Title :
An exploration of the relationship between the Baum-Welch algorithm and turbo decoding
Author :
Kim, Saejoon ; Wicker, Stephen B.
Author_Institution :
Sch. of Electr. Eng., Cornell Univ., Ithaca, NY, USA
Abstract :
It is shown that the problem of APP decoding of single and parallel concatenated codes can be recast in terms of parameter estimation for one or more nonstationary hidden Markov processes. It is then shown that simple extensions of the Baum-Welch algorithm provide solutions to these problems, including solutions that provide channel-matching turbo decoding algorithms
Keywords :
concatenated codes; decoding; hidden Markov models; parameter estimation; statistical analysis; turbo codes; APP decoding; Baum-Welch algorithm; channel-matching turbo decoding algorithms; nonstationary hidden Markov processes; parallel concatenated codes; parameter estimation; single concatenated codes; Concatenated codes; Hidden Markov models; Inference algorithms; Iterative algorithms; Markov processes; Maximum likelihood decoding; Parameter estimation; Polynomials; Probability distribution; Stochastic processes;
Conference_Titel :
Information Theory, 1998. Proceedings. 1998 IEEE International Symposium on
Conference_Location :
Cambridge, MA
Print_ISBN :
0-7803-5000-6
DOI :
10.1109/ISIT.1998.708708
