The Annals of Applied Probability

Improved bounds for sparse recovery from subsampled random convolutions

Abstract

We study the recovery of sparse vectors from subsampled random convolutions via $\ell_{1}$-minimization. We consider the setup in which both the subsampling locations as well as the generating vector are chosen at random. For a sub-Gaussian generator with independent entries, we improve previously known estimates: if the sparsity $s$ is small enough, that is, $s\lesssim\sqrt{n/\log(n)}$, we show that $m\gtrsim s\log(en/s)$ measurements are sufficient to recover $s$-sparse vectors in dimension $n$ with high probability, matching the well-known condition for recovery from standard Gaussian measurements. If $s$ is larger, then essentially $m\geq s\log^{2}(s)\log(\log(s))\log(n)$ measurements are sufficient, again improving over previous estimates. Our results are shown via the so-called robust null space property which is weaker than the standard restricted isometry property. Our method of proof involves a novel combination of small ball estimates with chaining techniques which should be of independent interest.

Article information

Source
Ann. Appl. Probab., Volume 28, Number 6 (2018), 3491-3527.

Dates
Revised: March 2018
First available in Project Euclid: 8 October 2018

https://projecteuclid.org/euclid.aoap/1538985627

Digital Object Identifier
doi:10.1214/18-AAP1391

Mathematical Reviews number (MathSciNet)
MR3861818

Zentralblatt MATH identifier
06994398

Subjects
Primary: 94A20: Sampling theory
Secondary: 60B20: Random matrices (probabilistic aspects; for algebraic aspects see 15B52)

Citation

Mendelson, Shahar; Rauhut, Holger; Ward, Rachel. Improved bounds for sparse recovery from subsampled random convolutions. Ann. Appl. Probab. 28 (2018), no. 6, 3491--3527. doi:10.1214/18-AAP1391. https://projecteuclid.org/euclid.aoap/1538985627

