Enhanced adaptive beamforming using LMMN algorithm with SMI initialization

The purpose of this paper is to develop an enhanced adaptive beamforming algorithm based on LMMN but with SMI initialization to ensure faster convergence. It is shown that judicious choice of the LMMN algorithm mixing parameter provides an algorithm with intermediate performance between the two special cases of least mean squares (LMS) and least mean fourth (LMF) algorithms. It is shown that the developed LMMN algorithm along with SMI initialization provides better steady state performance than the LMS algorithm and better stability properties than the LMF algorithm.