Sparse DOA estimation with polynomial rooting

Angeliki Xenaki, Peter Gerstoft, Efren Fernandez Grande

Research output: Chapter in Book/Report/Conference proceedingArticle in proceedingsResearchpeer-review

578 Downloads (Pure)

Abstract

Direction-of-arrival (DOA) estimation involves the localization of a few sources from a limited number of observations on an array of sensors. Thus, DOA estimation can be formulated as a sparse signal reconstruction problem and solved efficiently with compressive sensing (CS) to achieve highresolution imaging. Utilizing the dual optimal variables of the CS optimization problem, it is shown with Monte Carlo simulations that the DOAs are accurately reconstructed through polynomial rooting (Root-CS). Polynomial rooting is known to improve the resolution in several other DOA estimation methods. However, traditional methods involve the estimation of the cross-spectral matrix hence they require many snapshots and stationary incoherent sources and are suitable only for uniform linear arrays (ULA). Root-CS does not have these limitations as demonstrated on experimental towed array data from ocean acoustic measurements.
Original languageEnglish
Title of host publicationProceedings of IEEE International Workshop on Compressed Sensing Theory and its Applications to Radar, Sonar and Remote Sensing
Number of pages5
PublisherIEEE
Publication date2015
Article number7330273
ISBN (Print)9781479974207
DOIs
Publication statusPublished - 2015
Event2015 3rd International Workshop on Compressed Sensing Theory and its Applications to Radar, Sonar and Remote Sensing - Pisa, Italy
Duration: 17 Jun 201519 Jun 2015
Conference number: 3
https://ieeexplore.ieee.org/xpl/conhome/7312392/proceeding

Workshop

Workshop2015 3rd International Workshop on Compressed Sensing Theory and its Applications to Radar, Sonar and Remote Sensing
Number3
Country/TerritoryItaly
CityPisa
Period17/06/201519/06/2015
Internet address

Fingerprint

Dive into the research topics of 'Sparse DOA estimation with polynomial rooting'. Together they form a unique fingerprint.

Cite this