Quadratically constrained adaptive beamforming for coherent signals and interference

Severe signal cancellation often occurs in conventional adaptive beamforming if coherent interferers are present. The paper proposes adding quadratic constraints to a linearly constrained minimum variance (LCMV) adaptive beamformer to prevent signal cancellation in coherent interference environments. The additional constraint limits the beamformer output mean squared error to be less than a specified level. It is shown that if the quadratic constraint is properly chosen, then the original linearly constrained beamformer´s signal cancellation is reduced to an arbitrarily small level while the interference cancellation is unchanged. In practice, the quadratic constraint is constructed based on estimates of the interference parameters. Sensitivity analyses show the performance of the resulting beamformer is robust with respect to this approximate constraint. The effectiveness of the quadratically constrained adaptive beamformer is further illustrated through simulations