Nonlinear System Modelling Utilizing Second Order Augmented Statistics Complex Value Algorithm

The recently introduced augmented complex nonlinear gradient descent (ACNGD) algorithm for complex domain adaptive filtering which utilises the full second order statistical information is shown to be suitable for the processing of both circular and noncircular signals. By virtue of the underlying widely nonlinear model, the complex nonlinear gradient descent (CNGD) is shown to successfully model conventional system, however, unable to model the widely nonlinear system, and the ACNGD is capable to model both the conventional and widely nonlinear system. Simulations in adaptive modelling context for signals with different probability distributions and degrees of circularity support the analysis.

[1]  Danilo P. Mandic,et al.  Complex Valued Nonlinear Adaptive Filters , 2009 .

[2]  Danilo P. Mandic,et al.  Adaptive Convex Combination Approach for the Identification of Improper Quaternion Processes , 2014, IEEE Transactions on Neural Networks and Learning Systems.

[3]  Arie Feuer,et al.  Variable length stochastic gradient algorithm , 1991, IEEE Trans. Signal Process..

[4]  Yili Xia,et al.  A complex Echo State Network for nonlinear adaptive filtering , 2008, 2008 IEEE Workshop on Machine Learning for Signal Processing.

[5]  Danilo P. Mandic,et al.  A Quaternion Widely Linear Adaptive Filter , 2010, IEEE Transactions on Signal Processing.

[6]  Danilo P. Mandic,et al.  A Regularised Normalised Augmented Complex Least Mean Square algorithm , 2010, 2010 7th International Symposium on Wireless Communication Systems.

[7]  Akira Hirose,et al.  Complex-Valued Neural Networks: Theories and Applications , 2003 .

[8]  B. Widrow,et al.  The complex LMS algorithm , 1975, Proceedings of the IEEE.

[9]  Danilo P. Mandic,et al.  An Augmented Echo State Network for Nonlinear Adaptive Filtering of Complex Noncircular Signals , 2011, IEEE Transactions on Neural Networks.

[10]  D. Mandic,et al.  Complex Valued Nonlinear Adaptive Filters: Noncircularity, Widely Linear and Neural Models , 2009 .

[11]  Tülay Adali,et al.  Approximation by Fully Complex Multilayer Perceptrons , 2003, Neural Computation.

[12]  James L. Massey,et al.  Proper complex random processes with applications to information theory , 1993, IEEE Trans. Inf. Theory.