Overcomplete blind source separation by a simplified algorithm <Abstract> Blind separation of unknown sources from their mixtures is currently a timely research topic in statistical signal processing and unsupervised neural learning. Several source separation algorithms have been presented where it is assumed that there are at least as many sensors as sources. The algorithm has been introduced as Overcomplete Blind Source Separation has two steps, that is, to initialize and then estimate the mixture media. However, it has a drawback of time consumption. As the number of the sources becomes larger, it requires more computation time exponentially for the initialization. In this paper, we present an approach to accelerate convergence rate in the step of initialization when the solution is unique. Simulation results verify the proposed algorithm. |