Successive-erasure decoding of RS-coded MPSK schemes in a Rayleigh fading channel

Abstract

Earlier analysis has shown that softdecision decoding using the Viterbi algorithm significantly improves the performance of the bandwidth-efficient RS-coded MPSK schemes in a Rayleigh fading channel. The Viterbi decoding, however, is applicable to the decoding of short RS codes with a small amount of redundancy because of its higher complexity. Another soft-decision decoding algorithm, termed successive-erasure decoding (SED), is presented for decoding of RScoded MPSK schemes. This decoding procedure, originally introduced by G.D. Forney in connection with generalised minimum-distance decoding, uses channel measurement information in algebraic decoding algorithm. The performance evaluation of some examples of RS-coded MPSK schemes shows that the use of such information bridges the gap in performance between the harddecision decoding and maximum likelihood decoding (MLD). For example, for RS(7, 5)-coded 8PSK, SED provides 3.6-6 dB coding gain at Pb = 105, compared with errors-and-erasures and errors-only decodings, respectively. This coding gain is only 1.4 dB less than the achievable gain by MLD. The performance of the SED algorithm is improved by an additional 0.7 dB when the channel state information is used.