Sound Source Localization System Based on TDOA Algorithm

Abstract

Sound source localization holds significant value in various fields such as drone navigation, speech recognition, audio surveillance, and more. This paper presents a sound source localization system based on TDOA algorithm. The system is designed to meet the requirements of low-cost, simplicity, and reliability in sound localization devices, and it has been experimentally validated. Initially, the TDOA algorithm is employed to estimate the time delays of the target sound source arriving at different microphones, which are then used to determine the source's direction. The calculation of the source's direction relies on spatial geometric relationships, enabling precise triangulation using TDOA measurements. To enhance system performance, we introduce Kalman filtering to correct the state estimates of the sound source target, further refining the estimated source position. Through the operation and fine-tuning of Kalman filtering, we successfully improve the system's localization accuracy, achieving rapid and stable tracking of the sound source target. Experimental results demonstrate that the sound source localization system based on the TDOA algorithm offers significant advantages in terms of low cost and reliability, making it suitable for a wide range of applications.