Title :
A piecewise linear dynamical functional artificial neural network (PWL-DFANN) for nonlinear adaptive time series prediction
Author :
Figueroa, J.L. ; Cousseau, J.E. ; De Figueiredo, R. J P
Author_Institution :
Dept. of Electr. Eng., Univ. Nacional del Sur, Bahia Blanca, Argentina
fDate :
6/24/1905 12:00:00 AM
Abstract :
A nonlinear adaptive time series predictor has been developed using a PWL-DFANN network (piecewise linear dynamical functional artificial neural network) for its underlying model structure. This network is so called because it is a D-FANN the activation functions of which are piecewise linear. The network has been successfully used to model and predict an important class of highly dynamic and nonstationary signals, namely speech signals.
Keywords :
adaptive signal processing; neural nets; piecewise linear techniques; speech processing; time series; PWL-DFANN; activation functions; model structure; nonlinear adaptive time series prediction; nonstationary signals; piecewise linear dynamical functional artificial neural network; speech signals; Adaptive filters; Artificial neural networks; Discrete cosine transforms; Ear; Finite impulse response filter; Piecewise linear techniques; Predictive models; Recurrent neural networks; Signal processing; Speech processing;
Conference_Titel :
Circuits and Systems, 2002. ISCAS 2002. IEEE International Symposium on
Print_ISBN :
0-7803-7448-7
DOI :
10.1109/ISCAS.2002.1009769