Basic Digital
Passband Modulation
Reference : HA H. NGUYEN and ED SHWEDYK A First Course in Digital Communications , 2009
Binary amplitude-shift keying (BASK)
s(t) = m(t)c(t)
where m(t) is the modulating signal (the baseband signal, an NRZ
signal) and c(t) = V cos(2πfct) is the sinusoidal carrier.
Optimum Receiver
Threshold
BASK signaling scheme: (a) signal space plot; (b) optimum receiver
implementation; (c) decision regions
Error probability
Power Spectral Density (PSD)
PSD of BASK
Approximately 95% of the total transmitted
power lies in a band of 3/Tb (hertz),
centered at fc.
Binary phase-shift keying (BPSK)
A BPSK signal is generated by amplitude modulating
the sinusoidal carrier with a NRZ-L signal of amplitude
±1. The transmitted signal is s(t) = m(t)c(t) (where m(t) is
a NRZ-L signal) with a resultant phase that is either 0 or
π radians.
Signal space plot of BPSK.
PSD for the BPSK
The PSD of BPSK is similar to that of BASK except that
there are no impulse functions at ±fc, reflecting the fact
that there is no power at the carrier. This is reasonable
since BPSK is really a “double” sideband suppressed
carrier modulation.
Binary frequency-shift keying (BFSK)
Simple BFSK
modulators:
(a) by gating
two oscillators;
(b) using a
voltage
controlled
oscillator
(VCO).
Two oscillators
s1(t) and s2(t) are orthogonal
over the interval [0, Tb]
If the two phases are the same
If the two phases are different
VCO
where n and m are positive integers, and n m.
Signal space plot and decision regions of BFSK.
Performance comparison of BASK, BPSK, and
BFSK
BPSK is 3 dB more efficient than BFSK, which has the
same performance as BASK.
BFSK occupies a larger bandwidth than BPSK and BASK
(recall that BPSK and BASK occupy the same bandwidth).
Each of the three modulation techniques has a spectrum
that decays as 1/f ^2 for frequencies away from the
carrier, reflecting the fact that for each modulation the
transmitted signal has discontinuities.
Digital modulation techniques for
spectral efficiency
Quadrature phase-shift keying (QPSK)
QPSK signals and a mapping to the messages
An example of a QPSK signal
The basic idea behind QPSK exploits the fact that
cos(2πfct) and sin(2πfct) are orthogonal over the
interval [0, Tb] when fc = k/Tb, k integer.
Therefore only two orthonormal functions are needed to
represent the four signals
Signal space plot of QPSK modulation.
Optimum Receiver for QPSK
The optimum receiver is derived by expanding the received
signal r(t) = si(t) + w(t) over the interval of Ts seconds
into a series as follows:
The criterion will be to find a receiver that minimizes the
symbol (message) error probability.
Rather than minimizing the error, consider instead the
equivalent criterion, that of maximizing the probability of
a correct decision.
Assign the observation vector r = (r1, r2, . . . , rm)
in the m-dimensional signal space to the region
for which the integrand Pi f(r|si(t)) is the largest.
P1 = P2 = P3 = P4
Minimum-distance receiver
P[error] = P[error|si(t)] = 1 − P[correct|si(t)].
Decision regions of the minimum-distance receiver of
QPSK
Signal space diagram of QPSK: to compute P[correct|
s1(t)] one finds the volume of f(ˆr1, ˆr2|s1(t)) over the
shaded quadrant
same as that of BPSK
An alternative representation of QPSK
Inphase carrier V cos(2πfct)
Quadrature carrier V sin(2πfct),
A different block
diagram
of a QPSK
modulator
Signal space plots for inphase and quadrature bit streams
Receiver implementation for QPSK
Offset quadrature phase-shift keying (OQPSK)
To prevent the phase change of π in QPSK, offset (or
staggered) quadrature phase-shift keying (OQPSK) is
used so that signal amplification can be done more
efficiently.
OQPSK differs from QPSK only in that in OQPSK
the aI(t) and aQ(t) bit streams are offset by one
bit interval Tb.
Minimum shift keying (MSK)
The transmitted signals s(t) have sudden jumps at
multiples of symbol duration for QPSK), or multiples of bit
duration (for OQPSK).
The two carriers V cos(2πfct) and Vsin(2πfct) are
weighted by sinusoids of frequency 1/(4Tb).
The signals are orthogonal over the interval of Tb
seconds, or any integer multiple of Tb
The bit error probability of MSK is the same as that of
BPSK, QPSK, and OQPSK
The expression in (7.64) shows that s(t) not only has a
constant envelope, but also a continuous
phase. Furthermore, the transmitted signal is of either
frequency f2 = fc + 1/4Tb or frequency f1 = fc −
1/4Tb depending on the ratio aQ(t)/aI (t).
Thus the transmitted signal may be
considered to be a frequency shift keying
signal with continuous phase (CPFSK).
Note also that the frequency separation is
f2 − f1 = 1/2Tb, which is the minimum
separation possible for the two sinusoidal
carriers to be “coherently” orthogonal.
This explains the name “minimum shift
keying” of the modulation scheme.
