Computer Science – Information Theory
Scientific paper
2011-11-29
Computer Science
Information Theory
30 pages, 16 figures, submitted to IEEE Trans. on Signal Processing
Scientific paper
The paper focuses on minimum mean square error (MMSE) Bayesian estimation for a Gaussian source impaired by additive Middleton's Class-A impulsive noise. In order to reduce the implementation complexity associated with the expression of the optimum Bayesian estimator, the paper considers also two popular suboptimal estimators, which are the soft-limiter and the blanker. The optimum Bayesian thresholds for such suboptimal estimators are obtained by iteratively solving fixed point equations. Connections with the maximum SNR estimators are also established. Theoretical expressions for the MSE and the SNR of the suboptimal estimators are also derived, and the results confirmed by simulations. Noteworthy, the results can be applied to any noise characterized by a Gaussian mixture distribution.
No associations
LandOfFree
Bayesian Estimation of a Gaussian source in Middleton's Class-A Impulsive Noise does not yet have a rating. At this time, there are no reviews or comments for this scientific paper.
If you have personal experience with Bayesian Estimation of a Gaussian source in Middleton's Class-A Impulsive Noise, we encourage you to share that experience with our LandOfFree.com community. Your opinion is very important and Bayesian Estimation of a Gaussian source in Middleton's Class-A Impulsive Noise will most certainly appreciate the feedback.
Profile ID: LFWR-SCP-O-16540