An Algorithm for Blind Signal Extraction of Linear Mixture
Gao Ying①②; Xie Sheng-li②
①Dept. of Computer Science and Technology., Guangzhou University, Guangzhou 510405, China;②College of Electronic & Information Engineering, South China University of Technology, Guangzhou 510641, China
Abstract:In this paper, a measure of signal variability is defined. Given any set of statistically independent source signals, it is proved here that a linear mixture of those signals has the following property: the signal variability of any signal mixture is greater than (or equal to) minimal that of its component source signals, and is less than (or equal to) maximal that of its component source signals. Based on the property, an algorithm for linear blind signal extraction is proposed. In the proposed algorithm, the source signal is extracted one by one by using generalized eigenvalue theory and deflation approach. The presented algorithm has less computations. Simulation results illustrate the efficiency and the good performance of the algorithm.