Logo image
Doubly Adaptive Covariance Matrix Taper Universal Beamformer
Conference proceeding

Doubly Adaptive Covariance Matrix Taper Universal Beamformer

Savas Erdim, John R. Buck, Christopher Gravelle, C.J. Berg and Isaiah Lacombe
OCEANS 2022, Hampton Roads, pp.1-6
10/17/2022

Abstract

adaptive beamformer Arrays Covariance matrices covariance matrix taper Microphone arrays minimum variance distortionless response Oceans Roads universal algorithm
In a dynamic environment, interferers moving with different bearing rates transit the resolution cells of an array at different speeds. This condition degrades the MVDR beamformer's performance due to the lag between the calculated notch location and interferer's actual location. One approach to overcome that challenge is applying a covariance matrix taper (CMT) that broadens the beampattern notches. The CMT widens MVDR beampattern notches by a fixed amount, and the optimal notch width depends on the unknown time-varying bearing rate of each interferer. Thus, a single fixed notch width cannot suppress all the interferers in the environment perfectly. The need for different notch widths for different bearing rates leads us to design a universal algorithm for this parameter. This universal algorithm rivals the best beamformer in a set of fixed notch width beamformers without requiring any prior knowledge of the interferer motion or the environment. This paper evaluates the performance of the universal CMT beamformer with microphone array data containing multiple moving interferers.

Metrics

11 Record Views

Details

Logo image