Multi Snapshot Sparse Bayesian Learning for DOA Estimation

TL;DR

This paper introduces a Sparse Bayesian Learning approach for estimating directions of arrival from multi-snapshot sensor data, offering a probabilistic framework that promotes sparsity and improves estimation accuracy.

Contribution

The paper develops a novel SBL-based method for DOA estimation that automatically selects hyperparameters via evidence maximization, enhancing sparsity and estimation performance.

Findings

Competitive performance against LASSO, beamforming, and MUSIC.

Automatic hyperparameter selection improves sparsity and accuracy.

Probabilistic framework provides robust DOA estimates.

Abstract

The directions of arrival (DOA) of plane waves are estimated from multi-snapshot sensor array data using Sparse Bayesian Learning (SBL). The prior source amplitudes is assumed independent zero-mean complex Gaussian distributed with hyperparameters the unknown variances (i.e. the source powers). For a 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 SBL scheme for DOA estimation is discussed and evaluated competitively against LASSO ($\ell_1$-regularization), conventional beamforming, and MUSIC