Population Monte Carlo methodology a la Gibbs sampling

Population Monte Carlo (PMC) algorithms iterate on sets of samples and weights to approximate a stationary target distribution. The target distribution is often the a posteriori distribution of a set of unknowns of interest given observed data and the employed model. The accuracy of the estimation depends on many factors including the number and “quality” of the generated samples. In this paper, we propose a PMC algorithm that can be used for high-dimensional models and that is built in the spirit of the Gibbs sampling method. We demonstrate the proposed approach on the classical problem of estimating the frequencies of multiple sinusoids. The simulation results show the accuracy of the estimates and their comparison with the results of an alternative approach.