Sparse recovery in probability via l q -minimization with Weibull random matrices for 0 < q ≤ 1

  • Yi Gao
  • Ji-gen Peng
  • Shi-gang Yue


Although Gaussian random matrices play an important role of measurement matrices in compressed sensing, one hopes that there exist other random matrices which can also be used to serve as the measurement matrices. Hence, Weibull random matrices induce extensive interest. In this paper, we first propose the l2,q robust null space property that can weaken the D-RIP, and show that Weibull random matrices satisfy the l2,q robust null space property with high probability. Besides, we prove that Weibull random matrices also possess the l q quotient property with high probability. Finally, with the combination of the above mentioned properties, we give two important approximation characteristics of the solutions to the l q -minimization with Weibull random matrices, one is on the stability estimate when the measurement noise e ∈ ℝ n needs a priori ||e||2 ≤ є, the other is on the robustness estimate without needing to estimate the bound of ||e||2. The results indicate that the performance of Weibull random matrices is similar to that of Gaussian random matrices in sparse recovery.


compressed sensing lq-minimization Weibull matrices null space property quotient property 

MR Subject Classification

15A52 60E05 94A12 94A20 


Unable to display preview. Download preview PDF.

Unable to display preview. Download preview PDF.


  1. [1]
    A Aldroubi, X Chen, AM Powell. Perturbations of measurement matrices and dictionaries in compressed sensing, Appl Comput Harmon Anal, 2012, 33: 282–291.MathSciNetCrossRefzbMATHGoogle Scholar
  2. [2]
    R Baraniuk, M Davenport, R DeVore, M Wakin. A simple proof of the restricted isometry property for random matrices, Constr Approx, 2008, 28: 253–263.MathSciNetCrossRefzbMATHGoogle Scholar
  3. [3]
    K Bryan, T Leise. Making do with less: An introduction to compressed sensing, SIAM Rev, 2013, 55(3): 547–566.MathSciNetCrossRefzbMATHGoogle Scholar
  4. [4]
    T Cai, L Wang, G Xu. Shifting inequality and recovery of sparse signals, IEEE Trans Signal Process, 2010, 58: 1300–1308.MathSciNetCrossRefGoogle Scholar
  5. [5]
    T Cai, A Zhang. Sharp RIP bound for sparse signal and low-rank matrix recovery, Appl Comput Harmon Anal, 2013, 35: 74–93.MathSciNetCrossRefzbMATHGoogle Scholar
  6. [6]
    T Cai, A Zhang. Sparse representation of a polytope and recovery of sparse signal and low-rank matrices, IEEE Trans Inform Theory, 2014, 60(1): 122–132.MathSciNetCrossRefzbMATHGoogle Scholar
  7. [7]
    Y Cai, S Li. Compressed data separation via dual frames based split-analysis with Weibull matrices, Appl Math J Chinese Univ Ser B, 2013, 28(4): 427–437.MathSciNetCrossRefzbMATHGoogle Scholar
  8. [8]
    E Candès. The restricted isometry property and its implications for compressed sensing, C RMath Acad Sci Paris Ser I, 2008, 346: 589–592.MathSciNetCrossRefzbMATHGoogle Scholar
  9. [9]
    E Candès, Y Eldar, D Needell, P Randall. Compressed sensing with coherent and redundant dictionaries, Appl Comput Harmon Anal, 2011, 31: 59–73.MathSciNetCrossRefzbMATHGoogle Scholar
  10. [10]
    E Candès, J Romberg, T Tao. Robust uncertainty principles: exact signal reconstruction from highly incomplete frequency information, IEEE Trans Inform Theory, 2006, 52(2): 489–509.MathSciNetCrossRefzbMATHGoogle Scholar
  11. [11]
