ABSTRACT
This chapter discusses the definition of a neural network for signal processing and why it is important. It explores several modern neural network models that have found successful signal processing applications. The chapter describes how to adjust the weights of an Multilayer Perceptron (MLP) with a single layer of neurons. A MLP neural network model consists of a feed-forward, layered network of W. McCulloch and W. Pitts’ neurons. Each neuron in an MLP has a nonlinear activation function that is often continuously differentiable. The nonlinear nature of neural networks, the ability of neural networks to learn from their environments in supervised as well as unsupervised ways, as well as the universal approximation property of neural networks make them highly suited for solving difficult signal processing problems. From a signal processing perspective, it is imperative to develop a proper understanding of basic neural network structures and how they impact signal processing algorithms and applications.
