Bayesian compressive sensing using Monte Carlo methods

Ioannis Kyriakides, Radmila Pribic

Research output: Chapter in Book/Report/Conference proceedingConference contributionpeer-review


The problem of reconstructing a signal from compressively sensed measurements is solved in this work from a Bayesian perspective. The proposed reconstruction solution differs from previous Bayesian methods in that it numerically evaluates the posterior of the sparse solution. This allows the method to utilize any kind of information on the signal without the need to evaluate the posterior in closed form. Specifically, the method uses multi-stage sampling together with a greedy subroutine to efficiently draw information directly from the likelihood and any prior distribution on the signal, including a sparsity prior. The approach is shown to accurately represent the Bayesian belief on the sparse solution based on noisy compressively sensed signals.

Original languageEnglish
Title of host publication2013 Proceedings of the 21st European Signal Processing Conference, EUSIPCO 2013
PublisherEuropean Signal Processing Conference, EUSIPCO
ISBN (Print)9780992862602
Publication statusPublished - 2013
Event2013 21st European Signal Processing Conference, EUSIPCO 2013 - Marrakech, Morocco
Duration: 9 Sept 201313 Sept 2013


Other2013 21st European Signal Processing Conference, EUSIPCO 2013


  • Bayesian compressive sensing
  • Monte Carlo methods
  • sparse reconstruction


Dive into the research topics of 'Bayesian compressive sensing using Monte Carlo methods'. Together they form a unique fingerprint.

Cite this