Study of Direction of Arrival Estimation with a Differential Microphone Array using MVDR Algorithm
Shweta Pal, Shweta Kishan Yadav, Arun Kumar, Monika Agrawal · 2024
Directional reception of a radiating source using beamforming algorithms and microphone arrays is a popular approach used for a wide range of real-time applications such as speech acoustics, source localization, teleconferencing, hearing aids, etc. The compact microphone array is gaining increasing relevance within the realm of consumer electronics. However, the limited aperture of this compact array often constrains the effectiveness of traditional Direction of Arrival (DOA) estimation techniques. In contrast, there has been a recent surge of interest in the Differential Microphone Array (DMA). The DMA provides miniature dimension, high directionality and boasts frequency-independent beampatterns. In this paper, we present a comprehensive analysis of the minimum variance distortionless response (MVDR) technique for a linear DMA geometry in the context of DOA estimation. For narrowband tonal signals, a single source-single interference scenario has been studied with respect to a number of performance metrics. The inverse relationship between mean square error (MSE) and input SNR (iSNR) is discussed and validated through simulations. Furthermore, a case study for DOA estimation of a realistic broadband speech application using DMA geometry is presented which has not been studied earlier in our knowledge. Simulation results show that DMA offers a substantial reduction in overall array size as compared to linear arrays with similar DOA performance.