Nakmouche M.F.Allam A.M.M.A.Fawzy D.E.Bing Lin D.Abo Sree M.F.2023-06-162023-06-1620219.79E+12https://doi.org/10.23919/EuCAP51087.2021.9411213https://hdl.handle.net/20.500.14365/372915th European Conference on Antennas and Propagation, EuCAP 2021 -- 22 March 2021 through 26 March 2021 -- 168667The proposed work conducts a new approach for modeling a dual band monopole antenna design using DGS assisted by ANN. In the aim of efficient dual band antenna design with gain and optimal impedance matching, the Artificial Neural Networks technique (ANN) is used for the development process. This work presents a modeling for H-slotted Defected Ground Structure (DGS) based dual band antenna using ANN for 5G Sub-6 GHz applications. The designed antenna operates at 3.76 GHz and 6.1 GHz. The antenna gain is 2.18 dB and 2.75 dB at both frequencies, respectively. Firstly, a simulation is performed using CST EM simulator, then the predicted results in term of return losses and frequencies are fed into the ANN model. Secondly using a hybrid algorithm based on both feed-forward back-propagation and Levenberg-Marquart (LM) learning algorithm, the optimal position of the H-Slotted DGS in terms of 5G Sub-6 GHz band is extracted. Finally, the experimental validation is conducted and compared with the simulation results, a good agreement is obtained. © 2021 EurAAP.eninfo:eu-repo/semantics/closedAccess5G applicationArtificial Neural NetworksDefected Ground StructureDual-bandFinite Element MethodsMonopole AntennaSub-6 GHzAntenna feedersBackpropagationDefected ground structuresMicrowave antennasMonopole antennasNeural networksSlot antennasDevelopment processDual band antennasDual-band monopole antennasExperimental validationsFeed-forward back propagationHybrid algorithmsNew approachesOptimal position5G mobile communication systemsConference Object10.23919/EuCAP51087.2021.94112132-s2.0-85105468726