Title: Imperialist competitive algorithm for minimum bit error rate beamforming

Authors: Arash Khabbazi, Esmaeil Atashpaz-Gargari, Caro Lucas

Addresses: School of Electrical and Computer Engineering, University of Tehran, Tehran, Iran. ' Control and Intelligent Processing Centre of Excellence, School of Electrical and Computer Engineering, University of Tehran, Tehran, Iran. ' Control and Intelligent Processing Centre of Excellence, School of Electrical and Computer Engineering, University of Tehran, Tehran, Iran

Abstract: In this paper, the recently introduced optimisation strategy, imperialist competitive algorithm (ICA) is used to design an optimal antenna array which minimises the error probability for binary phase shift keying modulation, called minimum bit error rate (MBER) beamforming. ICA is used to deal with the high complexity and high dimensionality of this challenging problem which can not be easily solved by gradient-based methods. The results are compared to that of both a genetic algorithm (GA) and the optimal array obtained by a properly set gradient based algorithm. Comparison shows that in contrast with GA, ICA leads to the array with error probability which is very close to the optimal value. Also, being faster than GA, ICA minimises the cost function with more consistency. The total results show that ICA is a powerful and reliable tool for solving complex optimisation problems such as MBER beamforming.

Keywords: imperialist competitive algorithm; ICA; minimum bit error rate; beamforming; bio-inspired computation; optimisation; optimal antenna array; array design; binary phase shift keying modulation; genetic algorithms; GAs.

DOI: 10.1504/IJBIC.2009.022781

International Journal of Bio-Inspired Computation, 2009 Vol.1 No.1/2, pp.125 - 133

Published online: 26 Jan 2009 *

Full-text access for editors Full-text access for subscribers Purchase this article Comment on this article