Abstract: Wideband beamforming usually needs a large number of sensors or tapped delay lines, which lead to heavy hardware load and low beamforming efficiency. To alleviate this problem, a subband sensor delay lines (SDL) based wideband adaptive beamforming algorithm is proposed. Firstly, a subband SDL model is developed. Based on this novel model, received signal is decomposed by analysis filter banks, then the linearly constrained minimum variance (LCMV) algorithm is used in each subband beamformer. In the end, the output of subband beamformers are recombined by synthesis filter banks. Simulation results demonstrate that, comparing with fullband beamforming, subband beamforming has higher efficiency, better frequency invariance, stronger anti-jamming capability, and faster convergence speed. Moreover, subband beamforming can release heavy hardware load largely.