Neural Network training using FFA and its variants for Channel Equalization

Abstract

This Work is on the Artificial Neural Network (ANN) training and application in channel equalization. Here, we design a novel training strategy for neural networks. This training uses Firefly Algorithm (FFA) and its variants to train ANN. Then, these FFA trained ANNs are applied for equalization of nonlinear channels. As proved through simulations, the proposed methodology outperforms the existing ANN based equalization schemes.