MAP decoding in channels with memory

The expectation-maximization (EM) algorithm is popular in estimating parame ters of various statistical models. In this paper, we consider applications of the EM algorithm to the maximum a posteriori (MAP) sequence decoding as suming that sources and channels are described by hidden Markov models (HMM 's), HMM's call accurately approximate a large variety of communication cha nnels with memory and, in particular, wireless fading channels with noise. The direct maximization of the a posteriori probability (APP) is too comple x, The EM algorithm allows us to obtain the MAP sequence estimation iterati vely, Since each step of the EM algorithm increases the APP, the algorithm can improve performance of ally decoding procedure.