References

• [1] Ailon, N. and Liberty, E. (2009). Fast dimension reduction using Rademacher series on dual BCH codes. Discrete Comput. Geom. 42 615–630.
• [2] Ailon, N. and Rauhut, H. (2014). Fast and RIP-optimal transforms. Discrete Comput. Geom. 52 780–798.
• [3] Bajwa, W. U., Haupt, J. D., Raz, G. M., Wright, S. J. and Nowak, R. D. (2007). Toeplitz-structured compressed sensing matrices. In 2007 IEEE/SP 14th Workshop on Statistical Signal Processing 294–298. IEEE, New York.
• [4] Baraniuk, R., Davenport, M., DeVore, R. and Wakin, M. (2008). A simple proof of the restricted isometry property for random matrices. Constr. Approx. 28 253–263.
• [5] Bourgain, J. (2014). An improved estimate in the restricted isometry problem. In Geometric Aspects of Functional Analysis. Lecture Notes in Math. 2116 65–70. Springer, Cham.
• [6] Candès, E. J., Romberg, J. and Tao, T. (2006). Robust uncertainty principles: Exact signal reconstruction from highly incomplete frequency information. IEEE Trans. Inform. Theory 52 489–509.
• [7] Candes, E. J. and Tao, T. (2006). Near-optimal signal recovery from random projections: Universal encoding strategies? IEEE Trans. Inform. Theory 52 5406–5425.
• [8] Carl, B. (1985). Inequalities of Bernstein–Jackson-type and the degree of compactness of operators in Banach spaces. Ann. Inst. Fourier (Grenoble) 35 79–118.
• [9] Chandrasekaran, V., Recht, B., Parrilo, P. A. and Willsky, A. S. (2012). The convex geometry of linear inverse problems. Found. Comput. Math. 12 805–849.
• [10] Chen, S. S., Donoho, D. L. and Saunders, M. A. (1998). Atomic decomposition by basis pursuit. SIAM J. Sci. Comput. 20 33–61.
• [11] de la Peña, V. H. and Giné, E. (1999). Decoupling. From Dependence to Independence, Randomly Stopped Processes. $U$-Statistics and Processes. Martingales and Beyond. Springer, New York.
• [12] Dirksen, S., Jung, H. and Rauhut, H. (2017). One-bit compressed sensing with partial Gaussian circulant matrices. Preprint. Available at arXiv:1710.03287.
• [13] Dirksen, S., Lecué, G. and Rauhut, H. (2016). On the gap between restricted isometry properties and sparse recovery conditions. IEEE Trans. Inform. Theory.
• [14] Donoho, D. L. (2006). Compressed sensing. IEEE Trans. Inform. Theory 52 1289–1306.
• [15] Donoho, D. L. and Tanner, J. (2009). Counting faces of randomly projected polytopes when the projection radically lowers dimension. J. Amer. Math. Soc. 22 1–53.
• [16] Dorsch, D. and Rauhut, H. (2017). Refined analysis of sparse MIMO radar. J. Fourier Anal. Appl. 23 485–529.
• [17] Foucart, S., Pajor, A., Rauhut, H. and Ullrich, T. (2010). The Gelfand widths of $\ell_{p}$-balls for $0<p\leq1$. J. Complexity 26 629–640.
• [18] Foucart, S. and Rauhut, H. (2013). A Mathematical Introduction to Compressive Sensing. Birkhäuser, Basel.
• [19] Haupt, J., Bajwa, W. U., Raz, G. and Nowak, R. (2010). Toeplitz compressed sensing matrices with applications to sparse channel estimation. IEEE Trans. Inform. Theory 56 5862–5875.
• [20] Haviv, I. and Regev, O. (2016). The restricted isometry property of subsampled Fourier matrices. In Proceedings of the Twenty-Seventh Annual ACM-SIAM Symposium on Discrete Algorithms 288–297. ACM, New York.
• [21] Hügel, M., Rauhut, H. and Strohmer, T. (2014). Remote sensing via $\ell_{1}$-minimization. Found. Comput. Math. 14 115–150.
• [22] James, D. and Rauhut, H. (2015). Nonuniform sparse recovery with random convolutions. In Proc. SampTA 2015, Washington.
• [23] Kabanava, M. and Rauhut, H. (2015). Analysis $\ell_{1}$-recovery with frames and Gaussian measurements. Acta Appl. Math. 140 173–195.
• [24] Klartag, B. (2002). $5n$ Minkowski symmetrizations suffice to arrive at an approximate Euclidean ball. Ann. of Math. (2) 156 947–960.
• [25] Krahmer, F., Mendelson, S. and Rauhut, H. (2014). Suprema of chaos processes and the restricted isometry property. Comm. Pure Appl. Math. 67 1877–1904.
• [26] Krahmer, F. and Ward, R. (2011). New and improved Johnson–Lindenstrauss embeddings via the restricted isometry property. SIAM J. Math. Anal. 43 1269–1281.
• [27] Lecué, G. and Mendelson, S. (2017). Sparse recovery under weak moment assumptions. J. Eur. Math. Soc. (JEMS) 19 881–904.
• [28] Ledoux, M. and Talagrand, M. (1991). Probability in Banach Spaces: Isoperimetry and Processes. Ergebnisse der Mathematik und Ihrer Grenzgebiete (3) [Results in Mathematics and Related Areas (3)] 23. Springer, Berlin.
• [29] Mendelson, S., Pajor, A. and Tomczak-Jaegermann, N. (2007). Reconstruction and subgaussian operators in asymptotic geometric analysis. Geom. Funct. Anal. 17 1248–1282.
• [30] Mendelson, S., Pajor, A. and Tomczak-Jaegermann, N. (2008). Uniform uncertainty principle for Bernoulli and subgaussian ensembles. Constr. Approx. 28 277–289.
• [31] Mendelson, S. and Vershynin, R. (2004). Remarks on the geometry of coordinate projections in $\mathbb{R}^{n}$. Israel J. Math. 140 203–220.
• [32] Oliveira, R. I. (2013). The lower tail of random quadratic forms with applications to ordinary least squares and restricted eigenvalue properties. Preprint. Available at arXiv:1312.2903.
• [33] Pfander, G. E. and Rauhut, H. (2010). Sparsity in time-frequency representations. J. Fourier Anal. Appl. 16 233–260.
• [34] Pfander, G. E., Rauhut, H. and Tropp, J. A. (2013). The restricted isometry property for time-frequency structured random matrices. Probab. Theory Related Fields 156 707–737.
• [35] Rauhut, H. (2009). Circulant and Toeplitz matrices in compressed sensing. In Proc. SPARS ’09.
• [36] Rauhut, H. (2010). Compressive sensing and structured random matrices. In Theoretical Foundations and Numerical Methods for Sparse Recovery (M. Fornasier, ed.). Radon Ser. Comput. Appl. Math. 9 1–92. de Gruyter, Berlin.
• [37] Rauhut, H., Romberg, J. and Tropp, J. A. (2012). Restricted isometries for partial random circulant matrices. Appl. Comput. Harmon. Anal. 32 242–254.
• [38] Rauhut, H. and Ward, R. (2016). Interpolation via weighted $\ell_{1}$ minimization. Appl. Comput. Harmon. Anal. 40 321–351.
• [39] Romberg, J. (2009). Compressive sensing by random convolution. SIAM J. Imaging Sci. 2 1098–1128.
• [40] Rudelson, M. and Vershynin, R. (2008). On sparse reconstruction from Fourier and Gaussian measurements. Comm. Pure Appl. Math. 61 1025–1045.
• [41] Rudelson, M. and Vershynin, R. (2015). Small ball probabilities for linear images of high-dimensional distributions. Int. Math. Res. Not. IMRN 2015 9594–9617.
• [42] Talagrand, M. (2014). Upper and Lower Bounds for Stochastic Processes. Springer, Heidelberg.
• [43] Tropp, J. A., Wakin, M. B., Duarte, M. F., Baron, D. and Baraniuk, R. G. (2006). Random filters for compressive sampling and reconstruction. In 2006 IEEE International Conference on Acoustics Speech and Signal Processing Proceedings 3 III–III. IEEE, New York.
• [44] Vershynin, R. (2012). Introduction to the non-asymptotic analysis of random matrices. In Compressed Sensing 210–268. Cambridge Univ. Press, Cambridge.