Classification of underwater signals using a back-propagation neural network

Abstract

This thesis examines a number of underwater acoustic signals and the problem of classifying these signals using a back-propagation neural network. The neural network classifies the signals based upon features extracted from the original signals. The effect on classification by using an adaptive line enhancer for noise reduction is explored. Two feature extraction methods have been implemented; modeling by an autoregressive technique using the reduced-rank covariance method, and the discrete wavelet transformation. Both orthonormal and non-orthonormal transforms are considered in this study