On bilinear forms based on the resolvent of large random matrices

Walid Hachem; Philippe Loubaton; Jamal Najim; Pascal Vallet

Annales de l'I.H.P. Probabilités et statistiques (2013)

  • Volume: 49, Issue: 1, page 36-63
  • ISSN: 0246-0203

Abstract

top
Consider a N × n non-centered matrix 𝛴 n with a separable variance profile: 𝛴 n = D n 1 / 2 X n D ˜ n 1 / 2 n + A n . Matrices D n and D ˜ n are non-negative deterministic diagonal, while matrix A n is deterministic, and X n is a random matrix with complex independent and identically distributed random variables, each with mean zero and variance one. Denote by Q n ( z ) the resolvent associated to 𝛴 n 𝛴 n * , i.e. Q n ( z ) = 𝛴 n 𝛴 n * - z I N - 1 . Given two sequences of deterministic vectors ( u n ) and ( v n ) with bounded Euclidean norms, we study the limiting behavior of the random bilinear form: u n * Q n ( z ) v n ∀ z ∈ ℂ - ℝ + , as the dimensions of matrix 𝛴 n go to infinity at the same pace. Such quantities arise in the study of functionals of 𝛴 n 𝛴 n * which do not only depend on the eigenvalues of 𝛴 n 𝛴 n * , and are pivotal in the study of problems related to non-centered Gram matrices such as central limit theorems, individual entries of the resolvent, and eigenvalue separation.

How to cite

top

Hachem, Walid, et al. "On bilinear forms based on the resolvent of large random matrices." Annales de l'I.H.P. Probabilités et statistiques 49.1 (2013): 36-63. <http://eudml.org/doc/272084>.

@article{Hachem2013,
abstract = {Consider a $N\times n$ non-centered matrix $\varSigma _\{n\}$ with a separable variance profile: \[\varSigma \_\{n\}=\frac\{D\_\{n\}^\{1/2\}X\_\{n\}\tilde\{D\}\_\{n\}^\{1/2\}\}\{\sqrt\{n\}\}+A\_\{n\}.\] Matrices $D_\{n\}$ and $\tilde\{D\}_\{n\}$ are non-negative deterministic diagonal, while matrix $A_\{n\}$ is deterministic, and $X_\{n\}$ is a random matrix with complex independent and identically distributed random variables, each with mean zero and variance one. Denote by $Q_\{n\}(z)$ the resolvent associated to $\varSigma _\{n\}\varSigma _\{n\}^\{*\}$, i.e. \[Q\_\{n\}(z)=\bigl (\varSigma \_\{n\}\varSigma \_\{n\}^\{*\}-zI\_\{N\}\bigr )^\{-1\}.\] Given two sequences of deterministic vectors $(u_\{n\})$ and $(v_\{n\})$ with bounded Euclidean norms, we study the limiting behavior of the random bilinear form: \[u\_\{n\}^\{*\}Q\_\{n\}(z)v\_\{n\}\quad \forall z\in \mathbb \{C\} -\mathbb \{R\} ^\{+\},\] as the dimensions of matrix $\varSigma _\{n\}$ go to infinity at the same pace. Such quantities arise in the study of functionals of $\varSigma _\{n\}\varSigma _\{n\}^\{*\}$ which do not only depend on the eigenvalues of $\varSigma _\{n\}\varSigma _\{n\}^\{*\}$, and are pivotal in the study of problems related to non-centered Gram matrices such as central limit theorems, individual entries of the resolvent, and eigenvalue separation.},
author = {Hachem, Walid, Loubaton, Philippe, Najim, Jamal, Vallet, Pascal},
journal = {Annales de l'I.H.P. Probabilités et statistiques},
keywords = {random matrix; empirical distribution of the eigenvalues; Stieltjes transform; bilinear forms; spectral norm; wireless communication system; signal processing},
language = {eng},
number = {1},
pages = {36-63},
publisher = {Gauthier-Villars},
title = {On bilinear forms based on the resolvent of large random matrices},
url = {http://eudml.org/doc/272084},
volume = {49},
year = {2013},
}

