Abstract
It is well known that the complex-valued nonunitary joint diagonalization (NUJD) problem in blind source separation (BSS) can be linked to the trilinear models. In this paper, by exploring the special structure of the NUJD problem in BSS, we introduce a novel symmetric tri-quadratic cost function and then derive a simultaneous bi-iterative algorithm (s-BIA) for solving the NUJD problem. Finally, three experiments are conducted in order to show the comparatively good performance of the proposed BSS algorithm.
Original language | English |
---|---|
Pages (from-to) | 3636-3646 |
Number of pages | 11 |
Journal | IEEE Transactions on Signal Processing |
Volume | 59 |
Issue number | 8 |
DOIs | |
Publication status | Published - 2011 |
Keywords
- Bi-iterative algorithm
- blind source separation
- eigenmatrices
- joint diagonalization
- signal processing