Creation of Real Blocks for Neural Network Using Simulink
Hassan Jouni, Adnan Harb, Gilles Jacquemod, Yves Leduc · 2018
In this paper, we present the design of a multiplier, an activation function and its derivative blocks to realize a neural network based on the Multi-Layer Perceptron with back propagation (MLP) algorithm. The network and the building blocks are validated using Simulink simulations. This is to help in the future to build the equivalent blocks in Analog CMOS circuits.