Robust adaptive beamforming based on maximum likelihood estimation

The adaptive beamformer has higher resolution and stronger interference rejection than the data-independent beamformer. However, the adaptive beamformer is very sensitive to the direction-of-arrival (DOA) errors of signal sources, the imperfect array calibration and the error caused by the near-far effect. The constrained least mean square algorithm is the most widely used in adaptive beamforming technique, which will degrade dramatically when the mismatch in look direction exists. In order to overcome its weakness, an adaptive beamforming technique based on maximum likelihood estimation (MLE) is proposed in this paper. This new method discussed in this paper chooses the performance surface which comes from MLE, instead of the power surface which employed in constrained LMS algorithm. Simulation results show that this algorithm improves the robustness against uncertainties in array manifold. Meanwhile two variable scalars are provided to adjust the convergence speed together.