TY - JOUR
T1 - Convergence analyses and comparisons of Markov chain Monte Carlo algorithms in digital communications
AU - Chen, Rong
AU - Liu, Jun S.
AU - Wang, Xiaodong
N1 - Funding Information: Manuscript received January 31, 2001; revised October 3, 2001. This work was supported in part by the National Science Foundation under Grants DMS-9803649, CCR-9980599, DMS-0073601, DMS-0094613, and DMS-0073651. The associate editor coordinating the review of this paper and approving it for publication was Prof. Simon J. Godsill. R. Chen is with the Department of Information and Decision Science, University of Illinois at Chicago, Chicago, IL 6060 USA. J. S. Liu is with the Department of Statistics, Harvard University, Cambridge, MA 02138 USA (e-mail: [email protected]). X. Wang is with the Department of Electrical Engineering, Columbia University, New York, NY 10027 USA. Publisher Item Identifier S 1053-587X(02)00552-4.
PY - 2002/2
Y1 - 2002/2
N2 - Recently, Markov chain Monte Carlo (MCMC) methods have been applied to the design of blind Bayesian receivers in a number of digital communications applications. The salient features of these MCMC receivers include the following: a) They are optimal in the sense of achieving minimum symbol error rate; b) they do not require the knowledge of the channel states, nor do they explicitly estimate the channel by employing training signals or decision-feedback; and c) they are well suited for iterative (turbo) processing in coded systems. In this paper, we investigate the convergence behaviors of several MCMC algorithms (both existing and new ones) in digital communication applications. The geometric convergence property of these algorithms is established by considering only the chains or the marginal chains corresponding to the transmitted digital symbols, which take values from a finite discrete set. We then focus on three specific applications, namely, the MCMC decoders in AWGN channels, ISI channels, and CDMA channels. The convergence rates for these algorithms are computed for small simulated datasets. Different convergence behaviors are observed. It is seen that differential encoding, parameter constraining, collapsing, and grouping are efficient ways of accelerating the convergence of the MCMC algorithms, especially in the presence of channel phase ambiguity.
AB - Recently, Markov chain Monte Carlo (MCMC) methods have been applied to the design of blind Bayesian receivers in a number of digital communications applications. The salient features of these MCMC receivers include the following: a) They are optimal in the sense of achieving minimum symbol error rate; b) they do not require the knowledge of the channel states, nor do they explicitly estimate the channel by employing training signals or decision-feedback; and c) they are well suited for iterative (turbo) processing in coded systems. In this paper, we investigate the convergence behaviors of several MCMC algorithms (both existing and new ones) in digital communication applications. The geometric convergence property of these algorithms is established by considering only the chains or the marginal chains corresponding to the transmitted digital symbols, which take values from a finite discrete set. We then focus on three specific applications, namely, the MCMC decoders in AWGN channels, ISI channels, and CDMA channels. The convergence rates for these algorithms are computed for small simulated datasets. Different convergence behaviors are observed. It is seen that differential encoding, parameter constraining, collapsing, and grouping are efficient ways of accelerating the convergence of the MCMC algorithms, especially in the presence of channel phase ambiguity.
KW - Bayesian receiver
KW - Duality
KW - Eigenvalue
KW - Forward-backward algorithm
KW - Gibbs sampler
KW - Metropolis algorithm
KW - Transition function
UR - https://www.scopus.com/pages/publications/0036475631
UR - https://www.scopus.com/pages/publications/0036475631#tab=citedBy
U2 - 10.1109/78.978381
DO - 10.1109/78.978381
M3 - Article
SN - 1053-587X
VL - 50
SP - 255
EP - 270
JO - IEEE Transactions on Signal Processing
JF - IEEE Transactions on Signal Processing
IS - 2
ER -