Synthesis of aperiodic linear antenna arrays using genetic algorithm

In this paper, a genetic algorithm (GA) has been used to design the beampattern of the non-uniform linear array with minimum side lobe level and null control. The evaluation or cost function has been written for the array geometry with the goal of minimum side lobe level and null control in certain directions. Design examples employing the GA for finding the optimum element locations have been illustrated. Comparison with other existing techniques like, particle swarm optimization and quadrature programming method have been carried out.

1. INTRODUCTION

Antenna array synthesis consists of finding the complex excitation or physical layout of the array that produce the radiation pattern as close as possible to the desired one. For antenna array synthesis, many methods are available that use either variation of complex excitations i.e. of amplitude and phase or variation of element spacings, to shape the radiation pattern. Arrays with fewer elements than required by half wavelength condition are called sparse arrays. Grating lobes in the radiation pattern of a sparse array can be reduced by eliminating the periodicity of the array. Many researchers have investigated the design of antenna array using gradient method. These methods either require analytical formulae or evaluation of the gradient of some fitness function, which is sometimes not formidable. Thus the solution lies in the use of stochastic approaches: simulated annealing, genetic algorithms and particle swarm optimization. The simulated annealing [1]–[5], genetic algorithms [6]–[11] and Particle Swarm optimization [12]–[14] are currently being used for antenna array optimization.