Adaptive Particle Filter for Stable Distribution
Haroldo Fraga de Campos Velho, Helaine C. M. Furtado · Birkhäuser Boston eBooks · 2011
Particle filters are considered as the most robust method in the estimation theory. However, all techniques presented in the literature consider only cases where the data can be described by a probability distribution function (PDF) with all statistical moments well defined. The present chapter a novel particle filter is introduced for estimating a posteriori PDF using a Bayesian scheme. The key issue is to consider an adaptive likelihood function. The scheme generalizes the traditonal particle filter approaches. The scheme can be applied to inverse problem, control theory, image and signal processing, and data assimilation. These keywords were added by machine and not by the authors. This process is experimental and the keywords may be updated as the learning algorithm improves.