Detection of Direct-path Arrivals for Multi-narrowband Sequences (3-30 kHz) in Shallow Water

Abstract

In an effort to measure underwater acoustic transmission loss over direct‐path lengths ranging from a few hundred meters to ten kilometers in shallow water, a sequence of 16 gated pure tones (3–30 kHz) was transmitted every 10 s from a lowed source and received at moored sonobuoys. The magnitude of multipath arrivals often exceeded that of direct‐path arrivals, resulting in variable detection performance of simple matched filtering techniques. More reliable signal recognition was obtained via iterative least square time constraints on the arrival times across all frequencies in a sequence, based on the known time intervals between transmitted tones. Signal detection improvement was obtained also by searching for the direct‐path arrival near the global maximum of the sum of the rectified correlograms of the received sequences. These methods allowed detection in environments characterized by multipath interferences, as well as low signal‐to‐noise ratio and fading, and in the presence of other unrelated sonar signals that cause large detection errors. It also improved the direct‐path signal strength estimation, and associated transmission loss computation, by bounding the time interval over which to compute the signals’ autocorrelations and estimate their power. These algorithms were tested on a limited data set recorded in the Southern California Offshore Range, confirming that frequencies below 6 kHz suffered less direct‐path transmission losses than higher frequencies (7–30 kHz).

Department

Center for Coastal and Ocean Mapping

Publication Date

2004

Volume

728

Journal Title

AIP Conference Proceedings

Series

High Frequency Ocean Acoustics Conference

Pages

478-488

Conference Date

Mar 1 - Mar 5, 2004

Publisher Place

La Jolla, CA, USA

Rights

© 2004 American Institute of Physics

Publisher

AIP Publishing

Digital Object Identifier (DOI)

10.1063/1.1843043

Document Type

Conference Proceeding

Share

COinS