3D sound source localization with fiber optic sensor array based on genetic algorithm

Abstract

Sound source localization (SLL), which is an area of multi-point acoustic detection, has always been an important research topic since it has been employed in many applications. It has been notable to use fiber optic sensors in the SSL due to the advantages compared to traditional sensors. In this paper, a fiber optic sensor array is obtained with sensor tips formed with polymer-based diaphragm material, which is low cost and easy to manufacture. The SSL process has been experimentally verified in the measurement medium with a volume of 1 m3. Both constant frequency sound signals and speech signals were used in experimental measurements. The signals received from the sensors were processed by using the genetic algorithm, and the average distance to the source was calculated in 3D with an error of 0.0248 m. In the measurements, results were obtained at very close distances to the sound source, such as 0.0032 m, 0.0046 m, 0.0071 m. © 2020 Elsevier Inc.