A new computing paradigm for the optimization of parameters in adaptive beamforming using fractional processing

Abstract

The use of fractional calculus based novel adaptive algorithms to solve various applied physics and engineering problems is an emerging area of research. In the present study, the parameter estimation problem in adaptive beamforming is explored through the fractional least mean square (FrLMS) adaptive algorithm. The FrLMS algorithm uses the concept of the fractional order gradient in addition to the standard integer order gradient calculation in the recursive parameter update mechanism of optimization. The unknown parameters of adaptive beamforming networks are effectively estimated using FrLMS for various scenarios based on the number of antenna elements in a uniform linear array, the number of interference signals, the signal to noise ratios (SNRs) as well as fractional orders. A comparative study of the proposed FrLMS with standard LMS for different scenarios of adaptive beamforming shows the quality of the design scheme in terms of accuracy, convergence, robustness and stability.