Extraction of Periodic Signals Based on Maximum Likelihood Estimation

In many applications, such as biomedical signal processing, it is often required to obtain specific periodic source signals. In this paper, we propose a new method based on maximum likelihood estimation to extract periodic signals from linear mixtures. Maximum likelihood function is firstly given according to the periodic property of the desired signals. Then the updating rules for extraction vector are deduced. The derived algorithm is computationally simple, since it is based only on the second order statistical information. Simulation results are given to show its good performance.