PATTERN SYNTHESIS USING HYBRID FOURIER-NEURAL NETWORKS FOR IEEE 802.11 MIMO APPLICATION

Abstract

In this paper, the application of Articial Neural Network (ANN) with back-propagation algorithm and weighted Fourier method are used for the synthesis of antenna arrays. The neural networks facilitate the modelling of antenna arrays by estimating the phases. The most important synthesis problem is tond the weights of the linear antenna array elements that are optimum to provide the radiation pattern with maximum reduction in the sidelobe level. This technique is used to prove its effectiveness in improving the performance of the antenna array. To achieve this goal, antenna array for Wi-Fi IEEE 802.11a with frequency at 2 : 4 GHz to 2 : 5 GHz is implemented using Hybrid Fourier-Neural Networks method. To verify the validity of the technique, several illustrative examples of uniform excited array patterns with the main beam are placed in the direction of the useful signal. The neural network synthesis method not only allows to establish important analytical equations for the synthesis of antenna array, but also provides a great exibility between the system parameters in input and output which makes the synthesis possible due to the explicit relation given by them.