Variable step-size sign natural gradient algorithm for sequential blind source separation
A novel variable step-size sign natural gradient algorithm (VS-S-NGA) for online blind separation of independent sources is presented. A sign operator for the adaptation of the separation model is obtained from the derivation of a generalized dynamic separation model. A variable step size is also de...
Saved in:
Main Authors: | , , |
---|---|
Format: | Default Article |
Published: |
2005
|
Subjects: | |
Online Access: | https://hdl.handle.net/2134/5771 |
Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
id |
rr-article-9565538 |
---|---|
record_format |
Figshare |
spelling |
rr-article-95655382005-01-01T00:00:00Z Variable step-size sign natural gradient algorithm for sequential blind source separation Lianxi Yuan (7212854) Wenwu Wang (4352767) Jonathon Chambers (1251609) Mechanical engineering not elsewhere classified Adaptive step size Natural gradient Sequential blind source separation (BSS) Sign algorithm Mechanical Engineering not elsewhere classified A novel variable step-size sign natural gradient algorithm (VS-S-NGA) for online blind separation of independent sources is presented. A sign operator for the adaptation of the separation model is obtained from the derivation of a generalized dynamic separation model. A variable step size is also derived to better match the dynamics of the input signals and unmixing matrix. The proposed sign algorithm is appealing in practice due to its computational simplicity. Experimental results verify the superior convergence performance over conventional NGA in both stationary and nonstationary environments. 2005-01-01T00:00:00Z Text Journal contribution 2134/5771 https://figshare.com/articles/journal_contribution/Variable_step-size_sign_natural_gradient_algorithm_for_sequential_blind_source_separation/9565538 CC BY-NC-ND 4.0 |
institution |
Loughborough University |
collection |
Figshare |
topic |
Mechanical engineering not elsewhere classified Adaptive step size Natural gradient Sequential blind source separation (BSS) Sign algorithm Mechanical Engineering not elsewhere classified |
spellingShingle |
Mechanical engineering not elsewhere classified Adaptive step size Natural gradient Sequential blind source separation (BSS) Sign algorithm Mechanical Engineering not elsewhere classified Lianxi Yuan Wenwu Wang Jonathon Chambers Variable step-size sign natural gradient algorithm for sequential blind source separation |
description |
A novel variable step-size sign natural gradient algorithm (VS-S-NGA) for online blind separation of independent sources is presented. A sign operator for the adaptation of the separation model is obtained from the derivation of a generalized dynamic separation model. A variable step size is also derived to better match the dynamics of the input signals and unmixing matrix. The proposed sign algorithm is appealing in practice due to its computational simplicity. Experimental results verify the superior convergence performance over conventional NGA in both stationary and nonstationary environments. |
format |
Default Article |
author |
Lianxi Yuan Wenwu Wang Jonathon Chambers |
author_facet |
Lianxi Yuan Wenwu Wang Jonathon Chambers |
author_sort |
Lianxi Yuan (7212854) |
title |
Variable step-size sign natural gradient algorithm for sequential blind source separation |
title_short |
Variable step-size sign natural gradient algorithm for sequential blind source separation |
title_full |
Variable step-size sign natural gradient algorithm for sequential blind source separation |
title_fullStr |
Variable step-size sign natural gradient algorithm for sequential blind source separation |
title_full_unstemmed |
Variable step-size sign natural gradient algorithm for sequential blind source separation |
title_sort |
variable step-size sign natural gradient algorithm for sequential blind source separation |
publishDate |
2005 |
url |
https://hdl.handle.net/2134/5771 |
_version_ |
1797471048400109568 |