상세 보기
초록
In this paper, an improved FastICA algorithm is proposed for blind source separation by using a sixth-order Newton's method for finding the solutions of the nonlinear equations much faster than other Newton-type iterative methods. In particular, the proposed method has comparable separation performance, fewer iterations, and faster convergence speed.
키워드
independent component analysis; FastICA algorithm; sixth-order Newton's method; Blind source separation; Newton-Raphson method; Nonlinear equations; Independent component analysis; Faster convergence; FastICA algorithm; FastICA algorithms; Newton's methods; Separation performance; Sixth-order Newton's method
- 제목
- Improved FastICA algorithm using a sixth-order Newton's method
- 저자
- Kim, Keun Jun; Zhang, S.; Nam, Sang Won
- 발행일
- 2009-07
- 유형
- Article
- 권
- 6
- 호
- 13
- 페이지
- 904 ~ 909