Robust Inference for State-Space Models with Skewed Measurement Noise

Henri Nurminen; Tohid Ardeshiri; Robert Piché; Fredrik Gustafsson

doi:10.1109/LSP.2015.2437456

Robust Inference for State-Space Models with Skewed Measurement Noise

Henri Nurminen, Tohid Ardeshiri, Robert Piché, Fredrik Gustafsson

Research output: Contribution to journal › Article › Scientific › peer-review

113 Citations (Scopus)

53 Downloads (Pure)

Abstract

Filtering and smoothing algorithms for linear discrete-time state-space models with skewed and heavy-tailed measurement noise are presented. The algorithms use a variational Bayes approximation of the posterior distribution of models that have normal prior and skew-t-distributed measurement noise. The proposed filter and smoother are compared with conventional low-complexity alternatives in a simulated pseudorange positioning scenario. In the simulations the proposed methods achieve better accuracy than the alternative methods, the computational complexity of the filter being roughly 5 to 10 times that of the Kalman filter.

Original language	English
Pages (from-to)	1898-1902
Number of pages	5
Journal	IEEE Signal Processing Letters
Volume	22
Issue number	11
DOIs	https://doi.org/10.1109/LSP.2015.2437456
Publication status	Published - 1 Nov 2015
Publication type	A1 Journal article-refereed

Keywords

Kalman filter
robust filtering
RTS smoother
skew t
skewness
t-distribution
variational Bayes

Publication forum classification

Publication forum level 2

ASJC Scopus subject areas

Electrical and Electronic Engineering
Signal Processing
Applied Mathematics

Access to Document

10.1109/LSP.2015.2437456

SPL2015_final
Copyright (c) 2015 IEEE. Personal use of this material is permitted. However, permission to use this material for any other purposes must be obtained from the IEEE by sending a request to [email protected].
Accepted author manuscript, 475 KB

http://urn.fi/URN:NBN:fi:tty-201603173652

Cite this

@article{ed2602f448404cdb8e2d46528c299998,

title = "Robust Inference for State-Space Models with Skewed Measurement Noise",

abstract = "Filtering and smoothing algorithms for linear discrete-time state-space models with skewed and heavy-tailed measurement noise are presented. The algorithms use a variational Bayes approximation of the posterior distribution of models that have normal prior and skew-t-distributed measurement noise. The proposed filter and smoother are compared with conventional low-complexity alternatives in a simulated pseudorange positioning scenario. In the simulations the proposed methods achieve better accuracy than the alternative methods, the computational complexity of the filter being roughly 5 to 10 times that of the Kalman filter.",

keywords = "Kalman filter, robust filtering, RTS smoother, skew t, skewness, t-distribution, variational Bayes",

author = "Henri Nurminen and Tohid Ardeshiri and Robert Pich{\'e} and Fredrik Gustafsson",

year = "2015",

month = nov,

day = "1",

doi = "10.1109/LSP.2015.2437456",

language = "English",

volume = "22",

pages = "1898--1902",

journal = "IEEE Signal Processing Letters",

issn = "1070-9908",

publisher = "Institute of Electrical and Electronics Engineers Inc.",

number = "11",

}

TY - JOUR

T1 - Robust Inference for State-Space Models with Skewed Measurement Noise

AU - Nurminen, Henri

AU - Ardeshiri, Tohid

AU - Piché, Robert

AU - Gustafsson, Fredrik

PY - 2015/11/1

Y1 - 2015/11/1

N2 - Filtering and smoothing algorithms for linear discrete-time state-space models with skewed and heavy-tailed measurement noise are presented. The algorithms use a variational Bayes approximation of the posterior distribution of models that have normal prior and skew-t-distributed measurement noise. The proposed filter and smoother are compared with conventional low-complexity alternatives in a simulated pseudorange positioning scenario. In the simulations the proposed methods achieve better accuracy than the alternative methods, the computational complexity of the filter being roughly 5 to 10 times that of the Kalman filter.

AB - Filtering and smoothing algorithms for linear discrete-time state-space models with skewed and heavy-tailed measurement noise are presented. The algorithms use a variational Bayes approximation of the posterior distribution of models that have normal prior and skew-t-distributed measurement noise. The proposed filter and smoother are compared with conventional low-complexity alternatives in a simulated pseudorange positioning scenario. In the simulations the proposed methods achieve better accuracy than the alternative methods, the computational complexity of the filter being roughly 5 to 10 times that of the Kalman filter.

KW - Kalman filter

KW - robust filtering

KW - RTS smoother

KW - skew t

KW - skewness

KW - t-distribution

KW - variational Bayes

U2 - 10.1109/LSP.2015.2437456

DO - 10.1109/LSP.2015.2437456

M3 - Article

AN - SCOPUS:84933056939

SN - 1070-9908

VL - 22

SP - 1898

EP - 1902

JO - IEEE Signal Processing Letters

JF - IEEE Signal Processing Letters

IS - 11

ER -

Robust Inference for State-Space Models with Skewed Measurement Noise

Abstract

Keywords

Publication forum classification

ASJC Scopus subject areas

Access to Document

Fingerprint

Cite this