Modeling and FPGA Implementation of ANN Based Electronic Circuits

In this book, a new approach is proposed to build neural network architectures. Previous works are used back-propagation. The major limitation of this network that it can only learn an input ¿ output mapping which is static. Recurrent neural networks (RNNs) have features that well define dynamic systems which have attracted the attention of researches in this field. Generally, recurrent neural network requires less neurons in its structure and less computation time. Also, they show high immunity against external noise. In this book, a new approach is proposed to build neural network architectures. Previous works are used back-propagation. The major limitation of this network that it can only learn an input ¿ output mapping which is static. Recurrent neural networks (RNNs) have features that well define dynamic systems which have attracted the attention of researches in this field. Generally, recurrent neural network requires less neurons in its structure and less computation time. Also, they show high immunity against external noise.