Adaptive instrumental variable method for robust direction-of-arrival estimation

The authors consider the problem of estimating the directions of arrival (DOAs) of the signals impinging on an array of narrowband sensors with arbitrary geometry. By assuming that the noise covariance matrix has a block diagonal structure, an optimal spatial instrumental variable (OSIV) method is derived for DOA estimation. The OSIV method is computationally simple, yet it provides the minimum variance DOA estimates in a whole class of IV based estimators. In addition, OSIV is a robust method in the sense that it can work with an imperfectly calibrated subarray and, furthermore, it is fairly insensitive to the spatial noise correlation (within the assumed structure). The paper focuses on the adaptive implementation of OSIV, which should be of interest for DOA tracking applications. Numerical examples are included to illustrate the performance that can be achieved with the proposed adaptive OSIV algorithm.