Estimation and compensation of rotation perturbation in linear 2D acoustic vector sensor array

Most beamforming algorithms for vector sensor array are derived based on the assumption that the directivity pattern of all sensors are perfectly aligned. However this is not true in practice and calibration for rotational perturbations are required to avoid drastic degradation of beamforming performance. This paper presents two algorithms that are capable of estimating and compensating rotational perturbation angles. The first algorithm is based on steering vector error and the second algorithm is based on subspace criteria. Computer simulations show that performance of both algorithms approach theoretical bound, are consistent and able to restore distorted spatial spectral in the presence of single and multiple sources.