The delay-and-sum beamforming technique uses multiple microphones to localize sound sources. One disadvantage of this technique is that adjustments of the position or of the number of microphones change the quality nonlinearly. Additionally, due to the number of combinations possible, it is computationally hard to find the best configuration. This paper uses genetic algorithms to solve this problem. The algorithm searches for the microphone array configuration that provides the highest directivity for each steered orientation. The experiments showed that the genetic algorithm could find the best solution of a constrained search space comprising 33 554 431 solutions in a matter of seconds instead of days. Furthermore, this paper presents the used techniques to reduce the time of evolution, provides a web application and the source code of the implementation.
Lashi, D, Quevy, Q & Lemeire, J 2019, 'Optimizing Microphone Arrays for Delay-and-Sum Beamforming using Genetic Algorithms', Paper presented at 4th international conference on cloud computing technologies and applications , Brussels, Belgium, 26/11/18 - 28/11/18.
Lashi, D., Quevy, Q., & Lemeire, J. (2019). Optimizing Microphone Arrays for Delay-and-Sum Beamforming using Genetic Algorithms. Paper presented at 4th international conference on cloud computing technologies and applications , Brussels, Belgium.
@conference{94db450bcab64666add718b68f324c33,
title = "Optimizing Microphone Arrays for Delay-and-Sum Beamforming using Genetic Algorithms",
abstract = "The delay-and-sum beamforming technique uses multiple microphones to localize sound sources. One disadvantage of this technique is that adjustments of the position or of the number of microphones change the quality nonlinearly. Additionally, due to the number of combinations possible, it is computationally hard to find the best configuration. This paper uses genetic algorithms to solve this problem. The algorithm searches for the microphone array configuration that provides the highest directivity for each steered orientation. The experiments showed that the genetic algorithm could find the best solution of a constrained search space comprising 33 554 431 solutions in a matter of seconds instead of days. Furthermore, this paper presents the used techniques to reduce the time of evolution, provides a web application and the source code of the implementation.",
keywords = "beamforming, genetic algorithm, sound-source localization",
author = "Dugagjin Lashi and Quentin Quevy and Jan Lemeire",
year = "2019",
month = may,
day = "13",
language = "English",
note = "4th international conference on cloud computing technologies and applications : Cloudtech, Cloudtech ; Conference date: 26-11-2018 Through 28-11-2018",
}