Spectral analysis for recognition of acoustic fingerprint

Abstract

This article presents the results of the recognition process of acoustic fingerprints using the spectral characteristics of the signal. The Principal Component Analysis (PCA, for its acronym in English) is applied to reduce the size of the extracted features and then, based on the method of k-nearest neighbor (KNN), a classifier is implemented to identify the pattern of the audio signal. This classifier is compared with the Artificial Neural Network (ANN, for its acronym in English). It is necessary to implement a filtering system for the acquired signals in order to reduce the noise of 60 Hz generated by imperfections in the acquisition system. The methods described in this paper were used for recognition of marine vessels.