Subspace Pursuit Embedded in Orthogonal Matching Pursuit
Data(s) |
2012
|
---|---|
Resumo |
Orthogonal Matching Pursuit (OMP) is a popular greedy pursuit algorithm widely used for sparse signal recovery from an undersampled measurement system. However, one of the main shortcomings of OMP is its irreversible selection procedure of columns of measurement matrix. i.e., OMP does not allow removal of the columns wrongly estimated in any of the previous iterations. In this paper, we propose a modification in OMP, using the well known Subspace Pursuit (SP), to refine the subspace estimated by OMP at any iteration and hence boost the sparse signal recovery performance of OMP. Using simulations we show that the proposed scheme improves the performance of OMP in clean and noisy measurement cases. |
Formato |
application/pdf |
Identificador |
http://eprints.iisc.ernet.in/46340/1/IEEE_TENCON_amb_2012.pdf Ambat, Sooraj K and Chatterjee, Saikat and Hari, KVS (2012) Subspace Pursuit Embedded in Orthogonal Matching Pursuit. In: IEEE Region 10 Conference (TENCON) - Sustainable Development through Humanitarian Technology, NOV 19-22, 2012 , Cebu, PHILIPPINES . |
Publicador |
IEEE |
Relação |
http://dx.doi.org/ 10.1109/TENCON.2012.6412325 http://eprints.iisc.ernet.in/46340/ |
Palavras-Chave | #Electrical Communication Engineering |
Tipo |
Conference Proceedings NonPeerReviewed |