all out of band noise.
The output of the receiver filter is sampled at the symbol rate or
twice the symbol rate.
Sampled signal is applied to adaptive transversal filter equalizer.
Transversal filters are actually FIR discrete time filters.
The object is to adapt the coefficients to minimize the noise and
intersymbol interference (depending on the type of equalizer) at the
output.
The adaptation of the equalizer is driven by an error signal.

There are two modes that adaptive equalizers work;

1. Decision Directed Mode:

This means that the receiver decisions are used to
generate the error signal.

2. Decision directed equalizer adjustment is effective in tracking slow

variations in the channel response.
However, this approach is not effective during initial acquisition.

Training Mode: To make equalizer suitable in the initial acquisition

duration, a training signal is needed. In this mode of operation, the
transmitter generates a data symbol sequence known to the receiver.
The receiver therefore, substitutes this known training signal in place
of the slicer output. Once an agreed time has elapsed, the slicer
output is substituted and the actual data transmission begins.

Training mode
Initially, a known, fixed length training sequence is sent by the
transmitter so that the receiver‘s equalizer may average to a proper
setting. The training sequence is a pseudo random signal or a fixed,
prescribed bit pattern. Immediately following the training sequence,
the user data is sent.

Tracking mode
When the data of the users are received, the adaptive algorithm
of the equalizer tracks the changing channel. As a result of this, the
adaptive equalizer continuously changes the filter characteristics
over time. Equalizers are widely used in TDMA Systems .

The training sequence is designed to permit an equalizer at the

receiver to acquire the proper filter coefficients in the worst possible
channel conditions. Therefore when the training sequence is finished.
Therefore filter coefficients are near their optimal values for reception
of user data. An adaptive equalizer at the receiver uses a recursive
algorithm to evaluate the channel and estimate filter coefficients to
compensate for the channel.

Adaptive equalizer:

A Generic Adaptive Equalizer Transversal filter with N delay

elements, N+1 taps, and N+1 tunable complex weights .These
weights are updated continuously by an adaptive algorithm.
Algorithm for Adaptive Equalization
 Performance measures for an algorithm
 Rate of convergence
 Misadjustment
 Computational complexity
 Numerical properties

Factors dominate the choice of an equalization structure and its

algorithm
 The cost of computing platform
 The power budget
 The radio propagation characteristics

Algorithm for Adaptive Equalization

 The speed of the mobile unit determines the channel fading rate
and the Doppler spread, which is related to the coherent time of
the channel directly.
 The choice of algorithm, and its corresponding rate of
convergence, depends on the channel data rate and coherent
time.
 The number of taps used in the equalizer design depends on the
maximum expected time delay spread of the channel.
 The circuit complexity and processing time increases with the
number of taps and delay elements.

Algorithm for Adaptive Equalization

Three classic equalizer algorithms: zero forcing (ZF), least mean
squares (LMS), and recursive least squares (RLS) algorithms.

Linear equalizers: suffer from noise enhancement

DFE: Error propagation

MLSE
 Optimal method
 Viterbi equalizer implements MLSE with much lower
complexity.
Comparison of various algorithms for adaptive equalization

Adaptive Decision Feedback Equalizer • A decision feedback equalizer

(DFE) is a nonlinear equalizer that uses previous detector decisions
to eliminate the ISI on pulses that are currently being demodulated. •
The basic idea of a DFE is that if the values of the symbols previously
detected are known (past decisions are assumed to be correct), then
the ISI contributed by these symbols can be canceled out eactly the
output of the forward filter by subtracting past symbols values with
appropriate weighting.
If we look at Figure, we see that the estimated signal sequence
becomes,

{ci}s are coefficients of the precursor equalizer, {d i} are coefficients of

the postcursor equalizer. N is the number of precursor equalizer
coefficients and M is the number of postcursor equalizer coefficients.
Adaptive DFE algorithm is similar to stochastic gradient algorithm,
with the important difference that the input to the causal portion of
the filter is the decisions rather than the output of the precursor
equalizer filter. This difference will obviously change the desired tap
coefficients as well as reduce the noise enhancement due to
equalization.

The derivation of a stochastic gradient algorithm for the DFE is a

simple extension of the stochastic gradient algorithm for linear case.

An augmented vector of N +M coefficients

 5.5 Synchronization and carrier recovery of digital modulation.

In a digital communication system, the output of the

demodulator must be sampled periodically, once per symbol interval,
in order to recover the transmitted information. Since the
propagation delay from the transmitter to the receiver is generally
unknown at the receiver, symbol timing must be derived from the
received signal in order to synchronously sample the output of the
demodulator. The propagation delay in the transmitted signal also
results in a carrier offset, which must be estimated at the receiver if
the detector is phase-coherent.

