A New TURBO LMS Beamforming for Mobile Communication

Abstract

Array processing involves manipulation of signals induced on various antenna elements. Its capabilities of steering nulls to reduce co-channel interferences and pointing independent beams towards various users, as well as its ability to provide estimate of directions of radiating sources, make it attractive to the mobile communications system designer.In this paper existing Least Mean Square (LMS) algorithm is modified to obtain lesser Mean Square Error (MSE) by using leaky factor in the weight updation.A novel approach named as TURBO LMS is proposed which reduces the Mean Square Error and increases the convergence speed by a large amount as compared to existing beamforming algorithms. Simulation of Beamforming algorithms namely Least Mean Square (LMS), Leaky Least Mean Square (LLMS) and novel TURBO LMS algorithm are done for various look directions and jammer configurations and their MSE characteristics are compared. MATLAB simulation shows that the proposed TURBO LMS algorithm has improved convergence as compared to existing beamforming algorithms. Performance of TURBO LMS algorithm is measured by varying number of array elements. KeywordsAdaptive array beamforming, LMS algorithm, TURBO LMS and Mean Square Error.

A speed convergence Least Squares Constant Modulus Algorithm for smart antenna beamforming

“…This method proposed by Aitken is a root finding method. By defining a function and applying the method for finding the weights of the array, we will considerably increase the time needed to reach convergence [2].…”

“…What we propose in the end though, is the use of a speed convergence module based on a deviation of Steffensen's method of speeding convergence. The Aitken's delta squared process which is a deviation of Steffensen method uses a function that has already been previously used with LMS as stated in [2]. We thought of combining it with the LSCMA (Least Squares Constant Modulus Algorithm) to see the results obtained for an algorithm that does not need a training sequence.…”