Abstract
A widely linear affine projection algorithm (WL-APA) utilising the full second order statistical information in the complex domain ℂ is proposed. It is based on recent developments in augmented complex statistics, ℂℝ calculus, and the widely linear modelling in ℂ, making it suitable for the processing of noncircular complex valued signals. Simulations on circular and noncircular benchmark and real-world noncircular wind signals show the effectiveness of WL-APA as compared to standard APA. © 2009 IEEE.