A Note on Stability and Instance Optimality in Compressed Sensing

Abstract:

In this note, it is proved that every -sparse signal vector can be recovered stably from the measurement vector via minimization as soon as the restricted isometry constant of the measurement matrix is smaller than . Note that our results contain the case of noisy data, therefore previous known results in the literature are extent and improved. Also we obtain the results on the stability and instance optimality for some random measurement matrices.

• Periodical: Applied Mechanics and Materials (Volumes 130-134)
• Pages: 4194-4197
• DOI: https://doi.org/10.4028/www.scientific.net/AMM.130-134.4194
• Online since: October 2011
• Authors: Sheng Zhang, Pei Xin Ye
• Keywords: Compressed Sensing, Instance Optimality, Restricted Isometry Constant, Stability

References

[1] R. Baraniuk, M. Davenport, R. DeVore, M. Wakin: Constr. Approx. Vol. 28 (2008), p.253.

DOI: 10.1007/s00365-007-9003-x

[2] E. Candès, J. Romberg, T. Tao: IEEE Trans. Inf. Theory Vol. 52 (2006), p.489.

[3] E. Candès, J. Romberg, T. Tao: Commun. Pure Appl. Math. Vol. 59(2005), p.1207.

[4] E. Candès, T. Tao: IEEE Trans. Inf. Theory Vol. 51 (2005), p.4203.

[5] E. Candès: Compte Rendus de l'Academie des Sciences, Paris, Series I, Vol. 346 (2008), p.589.

[6] A. Cohen, W. Dahmen, R. DeVore: J. Amer. Math. Soc. Vol. 22 (2009), p.211.

[7] R. DeVore, G. Petrova, P. Wojtaszczyk: Appl. Comput. Harmon. Anal. Vol. 27(2009), p.275.

[8] D. Donoho: IEEE Trans. Inf. Theory Vol. 52(2006), p.1289.

[9] S. Foucart, M. Lai: Appl. Comput. Harmon. Anal. 26(2009), p.395.

[10] S. Foucart: Appl. Comput. Harmon. Anal. Vol. 29 (2010), p.97.

[11] P. Wojtaszczyk: Found Comput Math Vol. 10 (2010), p.1.

[12] S. Zhang, P. Ye, Stability and instance optimality in compressed sensing with regard to some measurement matrices, preprint.