Classification of Underwater Signals Using Neural Networks

摘要: ABSTRACT In this paper, four kinds of neural network classifiers have been used for the classification underwater passive sonar signals radiated by ships. Classification process can be divided into two stages. In preprocessing and feature extraction stage, Two-Pass Split-Windows (TPSW) algorithm is to extract tonal features from average power spectral density (APSD) input data. static are evaluate results, inclusive probabilistic based classifier-Probabilistic Neural Network (PNN), hyperplane classifier-Multilayer Perceptron (MLP), kernel classifierAdaptive Kernel Classifier (AKC), exemplar classifierLearning Vector Quantization (LVQ). For comparison, same but using Dyadic Wavelet Transform (DWT) in performance proposed method. Experimental results show that TPSW gives better than DWT, require more computation time. classifiers, give others both learning speed rate. Moreover, classifier LVQ families with data extracted DWT reach correction rate (100%) as various networks TPSW. Detail discussion experimental also included.