Fast Converging Layered Adaptive Beam Forming Algorithm

LMS algorithm is simple and is well suited for continuous transmission systems since it is a continuously adaptive algorithm. However, it is not known for its convergence speed in the presence of Gaussian, spatially white, of null mean and variance which has prompted people to use other complicated algorithms. In the above scenario LMS has maximum mean square error and minimum error stability. Hence, there is a need for an algorithm, which is simple to implement yet has a fast convergence rate and is not computationally intensive in the presence of noise. Thus, the algorithm featured in this paper is an attempt in achieving this and it will be referred to as the Fast Converging Layered-LMS algorithm. In the FCL-LMS algorithm the process of finding the optimum weights have been divided into two layers where both the layers have different convergence factors, the convergence factor of the upper layer is always greater than the lower layer, hence the larger value of the convergence factor helps in approaching the optimum weights and the smaller value of the convergence factor minimizes the misadjustments thus reducing the excess mean square error, this results in least mean square error, better error stability and faster convergence.