Skip to Main Content
Electronic devices that have a dynamic nonlinear behavior are generally modeled by the so-called Volterra series model. The construction of such analytical model, however, is a complex and time-consuming task. In this work we propose a new approach to help in the building of a Volterra series model for modeling an electronic device, using a very simple neural network model. As an initial case of study, the proposed approach is applied to the modeling of a diode and the simulation results are presented. Conclusions are drawn from the comparison of the results obtained with different kernels approximations and network topologies.
Neural Networks, 2004. Proceedings. 2004 IEEE International Joint Conference on (Volume:4 )
Date of Conference: 25-29 July 2004