Symbol synchronization is required in every digital communication

system which transmits information synchronously.
Carrier recovery is required if the signal is detected coherently.

Signal Parameter Estimation

We assume that the channel delays the signals transmitted

through it and corrupts them by the addition of Gaussian noise.
The received signal may be expressed as

where τ : propagation delay sl(t): the equivalent low-pass signal

The received signal may be expressed as:

where the carrier phase φ, due to the propagation delay τ, is

φ = -2πfcτ.

It may appear that there is only one signal parameter to be

estimated, the propagation delay, since one can determine φ from
knowledge of fc and τ.
However, the received carrier phase is not only dependent on the
time delay τ because:
The oscillator that generates the carrier signal for demodulation
at the receiver is generally not synchronous in phase with that at the
transmitter. The two oscillators may be drifting slowly with time.

The precision to which one must synchronize in time for the purpose
of demodulating the received signal depends on the symbol interval
T. Usually, the estimation error in estimating τ must be a relatively
small fraction of T.

±1 percent of T is adequate for practical applications. However, this

level of precision is generally inadequate for estimating the carrier
phase since fc is generally large.

In effect, we must estimate both parameters τ and φ in order to

demodulate and coherently detect the received signal.

Hence, we may express the received signal as

where φ and τ represent the signal parameters to be estimated.

To simplify the notation, we let ψ denote the parameter vector {φ, τ},
so that s(t; φ, τ) is simply denoted by s(t; ψ).

There are two criteria that are widely applied to signal parameter
estimation: the maximum-likelihood (ML) criterion and the maximum
a posteriori probability (MAP) criterion.

In the MAP criterion, ψ is modeled as random and characterized by

an a priori probability density function p(ψ).

In the ML criterion, ψ is treated as deterministic but unknown.

By performing an orthonormal expansion of r(t) using N orthonormal

functions {fn(t)}, we may represent r(t) by the vector of coefficients [r1
r2 ¸¸¸ rN] ≡ r.
The joint PDF of the random variables [r1 r2 ¸¸¸ rN] in the expansion
can be expressed as p(r| ψ).

The ML estimate of ψ is the value that maximizes p(r| ψ). The MAP
estimate is the value of ψ that maximizes the a posteriori probability
density function

If there is no prior knowledge of the parameter vector ψ, we may

assume that p(ψ) is uniform (constant) over the range of values of the
parameters.

In such a case, the value of ψ that maximizes p(r| ψ) also maximizes

p(ψ|r). Therefore, the MAP and ML estimates are identical. In our
treatment of parameter estimation given below, we view the
parameters φ and τ as unknown, but deterministic. Hence, we adopt
the ML criterion for estimating them.

In the ML estimation of signal parameters, we require that the

receiver extract the estimate by observing the received signal over a
time interval T0 ≥ T, which is called the observation interval.

Estimates obtained from a single observation interval are sometimes

called one-shot estimates.

In practice, the estimation is performed on a continuous basis by

using tracking loops (either analog or digital) that continuously
update the estimates.

The Likelihood Function Since the additive noise n(t) is white and
zero-mean Gaussian, the joint PDF p(r|ψ) may be expressed as
where T0 represents the integration interval in the expansion of r(t)
and s(t; ψ). By substituting from Equation (B) into Equation (A):

Now, the maximization of p(r|ψ) with respect to the signal

parameters ψ is equivalent to the maximization of the likelihood
function.

Below, we shall consider signal parameter estimation from the

viewpoint of maximizing Λ(ψ).

Carrier Recovery and Symbol Synchronization in Signal

Demodulation

Binary PSK signal demodulator and detector:

The carrier phase estimate is used in generating the ∧ reference
signal g(t)cos(2πfct+φ) for the correlator. The symbol synchronizer
controls the sampler and the output of the signal pulse generator. If
the signal pulse is rectangular, then the signal generator can be
eliminated.
QAM signal demodulator and detector

An AGC is required to maintain a constant average power signal at

the input to the demodulator. The demodulator is similar to a PSK
demodulator, in that both generate in-phase and quadrature signal
samples (X, Y) for the detector. The detector computes the Euclidean
distance between the received noise-corrupted signal point and the M
possible transmitted points, and selects the signal closest to the
received point.

Carrier Phase Estimation

Two basic approaches for dealing with carrier synchronization at the

receiver:
One is to multiplex, usually in frequency, a special signal, called a
pilot signal that allows the receiver to extract and to synchronize its
local oscillator to the carrier frequency and phase of the received
signal.
When an unmodulated carrier component is transmitted along with
the information-bearing signal, the receiver employs a phase-locked
loop (PLL) to acquire and track the carrier component. The PLL is
designed to have a narrow bandwidth so that it is not significantly
affected by the presence of frequency components from the
information-bearing signal.