    E Candès, T Tao. Decoding by linear programming, IEEE Trans Inform Theory, 2005, 51(12): 4203–4215.MathSciNetCrossRefzbMATHGoogle Scholar
  12. [12]
    E Candès, T Tao. Near-optimal signal recovery from random projections: universal encoding strategies, IEEE Trans Inform Theory, 2006, 52(12): 5406–5425.MathSciNetCrossRefzbMATHGoogle Scholar
  13. [13]
    R Chartrand. Exact reconstruction of sparse signals via nonconvex minimization, IEEE Signal Process Lett, 2007, 14(10): 707–710.CrossRefGoogle Scholar
  14. [14]
    R Chartrand, V Staneva. Restricted isometry properties and nonconvex compressive sensing, Inverse Problems, 2008, (24): 1–14.MathSciNetCrossRefzbMATHGoogle Scholar
  15. [15]
    A Cohen, W Dahmen, A DeVore. Compressed sensing and best k-term approximation, AmerMath Soc, 2009, 22(1): 211–231.MathSciNetzbMATHGoogle Scholar
  16. [16]
    M Davies, R Gribonval. Restricted isometry constants where lp sparse recovery can fail for 0 < p ≤ 1, IEEE Trans Inform Theory, 2009, 55(5): 2203–2214.MathSciNetCrossRefzbMATHGoogle Scholar
  17. [17]
    RD eVore, G Petrova, P Wojtaszczyk. Instance-optimality in probability with an l1-minimization decoder, Appl Comput Harmon Anal, 2009, 27: 275–288.MathSciNetCrossRefzbMATHGoogle Scholar
  18. [18]
    D Donoho. Compressed sensing, IEEE Trans Inform Theory, 2006, 52(4): 1289–1306.MathSciNetCrossRefzbMATHGoogle Scholar
  19. [19]
    D Donoho, X Huo. Uncertainty principles and ideal atomic decompositions, IEEE Trans Inform Theory, 2001, 47(4): 2845–2862.MathSciNetCrossRefzbMATHGoogle Scholar
  20. [20]
    M Elad. Sparse and Redundant Representations, from Theory to Applications in Signal and Image Processing, Springer, New York, 2010.Google Scholar
  21. [21]
    M Elad, A Bruckstein. A generalized uncertainty principle and sparse representation in pairs of bases, IEEE Trans Inform Theory, 2002, 48(9): 2558–2567.MathSciNetCrossRefzbMATHGoogle Scholar
  22. [22]
    M Elad, P Milanfar, R Rubinstein. Analysis versus synthesis in signal priors, Inverse Problems, 2007, 23(3): 947–968.MathSciNetCrossRefzbMATHGoogle Scholar
  23. [23]
    S Foucart. Stability and robustness of l1-minimizations with Weibull matrices and redundant dictionaries, Linear Algebra Appl, 2014, 441: 4–21.MathSciNetCrossRefzbMATHGoogle Scholar
  24. [24]
    S Foucart, M Lai. Sparsest solutions of underdetermined linear systems via lq-minimization for 0 < q ≤ 1, Appl Comput Harmon Anal, 2009, 26: 395–407.MathSciNetCrossRefzbMATHGoogle Scholar
  25. [25]
    S Foucart, M Lai. Sparse recovery with pre-Gaussian random matrices, Studia Math, 2010, 200: 91–102.MathSciNetCrossRefzbMATHGoogle Scholar
  26. [26]
    S Foucart, H Rauhut. A Mathematical Introduction to Compressive Sensing, Springer Science+ Business Media, New York, 2013.CrossRefzbMATHGoogle Scholar
  27. [27]
    R Gribonval, M Nielsen. Sparse representations in unions of bases, IEEE Trans Inform Theory, 2003, 49(5): 3320–3325.MathSciNetCrossRefzbMATHGoogle Scholar
  28. [28]
    R Gribonval, M Nielsen. Highly sparse representations from dictionaries are unique and independent of the sparseness measure, Appl Comput Harmon Anal, 2007, 22: 335–355.MathSciNetCrossRefzbMATHGoogle Scholar
  29. [29]
    R Handel. Structured Random Matrices, In: Convexity and Concentration, E Carlen, M Madiman, E Werner (eds), The IMA Volumes in Mathematics and its Applications, Vol 161, Springer, New York, 2017, 107–156.Google Scholar
  30. [30]
