Fast null-steering algorithm for broadband power inversion array

Abstract

A fast algorithm for broadband power inversion adaptive arrays is investigated. Essentially, in this algorithm, each broadband jammer is individually tracked by steering a broadband null formed from the use of a set of tapped delay line filters in a Davies beamformer. To steer a particular broadband null to its optimal position, the algorithm first swaps the associated set of null-controlling filters into the last stage of the beam former, and then employs the LMS algorithm with a simple time-invariant decorrelation preprocessor to adjust the filter parameters to minimise the output power for a number of iterations. Performing this procedure for each null in turn results in continuous tracking of all the broadband jammers in a cyclical manner. Although the algorithm requires implementation of the broadband tapped delay-line Davies beamformer, only one tapped delay-line filter is being adaptively updated at any instant. Thus, when the number of elements is small, the overall implementation complexity of the proposed algorithm is comparable to that of using the LMS algorithm directly. However, since each broadband null is independently steered, the proposed algorithm has a much faster convergence behaviour, particularly under situations of severe jamming.