Title :
On Filtering of Markov Chains in Strong Noise
Author :
Chigansky, Pavel
Author_Institution :
Dept. of Math., Weizmann Inst. of Sci., Rehovot
Abstract :
The filtering problem for finite-state Markov chains is revisited in the low signal-to-noise regime. We give a description of conditional measure concentration around the invariant distribution of the signal and derive asymptotic expressions for the performance indices of the minimum mean square error (MMSE) and minimum a posteriori probability (MAP) filtering estimates
Keywords :
Markov processes; filtering theory; least mean squares methods; maximum likelihood estimation; signal denoising; MAP filtering estimate; MMSE; asymptotic expression; conditional measure concentration; filtering theory; finite-state Markov chain; invariant signal distribution; minimum a posteriori probability; minimum mean square error; Atomic measurements; Density measurement; Equations; Filtering; Hidden Markov models; Loss measurement; Mean square error methods; Probability; Random variables; Recursive estimation; Error asymptotic; hidden Markov models; nonlinear filtering;
Journal_Title :
Information Theory, IEEE Transactions on
DOI :
10.1109/TIT.2006.880042
