Result: Non‐negative matrix factorization and its application in blind sparse source separation with less sensors than sources: Non-negative matrix factorization and its application in blind sparse source separation with less sensors than sources
Title:
Non‐negative matrix factorization and its application in blind sparse source separation with less sensors than sources: Non-negative matrix factorization and its application in blind sparse source separation with less sensors than sources
Authors:
Source:
COMPEL - The international journal for computation and mathematics in electrical and electronic engineering. 24:695-706
Publisher Information:
Emerald, 2005.
Publication Year:
2005
Subject Terms:
Technical applications of optics and electromagnetic theory, Positive matrices and their generalizations, cones of matrices, Numerical mathematical programming methods, Detection theory in information and communication theory, Other matrix algorithms, 0202 electrical engineering, electronic engineering, information engineering, 02 engineering and technology, Factorization of matrices
Document Type:
Academic journal
Article
File Description:
application/xml
Language:
English
ISSN:
0332-1649
DOI:
10.1108/03321640510571174
Access URL:
Rights:
Emerald Insight Site Policies
Accession Number:
edsair.doi.dedup.....db35d4c9770943e5965c3d652016b67d
Database:
OpenAIRE
Further Information
PurposeProposes a non‐negative matrix factorization method.Design/methodology approachPresents an algorithm for finding a suboptimal basis matrix. This is controlled by data cluster centers which can guarantee that the coefficient is very sparse. This leads to the proposition of an application of non‐matrix factorization for blind sparse source separation with less sensors than sources.FindingsTwo simulation examples reveal the validity and performance of the algorithm in this paper.Originality/valueUsing the approach in this paper, the sparse sources can be recovered even if the sources are overlapped to some degree.