Real-Time Implementation of Artificial Neural Network in FPGA Platform

Abstract

In this paper, we present the implementation of artificial neural networks in the FPGA embedded platform. The implementation is done by two different methods: a hardware implementation and a softcore implementation, in order to compare their performances and to choose the one that best approaches real-time systems and processes. For this, we have exploited the tools of this platform such as blocks Megafunctions and softcore NIOS II processor. The results obtained in terms of execution time have shown that the hardware implementation is much more efficient than that based on the NIOS II softcore.