Convergence analysis of least squares general modulus algorithm in beamforming applications

The convergence behavior of least squares general modulus algorithm (LSGMA), a generalized form of least squares constant modulus algorithm (LSCMA) in adaptive beamforming applications, is analyzed under several signal environments. Monte Carlo simulation results are also given in terms of output SINR gain and iteration number required for convergence. It is shown that fastest convergence speed can be achieved by LSGMA when modulus exponent equals -1 and the corresponding output signal quality does not deteriorate.