Share Email Print

Proceedings Paper

Acoustic classification of battlefield transient events using wavelet sub-band features
Author(s): M. R. Azimi-Sadjadi; Y. Jiang; S. Srinivasan
Format Member Price Non-Member Price
PDF $14.40 $18.00
cover GOOD NEWS! Your organization subscribes to the SPIE Digital Library. You may be able to download this paper for free. Check Access

Paper Abstract

Detection, localization and classification of battlefield acoustic transient events are of great importance especially for military operations in urban terrain (MOUT). Generally, there can be a wide variety of battlefield acoustic transient events such as different caliber gunshots, artillery fires, and mortar fires. The discrimination of different types of transient sources is plagued by highly non-stationary nature of these signals, which makes the extraction of representative features a challenging task. This is compounded by the variations in the environmental and operating conditions and existence of a wide range of possible interference. This paper presents new approaches for transient signal estimation and feature extraction from acoustic signatures collected by several distributed sensor nodes. A maximum likelihood (ML)-based method is developed to remove noise/interference and fading effects and restore the acoustic transient signals. The estimated transient signals are then represented using wavelets. The multi-resolution property of the wavelets allows for capturing fine details in the transient signals that can be utilized to successfully classify them. Wavelet sub-band higher order moments and energy-based features are used to characterize the transient signals. The discrimination ability of the subband features for transient signal classification has been demonstrated on several different caliber gunshots. Important findings and observations on these results are also presented.

Paper Details

Date Published: 11 May 2007
PDF: 8 pages
Proc. SPIE 6562, Unattended Ground, Sea, and Air Sensor Technologies and Applications IX, 656215 (11 May 2007); doi: 10.1117/12.722296
Show Author Affiliations
M. R. Azimi-Sadjadi, Information System Technologies, Inc. (United States)
Y. Jiang, Univ. of Colorado (United States)
S. Srinivasan, Information System Technologies, Inc. (United States)

Published in SPIE Proceedings Vol. 6562:
Unattended Ground, Sea, and Air Sensor Technologies and Applications IX
Edward M. Carapezza, Editor(s)

© SPIE. Terms of Use
Back to Top