Contents I. Introduction
Artificial Neural Networks (ANNs) emulate the adaptive learning mechanism used by the neurons of the human brain. The adaptive learning mechanism may use the aid of teacher (supervised learning) or without the aid of the teacher (unsupervised learning). The ANNs have been extensively used in modeling of various systems, time series analysis, pattern classification and recognition, signal processing and design of control systems. The reason for this popularity is the fact that they circumvent arduous calculations by learning a behavioral relationship between the input and output using approximation, pattern classification clustering and prediction [1]. All those problems and applications that do not have closed-form of a solution or proper analytical tool or the problems that require real-time performance along with faster convergence of optimization problems can be solved with much lesser effort [2].
