PATTERN SYNTHEISS OF LINEAR PHASE ARRAY USING ARTIFICIAL NEAURAL NETWORK BASED ON PARTICLE SWARM OPTIMIZATION

Abstract

 This paper focuses on the antenna synthesis of uniformly spaced linear phase array using artificial neural network (ANN) based on Particle Swarm Optimization (PSO). The weights of the Artificial Neural Networks (ANN) are trained by Particle Swarm Optimization (PSO). Subsequently the Particle Swarm Optimization (PSO) algorithm is applied in order to select the "global best" ANNs for the future investment decisions and to adapt the weights of other networks towards the weights of the best network. Chebyshev method is used to compare with this approach. Although, Chebyshev method is able to generate perfectly leveled side lobes, PSONN does not have the phenomena of up-swing in edges amplitude of the excitation and grating lobes does not appear in PSONN when the distances between elements are increased.  The basic rule is to alter the weights (current distributions of elements) such that the error between the output values and the target values (desired values) is minimized. In this paper, single layer feed forward neural network with PSO training is used.