Robust non-unitary joint block diagonalisation for non-square mixture.

The non-unitary joint block diagonalisation (JBD) of a set of given matrices in non-square mixing situation is addressed. To eliminate the degenerate solutions, the JBD problem is reformulated as a multicriteria optimisation model. The resulting algorithm is numerically robust for non-square mixing matrix estimation. Moreover, it outperforms the existing JBD algorithms in terms of convergence rate and stability, which has been verified by simulation results. [ABSTRACT FROM AUTHOR]