On quantifying the effects of noncircularity on the complex fastica algorithm
Mike Novey, Tülay Adalı · IEEE International Conference on Acoustics Speech and Signal Processing · 2008
The complex fast independent component analysis (c-FastICA) algorithm is one of the most popular methods for solving the ICA problem with complex-valued data. In this study, we extend the work of Bingham and Hyvarinen [1] by deriving conditions for local stability for the more general case of noncircular sources. We use the results of the analysis to quantify the effects of noncircularity on the performance of the algorithm using various nonlinearities and source distributions. Simulations are presented to demonstrate the results of our analysis.