Cryptography Reference
In-Depth Information
(a) Elementary encoder
(b) TTCM encoder
Figure 10.6 - Construction of an 8-PSK TTCM according to the method described in
[10.1]. Spectral eciency
η
=2
bit/s/Hz
.
can ideally reach the Hamming distanceof the code. Consequently, transmission
schemes using the BICM principle in practice have better performance on fading
channels than TTCMs have.
The code and the modulation not being jointly optimized, unlike a TTCM
scheme, we choose binary mapping of the constellation points which minimizes
the mean binary error rates at the input of the decoder. When it can be envis-
aged, Gray encoding satisfies this condition. For simplicity in implementing the
modulator and demodulator, in the case of square QAM (
q
even), the in-phase
and in-quadrature axes,
I
and
Q
, are mapped independently.
In Figure 10.9, the role of the "Multiplexing / symbol composition" block is
to distribute the encoded bits, after interleaving for fading channels, into modu-
lation symbols. This block, the meeting point between the code and the modu-
lation, enables a certain level of adjustment of the coded modulation according
to the performance targeted. This adjustment is possible since the code and the
modulation do not play the same role in relation to all the bits transmitted.
On the one hand, we can distinguish two distinct families of encoded bits
at the output of the encoder: systematic bits and redundancy bits. These two
families of bits play a different role in the decoding process: the systematic
bits, coming directly from the source, are used by the two elementary decoders
