A new broadband beamformer with fixed feedback weights

Along with the increase of signal´s bandwidth, large number of tapped delay lines (TDL) will be needed in the TDL-based beamformers to retain the desired signal to interference pulse noise ratio (SINR), which slows down the convergence and increases computational load. To solve the problem, a new broadband beamformer based on the generalized sidelobe canceller (GSC) beamformer is designed. Only one feedback branch with fixed feedback weights is introduced, thus it´s stable inherently. The unconstrained recursive Gauss-Newton (RGN) algorithm is applied to compute all the feedforward weights as a whole. Simulation results show comparable performance of the proposed method against those existing broadband beamformers having feedback branches with faster convergence rate, stronger interference suppression and less computational load.