SEQUENCE ESTIMATION OVER THE SLOW NONSELECTIVE RAYLEIGH FADING CHANNEL WITH DIVERSITY RECEPTION AND ITS APPLICATION TO VITERBI DECODING

We consider minimum error probability detection of a data sequence transmitted using linear-suppressed carrier modulations, specifically PSK, over the Gaussian channel with slow nonselective Rayleigh fading. Complete channel interleaving/deinterleaving and diversity reception are assumed. The problem is considered with application to Viterbi decoding, in particular. It is first shown that the two presently available receivers, namely the conventional ML receiver and the simultaneous estimation receiver, do not perform adequately for this problem. A new two-stage receiver is proposed in which the unknown channel fading gains are estimated in the first stage prior to data sequence estimation in the second stage. This receiver/decoder for Viterbi decoding of convolutionally/trellis coded sequences. The issue of optimum estimation of channel fading gains is clarified. The bit error probability of the receiver/decoder is analyzed, and numerical performance results are presented for the optimum rate-1/2 constraint-length-3 convolutional code with a Markov model for the channel fading processes.