Recovering sinusoids from data using bayesian inference with RJMCMC

We consider a problem of detecting and estimating of noisy sinusoids within a Bayesian probabilistic inferential framework in which inferences about signal parameters are drawn from posterior probability density function (PDF). However, this requires evaluation of some complicated high-dimensional integrals. Therefore, an efficient computational algorithm is implemented to draw samples from the posterior PDF of parameters under various proposal distributions. This algorithm, coded in Mathematica, is used for synthetic data sets. Simulations results support its effectiveness.