Aural classification for active sonar – Final report for TIF Project 11CQ11 "aural discrimination of true targets from geological clutter"


  1. Hines, P.C.
  2. Allen, N,
  3. Young, V.W.
Corporate Authors
Defence R&D Canada - Atlantic, Dartmouth NS (CAN)
Military sonars must detect, localize, classify, and track submarine threats from distances safely outside their circle of attack. Active sonars operating at low frequencies are favoured for the long ranges they afford against quiet targets. However, in littoral environments, echoes from geological features (clutter) are frequently mistaken for targets of interest which results in degraded performance. Conventional signal processing techniques for dealing with this have met with only limited success and ignore a potentially valuable discrimination tool – the human auditory system. For it is generally accepted within the sonar community that operators can aurally discriminate target echoes from echoes from naturally occurring objects; however, the performance improvement has never been quantified. This report presents the results from a DRDC Technology Investment Fund (TIF) project whose aim was to measure the ability of the human operator to discriminate clutter from true echoes, to identify specific aural features that the operator employs, and to integrate these features into an automatic aural classifier. Using the area under the Receiver Operator Characteristic (ROC) curve, Az, as a measure of performance, the results show that both the human operators and the automatic classifier (Az ≥ 0.98) score near to ideal performance (Az = 1) for a broadband impulsive source. Performance degrades somewhat when acoustic bandwidth is reduced but

Il y a un résumé en français ici.

Audition;auralization;active sonar;ASW;DLC;classification;clutter discrimination
Report Number
DRDC-ATLANTIC-TR-2007-361 — Technical Report
Date of publication
01 May 2008
Number of Pages

Permanent link

Document 1 of 1

Date modified: