Efficient FPGA Implementation of Sigmoid and Bipolar Sigmoid Activation Functions for Multilayer Perceptrons
Manish Panicker · IOSR Journal of Engineering · 2012
Artificial Neural Networks base their processing capabilities in a parallel architecture.This makes them extremely useful in pattern recognition, system identification and control problems.Multilayer Perceptron is an artificial neural network with one or more hidden layers.The Activation function determines the performance of a Multilayer Perceptron.In Multi Layer Perceptron, the most commonly used activation functions are sigmoid and bipolar sigmoid activation functions.In this paper we present a FPGA based digital hardware implementation of Sigmoid and Bipolar Sigmoid Activation function.The digital hardware was designed for 32 bit fixed point arithmetic and was modeled using Verilog HDL.The synthesis tool used was Xilinx 10.1 ISE and the design was implemented in Spartan 3 FPGA.