Direction estimation using acoustic dyadic sensor array

Microphone arrays can be used for high quality sound pickup in noisy, reverberating environments. In this paper, we present a novel approach for acoustic beamforming with an uniform linear dyadic sensor array. We introduced an acoustic model of reduced rank dyadic sensor to provide the theoretical background and guidelines for the miniature array beamforming algorithm development. We further developed an approach to approximate the dyadic sensor with a miniature circular array consisting of nine microphones. We modified the DOA estimation algorithm based upon the sensor model we created and defined the WNG (white noise gain) to benchmark the performance of the approach we have proposed. Numerical simulations are conducted to compare the performance of conventional, AVS (acoustic vector sensor) and Dyadic linear arrays.