Performance analysis of an adaptive algorithm for tracking dominant subspaces

  title={Performance analysis of an adaptive algorithm for tracking dominant subspaces},
  author={Jean Pierre Delmas and Jean-François Cardoso},
  journal={IEEE Trans. Signal Processing},
This paper provides a performance analysis of a least mean square (LMS) dominant invariant subspace algorithm. Based on an unconstrained minimization problem, this algorithm is a stochastic gradient algorithm driving the columns of a matrix W to an orthonormal basis of a dominant invariant subspace of a correlation matrix. We consider the stochastic algorithm governing the evolution of WW to the projection matrix onto this dominant invariant subspace and study its asymptotic distribution. A… CONTINUE READING
10 Citations
20 References
Similar Papers


Publications referenced by this paper.
Showing 1-10 of 20 references

Convergence studies of Thompson’s unbiased adaptative spectral estimator

  • M. G. Larimore, R. J. Calvert
  • Proc. 14th Asilomar Conf. Circuits Syst. Comput…
  • 1981
Highly Influential
3 Excerpts

Convergence of stochastic algorithms: From the the Kushner-Clark theorem to the Lyapounov functional method

  • J. C. Fort, G. Pag és
  • Adv. Appl. Prob. , vol. 28, pp. 1072–1094, 1996.
  • 1996
1 Excerpt

Similar Papers

Loading similar papers…