Data-dependent broad-band beamforming with many constraints using few sensors

A common technique for performing data-dependent beamforming is to minimize the beam former output power in a constrained manner. In real beamforming applications the number of available degrees of freedom is often limited due to the limited number of sensors. If the number of constraints is numerous the emphasis of the beamformer will be to fulfill the constraints and the interferers may therefore not be sufficiently suppressed. In this paper we propose a new approach where we include the constraints in the cost function for the optimization of the beamformer. It turns out that the beamformer obtained by this approach performs better than the standard beamformer in the case of few array sensors while performing equally well for more sensors in the array.