Robust Wideband Adaptive Beamforming With Adjustable Nulls in High Dynamic Scene

When the interference rapidly moves and exhibits non-stationary characteristics, the traditional space-time adaptive processing (STAP) algorithm cannot efficaciously solve the mismatch problem between null and interference. Covariance matrix tapering (CMT) is able to generate broad nulls in the beam-pattern and become an effective solution. Most CMT algorithms broaden each null symmetrically with the equal-width. However, symmetrical and equal-width broadened nulls are unnecessary since the non-stationarity of each interference is generally different. To cover the worst non-stationarity, the regular CMT methods must generate the broadest null, which wastes degrees of freedom (DOF). An asymmetric CMT strategy based on Simpson statistical prior is proposed, which is able to generate broad nulls with various widths and asymmetries. The asymmetry is achieved by adding the asymmetric virtual interference cluster that obeys the Simpson distribution, because the Simpson distribution can limit the spatial position of the virtual interference in the cluster. Based on the statistical information of interference contained in the interference space, the unequal null width is generated by tapering the covariance matrices of each interference spaces with different taper matrices. Moreover, the strategy also has the advantage of less DOF consumption. Experimental results demonstrate the practicability and benefits of the proposed strategy.