Non-Gaussian Bayesian filtering by density parametrization using power moments

DOI10.1016/J.AUTOMATICA.2023.111061zbMATH Open1520.93577arXiv2207.08519OpenAlexW4367839132MaRDI QIDQ6110312FDOQ6110312

Authors: Guangyu Wu, Anders Lindquist 

Publication date: 5 July 2023

Published in: Automatica (Search for Journal in Brave)

Abstract: Non-Gaussian Bayesian filtering is a core problem in stochastic filtering. The difficulty of the problem lies in parameterizing the state estimates. However the existing methods are not able to treat it well. We propose to use power moments to obtain a parameterization. Unlike the existing parametric estimation methods, our proposed algorithm does not require prior knowledge about the state to be estimated, e.g. the number of modes and the feasible classes of function. Moreover, the proposed algorithm is not required to store massive parameters during filtering as the existing nonparametric Bayesian filters, e.g. the particle filter. The parameters of the proposed parametrization can also be determined by a convex optimization scheme with moments constraints, to which the solution is proved to exist and be unique. A necessary and sufficient condition for all the power moments of the density estimate to exist and be finite is provided. The errors of power moments are analyzed for the density estimate being either light-tailed or heavy-tailed. Error upper bounds of the density estimate for the one-step prediction are proposed. Simulation results on different types of density functions of the state are given, including the heavy-tailed densities, to validate the proposed algorithm.

Full work available at URL: https://arxiv.org/abs/2207.08519

Recommendations

• Nonparametric estimation of a posteriori distributions in nonlinear filtering
• Gaussian filters for nonlinear filtering problems
• Filtering discrete time nonlinear systems with unknown parameters: a nonparametric approach.
• Particle Gaussian mixture filters. I.
• Gaussian progressive Bayesian filter

zbMATH Keywords

Bayesian methodsfilteringpower moments

Mathematics Subject Classification ID

Filtering in stochastic control theory (93E11)

Cites Work

• Matrix Analysis
• Title not available (Why is that?)
• A Mathematical Theory of Communication
• High-dimensional probability. An introduction with applications in data science
• Title not available (Why is that?)
• Title not available (Why is that?)
• On Kullback-Leibler loss and density estimation
• Nonlinear Bayesian estimation using Gaussian sum approximations
• Optimal Inequalities in Probability Theory: A Convex Optimization Approach
• A note on proximity of distributions in terms of coinciding moments.
• Kullback-leibler approximation of spectral density functions
• Digital synthesis of non-linear filters
• Recursive Bayesian estimation using Gaussian sums
• Monte Carlo techniques to estimate the conditional expectation in multi-stage non-linear filtering†
• Maximum entropy solutions and moment problem in unbounded domains
• The Kalman-Bucy Filter as a True Time-Varying Wiener Filter
• Optimal quadrature formula nonlinear estimators
• Choice of basis for Laplace approximation
• The moment problem
• A nonlinear filtering algorithm based on an approximation of the conditional distribution
• A note on the iterative application of Bayes' rule
• A state-space calculus for rational probability density functions and applications to non-Gaussian filtering
• A novel robust Kalman filter with unknown non-stationary heavy-tailed noise
• Interior point solutions of variational problems and global inverse function theorems

Cited In (3)

• A non-Gaussian Bayesian filter using power and generalized logarithmic moments
• A Bayes formula for nonlinear filtering with Gaussian and Cox noise
• A non-classical parameterization for density estimation using sample moments

This page was built for publication: Non-Gaussian Bayesian filtering by density parametrization using power moments

Report a bug (only for logged in users!)Click here to report a bug for this page (MaRDI item Q6110312)