article · 2010

A Blind Source Separation Algorithm Based on Whitening and Non-linear Decorrelation

Zhijun Li, Jianping An, Лэй Сун, Miao Yang

Cited by 3Abstract and details only

Abstract

Blind source separation (BSS) methods become more attractive targets in neural network and signal processing literature recently. And independent component analysis (ICA) methods are one kind of widely used and important solutions to blind source separation problems. Whitening is a very useful preparation step for blind separation of source and meanwhile non-linear decorrelation acts as a basic estimation theory upon ICA methods. Then an on-line real-time blind source separation algorithm based on the whitening and non-linear decorrelation is proposed in this paper. Performance of the proposed method is simulated and analyzed and then the results show that our proposed algorithm has better convergence speed and steady-state error than the normalized EASI, nature gradient and whitening-only methods.

Cite this paper

APA 7

Li, Z., An, J., Сун, Л., & Yang, M. (2010). A Blind Source Separation Algorithm Based on Whitening and Non-linear Decorrelation. https://doi.org/10.1109/iccms.2010.123

MLA 9

Li, Zhijun, et al. “A Blind Source Separation Algorithm Based on Whitening and Non-linear Decorrelation.” 2010, pp. 443–47, https://doi.org/10.1109/iccms.2010.123.

Chicago

Li, Zhijun, Jianping An, Лэй Сун, and Miao Yang. 2010. “A Blind Source Separation Algorithm Based on Whitening and Non-linear Decorrelation.” https://doi.org/10.1109/iccms.2010.123.

Harvard

Li, Z. et al. (2010) ‘A Blind Source Separation Algorithm Based on Whitening and Non-linear Decorrelation’. Available at: https://doi.org/10.1109/iccms.2010.123.

BibTeX

@article{li2010blind,
  title = {{A Blind Source Separation Algorithm Based on Whitening and Non-linear Decorrelation}},
  author = {Li, Zhijun and An, Jianping and Сун, Лэй and Yang, Miao},
  pages = {443--447},
  year = {2010},
  doi = {10.1109/iccms.2010.123}
}

Made from the index record. Check names, capitalisation and page numbers against the paper. Need another style? AI for Research has over 10,000.