Gibbs sampling based parameter estimation for RSC sub-codes of turbo codes
Authors:
- 1 October 2014
- conference paper
- Published by Institute of Electrical and Electronics Engineers (IEEE)
- p. 1-5
Abstract
This paper deals with the Bayesian parameter estimation for recursive convolutional sub-codes of turbo codes. A Monte Carlo method named Gibbs sampling is employed for this problem. The involved conditional probabilities of the encoder coefficients and the coded sequences are derived. Simulation results show that the proposed algorithm improves the estimation accuracy substantially.Keywords
This publication has 17 references indexed in Scilit:
- Novel Blind Identification of LDPC Codes Using Average LLR of Syndrome a Posteriori ProbabilityIEEE Transactions on Signal Processing, 2013
- A Fast Scheme for Blind Identification of Channel Codes2011
- Blind recovery of k/n rate convolutional encoders in a noisy environmentEURASIP Journal on Wireless Communications and Networking, 2011
- Bayesian Signal Processing2008
- Parameter Estimation of a Convolutional Encoder from Noisy Observations2007
- Block code reconstruction using iterative decoding techniques2006
- Maximum-likelihood binary shift-register synthesis from noisy observationsIEEE Transactions on Information Theory, 2002
- Detection and recognition of a binary linear codeDiscrete Applied Mathematics, 2001
- Blind turbo equalization in Gaussian and impulsive noiseIEEE Transactions on Vehicular Technology, 2001
- Asymptotic Behavior of the Gibbs SamplerJournal of the American Statistical Association, 1993