[Back]

C. Mecklenbräuker, P. Gerstoft, A. Xenaki:
"Multi Snapshot Sparse Bayesian Learning for DOA Estimation";
Talk: Gastvortrag Univ. Erlangen-Nürnberg, Erlangen (invited); 06-12-2017.

The directions of arrival (DOA) of plane waves are estimated from multi-snapshot
sensor array data using the Sparse Bayesian Learning (SBL) approach. Assuming as
prior information independent zero-mean complex Gaussian distributed source
amplitudes with hyperparameters the unknown variances (i.e. the source powers) and
complex Gaussian likelihood with hyperparameter the unknown noise variance, the
corresponding Gaussian posterior distribution is derived. For a given number of
DOAs, the hyperparameters are automatically selected by maximizing the evidence and
promote sparse DOA estimates. The resulting SBL scheme for DOA estimation is
discussed and evaluated competitively against LASSO (l1-regularization),
conventional beamforming, and MUSIC.

sparse recovery, compressive sensing, seismology, earthquake

http://publik.tuwien.ac.at/files/publik_260091.pdf