The Development and Analysis of Beamforming Algorithms Used for Designing an Acoustic Camera

Abstract

This paper discusses the beamforming algorithms used in developing an acoustic camera which can be used for the purpose of localizing sound sources. In order to design an acoustic camera which can display results in the form of an intensity map, it is necessary to determine the beamformed signal for all the desired incident angles, i.e. for all the desired pairs of azimuth and elevation angles. For the purpose of obtaining the beamformed signal required for localization of sound sources, two beamforming algorithms, which differ in the domain in which the beamforming is performed, were developed named respectively DASt.m and DASf.m. The aforementioned algorithms are implemented in the numerical computing environment MATLAB and furthermore compared in this paper. The beamforming was carried out using both of the aforementioned algorithms for all desired azimuth and elevation angle pairs and the obtained results were compared. In order to compare these two developed beamforming algorithms measurements were conducted in an open space using a Uniform Circular Array (UCA). The utilized UCA consists of 16 identical omnidirectional microphones which form a basic circular acoustic camera. The research showed that the DASf.m algorithm gives better results than the DASt.m algorithm, especially for the intensity maps of the average of the signal.