Comparison of Bayesian Methods for Recovering Sinusoids
INTERNATIONAL JOURNAL OF COMPUTERS AND COMMUNICATIONS, no.15, pp.36-43, 2021 (Peer-Reviewed Journal)
- Publication Type: Article / Article
- Publication Date: 2021
- Doi Number: 10.46300/91013.2021.15.7
- Journal Name: INTERNATIONAL JOURNAL OF COMPUTERS AND COMMUNICATIONS
- Journal Indexes: Index Copernicus, Other Indexes
- Page Numbers: pp.36-43
- Open Archive Collection: AVESIS Open Access Collection
- Istanbul University Affiliated: Yes
Abstract
In this paper, we study a problem of estimating parameters of sinusoids from noisy data within Bayesian inferential framework. In this context, three different computational schemes such as, Bretthorst’s integral method (BRETTHORST), Gibbs sampling (GIBBS) and parallel tempering (PT) are studied and modifications of their algorithms were tested on data generated from synthetic signals. In addition, our emphasis is given to a comparison of their performances with respect to Cramér-Rao lower bound (CRLB)