CP Decomposition and Its Application in Noise Reduction and Multiple Sources Identification

DOI

10.2991/csic-15.2015.12How to use a DOI?

Keywords

CP decomposition, Noise reduction, Multiple sources identification, identification, optimal number of sensors

Abstract

The CANDECOMP/PARAFAC (Canonical Decomposition / Parallel Factor Analysis, ab. CP) decomposition of a higher-order tensor is a powerful multilinear algebra, thus denoising observed data and identification of multiple sources can both be accomplished by the CP decomposition. In this paper, images are used to illustrate the denoising effect of CP decomposition. To identify sources, the number of senors affects the accuracy and convergence of algorithm greatly, especially, if the angle difference is smaller the problem would be worse. So, we concern to improve the accuracy of the identification of close sources with CP decomposition. By searching for the optimal number of sensors and denoising the original data by CP beforehand, the fewer number of sensors could be used and the accuracy is improved. Simulation results show that our technique outperforms good performances in both denoising data and identification of close sources.

Copyright

© 2015, the Authors. Published by Atlantis Press.

Open Access

This is an open access article distributed under the CC BY-NC license (http://creativecommons.org/licenses/by-nc/4.0/).

Download article (PDF)

TY  - CONF
AU  - Xuefeng Liu
AU  - Yuping Feng
AU  - Ximei Liu
AU  - Biao Huang
PY  - 2015/07
DA  - 2015/07
TI  - CP Decomposition and Its Application in Noise Reduction and Multiple Sources Identification
BT  - Proceedings of the 2015 International Conference on Computer Science and Intelligent Communication
PB  - Atlantis Press
SP  - 47
EP  - 51
SN  - 2352-538X
UR  - https://doi.org/10.2991/csic-15.2015.12
DO  - 10.2991/csic-15.2015.12
ID  - Liu2015/07
ER  -