UDC 621.391
COMPLEX RECURRENT NEURAL NETWORK
V. P. Kuznetsov, Ph.D; Associate Professor, the Department of Automation and Information Technologies in Control, RSREU, Ryazan, Russia;
orcid.org/0000-0000-0000-000X, e-mail: This email address is being protected from spambots. You need JavaScript enabled to view it.
The concept of complex feed-forward neural network is used to recurrent networks. The network con tains complex weights and complex activation functions as a complex variable function. Complex recurrent neural network may be used for identification and pre-distortion of dynamic objects with complex input and output signals. The aim of the work is to work out mathematical support for complex recurrent neural net work learning. The results of experimental research of complex recurrent neural network for identification of nonlinear power amplifier are presented.
Key words: : difference equation, neural network, weight coefficient, activation function, error signal, learning algorithm, target function, gradient, partial derivative.