LL Logo

SEVENTH ANNUAL
ASAP '99 WORKSHOP

 

 

Characterization of LMS and DMR Beamformers
in the Presence of Loud Periodic Interferers


Geoffrey C. Street
Lockheed Martin
9500 Godwin Drive
Manassas, VA 20110-4157
tel: (703) 367-6236,
email: geoff.street@lmco.com

Kristine L. Bell
George Mason University

Abstract The rapid expansion of deep-water offshore oil exploration has introduced loud impulsive periodic interferers with low duty cycles into sonar systems. Standard Least Mean Square (LMS) adaptive beamformers, designed for slowly varying noise fields, do not effectively eliminate this type of interference. LMS beamformers are slow to form nulls on the interferers because adaptation only occurs when the interferer is present in the data.  Even after convergence, null depth is related to the average interference level, rather than the maximum level, and nulls are often very shallow. In this paper we propose and examine the performance of several adaptive beamforming techniques designed to mitigate both periodic and static interference. The algorithms are variations on robust LMS and Dominant Mode Rejection (DMR) techniques. When the advent of a periodic interferer is detected, an estimate of its direction of arrival or array response vector is made. The first variation imposes first and second order null constraints in the direction of the interference. The second variation augments the array data vector with steering vectors pointed at each of the interferers. These steering vectors are spread in azimuth to account for DOA estimation inaccuracy. These algorithms have reasonable computational requirements and are robust to perturbations in the environment and array. Sets of simulations have been completed assessing performance in an idealized narrowband multi-interferer environment. The performance of each of the techniques in terms of interference suppression and computational complexity will be presented.

Presentation (pdf format)

____

 


LL Logo Disclaimer

Direct comments and questions to: webmaster@ll.mit.edu

MIT Lincoln Laboratory. All rights reserved.