Independent component analysis, a new concept?
Signal Processing - Special issue on higher order statistics
Approach and applications of constrained ICA
IEEE Transactions on Neural Networks
A Class of Complex ICA Algorithms Based on the Kurtosis Cost Function
IEEE Transactions on Neural Networks
Hi-index | 0.00 |
Semi-blind independent component analysis (ICA) incorporates some prior information into standard blind ICA, and thus solves some problems of blind ICA as well as provides improved performance. However, semi-blind algorithms thus far have been much focused on the separation of real-valued signals but little on separation of complex-valued signals. We propose in this paper a semi-blind complex ICA algorithm for extracting a complex-valued source of interest within the framework of constrained ICA. Specifically, magnitude information about the desired signal is utilized as inequality constraint to the cost function of kurtosis maximization algorithm, which is an efficient complex ICA algorithm for separating circular and noncircular sources. The simulation results demonstrate that the proposed algorithm can extract a desired complex signal with much improved performance and robustness.