TY - JOUR
AU - Hachem, Walid
AU - Loubaton, Philippe
AU - Najim, Jamal
AU - Vallet, Pascal
TI - On bilinear forms based on the resolvent of large random matrices
JO - Annales de l'I.H.P. Probabilités et statistiques
PY - 2013
PB - Gauthier-Villars
VL - 49
IS - 1
SP - 36
EP - 63
AB - Consider a $N\times n$ non-centered matrix $\varSigma _{n}$ with a separable variance profile: \[\varSigma _{n}=\frac{D_{n}^{1/2}X_{n}\tilde{D}_{n}^{1/2}}{\sqrt{n}}+A_{n}.\] Matrices $D_{n}$ and $\tilde{D}_{n}$ are non-negative deterministic diagonal, while matrix $A_{n}$ is deterministic, and $X_{n}$ is a random matrix with complex independent and identically distributed random variables, each with mean zero and variance one. Denote by $Q_{n}(z)$ the resolvent associated to $\varSigma _{n}\varSigma _{n}^{*}$, i.e. \[Q_{n}(z)=\bigl (\varSigma _{n}\varSigma _{n}^{*}-zI_{N}\bigr )^{-1}.\] Given two sequences of deterministic vectors $(u_{n})$ and $(v_{n})$ with bounded Euclidean norms, we study the limiting behavior of the random bilinear form: \[u_{n}^{*}Q_{n}(z)v_{n}\quad \forall z\in \mathbb {C} -\mathbb {R} ^{+},\] as the dimensions of matrix $\varSigma _{n}$ go to infinity at the same pace. Such quantities arise in the study of functionals of $\varSigma _{n}\varSigma _{n}^{*}$ which do not only depend on the eigenvalues of $\varSigma _{n}\varSigma _{n}^{*}$, and are pivotal in the study of problems related to non-centered Gram matrices such as central limit theorems, individual entries of the resolvent, and eigenvalue separation.
LA - eng
KW - random matrix; empirical distribution of the eigenvalues; Stieltjes transform; bilinear forms; spectral norm; wireless communication system; signal processing
UR - http://eudml.org/doc/272084
ER -

