Abstract
An attempt has been made to use efficient Neuron model for blind source separation. Generalized Harmonic Mean Neuron (GHMN) has been used as the neuron model. GHMN model is based on generalized harmonic mean of the inputs applied on it. Information-maximization approach has been used for training the neuron model. In this paper, it has been demonstrated how efficiently the GHMN model can be used for blind source separation. It has been shown on a generated mixture of finger prints and a real life mixture of finger prints (for blind source separation) that the new neuron model performs far superior as compared to the conventional neuron model.