Ecole d'ingénieur et centre de recherche en Sciences du numérique

A newton-type forward backward greedy method for multi-Snapshot compressed sensing

Bazzi, Ahmad; Slock, Dirk T.M; Meilhac, Lisa

ASILOMAR 2017, Asilomar Conference on Signals, Systels, and Computers, October 29th-November 1st, 2017, Pacific Grove, CA, USA

Parameter estimation has applications in many applications of signal processing, such as Angle-of-Arrival (AoA) estimation. Compressed sensing is a widely growing paradigm that can be applied to parameter estimation via sparse recovery. In this paper, we propose a Newton-type Forward Backward Greedy method that performs sparse recovery, given the observed data over multiple snapshots. This method is applied to the AoA estimation problem, where we have observed better performance, in terms of Mean-Squared Error and faster convergence when compared to existing methods. More information can be found in the conclusions section.

Document Doi Bibtex

Titre:A newton-type forward backward greedy method for multi-Snapshot compressed sensing
Ville:Pacific Grove
Département:Systèmes de Communication
Eurecom ref:5272
Copyright: © 2017 IEEE. Personal use of this material is permitted. However, permission to reprint/republish this material for advertising or promotional purposes or for creating new collective works for resale or redistribution to servers or lists, or to reuse any copyrighted component of this work in other works must be obtained from the IEEE.
Bibtex: @inproceedings{EURECOM+5272, doi = {}, year = {2017}, title = {{A} newton-type forward backward greedy method for multi-{S}napshot compressed sensing}, author = {{B}azzi, {A}hmad and {S}lock, {D}irk {T}.{M} and {M}eilhac, {L}isa}, booktitle = {{ASILOMAR} 2017, {A}silomar {C}onference on {S}ignals, {S}ystels, and {C}omputers, {O}ctober 29th-{N}ovember 1st, 2017, {P}acific {G}rove, {CA}, {USA}}, address = {{P}acific {G}rove, {\'{E}}{TATS}-{UNIS}}, month = {10}, url = {} }
Voir aussi: