Nonlinear Statistical Signal Processing [electronic resource] : A Particle Filtering Approach
- Washington, D.C. : United States. Dept. of Energy, 2007.
Oak Ridge, Tenn. : Distributed by the Office of Scientific and Technical Information, U.S. Dept. of Energy.
- Physical Description:
- PDF-file: 26 pages; size: 0.5 Mbytes
- Additional Creators:
- Lawrence Berkeley National Laboratory
United States. Department of Energy
United States. Department of Energy. Office of Scientific and Technical Information
- A introduction to particle filtering is discussed starting with an overview of Bayesian inference from batch to sequential processors. Once the evolving Bayesian paradigm is established, simulation-based methods using sampling theory and Monte Carlo realizations are discussed. Here the usual limitations of nonlinear approximations and non-gaussian processes prevalent in classical nonlinear processing algorithms (e.g. Kalman filters) are no longer a restriction to perform Bayesian inference. It is shown how the underlying hidden or state variables are easily assimilated into this Bayesian construct. Importance sampling methods are then discussed and shown how they can be extended to sequential solutions implemented using Markovian state-space models as a natural evolution. With this in mind, the idea of a particle filter, which is a discrete representation of a probability distribution, is developed and shown how it can be implemented using sequential importance sampling/resampling methods. Finally, an application is briefly discussed comparing the performance of the particle filter designs with classical nonlinear filter implementations.
- Published through SciTech Connect.
Imaging for Detection and Identification, vol. 7, no. 1, June 15, 2007, pp. 129-150 7 1 FT
- Funding Information:
View MARC record | catkey: 14677802