[Back]


Talks and Poster Presentations (without Proceedings-Entry):

C. Mecklenbräuker, P. Gerstoft:
"Sequential Bayesian Reconstruction of Sparse Source from Sensor Array Data";
Talk: Information Theory and Applications Workshop (ITA 2014), San Diego (A), USA (invited); 02-09-2014 - 02-14-2014.



English abstract:
The sequential Bayesian reconstruction of sparse source waveforms from sensor array data is analyzed. Acoustic waves are observed by a sensor array. The waves are emitted by a spatially-sparse set of sources. A weighted Laplace-like prior is assumed for the sources such that the maximum a posteriori source estimate at the current time step can be approximated from the weighted LASSO. The new weighting for time step k+1 is defined from a fit to the approximated posterior distribution at the previous time step k. Thus, a sequence of weighted LASSO problems is solved for estimating the temporal evolution of a sparse source field. Finally, we explore M. E. Tipping's approach to fast marginal likelihood maximization for sparse Bayesian models for sequential source waveform reconstruction.

Keywords:
sparsity, sequential, LASSO, convex optimization, duality,

Created from the Publication Database of the Vienna University of Technology.