Skip to content

Advertisement

Advanced Computational Methods for Bayesian Signal Processing

The problem of estimating some variables of interest from noisy observations is ubiquitous in different fields, such as signal processing, finance, oceanography, video tracking and so on. Computational methods are often required in Bayesian inference and nonlinear signal processing to deal with intractable posterior densities. For instance, Sequential Importance Sampling (a.k.a. particle filters) and Markov Chain Monte Carlo (MCMC) methods, which have been popular approaches within the statistical community for a long time, have been widely used in signal processing and communications applications. Over the last years, several extensions and variants of these two families of methods have been proposed in order to improve their performance (e.g., for the estimation of fixed parameters or dealing with multi-modal target densities): population Monte Carlo (PMC) schemes, particle MCMC (PMCMC), adaptive Monte Carlo approaches (i.e., MCMC with adaptive proposal functions), multiple try Metropolis (MTM) strategies, parallel Monte Carlo chains, etc. Some of these methods have found their way into the signal processing literature, but there are still many recent advanced Monte Carlo methods, developed within the statistical community, that are not so widely known by signal processing practitioners and which may be very useful for signal processing applications. This special issue intends to bridge the gap between both communities by presenting a collection of papers that describe recent advances in Monte Carlo methods with signal processing applications in mind.

Edited by: François Desbouvries, David Luengo, Monica Bugallo, Victor Elvira, Fredrik Lindsted, Luca Martino, Jimmy Olsson, Yohan Petetin, Branco Ristic, Simo Sarkka and François Septier

  1. Content type: Research

    Considerable effort has been recently devoted to the design of schemes for the parallel implementation of sequential Monte Carlo (SMC) methods for dynamical systems, also widely known as particle filters (PFs)...

    Authors: Dan Crisan, Joaquín Míguez and Gonzalo Ríos-Muñoz

    Citation: EURASIP Journal on Advances in Signal Processing 2018 2018:31

    Published on:

  2. Content type: Research

    This paper introduces a new algorithm to approximate smoothed additive functionals of partially observed diffusion processes. This method relies on a new sequential Monte Carlo method which allows to compute s...

    Authors: Pierre Gloaguen, Marie-Pierre Étienne and Sylvain Le Corff

    Citation: EURASIP Journal on Advances in Signal Processing 2018 2018:9

    Published on:

  3. Content type: Research

    Monte Carlo methods have become essential tools to solve complex Bayesian inference problems in different fields, such as computational statistics, machine learning, and statistical signal processing. In this ...

    Authors: Luca Martino, Roberto Casarin, Fabrizio Leisen and David Luengo

    Citation: EURASIP Journal on Advances in Signal Processing 2018 2018:5

    Published on:

  4. Content type: Research

    We consider the problem of sequential inference of latent time-series with innovations correlated in time and observed via nonlinear functions. We accommodate time-varying phenomena with diverse properties by ...

    Authors: Iñigo Urteaga, Mónica F. Bugallo and Petar M. Djurić

    Citation: EURASIP Journal on Advances in Signal Processing 2017 2017:84

    Published on:

  5. Content type: Research

    Fault detection and isolation is crucial for the efficient operation and safety of any industrial process. There is a variety of methods from all areas of data analysis employed to solve this kind of task, suc...

    Authors: Jerzy Baranowski, Piotr Bania, Indrajeet Prasad and Tian Cong

    Citation: EURASIP Journal on Advances in Signal Processing 2017 2017:79

    Published on:

  6. Content type: Research

    Particle filtering is a numerical Bayesian technique that has great potential for solving sequential estimation problems involving non-linear and non-Gaussian models. Since the estimation accuracy achieved by ...

    Authors: Jeyarajan Thiyagalingam, Lykourgos Kekempanos and Simon Maskell

    Citation: EURASIP Journal on Advances in Signal Processing 2017 2017:71

    Published on:

  7. Content type: Review

    The ensemble Kalman filter (EnKF) is a Monte Carlo-based implementation of the Kalman filter (KF) for extremely high-dimensional, possibly nonlinear, and non-Gaussian state estimation problems. Its ability to ...

    Authors: Michael Roth, Gustaf Hendeby, Carsten Fritsche and Fredrik Gustafsson

    Citation: EURASIP Journal on Advances in Signal Processing 2017 2017:56

    Published on:

  8. Content type: Research

    This paper focuses on sequential Monte Carlo approximations of smoothing distributions in conditionally linear and Gaussian state spaces. To reduce Monte Carlo variance of smoothers, it is typical in these mod...

    Authors: Ngoc Minh Nguyen, Sylvain Le Corff and Éric Moulines

    Citation: EURASIP Journal on Advances in Signal Processing 2017 2017:54

    Published on: