Abstract
In dynamic environments, fast-moving interferers traverse resolution cells quickly, challenging the minimum variance distortionless response (MVDR) beamformer to place accurate notches in the direction of interferers. The hybrid MVDR (HMVDR) beamformer generates serendipitous wide notches by factoring the beampattern into adaptive and fixed components. The adaptive MVDR factor implicitly places notches in the interferer direction for suppression and reduces some background noise, while the fixed conventional beamformer (CBF) factor improves white noise gain. The HMVDR product beampattern flexibly allocates degrees of freedom (DoFs) between the adaptive and fixed components. However, optimizing DoF allocation for the adaptive part remains a challenge, especially in environments with an unknown time-varying number of interferers. To address this challenge, we propose the universal HMVDR (UHMVDR) beamformer, which integrates a mixture of experts' universal algorithms into the HMVDR beamformer, blending different factorizations to dynamically allocate DoFs between the adaptive and fixed components. Simulations and microphone array experiments confirm that the UHMVDR beamformer achieves better array output power and white noise gain than competing beamformers. [Work supported by ONR 321US.]