We gratefully acknowledge support from
the Simons Foundation and member institutions.
Full-text links:

Download:

Current browse context:

cs

Change to browse by:

References & Citations

DBLP - CS Bibliography

Bookmark

(what is this?)
CiteULike logo BibSonomy logo Mendeley logo del.icio.us logo Digg logo Reddit logo ScienceWISE logo

Computer Science > Information Theory

Title: A remark on weaken restricted isometry property in compressed sensing

Authors: Hui Zhang
Abstract: The restricted isometry property (RIP) has become well-known in the compressed sensing community. Recently, a weaken version of RIP was proposed for exact sparse recovery under weak moment assumptions. In this note, we prove that the weaken RIP is also sufficient for \textsl{stable and robust} sparse recovery by linking it with a recently introduced robust width property in compressed sensing. Moreover, we show that it can be widely apply to other compressed sensing instances as well.
Comments: 7 pages
Subjects: Information Theory (cs.IT)
Cite as: arXiv:1504.00086 [cs.IT]
  (or arXiv:1504.00086v1 [cs.IT] for this version)

Submission history

From: Hui Zhang [view email]
[v1] Wed, 1 Apr 2015 02:42:02 GMT (7kb)

Link back to: arXiv, form interface, contact.