Electronic Theses and Dissertations

Date of Award

12-1-2005

Document Type

Thesis

Degree Name

M.S. in Engineering Science

First Advisor

William G. Frazier

Second Advisor

James P. Chambers

Third Advisor

Roger M. Waxler

Relational Format

Dissertation/Thesis

Abstract

Acoustic signature classification offers a unique capability of exploiting acoustic sensor data to computationally discriminate between detected sources. There are many variations of different techniques currently being used for acoustic classification. A comparison of some of these methods is presented and, performance analyses are documented for each test case within the matrix of presented techniques. This matrix consists of power spectral density and Prony type feature vectors combined with Bayesian maximum likelihood and backpropagation neural network classifiers. Combinations of each feature vector type and classifier type are tested with both simulated data and real data. The data represents short duration continuous sound of the various targets used for testing. Advantages, disadvantages, and recommendations are discussed for all the elements of each technique presented.

Share

COinS
 
 

To view the content in your browser, please download Adobe Reader or, alternately,
you may Download the file to your hard drive.

NOTE: The latest versions of Adobe Reader do not support viewing PDF files within Firefox on Mac OS and if you are using a modern (Intel) Mac, there is no official plugin for viewing PDF files within the browser window.