Check out RSS, or use RSS reader to subscribe this item
Confirmation
Authentication email has already been sent, please check your email box: and activate it as soon as possible.
You can login to My Profile and manage your email alerts.
Sponsored by the Center for Science and Technology Development of the Ministry of Education
Supervised by Ministry of Education of the People's Republic of China
In this paper, the convolutive blind source separation (CBSS) via joint block diagonalization (JBD) technique is revisited to reduce the rigorous constraint. As is known that the CBSS signal model under the JBD framework is equivalent to an instantaneous mixture model, where the mixing matrix is almost always non-square. This considerably restricts the applicability of existing JBD algorithms in CBSS problem for non-square mixing cases. In this paper the nonunitary JBD problem is reformulated as a multicriteria optimization model, where the mixing matrix can be non-square. Moreover, by optimizing the proposed model the resulting algorithm can eliminate the degenerate solutions in nonunitary JBD. The simulation results show that the proposed algorithm outperforms the existing JBD algorithms in terms of separation accuracy and stability.