Ändra sökning
RefereraExporteraLänk till posten
Permanent länk

Direktlänk
Referera
Referensformat
  • apa
  • ieee
  • modern-language-association-8th-edition
  • vancouver
  • Annat format
Fler format
Språk
  • de-DE
  • en-GB
  • en-US
  • fi-FI
  • nn-NO
  • nn-NB
  • sv-SE
  • Annat språk
Fler språk
Utmatningsformat
  • html
  • text
  • asciidoc
  • rtf
Approximate Bayesian Smoothing with Unknown Process and Measurement Noise Covariances
Linköpings universitet, Institutionen för systemteknik, Reglerteknik. Linköpings universitet, Tekniska fakulteten.
Linköpings universitet, Institutionen för systemteknik, Reglerteknik. Linköpings universitet, Tekniska fakulteten.
Department of Electrical and Electronics Engineering, Middle East Technical University, Ankara, Turkey.
Linköpings universitet, Institutionen för systemteknik, Reglerteknik. Linköpings universitet, Tekniska fakulteten.
2015 (Engelska)Ingår i: IEEE Signal Processing Letters, ISSN 1070-9908, E-ISSN 1558-2361, Vol. 22, nr 12, s. 2450-2454Artikel i tidskrift (Refereegranskat) Published
Abstract [en]

We present an adaptive smoother for linear state-space models with unknown process and measurement noise covariances. The proposed method utilizes the variational Bayes technique to perform approximate inference. The resulting smoother is computationally efficient, easy to implement, and can be applied to high dimensional linear systems. The performance of the algorithm is illustrated on a target tracking example.

Ort, förlag, år, upplaga, sidor
Institute of Electrical and Electronics Engineers (IEEE), 2015. Vol. 22, nr 12, s. 2450-2454
Nyckelord [en]
Adaptive smoothing, Kalman filtering, noise covariance, Rauch-Tung-Striebel smoother, sensor calibration, time-varying noiseco variances, variational Bayes
Nationell ämneskategori
Signalbehandling
Identifikatorer
URN: urn:nbn:se:liu:diva-121617DOI: 10.1109/LSP.2015.2490543ISI: 000364207300007OAI: oai:DiVA.org:liu-121617DiVA, id: diva2:857313
Anmärkning

At the time for thesis presentation publication was in status: Manuscript

Tillgänglig från: 2015-09-28 Skapad: 2015-09-28 Senast uppdaterad: 2018-03-09Bibliografiskt granskad
Ingår i avhandling
1. Analytical Approximations for Bayesian Inference
Öppna denna publikation i ny flik eller fönster >>Analytical Approximations for Bayesian Inference
2015 (Engelska)Doktorsavhandling, sammanläggning (Övrigt vetenskapligt)
Abstract [en]

Bayesian inference is a statistical inference technique in which Bayes’ theorem is used to update the probability distribution of a random variable using observations. Except for few simple cases, expression of such probability distributions using compact analytical expressions is infeasible. Approximation methods are required to express the a priori knowledge about a random variable in form of prior distributions. Further approximations are needed to compute posterior distributions of the random variables using the observations. When the computational complexity of representation of such posteriors increases over time as in mixture models, approximations are required to reduce the complexity of such representations.

This thesis further extends existing approximation methods for Bayesian inference, and generalizes the existing approximation methods in three aspects namely; prior selection, posterior evaluation given the observations and maintenance of computation complexity.

Particularly, the maximum entropy properties of the first-order stable spline kernel for identification of linear time-invariant stable and causal systems are shown. Analytical approximations are used to express the prior knowledge about the properties of the impulse response of a linear time-invariant stable and causal system.

Variational Bayes (VB) method is used to compute an approximate posterior in two inference problems. In the first problem, an approximate posterior for the state smoothing problem for linear statespace models with unknown and time-varying noise covariances is proposed. In the second problem, the VB method is used for approximate inference in state-space models with skewed measurement noise.

Moreover, a novel approximation method for Bayesian inference is proposed. The proposed Bayesian inference technique is based on Taylor series approximation of the logarithm of the likelihood function. The proposed approximation is devised for the case where the prior distribution belongs to the exponential family of distributions.

Finally, two contributions are dedicated to the mixture reduction (MR) problem. The first contribution, generalize the existing MR algorithms for Gaussian mixtures to the exponential family of distributions and compares them in an extended target tracking scenario. The second contribution, proposes a new Gaussian mixture reduction algorithm which minimizes the reverse Kullback-Leibler divergence and has specific peak preserving properties.

Ort, förlag, år, upplaga, sidor
Linköping: Linköping University Electronic Press, 2015. s. 79
Serie
Linköping Studies in Science and Technology. Dissertations, ISSN 0345-7524 ; 1710
Nationell ämneskategori
Signalbehandling
Identifikatorer
urn:nbn:se:liu:diva-121619 (URN)10.3384/diss.diva-121619 (DOI)978-91-7685-930-8 (ISBN)
Disputation
2015-11-06, Visionen, B-huset, Campus Valla, Linköping, 10:15 (Engelska)
Opponent
Handledare
Tillgänglig från: 2015-10-05 Skapad: 2015-09-28 Senast uppdaterad: 2015-10-07Bibliografiskt granskad

Open Access i DiVA

Fulltext saknas i DiVA

Övriga länkar

Förlagets fulltext

Sök vidare i DiVA

Av författaren/redaktören
Ardeshiri, TohidÖzkan, EmreGustafsson, Fredrik
Av organisationen
ReglerteknikTekniska fakulteten
I samma tidskrift
IEEE Signal Processing Letters
Signalbehandling

Sök vidare utanför DiVA

GoogleGoogle Scholar

doi
urn-nbn

Altmetricpoäng

doi
urn-nbn
Totalt: 429 träffar
RefereraExporteraLänk till posten
Permanent länk

Direktlänk
Referera
Referensformat
  • apa
  • ieee
  • modern-language-association-8th-edition
  • vancouver
  • Annat format
Fler format
Språk
  • de-DE
  • en-GB
  • en-US
  • fi-FI
  • nn-NO
  • nn-NB
  • sv-SE
  • Annat språk
Fler språk
Utmatningsformat
  • html
  • text
  • asciidoc
  • rtf