Skip to Main content Skip to Navigation
Conference papers

Joint block diagonalization algorithms for optimal separation of multidimensional components

Abstract : This paper deals with non-orthogonal joint block diagonalization. Two algorithms which minimize the Kullback-Leibler divergence between a set of real positive-definite matrices and a block-diagonal transformation thereof are suggested. One algorithm is based on the relative gradient, and the other is based on a quasi-Newton method. These algorithms allow for the optimal, in the mean square error sense, blind separation of multidimensional Gaussian components. Simulations demonstrate the convergence properties of the suggested algorithms, as well as the dependence of the criterion on some of the model parameters
Document type :
Conference papers
Complete list of metadatas

https://hal.telecom-paris.fr/hal-02286752
Contributor : Telecomparis Hal <>
Submitted on : Friday, September 13, 2019 - 4:08:35 PM
Last modification on : Friday, July 31, 2020 - 11:28:08 AM

Links full text

Identifiers

Collections

Citation

Dana Lahat, Jean-François Cardoso, Hagit Messer. Joint block diagonalization algorithms for optimal separation of multidimensional components. Latent variable analysis and signal separation, Mar 2012, Tel Aviv, Israel. pp.155-162, ⟨10.1007/978-3-642-28551-6_20⟩. ⟨hal-02286752⟩

Share

Metrics

Record views

33