The Oxford handbook of nonlinear filtering

摘要

The main purpose of filtering is to obtain optimal recursive estimates of the state of a dynamical system based on noisy observations. If the dynamical systems are nonlinear, the theory of nonlinear filtering (NLF) is an extension of the Bayesian framework to the prediction, interpolation and smoothing of nonlinear stochastic dynamics, which works by considering the evolution of the conditional distribution of the state given all the available observations.