In this paper, we analyze the forward-backward algorithm for the maximum a
posteriori (MAP) decoding of information transmitted over channels with mem
ory and propose a suboptimal forward-only algorithm, We assume that both th
e information source and channel are described using hidden Markov models (
HMMs), The algorithm lends itself to parallel implementation and pipelining
. We apply the algorithm to MAP decoding of symbols which were trellis-code
modulated and transmitted over channels with error bursts.