New algorithms for adaptive BSS

Blind Source Separation (BSS) is a vast field of research where a large number of BSS algorithms have already been proposed. However, derivation of adaptive BSS solutions with good trade-off between the computational cost and the convergence rate is still a challenging problem. In this work we consider low-cost adaptive BSS methods using statistical independence criteria. We introduce new algorithms using fast data whitening followed by source separation via Givens rotations. These algorithms are compared to other existing techniques and shown to present faster convergence rate and better separation performance.