K-EVD clustering and its applications to sparse component analysis

Authors:
Zhaoshui He;Andrzej Cichocki
Affiliations:
Laboratory for Advanced Brain Signal Processing, RIKEN Brain Science Institute, Wako-shi, Saitama, Japan;Laboratory for Advanced Brain Signal Processing, RIKEN Brain Science Institute, Wako-shi, Saitama, Japan
Venue:
ICA'06 Proceedings of the 6th international conference on Independent Component Analysis and Blind Signal Separation
Year:
2006

Citing 7
Cited 6

Dictionary learning algorithms for sparse representation

Neural Computation
Analysis of sparse representation and blind source separation

Neural Computation
Blind separation of speech mixtures via time-frequency masking

IEEE Transactions on Signal Processing
Performance analysis of minimum ℓ1-norm solutions for underdetermined source separation

IEEE Transactions on Signal Processing
Underdetermined blind source separation based on sparse representation

IEEE Transactions on Signal Processing
Sparse component analysis and blind source separation of underdetermined mixtures

IEEE Transactions on Neural Networks
Blind estimation of channel parameters and source components for EEG signals: a sparse factorization approach

IEEE Transactions on Neural Networks

Structure learning by pruning in independent component analysis

Neurocomputing
An Efficient K-Hyperplane Clustering Algorithm and Its Application to Sparse Component Analysis

ISNN '07 Proceedings of the 4th international symposium on Neural Networks: Part II--Advances in Neural Networks
Underdetermined Blind Source Separation Using SVM

ISNN '07 Proceedings of the 4th international symposium on Neural Networks: Advances in Neural Networks, Part III
K-hyperline clustering learning for sparse component analysis

Signal Processing
K-hyperplanes clustering and its application to sparse component analysis

ICONIP'06 Proceedings of the 13 international conference on Neural Information Processing - Volume Part I
Matrix estimation based on normal vector of hyperplane in sparse component analysis

ICSI'10 Proceedings of the First international conference on Advances in Swarm Intelligence - Volume Part II

Quantified Score

Hi-index	0.00

Visualization

Abstract

In this paper we introduce a simple distance measure from a m-dimensional point a hyper-line in the complex-valued domain. Based on this distance measure, the K-EVD clustering algorithm is proposed for estimating the basis matrix A in sparse representation model X= AS+ N Compared to existing clustering algorithms, the proposed one has advantages in two aspects: it is very fast; furthermore, when the number of basis vectors is overestimated, this algorithm can estimate and identify the significant basis vectors which represent a basis matrix, thus the number of sources can be also precisely estimated. We have applied the proposed approach for blind source separation. The simulations show that the proposed algorithm is reliable and of high accuracy, even when the number of sources is unknown and/or overestimated.