Abstract: To address the problems of slow weight convergence and shallow nulling in the recursive least square (RLS) algorithm, we propose an optimized recursive least square (ORLS) beamforming method. By adding the linear constraint part of the linearly constrained minimum variance algorithm to the RLS algorithm, we solve problems such as the slow convergence speed of the weights in the RLS algorithm. To address the sensitivity of the RLS algorithm to noise and its large convergence error under the conditions of a low signal-to-noise ratio and small forgetting factor, we propose an adaptive beamforming algorithm based on variational mode decomposition (VMD) and ORLS. We use VMD to reduce the noise of the received signal array, and then use the ORLS algorithm for beamforming. Simulation results show that compared with the traditional RLS algorithm, the proposed algorithm has a smaller mean-square error and faster convergence speed, as well as deeper nulling and a stronger ability to suppress interference.