Blind source separation of Fast ICA algorithm based on Negentropy and its optimization

Yi Wan; Yanghong Zhou; Zichen Yang

doi:10.1117/12.2672323

28 March 2023 Blind source separation of Fast ICA algorithm based on Negentropy and its optimization

Yi Wan, Yanghong Zhou, Zichen Yang

Proceedings Volume 12597, Second International Conference on Statistics, Applied Mathematics, and Computing Science (CSAMCS 2022); 1259737 (2023) https://doi.org/10.1117/12.2672323
Event: Second International Conference on Statistics, Applied Mathematics, and Computing Science (CSAMCS 2022), 2022, Nanjing, China

Abstract

Blind Source Separation (BSS) is the problem of dividing a set of source signals from a mixed one. At present, Independent Component Analysis (ICA) is the main method to solve such problems. In this paper, the ICA algorithm based on maximum likelihood estimation is introduced at first. On the basis of summarizing its principle, we implement the data processing with MATLAB code. Then we give the FastICA algorithm based on Negentropy as an upgraded ICA. Further, by optimizing the distribution function, we derive the optimizing form of FastICA. The experimental simulation of this optimization algorithm is carried out with the previous two algorithms. After we analyze the experimental data, we confirm the superiority of the FastICA algorithm and the effectiveness of the optimized FastICA algorithm, and finally solve the cocktail party problem mentioned in the introduction.

Citation Download Citation

Yi Wan, Yanghong Zhou, and Zichen Yang "Blind source separation of Fast ICA algorithm based on Negentropy and its optimization", Proc. SPIE 12597, Second International Conference on Statistics, Applied Mathematics, and Computing Science (CSAMCS 2022), 1259737 (28 March 2023); https://doi.org/10.1117/12.2672323

ACCESS THE FULL ARTICLE

INSTITUTIONAL
Select your institution to access the SPIE Digital Library.

SELECT YOUR INSTITUTION

PERSONAL
Sign in with your SPIE account to access your personal subscriptions or to use specific features such as save to my library, sign up for alerts, save searches, etc.

PERSONAL SIGN IN

No SPIE Account? Create one

PURCHASE THIS CONTENT

SUBSCRIBE TO DIGITAL LIBRARY

50 downloads per 1-year subscription

Members: $195

Non-members: $335 ADD TO CART

25 downloads per 1 - year subscription

Members: $145

Non-members: $250 ADD TO CART

PURCHASE SINGLE ARTICLE

Includes PDF, HTML & Video, when available