A new approach to robust beamforming against generalized phase errors

In this paper, an approach is proposed for robust beamforming in the presence of various array errors, such as steering error, array geometry error, sensor phase error, etc. The basic idea is to express all these errors as generalized array phase errors in the steering vector, then iteratively search the actual steering vector by maximizing the array output power. A gradient-based optimization algorithm is proposed for searching the optimal phases which are used to compensate the generalized phase errors. This new array processor has the advantage that it does not suffer any loss in the degrees of freedom for interference rejection. Computer simulation results are shown to illustrate the effectiveness of the proposed method.