Maximum-likelihood estimation of a class of chaotic signals

The chaotic sequences corresponding to tent map dynamics are potentially attractive in a range of engineering applications. Optimal estimation algorithms for signal filtering, prediction, and smoothing in the presence of white Gaussian noise are derived for this class of sequences based on the method of maximum likelihood. The resulting algorithms are highly nonlinear but have convenient recursive implementations that are efficient both in terms of computation and storage. Performance evaluations are also included and compared with the associated Cramer-Rao bounds.>