Page 320 - Satellite Communications, Fourth Edition

300 Chapter Ten

sample-and-hold circuit to be accurately synchronized to the bit rate,
which necessitates a bit timing recovery (BTR) section, as shown in
Fig. 10.13.
Thermal noise at the receiver will result in noise phase modulation
of the carrier, and so the demodulated waveform p′(t) will be accompa-
nied by noise. The noisy p′(t) signal is passed into the threshold detec-
tor which regenerates a noise-free output but one containing some bit
errors as a result of the noise already present on the waveform.
The QPSK signal has many features in common with BPSK and will
be examined before describing in detail the carrier and bit timing recov-
ery circuits and the effects of noise.

10.6.2 Quadrature phase-shift keying
With QPSK, the binary data are converted into 2-bit symbols which are
then used to phase modulate the carrier. Since four combinations con-
taining 2 bits are possible from a binary alphabet (logical 1s and 0s), the
carrier phase can be shifted to one of four states.
Figure 10.14a shows one way in which QPSK modulation can be
achieved. The incoming bit stream p(t) is converted in the serial-to-
parallel converter into two binary streams. The conversion is illustrated
by the waveforms of Fig. 10.14b. For illustration purposes, the bits in
the p(t) waveform are labeled a, b, c, d, e,and f. The serial-to-parallel
converter switches bit a to the I port and at the same time switches bit
b to the Q port. In the process, each bit duration is doubled, so the bit
rates at the I and Q outputs are half that of the input bit rate.
The p (t) bit stream is combined with a carrier cos t in a BPSK mod-
i
0
ulator, while the p (t) bit stream is combined with a carrier sin t, also
0
q
in a BPSK modulator. These two BPSK waveforms are added to give the
QPSK wave, the various combinations being shown in Table 10.1.
The phase-modulation angles are shown in the phasor diagram of
Fig. 10.15. Because the output from the I port modulates the carrier
directly, it is termed the in-phase component, and hence the designation
I. The output from the Q port modulates a quadrature carrier, one which
is shifted by 90° from the reference carrier, and hence the designation Q.
Because the modulation is carried out at half the bit rate of the incom-
ing data, the bandwidth required by the QPSK signal is exactly half that
required by a BPSK signal carrying the same input data. This is the
advantage of QPSK compared with BPSK modulation. The disadvantage
is that the modulator and demodulator circuits are more complicated,
being equivalent essentially to two BPSK systems in parallel.
Demodulation of the QPSK signal may be carried out by the circuit
shown in the block schematic of Fig. 10.16. With the incoming carrier
represented as p (t)cos 0 t p q (t) sin 0 t , it is easily shown that after
i