Yazar "Özek, Ahmet" seçeneğine göre listele

Listeleniyor 1 - 2 / 2
  • Küçük Resim Yok
    Öğe
    Neural network applications for optimization of microstrip linear phase filter with inset feeding
    (2010) Karpuz, Ceyhun; Özek, Ahmet; Görür, Adnan; Ba?, Nusret; Öztürk, Pinar
    ANN has a wide application areas. One of these applications areas is the optimization of microwave filters. In this study, of optimization ANN solution techniques have been applied to optimize RF and microwave structures Using Artificial neural networks (ANN) modeling technique in a filter design with the inset-feeding open-loop resonator is presented. With helping the Sonnet EM Simulator, to develop a new ANN model, the datas was produced. The modeling techniques which is in the Neuro Solutions Program were used for this process. The geometrical dimension of filter and coupling gap can be used to determine input output(I/O) relations of ANN model. As a result, for the -linear phase- microstrip filter, the results , which was obtained from the ANN model were compared with the experiment and simulation result. And provided a good agreement was observed.
  • Küçük Resim Yok
    Öğe
    Optimization of microstrip dual mode resonator filter structure modeling using artifical neural network applications
    (2010) Karpuz, Ceyhun; Özek, Ahmet; Görür, Adnan; Öztürk, Pinar; Ba?, Nusret
    Artificial Neural Network computational modules have recently gained recognition RF and microwave modeling. This situation is very important for computer aided tuning of microwave components. Neural networks can be trained to learn the behavior of passive/active components/circuits. This paper presents a design approach for a microstrip dual mode filter by using the artificial neural network modeling (ANN) technique. First to develop new models available in the programs a simple modeling technique were used. To these end important dimensions of the filter layout are used to capture critical input output relationships. Once fully developed, the ANN model has been shown to be as accurate as an EM simulator and much more efficient results in the filter optimization.