A fast squares power inversion array for tracking nonstationary environments

This correspondence presents and investigates a fast algorithm for the tracking of nonstationary environments in power inversion arrays. The algorithm is based on the use of iterative least squares techniques and the introduction of an additional gradient vector in the objective function so that both the optimal weight vector and its rate of change are estimated recursively. Although the algorithm is computationally more complex than conventional least squares methods, it has better performance in tracking moving sources as the velocities of the latter are also being estimated.