MAP decoding using the EM algorithm

Author

Turin, William

Author_Institution

Dept. of Commun. Res., AT&T Bell Labs., Florham Park, NJ, USA

Volume

3

fYear

1999

fDate

36342

Firstpage

1866

Abstract

The expectation-maximization (EM) algorithm is popular in estimating parameters of various statistical models. We consider applications of the EM algorithm to the maximum a posteriori (MAP) sequence decoding in channels described by hidden Markov models (HMMs). It is well known that HMMs can accurately approximate large variety of communication channels with memory and in particular, wireless fading channels with noise. The direct maximization of the a posteriori probability is too complex. The EM algorithm allows us to obtain the MAP sequence estimation iteratively

Keywords

Viterbi decoding; block codes; convolutional codes; fading channels; hidden Markov models; maximum likelihood decoding; noise; optimisation; sequential estimation; trellis coded modulation; EM algorithm; HMM; MAP decoding; TCM system; Viterbi algorithm; a posteriori probability; block codes; communication channels; convolutional codes; expectation-maximization algorithm; hidden Markov models; iterative MAP sequence estimation; maximum a posteriori sequence decoding; noise; parameter estimation; statistical models; wireless fading channels; Communication channels; Digital communication; Fading; Hidden Markov models; Iterative algorithms; Iterative decoding; Laboratories; Parameter estimation; Viterbi algorithm; Wireless communication;

fLanguage

English

Publisher

ieee

Conference_Titel

Vehicular Technology Conference, 1999 IEEE 49th

Conference_Location

Houston, TX

ISSN

1090-3038

Print_ISBN

0-7803-5565-2

Type

conf

DOI

10.1109/VETEC.1999.778361

Filename

778361