The uncorrelated Rayleigh channel can be regarded a worst case channel for digital radio communication. Crucial for the detection performance on this channel is the diversity order of the communication system, which shall be as large as possible. For trellis coded modulation (TCM) with symbol-interleaving the diversity order is the smallest number of distinct channel symbols along any error event. The diversity order can be further increased by bit-wise interleaving of the encoder output before mapping the bits onto channel symbols and is now the smallest number of distinct bits along any error event. The concept of bit-interleaving is generalized such that a convolutional code of lower rate (larger Hamming distance) can be used giving an even higher diversity order. The same spectral efficiency is maintained by an expansion of the channel symbol constellation. This concept is referred to as channel symbol expansion diversity (CSED). Simulation results indicate that this approach leads to coded modulation schemes with significantly better performance on the Rayleigh fading channel than the previously best known symbol- or bit-interleaved TCM-systems of comparable complexity. An interesting special case, referred to as repetition coded CSED, arises when the convolutional encoder is replaced by a repetition encoder. Theoretical analyses as well as simulation results indicate that this low-complexity system gives a large performance gain on the Rayleigh channel compared to uncoded signaling.
[1]
Seymour Stein,et al.
Fading Channel Issues in System Engineering
,
1987,
IEEE J. Sel. Areas Commun..
[2]
Tor Aulin,et al.
Channel symbol expansion diversity
,
1995
.
[3]
Ephraim Zehavi,et al.
8-PSK trellis codes for a Rayleigh channel
,
1992,
IEEE Trans. Commun..
[4]
Gottfried Ungerboeck,et al.
Channel coding with multilevel/phase signals
,
1982,
IEEE Trans. Inf. Theory.
[5]
Jun Du,et al.
Construction of new MPSK trellis codes for fading channels
,
1995,
IEEE Trans. Commun..
[6]
Tor Aulin,et al.
Theoretical Performance Evaluation of Different Soft-output Algorithms
,
1994
.
[7]
Daniel J. Costello,et al.
Bandwidth efficient coding for fading channels: code construction and performance analysis
,
1989,
IEEE Journal on Selected Areas in Communications.
[8]
B. D. Jelicic,et al.
Design of trellis coded QAM for flat fading and AWGN channels
,
1995
.
[9]
Jr. G. Forney,et al.
The viterbi algorithm
,
1973
.