The Bayesian Approach to Signal Modelling
Peter Julian Rayner, Bill Fitzgerald · Applied and numerical harmonic analysis · 1998
In this paper, an introduction to Bayesian methods in signal processing will be given. The paper starts by considering the important issues of model selection and parameter estimation. The important class of signal models, known as the General Linear Model, is introduced and the concept of marginal estimation of certain model parameter is developed. The techniques are illustrated for the problem of estimating sinusoidal frequency components in white Gaussian noise and for the general changepoint problem. Numerical integration methods are introduced based on Markov chain Monte Carlo techniques and the Gibbs sampler in particular and applications to audio restoration are presented. These keywords were added by machine and not by the authors. This process is experimental and the keywords may be updated as the learning algorithm improves.