Title :
Estimation for discrete Markov random fields observed in Gaussian noise
Author :
Elliott, Robert J. ; Aggoun, Lakhdar
Author_Institution :
Dept. of Math. Sci., Alberta Univ., Edmonton, Alta., Canada
fDate :
9/1/1994 12:00:00 AM
Abstract :
A finite state Markov random field is observed in Gaussian noise. Changes of measures are defined under which all random variables of the signal are independent and uniformly distributed over the finite state space and all random variables of the observation are independent and N(0,1). The problem of estimating the most likely signal given the observations is treated in a related form by introducing probabilities over the possible signals
Keywords :
Markov processes; parameter estimation; probability; random processes; signal detection; stochastic processes; Gaussian noise; discrete Markov random fields; finite state space; most likely signal; probabilities; random variables; Additive noise; Decision feedback equalizers; Gaussian noise; Lattices; Markov random fields; Maximum likelihood detection; Maximum likelihood estimation; Random variables; State-space methods; Viterbi algorithm;
Journal_Title :
Information Theory, IEEE Transactions on