References

top
  1. [1] C. Artigue and P. Loubaton. On the precoder design of flat fading MIMO systems equipped with MMSE receivers: A large-system approach. IEEE Trans. Inform. Theory57 (2011) 4138–4155. MR2840447
  2. [2] Z. D. Bai, B. Q. Miao and G. M. Pan. On asymptotics of eigenvectors of large sample covariance matrix. Ann. Probab.35 (2007) 1532–1572. Zbl1162.15012MR2330979
  3. [3] Z. D. Bai and J. W. Silverstein. No eigenvalues outside the support of the limiting spectral distribution of large-dimensional sample covariance matrices. Ann. Probab.26 (1998) 316–345. Zbl0937.60017MR1617051
  4. [4] Z. D. Bai and J. W. Silverstein. Exact separation of eigenvalues of large-dimensional sample covariance matrices. Ann. Probab.27 (1999) 1536–1555. Zbl0964.60041MR1733159
  5. [5] Z. Bai and J. W. Silverstein. No eigenvalues outside the support of the limiting spectral distribution of information-plus-noise type matrices. Random Matrices Theory Appl. 1 (2012) 1150004. Zbl1248.15028MR2930382
  6. [6] F. Benaych-Georges and R. N. Rao. The eigenvalues and eigenvectors of finite, low rank perturbations of large random matrices. Preprint, 2009. Available at arXiv:0910.2120. Zbl1226.15023
  7. [7] R. B. Dozier and J. W. Silverstein. On the empirical distribution of eigenvalues of large dimensional information-plus-noise-type matrices. J. Multivariate Anal.98 (2007) 678–694. Zbl1115.60035MR2322123
  8. [8] M. Capitaine, C. Donati-Martin and D. Féral. The largest eigenvalues of finite rank deformation of large Wigner matrices: Convergence and nonuniversality of the fluctuations. Ann. Probab.37 (2009) 1–47. Zbl1163.15026MR2489158
  9. [9] J. Dumont, W. Hachem, S. Lasaulce, P. Loubaton and J. Najim. On the capacity achieving covariance matrix for Rician MIMO channels: An asymptotic approach. IEEE Trans. Inform. Theory56 (2010) 1048–1069. MR2723661
  10. [10] L. Erdös, H.-T. Yau and J. Yin. Rigidity of eigenvalues of generalized Wigner matrices. Unpublished manuscript, 2010. Available at[4] http://arxiv.org/pdf/1007.4652. Zbl1238.15017MR2871147
  11. [11] V. L. Girko. An Introduction to Statistical Analysis of Random Arrays. VSP, Utrecht, 1998. Zbl0934.62055MR1694087
  12. [12] U. Haagerup and S. Thorbjørnsen. A new application of random matrices: Ext ( C red * ( F 2 ) ) is not a group. Ann. of Math. (2) 162 (2005) 711–775. Zbl1103.46032MR2183281
  13. [13] W. Hachem, M. Kharouf, J. Najim and J. Silverstein. A CLT for information-theoretic statistics of non-centered Gram random matrices. Random Matrices Theory Appl. 1 (2012) 1150010. Zbl1262.15039MR2934716
  14. [14] W. Hachem, P. Loubaton and J. Najim. The empirical distribution of the eigenvalues of a Gram matrix with a given variance profile. Ann. Inst. Henri Poincaré Probab. Stat.42 (2006) 649–670. Zbl1109.15021MR2269232
  15. [15] W. Hachem, P. Loubaton and J. Najim. Deterministic equivalents for certain functionals of large random matrices. Ann. Appl. Probab.17 (2007) 875–930. Zbl1181.15043MR2326235
  16. [16] R. A. Horn and C. R. Johnson. Topics in Matrix Analysis. Cambridge Univ. Press, Cambridge, 1994. Zbl0801.15001MR1288752
  17. [17] A. Kammoun, M. Kharouf, W. Hachem, J. Najim and A. El Kharroubi. On the fluctuations of the mutual information for non centered MIMO channels: The non Gaussian case. In Proc. IEEE International Workshop on Signal Processing Advances in Wireless Communications (SPAWC), 2010. 
  18. [18] V. A. Marčenko and L. A. Pastur. Distribution of eigenvalues in certain sets of random matrices. Mat. Sb. (N.S.) 72 (1967) 507–536. Zbl0152.16101MR208649
  19. [19] X. Mestre. Improved estimation of eigenvalues and eigenvectors of covariance matrices using their sample estimates. IEEE Trans. Inform. Theory54 (2008) 5113–5129. Zbl1318.62191MR2589886
  20. [20] X. Mestre. On the asymptotic behavior of the sample estimates of eigenvalues and eigenvectors of covariance matrices. IEEE Trans. Signal Process.56 (2008) 5353–5368. MR2472837
  21. [21] X. Mestre and M. A. Lagunas. Modified subspace algorithms for DOA estimation with large arrays. IEEE Trans. Signal Process.56 (2008) 598–614. MR2445537
  22. [22] W. Rudin. Real and Complex Analysis, 3rd edition. McGraw-Hill, New York, 1986. Zbl0278.26001
  23. [23] J. W. Silverstein. Strong convergence of the empirical distribution of eigenvalues of large-dimensional random matrices. J. Multivariate Anal.55 (1995) 331–339. Zbl0851.62015MR1370408
  24. [24] J. W. Silverstein and Z. D. Bai. On the empirical distribution of eigenvalues of a class of large-dimensional random matrices. J. Multivariate Anal.54 (1995) 175–192. Zbl0833.60038MR1345534
  25. [25] P. Vallet, P. Loubaton and X. Mestre. Improved subspace estimation for multivariate observations of high dimension: The deterministic signals case. IEEE Trans. Inform. Theory58 (2012) 1043–1068. MR2918009

NotesEmbed ?

top

You must be logged in to post comments.