Importance Densities for Particle Filtering Using Iterated Conditional Expectations
(2020) In IEEE Signal Processing Letters 27. p.211-215- Abstract
- In this letter, we consider Gaussian approximations of the optimal importance density in sequential importance sampling for nonlinear, non-Gaussian state-space models. The proposed method is based on generalized statistical linear regression and posterior linearization using conditional expectations. Simulation results show that the method outperforms the compared methods in terms of the effective sample size and provides a better local approximation of the optimal importance density.
Please use this url to cite or link to this publication:
https://lup.lub.lu.se/record/940c6217-9d62-4c53-aca6-962f63c0ab4d
- author
- Tronarp, Filip LU ; Hostettler, Roland ; Särkkä, Simo and Garcia-Fernandez, Angel F
- publishing date
- 2020
- type
- Contribution to journal
- publication status
- published
- subject
- keywords
- state estimation, particle filters, Monte Carlo methods, nonlinear systems, posterior linearization
- in
- IEEE Signal Processing Letters
- volume
- 27
- pages
- 5 pages
- publisher
- IEEE - Institute of Electrical and Electronics Engineers Inc.
- external identifiers
-
- scopus:85079094614
- ISSN
- 1070-9908
- DOI
- 10.1109/LSP.2020.2964531
- language
- English
- LU publication?
- no
- id
- 940c6217-9d62-4c53-aca6-962f63c0ab4d
- date added to LUP
- 2023-08-20 22:59:03
- date last changed
- 2023-10-13 15:39:51
@article{940c6217-9d62-4c53-aca6-962f63c0ab4d, abstract = {{In this letter, we consider Gaussian approximations of the optimal importance density in sequential importance sampling for nonlinear, non-Gaussian state-space models. The proposed method is based on generalized statistical linear regression and posterior linearization using conditional expectations. Simulation results show that the method outperforms the compared methods in terms of the effective sample size and provides a better local approximation of the optimal importance density.}}, author = {{Tronarp, Filip and Hostettler, Roland and Särkkä, Simo and Garcia-Fernandez, Angel F}}, issn = {{1070-9908}}, keywords = {{state estimation; particle filters; Monte Carlo methods; nonlinear systems; posterior linearization}}, language = {{eng}}, pages = {{211--215}}, publisher = {{IEEE - Institute of Electrical and Electronics Engineers Inc.}}, series = {{IEEE Signal Processing Letters}}, title = {{Importance Densities for Particle Filtering Using Iterated Conditional Expectations}}, url = {{http://dx.doi.org/10.1109/LSP.2020.2964531}}, doi = {{10.1109/LSP.2020.2964531}}, volume = {{27}}, year = {{2020}}, }