Hybrid evolutionary algorithm based beamforming for smart antenna system

Abstract

In this paper, a Hybrid evolutionary algorithm is applied to control the smart antenna patterns satisfying certain constrains like steering the main beam towards a signal of interest, placement of deep nulls in the directions of undesired signals, etc. In adaptive beamforming arrays, using Least Mean Square (LMS) algorithm alone suffered the problem of getting stuck at local minima and converging after large number of iterations. A Genetic algorithm (GA) is a global optimisation technique. The combination of LMS and GA can explore the large optimisation function surface than the individual resulting in faster convergence to global minimum. Simulation results are presented to illustrate the performance of the Hybrid approach.