    J Lin, S Li. Restriced q-isometry properties adapted to frames for nonconvex lq-analysis, IEEE Trans Inform Theory, 2016, 62(8): 4733–4747.MathSciNetzbMATHGoogle Scholar
  31. [31]
    Y Liu, T Mi, S Li. Compressed sensing with general frames via optimal-dual-based l1-analysis, IEEE Trans Inform Theory, 2012, 58(7): 4201–4214.MathSciNetCrossRefzbMATHGoogle Scholar
  32. [32]
    Q Mo, S Li. New bounds on the restricted isometry constant, Appl Comput Harmon Anal, 2011, 31: 460–468.MathSciNetCrossRefzbMATHGoogle Scholar
  33. [33]
    BK Natarajan. Sparse approximate solutions to linear systems, SIAM J Comput, 1995, 24: 227–234.MathSciNetCrossRefzbMATHGoogle Scholar
  34. [34]
    J Peng, S Yue, H Li. NP/CMP equivalence: a phenomenon hidden among sparsity models l0 minimization and lp minimization for information processing, IEEE Trans Inform Theory, 2015, 61(7): 4028–4033.MathSciNetCrossRefzbMATHGoogle Scholar
  35. [35]
    GE Pfander, H Rauhut, JA Tropp. The restricted isometry property for timeCfrequency structured random matrices, Probab Theory Related Fields, 2013, 156(3-4): 707–737.MathSciNetCrossRefzbMATHGoogle Scholar
  36. [36]
    H Rauhut, K Schnass, P Vandergheynst. Compressed sensing and redundant dictionaries, IEEE Trans Inform Theory, 2008, 54(5): 2210–2219.MathSciNetCrossRefzbMATHGoogle Scholar
  37. [37]
    R Saab, Ö Yilmaz. Sparse recovery by non-convex optimization-instance optimality, Appl Comput Harmon Anal, 2010, 29: 30–48.MathSciNetCrossRefzbMATHGoogle Scholar
  38. [38]
    Y Shen, S Li. Restricted p-isometry property and its application for nonconvex compressive sensing, Adv Comput Math, 2012, 37: 441–452.MathSciNetCrossRefzbMATHGoogle Scholar
  39. [39]
    Q Sun. Recovery of sparsest signals via lq-minimization, Appl Comput Harmon Anal, 2012, 32: 329–341.MathSciNetCrossRefzbMATHGoogle Scholar
  40. [40]
    P Wojtaszczyk. Stability and instance optimality for Gaussian measurements in compressed sensing, Found Comput Math, 2010, 10: 1–13.MathSciNetCrossRefzbMATHGoogle Scholar
  41. [41]
    P Wojtaszczyk. l1 minimization with noisy data, SIAM J Numer Anal, 2012, 50(2): 458–467.MathSciNetCrossRefzbMATHGoogle Scholar
  42. [42]
    R Wu, D Chen. The improved bounds of restricted isometry constant for recovery via lpminimization, IEEE Trans Inform Theory, 2013, 59(9): 6142–6147.MathSciNetCrossRefzbMATHGoogle Scholar
  43. [43]
    Z Xu. Compressed sensing: a survey (in Chinese), Sci Sin Math, 2012, 42(9): 865–877.CrossRefGoogle Scholar
  44. [44]
    R Zhang, S Li. A proof of conjecture on restricted isometry property constants δtk(0 < t < 4 3 ), IEEE Trans Inform Theory, 2017, 63: 10.1109/TIT.2017.2705741.Google Scholar
  45. [45]
    X Zhang, S Li. Compressed sensing via dual frame based l1-analysis with Weibull matrices, IEEE Signal Process Lett, 2013, 20(3): 265–268.CrossRefGoogle Scholar
  46. [46]
    S Zhou, L Kong, N Xiu. New bounds for RIC in compressed sensing, J Oper Res Soc China, 2013, 1: 227–237.CrossRefzbMATHGoogle Scholar

Copyright information

© Editorial Committee of Applied Mathematics-A Journal of Chinese Universities and Springer-Verlag GmbH Germany, part of Springer Nature 2018

Authors and Affiliations

  1. 1.School of Mathematics and StatisticsXi’an Jiaotong UniversityXi’anChina
  2. 2.School of Mathematics and Information ScienceNorth Minzu UniversityYinchuanChina
  3. 3.School of Mathematics and Information ScienceGuangzhou UniversityGuangzhouChina
  4. 4.School of Computer ScienceUniversity of LincolnLincolnUK

Personalised recommendations