Atec Rohde-Schwarz-SFQ Series - Manual (Page 6)

6 TV Test Transmitter R&S ® SFQ
DVB: coding and mapping for
antenna, satellite and cable
The I/Q coders of the TV Test Transmitter
R&S SFQ encode the applied transport
stream for terrestrial transmission via
antenna or for satellite or cable transmis-
sion in line with standards and condition
it so that I and Q (inphase and quadra-
ture) signals are obtained. The R&S SFQ
accepts MPEG transport streams with a
packet length of 188 or 204 bytes. The
input interfaces are synchronous parallel
(TS parallel, SPI) and asynchronous serial
(ASI). The input data rate and the symbol
rate for DVB-C, DVB-S and DVB-DSNG
modulation are selectable. With DVB-T
modulation, the channel bandwidths of
6 MHz, 7 MHz and 8 MHz can be selected;
their default settings can be varied.
Instead of the external transport data
stream (DATA) being used, an internal
data source can generate null transport
stream packets (NULL TS PACKET, as
defined in the DVB Measurement
Guidelines), or an unpacketed random
sequence (PRBS). The PRBS sequence is
also available in packeted form in the null
transport stream packets (NULL PRBS
PACKET). The R&S SFQ warns the user if
the input signal fails, the set data rate
does not match the incoming one or the
USEFUL DATA RATE is too high.
The input data stream is linked to a ran-
dom sequence, ensuring that the signal
energy is evenly distributed (energy
dispersal). Energy dispersal can be
switched off. The same applies to SYNC
BYTE inversion.
Following energy dispersal, a Reed-
Solomon coder (204,188) is provided as an
outer encoder for forward error correction
(FEC). 16 parity bytes are added to the
unchanged 188 data bytes of each trans-
port stream packet. These 16 parity bytes
form the redundancy that allows eight
errored bytes of a frame to be corrected by
the receiver. A convolutional interleaver
distributes the data so that consecutive
bits are separated. Burst errors occurring
on the transmission path are split up by the
de-interleaver into single errors that can be
corrected by the Reed-Solomon decoder.
The interleaver, too, can be disabled.
Up to and including the convolutional
interleaver, coding is identical for antenna
(COFDM), satellite (QPSK, 8PSK, 16QAM)
and cable (QAM) transmission. No further
FEC coding is provided for cable transmis-
sions, as in this case interference due to
noise, nonlinearities and interruptions is
less likely than on satellite links or with
antenna transmissions. With cable trans-
missions, mapping into the I and Q paths is
performed next.
For terrestrial transmissions via antenna
and for satellite transmissions, additional
inner FEC coding is performed after the
convolutional interleaver. The procedure,
which is known as convolutional encoding,
doubles the data rate. Puncturing is carried
out next, i.e. certain bits are left out in the
transmission according to a defined algo-
rithm, so that the data rate is reduced again.
With DVB-S satellite transmissions, map-
ping into the I and Q paths is performed at
this point. Instead of the convolutional
encoder (DVB-S), a pragmatic trellis coding
type is used for DVB-DSNG satellite trans-
mission.
A satellite turbo provides inner error cor-
rection for turbocoding and allows opera-
tion at considerably lower C/N ratios at
the same BER.
DVB-C DVB-S