Robust adaptive beamformer based on convex optimization

A new adaptive beamform method is proposed. In this method, two inequality constraints have been used, which restrict not only the minimum gain in the interesting look direction but also the maximum value of the squared norm of the weight vector. The interior-point method is used to acquire the optimal solution. Simulation results show that this new method not only can avoid the signal cancellation phenomenon when the signal of interesting is contained in the beamformer training cell data, but also can be fitted with the small steering vector errors and the small number of snapshots.