Estimation of steering vector errors for adaptive beamforming

Steering vector errors can severely degrade the performance of adaptive beamforming. For the case of a platformmounted array, unknown scattering from the platform can be a major source of bearing and frequency dependent errors. These errors can be estimated using a technique based on maximizing the signal-to-interference-plus-noise ratio (SINR) in the spatial spectrum computed using the minimum power distortionless response beamformer with sample matrix inverse (MPDR SMI). This technique is simple compared to some other techniques in the literature, and can be used if the noise is spatially correlated and weak interferers are present. We use simulations to show that good results are obtained if the uncertainty in the signal bearing is not too large, and interferers are sufficiently weak compared to the calibration signal.