Authors:
Vera Behar
1
and
Ivan Garvanov
2
Affiliations:
1
Institute of Information & Communication Technologie, Bulgaria
;
2
University of Library Studies & Information Technologies,, Bulgaria
Keyword(s):
Adaptive array processing, microphone arrays, sound signal processing, DOA estimation
Abstract:
A possible algorithm for sound source localization in a security system that is based on beamforming of a microphone array is described in this paper. It is shown that the adaptive beamforming algorithm, Minimum Variance Distortionless Response (MVDR), can be a part of the signal processing implemented in a security system. This signal processing includes the following stages: sound source localization, signal parameter estimation, signal priority analysis and, finally, control of protective and warning means (for example, video camera). The adaptive beamforming method MVDR is used for estimating the direction-of arrival (DOA) of signals generated by different sound sources, which arrive at the microphone array from different directions of the protected area. The scenario, in which four sound sources located at different
points of the protected area generate different sound signals (warning, alarm, emergency and natural noise), is simulated in order to verify the algorithm for DOA es
timation. The simulation results show that an adaptive microphone array can be successfully used for accurate localization of all sound sources in the observation area. The parallel version of the described algorithm is tested in Blue Gene environment using the interface MPI.
(More)