| Article ID | Journal | Published Year | Pages | File Type |
|---|---|---|---|---|
| 521929 | Journal of Computational Physics | 2012 | 27 Pages |
Abstract
⺠We cast the probabilistic identification problem in a linear Bayesian setting based on a functional approximation. ⺠The update procedure does not involve sampling at any stage of the computation. ⺠The method can handle non-Gaussian random variables. ⺠It is applicable to nonlinear systems. ⺠The method is compared with the Ensemble Kalman Filter (EnKF).
Related Topics
Physical Sciences and Engineering
Computer Science
Computer Science Applications
Authors
Bojana V. RosiÄ, Alexander Litvinenko, Oliver Pajonk, Hermann G. Matthies,