Classification of underwater signals using neural networks

The four kinds of neural network classifiers are used for the classification of underwater passive sonar signals radiated by ships. Classification process is divided into two stages: preprocessing and feature extraction stage and the classification stage. In the first stage, two-pass split-windows (TPSW) algorithm is used to extract tonal features from the average power spectral density (APSD) of the input data. In the second stage, four kinds of static neural network classifiers are used to evaluate the classification results, inclusive of the probabilistic based classifier, the hyperplane based classifier, the kernel based classifier and the exemplar based classifier.